User login
The Medical Roundtable: Bone Density: Diagnosis and Management—What Is the True Burden of Disease?
Dr. Maricic, what are the different types of osteopenia that primary care clinicians encounter in their practice?
DR. MARICIC: Osteopenia has several connotations, depending on the setting in which it is diagnosed. Radiologists will describe a patient as being osteopenic based on the subjective assessment that the patient has low bone density on radiography. With particular reference to osteoporosis or bone densitometry, osteopenia refers to a dual-energy X-ray absorptiometry (DEXA) T-score, which is a comparison of the bone mass of a patient to that of a young, ideal patient of the same sex and ethnicity.
DR. GALL: Is osteoporosis the only disease or are there other diseases that we may see associated with osteopenia?
DR. MARICIC: Neither osteopenia nor osteoporosis diagnosed by the DEXA score is necessarily caused by estrogen deficiency. The DEXA report tells you whether a patient has low bone density but not about the etiology. Although estrogen deficiency is a common cause of osteopenia or osteoporosis, other conditions such as primary hyperparathyroidism, multiple myeloma, vitamin D deficiency, osteomalacia and the conditions associated with it, chronic renal failure with osteitis, fibrosis cystica, and hyperparathyroidism should be considered by the primary care physician.
It’s typically recommended that all patients with low bone density diagnosed by their DEXA score have a chemistry screening done to determine the calcium and alkaline phosphatase levels, as well as 25-hydroxy-vitamin D level, in order to ensure that the patient is not vitamin D deficient. A 24-hour urine test is also recommended to ensure that the vitamin D level is not low, suggesting malabsorption, such as what we see in vitamin D deficiency or celiac disease, or that it’s not high, suggesting idiopathic hypercalciuria.
Those are the main, most common disorders that we’d like to exclude in patients with osteopenia or especially osteoporosis. We would also want to exclude secondary hyperparathyroidism; vitamin D deficiency; and, when suspected, celiac disease, idiopathic hypercalciuria, and multiple myeloma. Those are some of the conditions to be considered, especially when the T-score does not accord with the clinical history.
DR. GALL: Dr. McClung, can you please discuss the use of X-ray and DEXA scan as diagnostic tools?
DR. MCCLUNG: These 2 tools have completely different roles. We use the DEXA scan to measure bone density, primarily in the hip and the spine, in postmenopausal women and older men to obtain a T-score that is, a value comparing the patient’s bone density to the average bone density value of young adults of the same sex. A T-score of −2.5 is thought to be consistent with osteoporosis in postmenopausal women and older men, but, as Dr. Maricic pointed out, a T-score of −2.5 only tells us that the patient has low bone density. Before we conclude that the patient has osteoporosis, all of the other causes of low bone density, including osteomalacia, osteogenesis imperfecta, and any of the secondary causes that were outlined, need to be excluded.
I believe that osteopenia, which is defined as a T-score value between −1 and −2.5, is not a very useful diagnostic term for clinical purposes. We know that low bone density is one of several risk factors for fracture. Within the group of patients with osteopenia are those who are at very high risk for fracture, including older people and those who have had fractures, as well as many patients who are not at high risk for fracture. In response to this problem, the World Health Organization developed the Fracture Risk Assessment Tool (FRAX), a fracture risk assessment tool used to help distinguish patients with high or low fracture risk.1
DR. GALL: Clinicians often find the terms used in the DEXA report confusing. Could you define absolute bone mineral density (BMD) and Z-score for the clinician?
DR. MCCLUNG: The BMD test measures the bone mineral content of a specific anatomic region and the area of that region and divides those 2 numbers to get an estimate of the BMD. It is not a true bone density test, but rather a measurement of calcium content within a region of bone.
The BMD value is translated into a T-score by comparing it to the average value for healthy young adults of the same sex and is then used for diagnosis in postmenopausal women. The Z-score is obtained by comparing the patient’s BMD value with the expected average value for healthy adults of the same age and sex and is used to determine whether a low BMD value for that patient is unexpected. A low T-score for a 55-year-old woman may be surprising, whereas a low T-score for a 90-year-old woman would be expected. A Z-score lets us know how concerned we should be.
DR. GALL: Dr. Lisse, you read a lot of DEXA scans. What can give a false value? Can you tell us about that and discuss the FRAX?
DR. LISSE: Dr. McClung wrote a very nice treatise on the FRAX,2 a tool that assesses the risk of fracture in bisphosphonate-naïve patients based on the consideration of risk factors, in addition to the BMD value. The DEXA measures bone quantity, but does not assess the risk factors for osteoporosis, of which I think prior fracture and patient age are the most important.
By considering the risk factors of smoking history, advanced age, prior fracture, and family history of fracture, and by adjusting for the country and racial origin of the patient, the FRAX provides a better measure of fracture risk for the osteopenic patient compared to BMD alone. It also provides a value, the 10-year fracture risk, that’s easier for the primary doctor to interpret compared to T-scores and Z-scores, which require comparison in terms of standard deviations.
DR. GALL: How can we use the FRAX if we are interested in doing so?
DR. LISSE: The tool is available online and very easy to use. Usually, you complete the assay using the BMD value. You can actually perform the assay without a BMD value, but I think that including the BMD value improves the precision and predictive value of the results. You just plug in the information online and are given a reading. Recent literature suggests that you can use it for patients who are taking bisphosphonates, but I think that’s a little bit further down the line in our discussion.
DR. GALL: A FRAX result states that a patient has a 10-year risk of major fracture of 8%. How do I use that result? Should I start treating the patient?
DR. LISSE: Several studies that examined local socioeconomic factors concluded that using a 20% risk of major osteoporotic fracture and a 3% risk of hip fracture are cost-effective thresholds for the diagnosis of patients in the United States. Although they should not be considered absolute, they are the current thresholds, and I think physicians can use them as criteria to make treatment decisions.
DR. MCCLUNG: That description is correct. FRAX has a scientific base and is a very highly validated tool that accurately predicts fracture probability in healthy middle-aged and older adults.1 Using the tool is a good first step while considering whether osteoporosis treatment is justified. The threshold values of a 10-year probability of hip fracture of 3% and a 10-year probability of major osteoporotic fracture of 20% are specific to the United States, which is based on the United States healthcare policy and where osteoporosis fits into the United States medical priorities. Other countries have their own thresholds that are very different.
DR. GALL: Dr. Maricic, we’ve talked about the DEXA and the FRAX. What other imaging tools are available and what are their roles?
DR. LISSE: The other issue is that the FRAX uses femoral neck density, which is a good predictor; however, in some patients, especially patients taking glucocorticoids, the spine density may be lower than the hip density. A low spine T-score should also be taken into account. We must individualize patients and not become overly focused on any one tool, although the use of the FRAX is certainly a major step forward from the use of only the T-score, for which, former guidelines recommended treatment for patients with scores within the wide range of −1 and −2.5.
DR. GALL: Another issue regarding DEXA is that we often get a false high BMD in patients with fracture of the lumbar spine, osteoarthritis, or issues with alignment of the spine. In such cases, it is important to examine the image, as the presence of bone abnormalities may render the spine DEXA score and the spine T-score less important.
DR. MARICIC: The role of other modalities in diagnosing osteoporosis in the United States is limited because of the widespread availability of DEXA in most communities. A few communities in Arizona have access to only quantitative computed tomography, which, although tends to overdiagnosise osteoporosis, can be helpful if bone density is found to be low and the clinical setting is appropriate.
Likewise, quantitative ultrasound in patients older than 65 years has a very valid role in estimating risk of fracture, but because DEXA is so widely available, we do not or usually should not use ultrasound instead of a DEXA scan and ultrasound cannot be used to monitor or follow-up patients with treatment. The only modality that should be used is the DEXA scan because of its very high precision.
DR. GALL: I have seen the use of ultrasound in both remote areas and nursing homes and other areas where it is difficult for patients to visit an office at which a DEXA scan performed, but it is not as accurate as the DEXA scan.
DR. MCCLUNG: Yes, but my understanding is that you use the ultrasound to determine whether you need a DEXA scan.
DR. MARICIC: Yes, but you shouldn’t use the ultrasound to start therapy or monitor therapy.
DR. MCCLUNG: FRAX, which can be calculated without a bone density test, provides a much more accurate assessment.
DR. MARICIC: And FRAX is probably cheaper and easier.
DR. GALL: Dr. McClung, would you talk a little bit about bone markers? I once heard you speak about their role in making a decision about when to treat patients with borderline DEXA and FRAX values.
DR. MCCLUNG: Bone markers are values obtained from blood and urine tests that reflect the activity of the bone cells. Generally, high levels of bone markers reflecting high levels of bone absorption are predictive of more rapid rates of bone loss. So, when a physician is deciding whether to treat, measuring a bone marker can sometimes provide important information to help make the decision. A bone turnover rate in the low–normal range in a bisphosphonate-naïve patient indicates that his/her bone density is stable and that he/she may benefit less from treatment compared to a patient with a high rate.
You can also use bone marker values to determine whether treatment has been effective, particularly if you use antiresorptive agents such as bisphosphonates. Patients with celiac disease or those who have had a gastrectomy may not absorb oral bisphosphonates, so their markers may remain high even if they are under treatment. Although there are certain patients for whom bone markers are helpful, the use of these markers is not routine in the management of most patients with osteoporosis.
DR. GALL: I would say that bone markers are generally overused. Another area in which the use of bone marker values may be helpful is compliance. I’ve used bone biomarker values in cases in which, for example, I start a patient on a weekly bisphosphonate and the patient comes back 12 weeks later saying that he/she doesn’t need a prescription. In such cases, I may perform bone marker testing to see whether the patient has been compliant.
Is there ever a time when we need to do a bone biopsy?
DR. MARICIC: I think a bone biopsy is only helpful for patients with long-term renal failure and dialysis to separate out the various forms of renal osteodystrophy. Although we used to perform a biopsy to help us understand bone turnover or detect osteomalacia, we can now do both with a biochemical test.
In some cases, we want to perform a bone marrow biopsy to look for marrow-based diseases such as mastocytosis or myeloma.
DR. GALL: I’d now like to discuss the prevention of osteoporosis and the risk factors that indicate testing and treatment of patients who are not postmenopausal.
DR. LISSE: There are risk factors that are nonmodifiable, including family history and genetics; modifiable risk factors, including level of exercise, especially weight-bearing exercise, which can be decreased; smoking, which can be discontinued or at least attenuated; and alcohol intake, which can be decreased. I think the role of caffeine is still controversial, although I’d love to hear from the other members of the panel about that.
I think we have overemphasized the importance of calcium and vitamin D supplementation. However, I measure the serum 25-hydroxy-vitamin D level, which according to me, is becoming more and more popular, and I try to keep it at around 30 ng/mL. I think all female patients should obtain some calcium supplementation, especially if their dietary intake is not adequate.
It was surprising to know that there’s probably a higher prevalence of vitamin D insufficiency in Arizona than in Minnesota in spite of the abundance of sunshine in the former.
DR. GALL: I consider patients who are thin, have poor bone stock, are malnourished, did not drink milk in their childhood, and have a family history of hip fracture in a close relative to be at higher risk, and I tend to test them early and consider either preventative treatment or just treatment for them.
I think an appropriate target for daily intake of calcium is between 800 and 1 200 mg, including that obtained from diet. The appropriate target for vitamin D intake in the absence of regular sun exposure is between 600 and 2 000 units per day.
DR. LISSE: Most recommendations that I’ve seen for calcium are about 800 mg, but they vary widely.
DR. MCCLUNG: Let’s get back to the management of premenopausal women with risk factors for low bone mass such as small body size, family history of osteoporosis, and medical problems that may lead to impaired bone health such as amenorrhea during early adolescence. For these women, maintaining adequate nutrition and physical activity and avoiding harmful lifestyle factors should be encouraged, but bone density testing is rarely necessary. If premenopausal women are estrogen replete, even if they have low bone density, treatment with osteoporosis drugs is not recommended. Individuals with risk factors for low bone density should be screened at the time of menopause to decide whether aggressive prevention of the rapid bone loss that occurs at that particularly important time is warranted.
DR. GALL: Dr. Maricic, there are 2 topics I’d like you to discuss. The first is comorbidities, specifically conditions such as thyroid disease, that we need to pay attention to, and the second is hormone replacement therapy and its current role.
DR. MARICIC: The NOF guidelines for the diagnosis and treatment of osteoporosis5 list a number of comorbid disorders and medications that may be responsible for a patient’s low bone density. We should consider the potential for osteoporosis in all men and women over the age of 50 years, especially if they have other disorders or medications known to affect bone mineral metabolism.
The NOF guidelines list about 30 diseases implicated in low bone density, including type 2 diabetes, hemochromatosis, and androgen- and estrogen-deficiency hypogonadism. Low bone density should be considered in patients with gastrointestinal (GI) disorders, of which celiac disease is the most important and common; those who have undergone GI bypass or other types of GI surgery; and those with hematological disorders such as mastocytosis and myeloma. Because chronic inflammatory production of cytokines and calcium immobilization lead to accelerated bone loss, low bone density should be considered in patients with rheumatic disorders, including systemic lupus; ankylosing spondylitis; and especially, rheumatoid arthritis, which FRAX identifies as a major risk factor for accelerated bone loss. Finally, low bone density should be considered in patients with a number of miscellaneous conditions, including chronic obstructive pulmonary disease, emphysema, chronic asthma, and sarcoidosis.
Regarding medications, the agents most commonly associated with low bone density are glucocorticoids. Low bone density should also be considered in patients taking certain anticonvulsants, especially phenobarbital and phenytoin, which can lead to accelerated degradation of 25-hydroxy-vitamin D and very low vitamin D levels and consequent osteomalacia, and in those taking aromatase inhibitors, which are now very important in the treatment of breast cancer.
DR. GALL: I’d just like to mention again the association between low bone density and both thyroid disease and overmedication in thyroid replacement therapy.
DR. MCCLUNG: We should probably also include proton pump inhibitors and selective serotonin reuptake inhibitors, which have been correlated with increased rates of fracture, although the mechanism underlying that correlation is unclear.
DR. LISSE: Do you mean that their wide use poses a concern?
DR. MCCLUNG: Yes, but their effect does not seem to be reflected in bone density testing. So, while recognizing their use as an independent risk factor for fracture is important, it is still unclear whether it should be considered while deciding whether to treat with an osteoporosis drug.
DR. MARICIC: Regarding hormone replacement therapy, we must first distinguish between hormone replacement, which is a combination of estrogen and progesterone for women with an intact uterus, and estrogen replacement or estrogen therapy, which is estrogen alone for women who have undergone a hysterectomy. Definitely, there are differences in the potential for adverse effects between these 2 groups.
I believe the data from the Women’s Health Initiative (WHI) suggested that increased risk of thrombosis and breast cancer is more commonly seen with hormone therapy and may be due more to progesterone than to estrogen replacement.6 There are also some differences in the risk with regard to the age of starting hormone therapy in our patients.
From the WHI, we know that hormone therapy prevents bone loss and reduces the risk of hip and other fractures. I think hormone therapy, especially estrogen therapy for women with an intact uterus, is an option to protect the skeleton in women with low bone density, especially those who are perimenopausal or early postmenopausal.
I personally avoid starting hormone therapy in patients above the age of 70 years and sometimes even above 65 years, but I believe it can be an option for women between the ages of 50 and 60 years with low bone density and no contraindications. For women with a strong family history of breast cancer, I usually recommend raloxifene. We’re now trying to limit the lifetime exposure of women to bisphosphonates, so I tend to reserve them for older women at very high risk of fracture, and almost never administer them to women aged 50 to 60 years.
I think that hormone therapy is the first option for women between 50 and 60 years with no strong family history of breast cancer. If there is, I start them on Evista, when possible.
DR. LISSE: I think that bisphosphonate therapy has been the primary choice of osteoporosis therapy since at least 2 000, which is when alendronate came into the market under its brand name. These drugs have been shown to reliably decrease fracture and improve the quality of life. Some drugs, particularly zoledronic acid, actually decrease mortality in patients with hip fracture.7
There are some differences among them. The first one on the market obviously was alendronate, so there’s probably the most data on it. Risedronate, alendronate, zoledronic acid, denosumab, and teriparatide have been shown to reduce vertebral and nonvertebral fracture; ibandronate, to reduce spinal but not nonvertebral fracture; alendronate, risedronate, denosumab, and zoledronic acid, to prevent hip fracture in postmenopausal women; and risedronate, to decrease vertebral and nonvertebral fracture in men.7
The original studies on the effect of many of these drugs on fracture risk were conducted with daily dosing, and equivalent dosing in weekly and monthly regimens has been used and approved, sometimes without fracture data. The drugs work, their safety-to-risk ratio is quite good, and we have long-term experience with them. I think it’s not surprising that a small percentage of patients who have been on them for a decade or more have issues and side effects. There are risks to taking these drugs, as there are with any other drug.
Probably, the most common risk is GI complications, ranging from irritation to esophageal perforation, which surfaced shortly after alendronate came on the market. To prevent GI complications, strict regimens have been developed for oral bisphosphonate use. To some extent, I think that they limit our patient population because many cannot or will not comply with the requirement of taking the drug first thing in the morning on an empty stomach with approximately 230 mL of water and remain upright for 30 to 60 minutes, depending on the medication, to realize any benefit.
Gastrointestinal complications, including those affecting the esophagus, are, by far, the most common issues observed with bisphosphonate therapy. Gastric ulcerations in certain patients, especially in those who use risedronate in coated form, have also been observed.
There are other probably less common but controversial issues, including a possible association with atrial fibrillation, which was identified in the Health Outcomes and Reduced Incidence with Zoledronic Acid Once Yearly (HORIZON) study.8 An association with alendronate was also suspected, but a recent meta-analysis does not support this association.9
An association with esophageal cancer has been anecdotally described in several populations, but I think that the data are mixed at this point. Then, there is the association with osteonecrosis of the jaw and atypical fracture, both of which are quite rare and occur in between 1 in 10 000 to 1 in 100 000 patients.10 I think that the wide coverage of this association in the lay press may have caused many patients who probably shouldn’t have gone off bisphosphonates, to go off them.
A positive byproduct of these associations being reported could be increased research into whether there should be a time or exposure limit to bisphosphonate use and whether a drug holiday from them is warranted.
DR. GALL: Dr. McClung, could you please describe how you monitor a patient after initiation of bisphosphonate therapy and then discuss an atypical fracture?
DR. MCCLUNG: Treatment with any osteoporosis drug is generally monitored by measuring BMD 2 years after the patient has started therapy. The changes in BMD with treatment are quite modest, so we expect to see either a modest increase or at least stability. If a clear decrease in BMD is observed, we consider noncompliance, secondary causes of bone loss, and other reasons for treatment failure.
In theory, biochemical testing of bone turnover should be useful in monitoring therapy. However, because of limitations in the quality and the reproducibility of the assays in the clinical setting, it is difficult to use the results. I think that marker measurement is not ready for use in the everyday clinic, although it has been very valuable in our research.
Atypical fracture is important. As mentioned earlier, there has been concern right from the time when bisphosphonate research began about the potential for skeletal consequences with long-term therapy because of the pharmacology of bisphosphonates. We know that these drugs accumulate in the skeleton over time, and there were concerns about whether we would see progressive inhibition of bone turnover with long-term therapy. Although we were pleased to observe persistent, but not progressive, reduction in bone turnover with long-term treatment, at least for up to 10 years,11 it appears that some individuals, or perhaps, a unique subset of patients, experience too much inhibition of bone metabolism for prolonged durations. When this happens, bone brittleness occurs and is manifested by atypical chalk stick-like fractures of the femur. It is clear that there is a duration-dependent increase in risk, with fractures occurring in roughly 1 in 1 000 patients who have completed at least 10 years of alendronate therapy.12 If we properly decide whom to treat and how to treat, it is important to recognize that this risk is much, much lower than the risk of important fracture, such as that of the hip and spine, in patients who are not treated with bisphosphonate therapy.13 We should reserve bisphosphonates for patients who clearly have osteoporosis or are at high risk for fracture.
DR. GALL: Dr. Maricic, can you conclude on the use of calcitonin, denosumab, and teriparatide, ie, some of the other drugs approved for the treatment of osteoporosis?
DR. MARICIC: Calcitonin is extremely safe but it isn’t very effective and has been shown to reduce spinal fracture risk by 33% and have no effect on hip fracture risk,14 so I personally haven’t prescribed it in almost 10 years.
Denosumab, a receptor activator of nuclear factor-kB (RANK)-ligand inhibitor and monoclonal antibody, is probably among the most effective medications we have. Because it bypasses the GI tract, it’s an excellent choice for patients who develop GI problems with bisphosphonates. Nevertheless, because it is a very potent osteoclast inhibitor, it must be used carefully in patients prone to hypocalcemia, with renal failure, or who are extremely vitamin D deficient. Although hypocalcemia must be considered before initiating its use, I think it is an extremely effective agent.
Teriparatide is the only anabolic agent currently approved for use in patients at high risk of fracture, and that’s for whom we mainly use it. Because its mode of action is completely different from that of bisphosphonates, we use it for patients with either very low bone density or with multiple fractures who have not responded to bisphosphonate therapy. The Food and Drug Administration (FDA) has approved teriparatide therapy for only 2 years, after which it must be followed by antiresorptive therapy. Unfortunately, I occasionally see patients who had not been given antiresorptives after 2 years of teriparatide, causing them to lose any benefit they may have derived from being on this very expensive drug for 2 years. It must always be followed up by an antiresorptive drug.
DR. MCCLUNG: Regarding a drug holiday, you can consider it if you observe a duration-dependent increase in the risk of atypical fracture.15,16 The FDA has provided us with at least a general template for when it should be considered.17 Based on the limited information we have, it is suggested that fracture risk be reassessed after 3 to 5 years of therapy. For patients with a modest risk for fracture at that point, ie, they no longer have osteoporosis according to the results of bone density testing, it is suggested that therapy be discontinued for 1 to 2 years.
For patients found to remain at high risk for fracture, especially spine fracture, continuation of therapy is recommended. The evidence for this recommendation came from 2 large extension studies of the alendronate and zoledronic acid clinical trials.18,19
If treatment is discontinued, the question of when, if ever, to restart therapy arises. The data from the clinical trials suggest that neither the measurement of bone density nor bone markers are helpful in that regard. So, in my clinic, we treat the holiday like a vacation that has a finite end.13 We take patients off alendronate for 2 or 3 years but off risedronate for only 1 year because its effects wear off more quickly. At that point, we reassess the patient. If he/she meets the criteria for treatment with an osteoporosis drug again, we decide on the best treatment option.
DR. GALL: Dr. Lisse, please describe how treating steroid-induced male osteoporosis differs from treating postmenopausal and female osteoporosis.
DR. LISSE: Before I do so, I wanted to add that a flu-like syndrome sometimes occurs with bisphosphonate use. In a corollary to what Dr. McClung just discussed, the European registries provide at least some reassuring evidence that when patients stop taking bisphosphonates, the risk of atypical fracture appears to decrease fairly rapidly, whereas bisphosphonate efficacy doesn’t seem to tail off as quickly.15 It should be reassuring that if patients stop taking a drug, they may have some protection during that period in terms of osteoporosis coverage, unlike when they stop taking estrogen or teriparatide, whose effects tail off rather rapidly.
With glucocorticoid-induced osteoporosis, the biggest issue is probably the fact that fractures tend to occur early and at a higher bone density than what we’re used to seeing with traditional postmenopausal osteoporosis. So it has thrown the old paradigm into disarray. Before the introduction of FRAX, treatment was recommended for patients with a T-score of −1. As the FRAX includes glucocorticoids as a risk factor, its use is very helpful in this regard.
FRAX does have its limitations, and there are physicians who are concerned that it does not include consideration of steroid duration or dose. I think the American College of Rheumatology’s recommendation is to measure bone density and then subdivide the patient population into low risk, moderate risk, and high risk according to the risk for major osteoporotic fracture of ≤10%, 10% to 20%, and ≥20%, respectively.20 Treatment should be considered based on the risk and dosage. Patients taking a dose of 7.5 mg or higher probably need treatment, even if they’re at low risk. Patients on lower doses probably need treatment if they remain on glucocorticoid therapy for more than 3 months.
Although most fractures due to steroid use are of the spine, not all are. I’m sure all of us on the panel are old enough to remember the old asthmatics who used to take large doses of steroids. Certainly, one of the easiest things to do is to taper patients off steroids or reduce the dose down to the lowest possible dose. The use of topical steroids is preferred to that of oral steroids or inhaled steroids.
The NOF guidelines for testing for male osteoporosis differ from those for testing female osteoporosis. Although they recommend that all women older than 65 years be tested, they do not recommend routine testing in men older than 70 years. I think that this latter recommendation, which is based on a US Preventative Health Service Task Force recommendation,21 is probably a mistake and rather confusing.
Even though there has been a study of raloxifene in men, the guidelines are more limited regarding how men who are hypogonadal should be treated. Testosterone should be measured in all men, and those found to be hypogonadal deserve testosterone replacement. Again, the backbone of therapy is the use of bisphosphonates for the most part. Teriparatide has also been approved for use in men, although it has not been proven to provide protection against hip fracture.
DR. GALL: We’ve had an extensive discussion today on low bone density diseases, focusing mostly on osteoporosis. We have discussed at some length the risk factors and the diagnostic tools that we have, particularly the DEXA scan and FRAX, to assess the risk of fracture. We then discussed the risk factors, including comorbidities; means of prevention; and the different treatment modalities, including physical therapy and several nonpharmacologic therapies, in addition to the various medical therapies, including hormone replacement, bisphosphonates, calcitonin, denosumab, and teriparatide. We’ve also briefly touched on the differences between postmenopausal osteoporosis and steroid-induced osteoporosis and male osteoporosis.
FoxP2 Media LLC is the publisher of The Medical Roundtable.
Dr. Maricic, what are the different types of osteopenia that primary care clinicians encounter in their practice?
DR. MARICIC: Osteopenia has several connotations, depending on the setting in which it is diagnosed. Radiologists will describe a patient as being osteopenic based on the subjective assessment that the patient has low bone density on radiography. With particular reference to osteoporosis or bone densitometry, osteopenia refers to a dual-energy X-ray absorptiometry (DEXA) T-score, which is a comparison of the bone mass of a patient to that of a young, ideal patient of the same sex and ethnicity.
DR. GALL: Is osteoporosis the only disease or are there other diseases that we may see associated with osteopenia?
DR. MARICIC: Neither osteopenia nor osteoporosis diagnosed by the DEXA score is necessarily caused by estrogen deficiency. The DEXA report tells you whether a patient has low bone density but not about the etiology. Although estrogen deficiency is a common cause of osteopenia or osteoporosis, other conditions such as primary hyperparathyroidism, multiple myeloma, vitamin D deficiency, osteomalacia and the conditions associated with it, chronic renal failure with osteitis, fibrosis cystica, and hyperparathyroidism should be considered by the primary care physician.
It’s typically recommended that all patients with low bone density diagnosed by their DEXA score have a chemistry screening done to determine the calcium and alkaline phosphatase levels, as well as 25-hydroxy-vitamin D level, in order to ensure that the patient is not vitamin D deficient. A 24-hour urine test is also recommended to ensure that the vitamin D level is not low, suggesting malabsorption, such as what we see in vitamin D deficiency or celiac disease, or that it’s not high, suggesting idiopathic hypercalciuria.
Those are the main, most common disorders that we’d like to exclude in patients with osteopenia or especially osteoporosis. We would also want to exclude secondary hyperparathyroidism; vitamin D deficiency; and, when suspected, celiac disease, idiopathic hypercalciuria, and multiple myeloma. Those are some of the conditions to be considered, especially when the T-score does not accord with the clinical history.
DR. GALL: Dr. McClung, can you please discuss the use of X-ray and DEXA scan as diagnostic tools?
DR. MCCLUNG: These 2 tools have completely different roles. We use the DEXA scan to measure bone density, primarily in the hip and the spine, in postmenopausal women and older men to obtain a T-score that is, a value comparing the patient’s bone density to the average bone density value of young adults of the same sex. A T-score of −2.5 is thought to be consistent with osteoporosis in postmenopausal women and older men, but, as Dr. Maricic pointed out, a T-score of −2.5 only tells us that the patient has low bone density. Before we conclude that the patient has osteoporosis, all of the other causes of low bone density, including osteomalacia, osteogenesis imperfecta, and any of the secondary causes that were outlined, need to be excluded.
I believe that osteopenia, which is defined as a T-score value between −1 and −2.5, is not a very useful diagnostic term for clinical purposes. We know that low bone density is one of several risk factors for fracture. Within the group of patients with osteopenia are those who are at very high risk for fracture, including older people and those who have had fractures, as well as many patients who are not at high risk for fracture. In response to this problem, the World Health Organization developed the Fracture Risk Assessment Tool (FRAX), a fracture risk assessment tool used to help distinguish patients with high or low fracture risk.1
DR. GALL: Clinicians often find the terms used in the DEXA report confusing. Could you define absolute bone mineral density (BMD) and Z-score for the clinician?
DR. MCCLUNG: The BMD test measures the bone mineral content of a specific anatomic region and the area of that region and divides those 2 numbers to get an estimate of the BMD. It is not a true bone density test, but rather a measurement of calcium content within a region of bone.
The BMD value is translated into a T-score by comparing it to the average value for healthy young adults of the same sex and is then used for diagnosis in postmenopausal women. The Z-score is obtained by comparing the patient’s BMD value with the expected average value for healthy adults of the same age and sex and is used to determine whether a low BMD value for that patient is unexpected. A low T-score for a 55-year-old woman may be surprising, whereas a low T-score for a 90-year-old woman would be expected. A Z-score lets us know how concerned we should be.
DR. GALL: Dr. Lisse, you read a lot of DEXA scans. What can give a false value? Can you tell us about that and discuss the FRAX?
DR. LISSE: Dr. McClung wrote a very nice treatise on the FRAX,2 a tool that assesses the risk of fracture in bisphosphonate-naïve patients based on the consideration of risk factors, in addition to the BMD value. The DEXA measures bone quantity, but does not assess the risk factors for osteoporosis, of which I think prior fracture and patient age are the most important.
By considering the risk factors of smoking history, advanced age, prior fracture, and family history of fracture, and by adjusting for the country and racial origin of the patient, the FRAX provides a better measure of fracture risk for the osteopenic patient compared to BMD alone. It also provides a value, the 10-year fracture risk, that’s easier for the primary doctor to interpret compared to T-scores and Z-scores, which require comparison in terms of standard deviations.
DR. GALL: How can we use the FRAX if we are interested in doing so?
DR. LISSE: The tool is available online and very easy to use. Usually, you complete the assay using the BMD value. You can actually perform the assay without a BMD value, but I think that including the BMD value improves the precision and predictive value of the results. You just plug in the information online and are given a reading. Recent literature suggests that you can use it for patients who are taking bisphosphonates, but I think that’s a little bit further down the line in our discussion.
DR. GALL: A FRAX result states that a patient has a 10-year risk of major fracture of 8%. How do I use that result? Should I start treating the patient?
DR. LISSE: Several studies that examined local socioeconomic factors concluded that using a 20% risk of major osteoporotic fracture and a 3% risk of hip fracture are cost-effective thresholds for the diagnosis of patients in the United States. Although they should not be considered absolute, they are the current thresholds, and I think physicians can use them as criteria to make treatment decisions.
DR. MCCLUNG: That description is correct. FRAX has a scientific base and is a very highly validated tool that accurately predicts fracture probability in healthy middle-aged and older adults.1 Using the tool is a good first step while considering whether osteoporosis treatment is justified. The threshold values of a 10-year probability of hip fracture of 3% and a 10-year probability of major osteoporotic fracture of 20% are specific to the United States, which is based on the United States healthcare policy and where osteoporosis fits into the United States medical priorities. Other countries have their own thresholds that are very different.
DR. GALL: Dr. Maricic, we’ve talked about the DEXA and the FRAX. What other imaging tools are available and what are their roles?
DR. LISSE: The other issue is that the FRAX uses femoral neck density, which is a good predictor; however, in some patients, especially patients taking glucocorticoids, the spine density may be lower than the hip density. A low spine T-score should also be taken into account. We must individualize patients and not become overly focused on any one tool, although the use of the FRAX is certainly a major step forward from the use of only the T-score, for which, former guidelines recommended treatment for patients with scores within the wide range of −1 and −2.5.
DR. GALL: Another issue regarding DEXA is that we often get a false high BMD in patients with fracture of the lumbar spine, osteoarthritis, or issues with alignment of the spine. In such cases, it is important to examine the image, as the presence of bone abnormalities may render the spine DEXA score and the spine T-score less important.
DR. MARICIC: The role of other modalities in diagnosing osteoporosis in the United States is limited because of the widespread availability of DEXA in most communities. A few communities in Arizona have access to only quantitative computed tomography, which, although tends to overdiagnosise osteoporosis, can be helpful if bone density is found to be low and the clinical setting is appropriate.
Likewise, quantitative ultrasound in patients older than 65 years has a very valid role in estimating risk of fracture, but because DEXA is so widely available, we do not or usually should not use ultrasound instead of a DEXA scan and ultrasound cannot be used to monitor or follow-up patients with treatment. The only modality that should be used is the DEXA scan because of its very high precision.
DR. GALL: I have seen the use of ultrasound in both remote areas and nursing homes and other areas where it is difficult for patients to visit an office at which a DEXA scan performed, but it is not as accurate as the DEXA scan.
DR. MCCLUNG: Yes, but my understanding is that you use the ultrasound to determine whether you need a DEXA scan.
DR. MARICIC: Yes, but you shouldn’t use the ultrasound to start therapy or monitor therapy.
DR. MCCLUNG: FRAX, which can be calculated without a bone density test, provides a much more accurate assessment.
DR. MARICIC: And FRAX is probably cheaper and easier.
DR. GALL: Dr. McClung, would you talk a little bit about bone markers? I once heard you speak about their role in making a decision about when to treat patients with borderline DEXA and FRAX values.
DR. MCCLUNG: Bone markers are values obtained from blood and urine tests that reflect the activity of the bone cells. Generally, high levels of bone markers reflecting high levels of bone absorption are predictive of more rapid rates of bone loss. So, when a physician is deciding whether to treat, measuring a bone marker can sometimes provide important information to help make the decision. A bone turnover rate in the low–normal range in a bisphosphonate-naïve patient indicates that his/her bone density is stable and that he/she may benefit less from treatment compared to a patient with a high rate.
You can also use bone marker values to determine whether treatment has been effective, particularly if you use antiresorptive agents such as bisphosphonates. Patients with celiac disease or those who have had a gastrectomy may not absorb oral bisphosphonates, so their markers may remain high even if they are under treatment. Although there are certain patients for whom bone markers are helpful, the use of these markers is not routine in the management of most patients with osteoporosis.
DR. GALL: I would say that bone markers are generally overused. Another area in which the use of bone marker values may be helpful is compliance. I’ve used bone biomarker values in cases in which, for example, I start a patient on a weekly bisphosphonate and the patient comes back 12 weeks later saying that he/she doesn’t need a prescription. In such cases, I may perform bone marker testing to see whether the patient has been compliant.
Is there ever a time when we need to do a bone biopsy?
DR. MARICIC: I think a bone biopsy is only helpful for patients with long-term renal failure and dialysis to separate out the various forms of renal osteodystrophy. Although we used to perform a biopsy to help us understand bone turnover or detect osteomalacia, we can now do both with a biochemical test.
In some cases, we want to perform a bone marrow biopsy to look for marrow-based diseases such as mastocytosis or myeloma.
DR. GALL: I’d now like to discuss the prevention of osteoporosis and the risk factors that indicate testing and treatment of patients who are not postmenopausal.
DR. LISSE: There are risk factors that are nonmodifiable, including family history and genetics; modifiable risk factors, including level of exercise, especially weight-bearing exercise, which can be decreased; smoking, which can be discontinued or at least attenuated; and alcohol intake, which can be decreased. I think the role of caffeine is still controversial, although I’d love to hear from the other members of the panel about that.
I think we have overemphasized the importance of calcium and vitamin D supplementation. However, I measure the serum 25-hydroxy-vitamin D level, which according to me, is becoming more and more popular, and I try to keep it at around 30 ng/mL. I think all female patients should obtain some calcium supplementation, especially if their dietary intake is not adequate.
It was surprising to know that there’s probably a higher prevalence of vitamin D insufficiency in Arizona than in Minnesota in spite of the abundance of sunshine in the former.
DR. GALL: I consider patients who are thin, have poor bone stock, are malnourished, did not drink milk in their childhood, and have a family history of hip fracture in a close relative to be at higher risk, and I tend to test them early and consider either preventative treatment or just treatment for them.
I think an appropriate target for daily intake of calcium is between 800 and 1 200 mg, including that obtained from diet. The appropriate target for vitamin D intake in the absence of regular sun exposure is between 600 and 2 000 units per day.
DR. LISSE: Most recommendations that I’ve seen for calcium are about 800 mg, but they vary widely.
DR. MCCLUNG: Let’s get back to the management of premenopausal women with risk factors for low bone mass such as small body size, family history of osteoporosis, and medical problems that may lead to impaired bone health such as amenorrhea during early adolescence. For these women, maintaining adequate nutrition and physical activity and avoiding harmful lifestyle factors should be encouraged, but bone density testing is rarely necessary. If premenopausal women are estrogen replete, even if they have low bone density, treatment with osteoporosis drugs is not recommended. Individuals with risk factors for low bone density should be screened at the time of menopause to decide whether aggressive prevention of the rapid bone loss that occurs at that particularly important time is warranted.
DR. GALL: Dr. Maricic, there are 2 topics I’d like you to discuss. The first is comorbidities, specifically conditions such as thyroid disease, that we need to pay attention to, and the second is hormone replacement therapy and its current role.
DR. MARICIC: The NOF guidelines for the diagnosis and treatment of osteoporosis5 list a number of comorbid disorders and medications that may be responsible for a patient’s low bone density. We should consider the potential for osteoporosis in all men and women over the age of 50 years, especially if they have other disorders or medications known to affect bone mineral metabolism.
The NOF guidelines list about 30 diseases implicated in low bone density, including type 2 diabetes, hemochromatosis, and androgen- and estrogen-deficiency hypogonadism. Low bone density should be considered in patients with gastrointestinal (GI) disorders, of which celiac disease is the most important and common; those who have undergone GI bypass or other types of GI surgery; and those with hematological disorders such as mastocytosis and myeloma. Because chronic inflammatory production of cytokines and calcium immobilization lead to accelerated bone loss, low bone density should be considered in patients with rheumatic disorders, including systemic lupus; ankylosing spondylitis; and especially, rheumatoid arthritis, which FRAX identifies as a major risk factor for accelerated bone loss. Finally, low bone density should be considered in patients with a number of miscellaneous conditions, including chronic obstructive pulmonary disease, emphysema, chronic asthma, and sarcoidosis.
Regarding medications, the agents most commonly associated with low bone density are glucocorticoids. Low bone density should also be considered in patients taking certain anticonvulsants, especially phenobarbital and phenytoin, which can lead to accelerated degradation of 25-hydroxy-vitamin D and very low vitamin D levels and consequent osteomalacia, and in those taking aromatase inhibitors, which are now very important in the treatment of breast cancer.
DR. GALL: I’d just like to mention again the association between low bone density and both thyroid disease and overmedication in thyroid replacement therapy.
DR. MCCLUNG: We should probably also include proton pump inhibitors and selective serotonin reuptake inhibitors, which have been correlated with increased rates of fracture, although the mechanism underlying that correlation is unclear.
DR. LISSE: Do you mean that their wide use poses a concern?
DR. MCCLUNG: Yes, but their effect does not seem to be reflected in bone density testing. So, while recognizing their use as an independent risk factor for fracture is important, it is still unclear whether it should be considered while deciding whether to treat with an osteoporosis drug.
DR. MARICIC: Regarding hormone replacement therapy, we must first distinguish between hormone replacement, which is a combination of estrogen and progesterone for women with an intact uterus, and estrogen replacement or estrogen therapy, which is estrogen alone for women who have undergone a hysterectomy. Definitely, there are differences in the potential for adverse effects between these 2 groups.
I believe the data from the Women’s Health Initiative (WHI) suggested that increased risk of thrombosis and breast cancer is more commonly seen with hormone therapy and may be due more to progesterone than to estrogen replacement.6 There are also some differences in the risk with regard to the age of starting hormone therapy in our patients.
From the WHI, we know that hormone therapy prevents bone loss and reduces the risk of hip and other fractures. I think hormone therapy, especially estrogen therapy for women with an intact uterus, is an option to protect the skeleton in women with low bone density, especially those who are perimenopausal or early postmenopausal.
I personally avoid starting hormone therapy in patients above the age of 70 years and sometimes even above 65 years, but I believe it can be an option for women between the ages of 50 and 60 years with low bone density and no contraindications. For women with a strong family history of breast cancer, I usually recommend raloxifene. We’re now trying to limit the lifetime exposure of women to bisphosphonates, so I tend to reserve them for older women at very high risk of fracture, and almost never administer them to women aged 50 to 60 years.
I think that hormone therapy is the first option for women between 50 and 60 years with no strong family history of breast cancer. If there is, I start them on Evista, when possible.
DR. LISSE: I think that bisphosphonate therapy has been the primary choice of osteoporosis therapy since at least 2 000, which is when alendronate came into the market under its brand name. These drugs have been shown to reliably decrease fracture and improve the quality of life. Some drugs, particularly zoledronic acid, actually decrease mortality in patients with hip fracture.7
There are some differences among them. The first one on the market obviously was alendronate, so there’s probably the most data on it. Risedronate, alendronate, zoledronic acid, denosumab, and teriparatide have been shown to reduce vertebral and nonvertebral fracture; ibandronate, to reduce spinal but not nonvertebral fracture; alendronate, risedronate, denosumab, and zoledronic acid, to prevent hip fracture in postmenopausal women; and risedronate, to decrease vertebral and nonvertebral fracture in men.7
The original studies on the effect of many of these drugs on fracture risk were conducted with daily dosing, and equivalent dosing in weekly and monthly regimens has been used and approved, sometimes without fracture data. The drugs work, their safety-to-risk ratio is quite good, and we have long-term experience with them. I think it’s not surprising that a small percentage of patients who have been on them for a decade or more have issues and side effects. There are risks to taking these drugs, as there are with any other drug.
Probably, the most common risk is GI complications, ranging from irritation to esophageal perforation, which surfaced shortly after alendronate came on the market. To prevent GI complications, strict regimens have been developed for oral bisphosphonate use. To some extent, I think that they limit our patient population because many cannot or will not comply with the requirement of taking the drug first thing in the morning on an empty stomach with approximately 230 mL of water and remain upright for 30 to 60 minutes, depending on the medication, to realize any benefit.
Gastrointestinal complications, including those affecting the esophagus, are, by far, the most common issues observed with bisphosphonate therapy. Gastric ulcerations in certain patients, especially in those who use risedronate in coated form, have also been observed.
There are other probably less common but controversial issues, including a possible association with atrial fibrillation, which was identified in the Health Outcomes and Reduced Incidence with Zoledronic Acid Once Yearly (HORIZON) study.8 An association with alendronate was also suspected, but a recent meta-analysis does not support this association.9
An association with esophageal cancer has been anecdotally described in several populations, but I think that the data are mixed at this point. Then, there is the association with osteonecrosis of the jaw and atypical fracture, both of which are quite rare and occur in between 1 in 10 000 to 1 in 100 000 patients.10 I think that the wide coverage of this association in the lay press may have caused many patients who probably shouldn’t have gone off bisphosphonates, to go off them.
A positive byproduct of these associations being reported could be increased research into whether there should be a time or exposure limit to bisphosphonate use and whether a drug holiday from them is warranted.
DR. GALL: Dr. McClung, could you please describe how you monitor a patient after initiation of bisphosphonate therapy and then discuss an atypical fracture?
DR. MCCLUNG: Treatment with any osteoporosis drug is generally monitored by measuring BMD 2 years after the patient has started therapy. The changes in BMD with treatment are quite modest, so we expect to see either a modest increase or at least stability. If a clear decrease in BMD is observed, we consider noncompliance, secondary causes of bone loss, and other reasons for treatment failure.
In theory, biochemical testing of bone turnover should be useful in monitoring therapy. However, because of limitations in the quality and the reproducibility of the assays in the clinical setting, it is difficult to use the results. I think that marker measurement is not ready for use in the everyday clinic, although it has been very valuable in our research.
Atypical fracture is important. As mentioned earlier, there has been concern right from the time when bisphosphonate research began about the potential for skeletal consequences with long-term therapy because of the pharmacology of bisphosphonates. We know that these drugs accumulate in the skeleton over time, and there were concerns about whether we would see progressive inhibition of bone turnover with long-term therapy. Although we were pleased to observe persistent, but not progressive, reduction in bone turnover with long-term treatment, at least for up to 10 years,11 it appears that some individuals, or perhaps, a unique subset of patients, experience too much inhibition of bone metabolism for prolonged durations. When this happens, bone brittleness occurs and is manifested by atypical chalk stick-like fractures of the femur. It is clear that there is a duration-dependent increase in risk, with fractures occurring in roughly 1 in 1 000 patients who have completed at least 10 years of alendronate therapy.12 If we properly decide whom to treat and how to treat, it is important to recognize that this risk is much, much lower than the risk of important fracture, such as that of the hip and spine, in patients who are not treated with bisphosphonate therapy.13 We should reserve bisphosphonates for patients who clearly have osteoporosis or are at high risk for fracture.
DR. GALL: Dr. Maricic, can you conclude on the use of calcitonin, denosumab, and teriparatide, ie, some of the other drugs approved for the treatment of osteoporosis?
DR. MARICIC: Calcitonin is extremely safe but it isn’t very effective and has been shown to reduce spinal fracture risk by 33% and have no effect on hip fracture risk,14 so I personally haven’t prescribed it in almost 10 years.
Denosumab, a receptor activator of nuclear factor-kB (RANK)-ligand inhibitor and monoclonal antibody, is probably among the most effective medications we have. Because it bypasses the GI tract, it’s an excellent choice for patients who develop GI problems with bisphosphonates. Nevertheless, because it is a very potent osteoclast inhibitor, it must be used carefully in patients prone to hypocalcemia, with renal failure, or who are extremely vitamin D deficient. Although hypocalcemia must be considered before initiating its use, I think it is an extremely effective agent.
Teriparatide is the only anabolic agent currently approved for use in patients at high risk of fracture, and that’s for whom we mainly use it. Because its mode of action is completely different from that of bisphosphonates, we use it for patients with either very low bone density or with multiple fractures who have not responded to bisphosphonate therapy. The Food and Drug Administration (FDA) has approved teriparatide therapy for only 2 years, after which it must be followed by antiresorptive therapy. Unfortunately, I occasionally see patients who had not been given antiresorptives after 2 years of teriparatide, causing them to lose any benefit they may have derived from being on this very expensive drug for 2 years. It must always be followed up by an antiresorptive drug.
DR. MCCLUNG: Regarding a drug holiday, you can consider it if you observe a duration-dependent increase in the risk of atypical fracture.15,16 The FDA has provided us with at least a general template for when it should be considered.17 Based on the limited information we have, it is suggested that fracture risk be reassessed after 3 to 5 years of therapy. For patients with a modest risk for fracture at that point, ie, they no longer have osteoporosis according to the results of bone density testing, it is suggested that therapy be discontinued for 1 to 2 years.
For patients found to remain at high risk for fracture, especially spine fracture, continuation of therapy is recommended. The evidence for this recommendation came from 2 large extension studies of the alendronate and zoledronic acid clinical trials.18,19
If treatment is discontinued, the question of when, if ever, to restart therapy arises. The data from the clinical trials suggest that neither the measurement of bone density nor bone markers are helpful in that regard. So, in my clinic, we treat the holiday like a vacation that has a finite end.13 We take patients off alendronate for 2 or 3 years but off risedronate for only 1 year because its effects wear off more quickly. At that point, we reassess the patient. If he/she meets the criteria for treatment with an osteoporosis drug again, we decide on the best treatment option.
DR. GALL: Dr. Lisse, please describe how treating steroid-induced male osteoporosis differs from treating postmenopausal and female osteoporosis.
DR. LISSE: Before I do so, I wanted to add that a flu-like syndrome sometimes occurs with bisphosphonate use. In a corollary to what Dr. McClung just discussed, the European registries provide at least some reassuring evidence that when patients stop taking bisphosphonates, the risk of atypical fracture appears to decrease fairly rapidly, whereas bisphosphonate efficacy doesn’t seem to tail off as quickly.15 It should be reassuring that if patients stop taking a drug, they may have some protection during that period in terms of osteoporosis coverage, unlike when they stop taking estrogen or teriparatide, whose effects tail off rather rapidly.
With glucocorticoid-induced osteoporosis, the biggest issue is probably the fact that fractures tend to occur early and at a higher bone density than what we’re used to seeing with traditional postmenopausal osteoporosis. So it has thrown the old paradigm into disarray. Before the introduction of FRAX, treatment was recommended for patients with a T-score of −1. As the FRAX includes glucocorticoids as a risk factor, its use is very helpful in this regard.
FRAX does have its limitations, and there are physicians who are concerned that it does not include consideration of steroid duration or dose. I think the American College of Rheumatology’s recommendation is to measure bone density and then subdivide the patient population into low risk, moderate risk, and high risk according to the risk for major osteoporotic fracture of ≤10%, 10% to 20%, and ≥20%, respectively.20 Treatment should be considered based on the risk and dosage. Patients taking a dose of 7.5 mg or higher probably need treatment, even if they’re at low risk. Patients on lower doses probably need treatment if they remain on glucocorticoid therapy for more than 3 months.
Although most fractures due to steroid use are of the spine, not all are. I’m sure all of us on the panel are old enough to remember the old asthmatics who used to take large doses of steroids. Certainly, one of the easiest things to do is to taper patients off steroids or reduce the dose down to the lowest possible dose. The use of topical steroids is preferred to that of oral steroids or inhaled steroids.
The NOF guidelines for testing for male osteoporosis differ from those for testing female osteoporosis. Although they recommend that all women older than 65 years be tested, they do not recommend routine testing in men older than 70 years. I think that this latter recommendation, which is based on a US Preventative Health Service Task Force recommendation,21 is probably a mistake and rather confusing.
Even though there has been a study of raloxifene in men, the guidelines are more limited regarding how men who are hypogonadal should be treated. Testosterone should be measured in all men, and those found to be hypogonadal deserve testosterone replacement. Again, the backbone of therapy is the use of bisphosphonates for the most part. Teriparatide has also been approved for use in men, although it has not been proven to provide protection against hip fracture.
DR. GALL: We’ve had an extensive discussion today on low bone density diseases, focusing mostly on osteoporosis. We have discussed at some length the risk factors and the diagnostic tools that we have, particularly the DEXA scan and FRAX, to assess the risk of fracture. We then discussed the risk factors, including comorbidities; means of prevention; and the different treatment modalities, including physical therapy and several nonpharmacologic therapies, in addition to the various medical therapies, including hormone replacement, bisphosphonates, calcitonin, denosumab, and teriparatide. We’ve also briefly touched on the differences between postmenopausal osteoporosis and steroid-induced osteoporosis and male osteoporosis.
FoxP2 Media LLC is the publisher of The Medical Roundtable.
Dr. Maricic, what are the different types of osteopenia that primary care clinicians encounter in their practice?
DR. MARICIC: Osteopenia has several connotations, depending on the setting in which it is diagnosed. Radiologists will describe a patient as being osteopenic based on the subjective assessment that the patient has low bone density on radiography. With particular reference to osteoporosis or bone densitometry, osteopenia refers to a dual-energy X-ray absorptiometry (DEXA) T-score, which is a comparison of the bone mass of a patient to that of a young, ideal patient of the same sex and ethnicity.
DR. GALL: Is osteoporosis the only disease or are there other diseases that we may see associated with osteopenia?
DR. MARICIC: Neither osteopenia nor osteoporosis diagnosed by the DEXA score is necessarily caused by estrogen deficiency. The DEXA report tells you whether a patient has low bone density but not about the etiology. Although estrogen deficiency is a common cause of osteopenia or osteoporosis, other conditions such as primary hyperparathyroidism, multiple myeloma, vitamin D deficiency, osteomalacia and the conditions associated with it, chronic renal failure with osteitis, fibrosis cystica, and hyperparathyroidism should be considered by the primary care physician.
It’s typically recommended that all patients with low bone density diagnosed by their DEXA score have a chemistry screening done to determine the calcium and alkaline phosphatase levels, as well as 25-hydroxy-vitamin D level, in order to ensure that the patient is not vitamin D deficient. A 24-hour urine test is also recommended to ensure that the vitamin D level is not low, suggesting malabsorption, such as what we see in vitamin D deficiency or celiac disease, or that it’s not high, suggesting idiopathic hypercalciuria.
Those are the main, most common disorders that we’d like to exclude in patients with osteopenia or especially osteoporosis. We would also want to exclude secondary hyperparathyroidism; vitamin D deficiency; and, when suspected, celiac disease, idiopathic hypercalciuria, and multiple myeloma. Those are some of the conditions to be considered, especially when the T-score does not accord with the clinical history.
DR. GALL: Dr. McClung, can you please discuss the use of X-ray and DEXA scan as diagnostic tools?
DR. MCCLUNG: These 2 tools have completely different roles. We use the DEXA scan to measure bone density, primarily in the hip and the spine, in postmenopausal women and older men to obtain a T-score that is, a value comparing the patient’s bone density to the average bone density value of young adults of the same sex. A T-score of −2.5 is thought to be consistent with osteoporosis in postmenopausal women and older men, but, as Dr. Maricic pointed out, a T-score of −2.5 only tells us that the patient has low bone density. Before we conclude that the patient has osteoporosis, all of the other causes of low bone density, including osteomalacia, osteogenesis imperfecta, and any of the secondary causes that were outlined, need to be excluded.
I believe that osteopenia, which is defined as a T-score value between −1 and −2.5, is not a very useful diagnostic term for clinical purposes. We know that low bone density is one of several risk factors for fracture. Within the group of patients with osteopenia are those who are at very high risk for fracture, including older people and those who have had fractures, as well as many patients who are not at high risk for fracture. In response to this problem, the World Health Organization developed the Fracture Risk Assessment Tool (FRAX), a fracture risk assessment tool used to help distinguish patients with high or low fracture risk.1
DR. GALL: Clinicians often find the terms used in the DEXA report confusing. Could you define absolute bone mineral density (BMD) and Z-score for the clinician?
DR. MCCLUNG: The BMD test measures the bone mineral content of a specific anatomic region and the area of that region and divides those 2 numbers to get an estimate of the BMD. It is not a true bone density test, but rather a measurement of calcium content within a region of bone.
The BMD value is translated into a T-score by comparing it to the average value for healthy young adults of the same sex and is then used for diagnosis in postmenopausal women. The Z-score is obtained by comparing the patient’s BMD value with the expected average value for healthy adults of the same age and sex and is used to determine whether a low BMD value for that patient is unexpected. A low T-score for a 55-year-old woman may be surprising, whereas a low T-score for a 90-year-old woman would be expected. A Z-score lets us know how concerned we should be.
DR. GALL: Dr. Lisse, you read a lot of DEXA scans. What can give a false value? Can you tell us about that and discuss the FRAX?
DR. LISSE: Dr. McClung wrote a very nice treatise on the FRAX,2 a tool that assesses the risk of fracture in bisphosphonate-naïve patients based on the consideration of risk factors, in addition to the BMD value. The DEXA measures bone quantity, but does not assess the risk factors for osteoporosis, of which I think prior fracture and patient age are the most important.
By considering the risk factors of smoking history, advanced age, prior fracture, and family history of fracture, and by adjusting for the country and racial origin of the patient, the FRAX provides a better measure of fracture risk for the osteopenic patient compared to BMD alone. It also provides a value, the 10-year fracture risk, that’s easier for the primary doctor to interpret compared to T-scores and Z-scores, which require comparison in terms of standard deviations.
DR. GALL: How can we use the FRAX if we are interested in doing so?
DR. LISSE: The tool is available online and very easy to use. Usually, you complete the assay using the BMD value. You can actually perform the assay without a BMD value, but I think that including the BMD value improves the precision and predictive value of the results. You just plug in the information online and are given a reading. Recent literature suggests that you can use it for patients who are taking bisphosphonates, but I think that’s a little bit further down the line in our discussion.
DR. GALL: A FRAX result states that a patient has a 10-year risk of major fracture of 8%. How do I use that result? Should I start treating the patient?
DR. LISSE: Several studies that examined local socioeconomic factors concluded that using a 20% risk of major osteoporotic fracture and a 3% risk of hip fracture are cost-effective thresholds for the diagnosis of patients in the United States. Although they should not be considered absolute, they are the current thresholds, and I think physicians can use them as criteria to make treatment decisions.
DR. MCCLUNG: That description is correct. FRAX has a scientific base and is a very highly validated tool that accurately predicts fracture probability in healthy middle-aged and older adults.1 Using the tool is a good first step while considering whether osteoporosis treatment is justified. The threshold values of a 10-year probability of hip fracture of 3% and a 10-year probability of major osteoporotic fracture of 20% are specific to the United States, which is based on the United States healthcare policy and where osteoporosis fits into the United States medical priorities. Other countries have their own thresholds that are very different.
DR. GALL: Dr. Maricic, we’ve talked about the DEXA and the FRAX. What other imaging tools are available and what are their roles?
DR. LISSE: The other issue is that the FRAX uses femoral neck density, which is a good predictor; however, in some patients, especially patients taking glucocorticoids, the spine density may be lower than the hip density. A low spine T-score should also be taken into account. We must individualize patients and not become overly focused on any one tool, although the use of the FRAX is certainly a major step forward from the use of only the T-score, for which, former guidelines recommended treatment for patients with scores within the wide range of −1 and −2.5.
DR. GALL: Another issue regarding DEXA is that we often get a false high BMD in patients with fracture of the lumbar spine, osteoarthritis, or issues with alignment of the spine. In such cases, it is important to examine the image, as the presence of bone abnormalities may render the spine DEXA score and the spine T-score less important.
DR. MARICIC: The role of other modalities in diagnosing osteoporosis in the United States is limited because of the widespread availability of DEXA in most communities. A few communities in Arizona have access to only quantitative computed tomography, which, although tends to overdiagnosise osteoporosis, can be helpful if bone density is found to be low and the clinical setting is appropriate.
Likewise, quantitative ultrasound in patients older than 65 years has a very valid role in estimating risk of fracture, but because DEXA is so widely available, we do not or usually should not use ultrasound instead of a DEXA scan and ultrasound cannot be used to monitor or follow-up patients with treatment. The only modality that should be used is the DEXA scan because of its very high precision.
DR. GALL: I have seen the use of ultrasound in both remote areas and nursing homes and other areas where it is difficult for patients to visit an office at which a DEXA scan performed, but it is not as accurate as the DEXA scan.
DR. MCCLUNG: Yes, but my understanding is that you use the ultrasound to determine whether you need a DEXA scan.
DR. MARICIC: Yes, but you shouldn’t use the ultrasound to start therapy or monitor therapy.
DR. MCCLUNG: FRAX, which can be calculated without a bone density test, provides a much more accurate assessment.
DR. MARICIC: And FRAX is probably cheaper and easier.
DR. GALL: Dr. McClung, would you talk a little bit about bone markers? I once heard you speak about their role in making a decision about when to treat patients with borderline DEXA and FRAX values.
DR. MCCLUNG: Bone markers are values obtained from blood and urine tests that reflect the activity of the bone cells. Generally, high levels of bone markers reflecting high levels of bone absorption are predictive of more rapid rates of bone loss. So, when a physician is deciding whether to treat, measuring a bone marker can sometimes provide important information to help make the decision. A bone turnover rate in the low–normal range in a bisphosphonate-naïve patient indicates that his/her bone density is stable and that he/she may benefit less from treatment compared to a patient with a high rate.
You can also use bone marker values to determine whether treatment has been effective, particularly if you use antiresorptive agents such as bisphosphonates. Patients with celiac disease or those who have had a gastrectomy may not absorb oral bisphosphonates, so their markers may remain high even if they are under treatment. Although there are certain patients for whom bone markers are helpful, the use of these markers is not routine in the management of most patients with osteoporosis.
DR. GALL: I would say that bone markers are generally overused. Another area in which the use of bone marker values may be helpful is compliance. I’ve used bone biomarker values in cases in which, for example, I start a patient on a weekly bisphosphonate and the patient comes back 12 weeks later saying that he/she doesn’t need a prescription. In such cases, I may perform bone marker testing to see whether the patient has been compliant.
Is there ever a time when we need to do a bone biopsy?
DR. MARICIC: I think a bone biopsy is only helpful for patients with long-term renal failure and dialysis to separate out the various forms of renal osteodystrophy. Although we used to perform a biopsy to help us understand bone turnover or detect osteomalacia, we can now do both with a biochemical test.
In some cases, we want to perform a bone marrow biopsy to look for marrow-based diseases such as mastocytosis or myeloma.
DR. GALL: I’d now like to discuss the prevention of osteoporosis and the risk factors that indicate testing and treatment of patients who are not postmenopausal.
DR. LISSE: There are risk factors that are nonmodifiable, including family history and genetics; modifiable risk factors, including level of exercise, especially weight-bearing exercise, which can be decreased; smoking, which can be discontinued or at least attenuated; and alcohol intake, which can be decreased. I think the role of caffeine is still controversial, although I’d love to hear from the other members of the panel about that.
I think we have overemphasized the importance of calcium and vitamin D supplementation. However, I measure the serum 25-hydroxy-vitamin D level, which according to me, is becoming more and more popular, and I try to keep it at around 30 ng/mL. I think all female patients should obtain some calcium supplementation, especially if their dietary intake is not adequate.
It was surprising to know that there’s probably a higher prevalence of vitamin D insufficiency in Arizona than in Minnesota in spite of the abundance of sunshine in the former.
DR. GALL: I consider patients who are thin, have poor bone stock, are malnourished, did not drink milk in their childhood, and have a family history of hip fracture in a close relative to be at higher risk, and I tend to test them early and consider either preventative treatment or just treatment for them.
I think an appropriate target for daily intake of calcium is between 800 and 1 200 mg, including that obtained from diet. The appropriate target for vitamin D intake in the absence of regular sun exposure is between 600 and 2 000 units per day.
DR. LISSE: Most recommendations that I’ve seen for calcium are about 800 mg, but they vary widely.
DR. MCCLUNG: Let’s get back to the management of premenopausal women with risk factors for low bone mass such as small body size, family history of osteoporosis, and medical problems that may lead to impaired bone health such as amenorrhea during early adolescence. For these women, maintaining adequate nutrition and physical activity and avoiding harmful lifestyle factors should be encouraged, but bone density testing is rarely necessary. If premenopausal women are estrogen replete, even if they have low bone density, treatment with osteoporosis drugs is not recommended. Individuals with risk factors for low bone density should be screened at the time of menopause to decide whether aggressive prevention of the rapid bone loss that occurs at that particularly important time is warranted.
DR. GALL: Dr. Maricic, there are 2 topics I’d like you to discuss. The first is comorbidities, specifically conditions such as thyroid disease, that we need to pay attention to, and the second is hormone replacement therapy and its current role.
DR. MARICIC: The NOF guidelines for the diagnosis and treatment of osteoporosis5 list a number of comorbid disorders and medications that may be responsible for a patient’s low bone density. We should consider the potential for osteoporosis in all men and women over the age of 50 years, especially if they have other disorders or medications known to affect bone mineral metabolism.
The NOF guidelines list about 30 diseases implicated in low bone density, including type 2 diabetes, hemochromatosis, and androgen- and estrogen-deficiency hypogonadism. Low bone density should be considered in patients with gastrointestinal (GI) disorders, of which celiac disease is the most important and common; those who have undergone GI bypass or other types of GI surgery; and those with hematological disorders such as mastocytosis and myeloma. Because chronic inflammatory production of cytokines and calcium immobilization lead to accelerated bone loss, low bone density should be considered in patients with rheumatic disorders, including systemic lupus; ankylosing spondylitis; and especially, rheumatoid arthritis, which FRAX identifies as a major risk factor for accelerated bone loss. Finally, low bone density should be considered in patients with a number of miscellaneous conditions, including chronic obstructive pulmonary disease, emphysema, chronic asthma, and sarcoidosis.
Regarding medications, the agents most commonly associated with low bone density are glucocorticoids. Low bone density should also be considered in patients taking certain anticonvulsants, especially phenobarbital and phenytoin, which can lead to accelerated degradation of 25-hydroxy-vitamin D and very low vitamin D levels and consequent osteomalacia, and in those taking aromatase inhibitors, which are now very important in the treatment of breast cancer.
DR. GALL: I’d just like to mention again the association between low bone density and both thyroid disease and overmedication in thyroid replacement therapy.
DR. MCCLUNG: We should probably also include proton pump inhibitors and selective serotonin reuptake inhibitors, which have been correlated with increased rates of fracture, although the mechanism underlying that correlation is unclear.
DR. LISSE: Do you mean that their wide use poses a concern?
DR. MCCLUNG: Yes, but their effect does not seem to be reflected in bone density testing. So, while recognizing their use as an independent risk factor for fracture is important, it is still unclear whether it should be considered while deciding whether to treat with an osteoporosis drug.
DR. MARICIC: Regarding hormone replacement therapy, we must first distinguish between hormone replacement, which is a combination of estrogen and progesterone for women with an intact uterus, and estrogen replacement or estrogen therapy, which is estrogen alone for women who have undergone a hysterectomy. Definitely, there are differences in the potential for adverse effects between these 2 groups.
I believe the data from the Women’s Health Initiative (WHI) suggested that increased risk of thrombosis and breast cancer is more commonly seen with hormone therapy and may be due more to progesterone than to estrogen replacement.6 There are also some differences in the risk with regard to the age of starting hormone therapy in our patients.
From the WHI, we know that hormone therapy prevents bone loss and reduces the risk of hip and other fractures. I think hormone therapy, especially estrogen therapy for women with an intact uterus, is an option to protect the skeleton in women with low bone density, especially those who are perimenopausal or early postmenopausal.
I personally avoid starting hormone therapy in patients above the age of 70 years and sometimes even above 65 years, but I believe it can be an option for women between the ages of 50 and 60 years with low bone density and no contraindications. For women with a strong family history of breast cancer, I usually recommend raloxifene. We’re now trying to limit the lifetime exposure of women to bisphosphonates, so I tend to reserve them for older women at very high risk of fracture, and almost never administer them to women aged 50 to 60 years.
I think that hormone therapy is the first option for women between 50 and 60 years with no strong family history of breast cancer. If there is, I start them on Evista, when possible.
DR. LISSE: I think that bisphosphonate therapy has been the primary choice of osteoporosis therapy since at least 2 000, which is when alendronate came into the market under its brand name. These drugs have been shown to reliably decrease fracture and improve the quality of life. Some drugs, particularly zoledronic acid, actually decrease mortality in patients with hip fracture.7
There are some differences among them. The first one on the market obviously was alendronate, so there’s probably the most data on it. Risedronate, alendronate, zoledronic acid, denosumab, and teriparatide have been shown to reduce vertebral and nonvertebral fracture; ibandronate, to reduce spinal but not nonvertebral fracture; alendronate, risedronate, denosumab, and zoledronic acid, to prevent hip fracture in postmenopausal women; and risedronate, to decrease vertebral and nonvertebral fracture in men.7
The original studies on the effect of many of these drugs on fracture risk were conducted with daily dosing, and equivalent dosing in weekly and monthly regimens has been used and approved, sometimes without fracture data. The drugs work, their safety-to-risk ratio is quite good, and we have long-term experience with them. I think it’s not surprising that a small percentage of patients who have been on them for a decade or more have issues and side effects. There are risks to taking these drugs, as there are with any other drug.
Probably, the most common risk is GI complications, ranging from irritation to esophageal perforation, which surfaced shortly after alendronate came on the market. To prevent GI complications, strict regimens have been developed for oral bisphosphonate use. To some extent, I think that they limit our patient population because many cannot or will not comply with the requirement of taking the drug first thing in the morning on an empty stomach with approximately 230 mL of water and remain upright for 30 to 60 minutes, depending on the medication, to realize any benefit.
Gastrointestinal complications, including those affecting the esophagus, are, by far, the most common issues observed with bisphosphonate therapy. Gastric ulcerations in certain patients, especially in those who use risedronate in coated form, have also been observed.
There are other probably less common but controversial issues, including a possible association with atrial fibrillation, which was identified in the Health Outcomes and Reduced Incidence with Zoledronic Acid Once Yearly (HORIZON) study.8 An association with alendronate was also suspected, but a recent meta-analysis does not support this association.9
An association with esophageal cancer has been anecdotally described in several populations, but I think that the data are mixed at this point. Then, there is the association with osteonecrosis of the jaw and atypical fracture, both of which are quite rare and occur in between 1 in 10 000 to 1 in 100 000 patients.10 I think that the wide coverage of this association in the lay press may have caused many patients who probably shouldn’t have gone off bisphosphonates, to go off them.
A positive byproduct of these associations being reported could be increased research into whether there should be a time or exposure limit to bisphosphonate use and whether a drug holiday from them is warranted.
DR. GALL: Dr. McClung, could you please describe how you monitor a patient after initiation of bisphosphonate therapy and then discuss an atypical fracture?
DR. MCCLUNG: Treatment with any osteoporosis drug is generally monitored by measuring BMD 2 years after the patient has started therapy. The changes in BMD with treatment are quite modest, so we expect to see either a modest increase or at least stability. If a clear decrease in BMD is observed, we consider noncompliance, secondary causes of bone loss, and other reasons for treatment failure.
In theory, biochemical testing of bone turnover should be useful in monitoring therapy. However, because of limitations in the quality and the reproducibility of the assays in the clinical setting, it is difficult to use the results. I think that marker measurement is not ready for use in the everyday clinic, although it has been very valuable in our research.
Atypical fracture is important. As mentioned earlier, there has been concern right from the time when bisphosphonate research began about the potential for skeletal consequences with long-term therapy because of the pharmacology of bisphosphonates. We know that these drugs accumulate in the skeleton over time, and there were concerns about whether we would see progressive inhibition of bone turnover with long-term therapy. Although we were pleased to observe persistent, but not progressive, reduction in bone turnover with long-term treatment, at least for up to 10 years,11 it appears that some individuals, or perhaps, a unique subset of patients, experience too much inhibition of bone metabolism for prolonged durations. When this happens, bone brittleness occurs and is manifested by atypical chalk stick-like fractures of the femur. It is clear that there is a duration-dependent increase in risk, with fractures occurring in roughly 1 in 1 000 patients who have completed at least 10 years of alendronate therapy.12 If we properly decide whom to treat and how to treat, it is important to recognize that this risk is much, much lower than the risk of important fracture, such as that of the hip and spine, in patients who are not treated with bisphosphonate therapy.13 We should reserve bisphosphonates for patients who clearly have osteoporosis or are at high risk for fracture.
DR. GALL: Dr. Maricic, can you conclude on the use of calcitonin, denosumab, and teriparatide, ie, some of the other drugs approved for the treatment of osteoporosis?
DR. MARICIC: Calcitonin is extremely safe but it isn’t very effective and has been shown to reduce spinal fracture risk by 33% and have no effect on hip fracture risk,14 so I personally haven’t prescribed it in almost 10 years.
Denosumab, a receptor activator of nuclear factor-kB (RANK)-ligand inhibitor and monoclonal antibody, is probably among the most effective medications we have. Because it bypasses the GI tract, it’s an excellent choice for patients who develop GI problems with bisphosphonates. Nevertheless, because it is a very potent osteoclast inhibitor, it must be used carefully in patients prone to hypocalcemia, with renal failure, or who are extremely vitamin D deficient. Although hypocalcemia must be considered before initiating its use, I think it is an extremely effective agent.
Teriparatide is the only anabolic agent currently approved for use in patients at high risk of fracture, and that’s for whom we mainly use it. Because its mode of action is completely different from that of bisphosphonates, we use it for patients with either very low bone density or with multiple fractures who have not responded to bisphosphonate therapy. The Food and Drug Administration (FDA) has approved teriparatide therapy for only 2 years, after which it must be followed by antiresorptive therapy. Unfortunately, I occasionally see patients who had not been given antiresorptives after 2 years of teriparatide, causing them to lose any benefit they may have derived from being on this very expensive drug for 2 years. It must always be followed up by an antiresorptive drug.
DR. MCCLUNG: Regarding a drug holiday, you can consider it if you observe a duration-dependent increase in the risk of atypical fracture.15,16 The FDA has provided us with at least a general template for when it should be considered.17 Based on the limited information we have, it is suggested that fracture risk be reassessed after 3 to 5 years of therapy. For patients with a modest risk for fracture at that point, ie, they no longer have osteoporosis according to the results of bone density testing, it is suggested that therapy be discontinued for 1 to 2 years.
For patients found to remain at high risk for fracture, especially spine fracture, continuation of therapy is recommended. The evidence for this recommendation came from 2 large extension studies of the alendronate and zoledronic acid clinical trials.18,19
If treatment is discontinued, the question of when, if ever, to restart therapy arises. The data from the clinical trials suggest that neither the measurement of bone density nor bone markers are helpful in that regard. So, in my clinic, we treat the holiday like a vacation that has a finite end.13 We take patients off alendronate for 2 or 3 years but off risedronate for only 1 year because its effects wear off more quickly. At that point, we reassess the patient. If he/she meets the criteria for treatment with an osteoporosis drug again, we decide on the best treatment option.
DR. GALL: Dr. Lisse, please describe how treating steroid-induced male osteoporosis differs from treating postmenopausal and female osteoporosis.
DR. LISSE: Before I do so, I wanted to add that a flu-like syndrome sometimes occurs with bisphosphonate use. In a corollary to what Dr. McClung just discussed, the European registries provide at least some reassuring evidence that when patients stop taking bisphosphonates, the risk of atypical fracture appears to decrease fairly rapidly, whereas bisphosphonate efficacy doesn’t seem to tail off as quickly.15 It should be reassuring that if patients stop taking a drug, they may have some protection during that period in terms of osteoporosis coverage, unlike when they stop taking estrogen or teriparatide, whose effects tail off rather rapidly.
With glucocorticoid-induced osteoporosis, the biggest issue is probably the fact that fractures tend to occur early and at a higher bone density than what we’re used to seeing with traditional postmenopausal osteoporosis. So it has thrown the old paradigm into disarray. Before the introduction of FRAX, treatment was recommended for patients with a T-score of −1. As the FRAX includes glucocorticoids as a risk factor, its use is very helpful in this regard.
FRAX does have its limitations, and there are physicians who are concerned that it does not include consideration of steroid duration or dose. I think the American College of Rheumatology’s recommendation is to measure bone density and then subdivide the patient population into low risk, moderate risk, and high risk according to the risk for major osteoporotic fracture of ≤10%, 10% to 20%, and ≥20%, respectively.20 Treatment should be considered based on the risk and dosage. Patients taking a dose of 7.5 mg or higher probably need treatment, even if they’re at low risk. Patients on lower doses probably need treatment if they remain on glucocorticoid therapy for more than 3 months.
Although most fractures due to steroid use are of the spine, not all are. I’m sure all of us on the panel are old enough to remember the old asthmatics who used to take large doses of steroids. Certainly, one of the easiest things to do is to taper patients off steroids or reduce the dose down to the lowest possible dose. The use of topical steroids is preferred to that of oral steroids or inhaled steroids.
The NOF guidelines for testing for male osteoporosis differ from those for testing female osteoporosis. Although they recommend that all women older than 65 years be tested, they do not recommend routine testing in men older than 70 years. I think that this latter recommendation, which is based on a US Preventative Health Service Task Force recommendation,21 is probably a mistake and rather confusing.
Even though there has been a study of raloxifene in men, the guidelines are more limited regarding how men who are hypogonadal should be treated. Testosterone should be measured in all men, and those found to be hypogonadal deserve testosterone replacement. Again, the backbone of therapy is the use of bisphosphonates for the most part. Teriparatide has also been approved for use in men, although it has not been proven to provide protection against hip fracture.
DR. GALL: We’ve had an extensive discussion today on low bone density diseases, focusing mostly on osteoporosis. We have discussed at some length the risk factors and the diagnostic tools that we have, particularly the DEXA scan and FRAX, to assess the risk of fracture. We then discussed the risk factors, including comorbidities; means of prevention; and the different treatment modalities, including physical therapy and several nonpharmacologic therapies, in addition to the various medical therapies, including hormone replacement, bisphosphonates, calcitonin, denosumab, and teriparatide. We’ve also briefly touched on the differences between postmenopausal osteoporosis and steroid-induced osteoporosis and male osteoporosis.
FoxP2 Media LLC is the publisher of The Medical Roundtable.
FoxP2 Media LLC
The Medical Roundtable: Osteoarthritis Coping Guidelines for the Elderly
DR. BLOCK: Throughout the 20th century, we always considered OA to be primarily a degenerative disease of cartilage, but during the past decade or more, 2 things have become apparent: First, from a clinical perspective, OA is not just a structural degenerative process, it is primarily a very painful disease with pain being its most prominent feature. Second, the degenerative process involves not only the cartilage but the entire joint and all the involved structures within it.
A reasonable working definition of OA right now would be a painful degenerative process involving all the joint structures and is not primarily inflammatory but involves progressive deterioration of joint structures including articular cartilage. It’s important to keep in mind that pain itself is critical because as individuals age, everybody has some degenerative processes in their joints, and if we look hard enough, we can find the pathological features of OA in all elderly people. However, not all elderly people actually have the clinical disease.
DR. GALL: In a moment, we’ll talk about the causes of this, but, Dr. Moskowitz, would you tell us what the burden of this disease is here in the United States and worldwide?
DR. MOSKOWITZ: Well, it’s the most common form of arthritis and affects nearly 27 million or more Americans.1 It’s the most common disabling disease of the rheumatologic group, and it’s very disabling among all diseases.
DR. GALL: In the United States, you say that there’s about—
DR. MOSKOWITZ: About 27 million people with OA. And, in a decade, it is anticipated that the number will significantly increase.
DR. GALL: How many of those people are affected to the point that they’re unable to do their normal activities of daily living (ADLs)?
DR. MOSKOWITZ: Well, in terms of disability, about 80% of patients with OA have some degree of movement limitation, 25% cannot perform major ADLs, 11% of adults with knee OA need help with personal care, and 14% require help with routine needs.2
DR. GALL: Dr. Block, you talked about pain as a major manifestation of this disease. Does everybody who has OA present with pain?
DR. BLOCK: Again, it’s a definitional question; there is a distinction between clinically evident OA and asymptomatic structural degeneration of the joint. The population that Dr. Moskowitz was just describing, which comprises 20 to 30 million people, has doctor-diagnosed symptomatic OA, which is a painful disease. A much higher number of people who have structural degenerative disease would be diagnosed by radiography and said to fulfill a radiographic definition of OA, but all of the people who Dr. Moskowitz refer to have pain.
DR. MOSKOWITZ: Pain is what brings a patient to the doctor. I’ll often ask my patients, “How are you doing and what can I do for you?” They have limited motion and can’t get around. They say, “Dr. Moskowitz, I just want to get rid of the pain. I’ll worry about the swelling and getting around later.” And, that’s what really motivates them to get medical attention.
DR. GALL: Often, we will get consultations from primary care physicians for a patient who has OA that’s been picked up by radiography, and the patients will come and really not complain of pain in that area or possibly might have another underlying disease that’s causing the pain that’s not related to the OA. Dr. Block, do you want to say a word about that?
DR. BLOCK: Sure. By the age of 80 years, essentially, 100% of the population will have evidence of radiographic OA in at least one of their large joints. If you start with that number and you’re then referred an elderly patient merely for the radiographic appearance of OA, that doesn’t tell you very much about what’s going on clinically, and so, most of us would not even describe that as a clinical disease.
They may have a degenerative process in their knee, hip, etc, but of that group, say at the age of 80 years where almost 100% of people will have radiographic evidence of OA, only about 15% to 20% have symptomatic disease.3 That’s the culprit that I think we’re talking about clinically—people who not only have degenerative structural disease but also have symptoms that are interfering with their quality of life or their daily activities.
DR. MOSKOWITZ: I think that’s an important point, Dr. Block. For example, we all see patients who come in complaining of hip pain and you examine their knees as well and they have terrible genu varum (bowleg) and you say, “Do your knees bother you?” and they say, “Doctor, my knees are fine.” You think, my goodness, how can you walk? It isn’t necessarily true that if you have a deformity or involvement of a joint that you’re going to have pain, but if you have painful joints, that brings you to the doctor.
DR. GALL: Dr. Block, once again, I would like you to summarize the etiology of OA. Obviously, there has been much research and much information that has come to light in recent years regarding the causes of this disease.
DR. BLOCK: As I said, throughout the 20th century, people interested in the pathophysiology of OA were quite convinced that it was a degenerative process of the articular cartilage. As people age, the articular cartilage starts degenerating, and when there is absence of intact cartilage that ought to be providing essentially frictionless articulation of the bones during motion, there is deterioration, and that’s what we always considered to be OA.
Of course, the missing link in that paradigm is that cartilage itself has no nerve or blood supply and is therefore painless, and yet, people have horrible pain with this disease. So, our understanding over the past 10 or 15 years has really dramatically progressed, and we have begun to appreciate both degenerative and reactive processes in the subchondral bone, the ligaments, and the muscle.
The way we look at the pathophysiology of the onset and progression of the disease today is that there is primarily a degenerative process that stimulates reactive processes throughout the joint and even in the immune and inflammatory pathways. So, the reaction to that degenerative process sometimes causes stimulation of nociceptors and a lot of pain because of either local inflammation or mechanical alterations.
The degeneration itself, as I said, is painless. Often, people can have very degenerative joints, as Dr. Moskowitz just mentioned, and have no pain and therefore no clinical symptoms, so the disease is a combination of the degeneration and the secondary stimulated pain.
DR. GALL: So, is this primarily a biochemical or a cellular disease? Is it a genetic disease?
DR. BLOCK: Yes to all.
DR. MOSKOWITZ: There are a number of factors, Dr. Gall, involved in this condition. First, for example, is aging. We see that the frequency of OA increases logarithmically as we age, and there are pathologic changes with age that appear to lead to OA. For example, pigmented collections in the joint, comprising advanced glycation end products, occur that may predispose a patient to OA. You’re going to find more OA in an aging population.
The second factor is trauma. Professional football and basketball players, for example, who have chronic trauma, meniscal injuries, or cruciate ligament injuries are predisposed to OA. Third—the big risk factor, of course—is being overweight or obese. Patients who are overweight have a much higher frequency of OA, particularly in the knees. Last, there are genetic factors. For example, OA in the hands, which is more often seen in women, is genetically oriented so that if your mother had it or your aunt had it, you’re more likely to get it. Accordingly, there are multifactorial etiological factors that we have to try to address.
DR. GALL: Can you say a word about the different areas of the body that are associated with OA and talk a little bit about the difference between OA of the hands that we see frequently in women in relation to OA of the knee, hips, and spine, which can be devastating for individuals?
DR. MOSKOWITZ: Any joint can have it. It can be in the ankle, for example, if there has been trauma, but we’re looking particularly at the knees, the hips, the hands, and the spine. It’s interesting—certain joints in the hands are predisposed to OA including the distal and proximal interphalangeal joints. It tends not to involve the metacarpophalangeal joints, although it can in severe cases. Of course, the first carpometacarpal joint is frequently involved, and OA in this joint can be very disabling.
I have patients, for example, who are pianists or people who use their hands in their daily occupation. Their problem is often unfortunately minimized, with the doctor saying, “Oh, you’ve got a touch of OA, you can live okay with it.” However, this is hard to do because of pain and decreased function—OA can be a very disabling disease. In a number of women, the joints of the hands can be very inflamed. There’s a form of hand OA called erosive inflammatory OA, which can be especially troublesome because it’s very destructive. It’s almost like a low-grade rheumatoid arthritis where you actually get a lot of joint breakdown and deformity—it can be very disabling. OA of the knees or hips obviously can be a problem because of difficulties with ambulation.
DR. BLOCK: Our medical model trains us that, in general, we should look for an abnormality on a laboratory test to make a diagnosis. We do look for radiographic evidence of OA because, as I have said several times already, almost every elderly patient will have radiographic evidence of OA-like degeneration; however, we primarily use radiography adjunctively to make sure nothing else is going on or to assess the severity of the degeneration.
We make a diagnosis purely clinically. We look for someone who has the appropriate degree of pain in specific joints and does not have evidence of a systemic inflammatory source of that pain or a systemic rheumatic disease and in whom the pain appears to be articular rather than extra-articular. Of course, clinically, we feel for crepitus when we move the joint because of the degenerative processes, and we feel for bony enlargements or osteophytes around the joint to assist in making the diagnosis of OA.
Radiography is helpful in the sense that it can tell us about the severity of the degeneration and help us to make sure that we’re not missing something else such as a malignancy; however, radiography itself is not critical for the clinical diagnosis of OA.
DR. GALL: Dr. Block, how can the clinician who first sees a patient with OA differentiate it from a common type of inflammatory arthritis such as rheumatoid arthritis?
DR. MOSKOWITZ: If I may comment on that; you’re looking at 2 things. First, you are looking for symptoms and signs that you expect to be associated with OA, and then, you are looking for signs and symptoms that are not consistent with OA. As we said earlier, you may get some inflammation in the knee. You may get some swelling, and you can have evidence of increased synovial fluid, but you wouldn’t have the diffuse inflammatory reaction that you would have in rheumatoid arthritis, and there are different joints involved. For example, in rheumatoid arthritis, there may be involvement of the proximal interphalangeal and metacarpophalangeal joints of the hands, wrists, elbows, or shoulders. You would have involvement of peripheral joints that are different from those that are characteristically involved with OA.
In OA, the patient doesn’t have systemic findings such as fever, weight loss, or generalized prolonged stiffness. The patient often has localized stiffness in the involved joint but not generalized stiffness like that seen in rheumatoid arthritis. In OA, you can have multiple joints involved, such as 2 knees and 1 hip, but involvement doesn’t come on all at once and with the same severity.
DR. BLOCK: I completely agree with everything Dr. Moskowitz said. Additionally, if one is still confused after all of the clinical signs, this is where radiography may be useful because the radiographic appearance of OA is really different from that of the inflammatory arthritides.
In OA, one expects asymmetric narrowing in the joint itself. Additionally, OA results in subchondral sclerosis instead of periarticular osteopenia, which is seen in inflammatory arthritis. Finally, often, in OA, there is osteophyte formation, which is in contrast to the erosions that may be seen in the inflammatory arthritides. This is one area where radiography can really help.
DR. MOSKOWITZ: You have to be careful with respect to diagnosis in older people because if you do a serum rheumatoid factor study, the rheumatoid factor may be positive, but this may be unrelated to the patient’s symptoms. Studies have shown that a positive rheumatoid factor may be nonspecifically related to aging. So, you have to be careful—this finding can be a red herring in the diagnosis.
DR. BLOCK: I agree completely, and because OA is so common, one needs to bear in mind that a patient might have both inflammatory arthritis and underlying OA at the same time.
DR. MOSKOWITZ: Great point. We all see patients who come in to the office with OA of the hands, and then they develop rheumatoid arthritis or lupus engrafted on that previous involvement.
DR. GALL: I’d like to move on now to the management of the disease and, Dr. Moskowitz, would you talk to us about the first approach, which would be prevention? This is an approach that has caught the attention of the Centers for Disease Control and the Arthritis Foundation, and there is a major ongoing campaign at this time about what we can do to prevent this disease. Could you summarize that for us, Dr. Moskowitz?
DR. MOSKOWITZ: Absolutely. It’s generally thought that there’s nothing you can do to prevent OA. Well, we now think that you can slow it down and perhaps prevent it. Number one, we talked about weight. If a patient were to lose just 10 or 12 pounds, that could have a significant impact on the destructive effect of weight on the progression or the development of the disease. Programmed exercise is also very important. People say, “Well, I have arthritis. I can’t exercise,” and this is a problem because if you’re overweight and you have OA of the knees, joint overuse may temporarily lead to more symptoms. But walking at a reasonable pace and duration can be helpful with respect to symptoms and disability—the body puts out pain-relieving endorphins, muscles are strengthened, and joint range-of-motion is improved. Walking on a treadmill at a reasonable pace helps to achieve weight loss.
The patient can walk on a prescribed basis according to his or her capabilities. It’s important not to ask the patient to do something you know they’re not going to be able to do. We wouldn’t ask them to lose 20 pounds in the next 2 months. We have to make the program practical—these things are not easy to do, so we’ve got to be very supportive.
While trying to prevent disease progression, you want to have the patient strengthen their joint-related muscles. You want them to do muscle-building exercises because if you have joint stability, you’re likely going to decrease the osteoarthritic process as well. Losing weight is very important. Exercise is important. Using hot and cold applications for local therapy, which is safer than medication, is helpful.
Importantly, you don’t want to limit the patient’s activities any more than necessary. The way that I’ve always practiced medicine is that you don’t want to take away the patient’s ability to live his or her life as normally as possible. You don’t want to prescribe “don’ts,” like “don’t walk” and “don’t climb steps.” You want the patient to walk and do exercises, but the exercises have to be programmed so that the patient is able to do them.
DR. GALL: This is actually a very common question that primary care physicians are faced with when a patient with this disease comes in and is afraid to exercise and has a difficult time losing weight. But, I think it’s important that we understand that the cartilage gets its nourishment by being compressed and released and it has no blood supply of its own, so a lack of exercise is going to further cartilage degeneration.
DR. MOSKOWITZ: There are other non-weight bearing activities such as aquatic exercises that most people can do. You don’t have to be able to swim. You can do water exercises where you have the buoyancy of the water allowing for passive assistive exercise.
DR. BLOCK: Let me just add one other thing to the exercise discussion. In addition to there being a structural benefit to controlled exercise, every single systematic evaluation that has been performed has demonstrated that there is a really substantial and sustained pain relief component to exercise. People who have OA who are able to exercise regularly get substantial pain relief just from the exercise, and this is really important because the pain is what’s slowing them down in the first place.
The problem, of course, is maintaining an exercise regimen for a long term. Remember, this is a disease of decades, and we are not that good at behavioral modification strategies. Patients often give up on the exercise after a while. But, if you can maintain it, there are structural benefits as well as really substantial pain relief benefits.
DR. MOSKOWITZ: Yes, I agree. The other thing is that it’s hard for the busy physician in the office to start teaching the patient how to exercise, so we need to use arthritis health professionals such as occupational therapists and physical therapists to create exercise programs and teach patients how to exercise more effectively.
DR. BLOCK: Can I add one more thing as long as we’re discussing prevention? One of the major societal risk factors right now is recreational trauma. We have a population of young individuals who are being brought up in various organized sports, and there’s a very substantial risk factor for adolescent and young women who are, for example, playing soccer. They are getting knee and anterior cruciate ligament injuries, and over the upcoming 10 to 15 years, these people are at an enormous risk for developing knee OA. So, we need to develop some strategies to reduce the recreational trauma that results in early OA in this population.
DR. GALL: So, now we realize that weight control and exercise are important. Can we now talk about the medical management of both the pain and the arthritis?
DR. MOSKOWITZ: I know we don’t want to take too much time with this, but before we finish, I wanted to mention that I keep getting asked about tai chi or acupuncture for treatment. There are data suggesting that these programs can be helpful, but they should be adjunctive considerations for patients not responding to more routine therapeutic programs.4
DR. GALL: Yes, I would certainly agree with that. These are adjunctive therapies that may be helpful in select patients, but we need to focus on the more proven aspects of the prevention of the disease. Let’s now divide our talk about treatment into pain management and disease management. Can we start with pain management?
DR. BLOCK: I think that, first of all, it’s really important not to neglect the adjunctive measures that we’ve already discussed, and so, pain management is multifactorial, but one needs to include a physical and occupational therapist because although regular exercise dramatically reduces pain, local measures such as heat or ice are often very useful.
When they don’t get sufficient relief from adjunctive measures, symptomatic patients with OA will need to move to the pharmacologic arena sooner or later. From my practice and from the literature, I would say that there are 2 kinds of pain that need to be treated in OA: Pain that occurs during a painful flare that may last a couple of weeks and chronic ongoing pain that may develop as the disease progresses.
During short painful flares, most of the organizations that have published guidelines over the years have suggested that acetaminophen is a good place to start.5,6 I think that the systematic literature at this point suggests that acetaminophen might be very good for short-term pain relief—meaning, a few weeks at most. It’s probably not very effective for long-term pain relief, and it has a fair amount of potential complications, especially in the elderly. I don’t tend to use a lot of acetaminophen in people who are having chronic pain from OA, but during short-term pain flares, it can be very useful.
One can choose various analgesics including topical nonsteroidal agents, which can be very effective for local and monoarticular pain, for example, pain in superficial joints such as the knees or fingers. However, sooner or later, I believe that if patients don’t have a contraindication, they will end up taking either nonsteroidal anti-inflammatory agents (NSAIDs) or COX-2 inhibitors (coxibs), and that’s probably for a good reason. The nonsteroidals and coxibs have been demonstrated repeatedly to retain pain relief during 2-year studies, and they’re almost unique among the pharmacologic agents in that they retain that pain relief over a couple of years.7,8
Of course, if people are undergoing adequate pharmacologic therapy and they have painful flares in single joints, then intra-articular therapies such as glucocorticoids or hyaluronans are often very effective.
DR. GALL: I’d like to just go back to the NSAIDs for a moment. There is increasing emphasis on the risk of NSAIDs, particularly in the elderly. How do we balance the benefits of NSAID use for chronic therapy versus the risks, and how would you monitor the patient?
DR. MOSKOWITZ: Well, I agree with the premises that Dr. Block mentioned. I think acetaminophen has varying efficacy, but I think it’s worth trying. If I am right, I think it was stated in the past that the acetaminophen dose could go up to 4 g/d.9 The current recommendation is that doses of acetaminophen should not exceed 3 g/d so as not to incur hepatic or renal toxicity. The problem is that acetaminophen is often present in other medicines that patients are taking, and so an excess total intake is not uncommon.10
I think a trial of acetaminophen is worthwhile because even if it doesn’t relieve all the pain, it’s a floor of analgesia that you can add to with nonsteroidals.
With regard to the question about whether nonsteroidals are dangerous in older people, I think that they can be. Most things we use are unfortunately associated with some risk, although pain itself is also a risk factor. Pain increases blood pressure and pulse rate. If you relieve the pain, you’re probably relieving more of the stress on the patient overall as opposed to the risk of the NSAID itself. If you use nonselective NSAIDs, I would suggest adding a proton pump inhibitor, especially if the patient has an increased risk of peptic ulcer disease. If you use a COX-2 selective inhibitor, data suggest there is a decreased gastrointestinal (GI) risk. Overall, I think that NSAIDs can be effective and reasonably safe. I’m more careful about using NSAIDs in individuals older than 75 or 80 years; however, many of these patients are physiologically younger than their numerical age, and they can more safely tolerate these agents.
Intra-articular steroids and intra-articular hyaluronans can be very helpful in the management of knee OA, and they have comfortable safety factors.
DR. GALL: I’ll just make a comment about my own approach to that. When I put a patient on an NSAID, and I do use them frequently, I mark it on a problem list. I’m careful to ask the patients about bleeding and to warn them about the signs of GI bleeding, and I do periodic blood counts to look at hemoglobin and creatinine levels. I also do a urinalysis to rule out silent GI or renal effects of these drugs, but I do use NSAIDs frequently.
Additionally, with the recent attention to the pain component of OA over the past decade, there has been a lot of systematic investigation into various neuroactive pain-directed medications. In fact, duloxetine, which is a neuroactive agent, was approved for use a couple of years ago. There are non–anti-inflammatory, non-purely analgesic medicines that can be very helpful for the treatment of the pain component of OA.
DR. GALL: Thank you. Could we just say a word about opioids and tramadol as adjunctive therapies and what the pluses and minuses of using these drugs are?
DR. MOSKOWITZ: I think there is not only a rationale but a place for these agents in treating OA, but we’ve got to be careful. I’m a little less concerned about using tramadol than using the classic opioids, per se, and I think tramadol can be a reasonable bridge when people are not responding to NSAIDS. A number of years ago, when I was writing a piece for the Arthritis Foundation pamphlet, I stated that there is never a place for opioids in the treatment of OA, and I can’t tell you how many of my colleagues wrote and said, “Dr. Moskowitz, there are times when we feel they need to be available.” I agree, but I think my guidance is that, if you use an opioid, use the lowest dose possible and for the shortest time possible.
I think if someone is in a lot of pain and they’re having a flare, particularly an older individual, it may be safer to use the opioids than to put them on NSAIDs, particularly if they have had past trouble with NSAIDS. I don’t think that we should never use opioids when treating OA. What are your thoughts on that, Dr. Block?
DR. BLOCK: I agree. I think that there is good evidence that the opiates provide really substantial pain relief, so they’re effective in OA. The problem, of course, is that they have very high side effect profiles, and the people that they’re the most dangerous for with regard to falling and injuring oneself are actually the same people who are at high risk with nonsteroidals—the very elderly with congestive heart failure. It really does put us in a bind.
I think that I agree with what you just said, which is that they are effective, and at times, I think that there is clearly a place for opiates in OA, but one needs to be judicious and careful when using them. Tramadol is, I think, much safer. It’s a weak opiate. On the other hand, it also has much less pain-relieving potential, so this is the bind we face.
Well, I think what we haven’t mentioned yet are the surgical approaches, which are critically important.
DR. MOSKOWITZ: Should we discuss glucosamine and chondroitin sulfate?
DR. BLOCK: So, more than half of patients who have symptomatic OA use various complementary medicine approaches. They are widely used because they’re widely believed to be beneficial, and I think that there has been a fair amount of systematic evaluation, specifically of glucosamine and chondroitin sulfate, over the past several years.
When we look at the independently sponsored studies, the evidence suggests that the side effect profile for both agents is very good. As long as they were manufactured in a safe manner, they are very safe and one shouldn’t worry about using them.
On the other hand, the efficacy above placebo is very low, but that doesn’t mean that they’re not effective in individual patients. The thing that’s really important to note here is that whenever there is a placebo-controlled study with pain as an outcome in OA, the placebo response is huge. More than half of the people who get placebo get very substantial clinically significant pain relief, and more importantly, it’s durable. In all of these placebo-controlled studies that go on for 2 years, the placebo group with pain relief had sustainability over 2 years,11,12 and so, if one gets more than a placebo response from glucosamine, and if it’s providing pain relief, I’m happy for them to use it.
The Glucosamine/Chondroitin Arthritis Intervention Trial (GAIT),7 the large National Institutes of Health multicenter study, which was a null study, showed that fundamentally, there was no systematic advantage of a combination glucosamine plus chondroitin sulfate. Even when that came out as a null study and was publicized as such, the market for glucosamine and chondroitin sulfate in the following year didn’t decrease in the United States.13,14
People who are getting a benefit, whether it’s due to an endorphin effect from placebo or the agent, are going to continue using the agent. As long as the agents are safe, I’m comfortable with my patients using them if they really are feeling like they are getting relief.
DR. MOSKOWITZ: Yes, I think that there are conflicting data, but there are a number of studies that seem to show that these agents do help some people. For example, this question about the GAIT study7 not showing an effect, if you’re using glucosamine and chondroitin sulfate in individuals who have high pain to start with—if they had a score of more than 300 mm on the Western Ontario and McMaster Universities Arthritis Index (WOMAC)—they appeared to do better, but there wasn’t a predefined endpoint, so the result has to be substantiated in a repeat study. However, I see enough people—like you do, Dr. Block—who seem to respond. They’re safe agents, and I think that they do have merit in patients who are not getting a benefit from other agents.
DR. GALL: Before we move on to surgery and the future of therapy, I just want to discuss intra-articular steroids. What are your thoughts on these?
DR. MOSKOWITZ: I would be lost without them. I remember when they were first used by Dr. Joseph Hollander in Philadelphia, and they failed to help because he used cortisone acetate instead of the hydrocortisone derivative, and cortisone has to be converted to hydrocortisone before it works.15 I use intra-articular corticosteroids a lot. The general recommendation is that you shouldn’t use it more than 4 times a year in the knee. I think maybe if somebody is older—for example, 88 years—and they cannot tolerate other therapy, you might use them that often. Studies have shown that an intra-articular corticosteroid injection every 3 months did not lead to deleterious anatomic effects.16 I think using them 2 or 3 times a year, if indicated, can be symptomatically effective and safe, and I don’t think that will lead to joint breakdown if the patient is careful about doing their exercises and other management.
DR. GALL: Let’s move on to the use of surgery, and without going into all the surgical procedures, when is it appropriate for the clinician and the patient to consider surgical intervention in OA?
DR. BLOCK: It is entirely patient-oriented. You can’t tell by structure or by X-ray. When the patient’s OA is painful and severe enough, when it cannot be controlled with medication and medical regimens, or when it’s interfering with their lifestyle and they can’t live with it, they’ll tell you they need surgery.
DR. MOSKOWITZ: I completely agree. I think you can’t tell them. They’ll let you know. They’ll say, “I just have so much pain and disability. I need something done.”
DR. GALL: I think it’s important also to bring the surgeon in, not as a plumber to fix something, but as a partner in making the decision. Certainly the primary care physician, often the rheumatologist, can also provide guidance and work with the patients and the surgeons in making that decision.
DR. MOSKOWITZ: Yes. Let me just add one thing: I don’t think the patient should be in an extreme state of pain and disability in order to have surgery. If the condition interferes with their ADLs and the pain is really encumbering their daily activities, then it’s not unreasonable to consider surgery. I don’t think they have to be so bad that they’re limping and in terrible distress.
DR. BLOCK: Having said that, we should mention that joint replacement, at least in the knees and the hips, is extraordinarily effective in relieving pain.
DR. GALL: Yes. I would agree and these advances have continued to increase over the years, and these replacements last for a longer period of time, so the amount of satisfaction has increased, and the number of complications in experienced centers has certainly decreased.
Let’s end by talking about the future of OA. There are 2 areas that I’d like to touch on: One is the disease-modifying OA drugs (DMOADs) and the other is cartilage regeneration with stem cells and other techniques. Dr. Moskowitz, maybe you can talk about the DMOADs, and Dr. Block, you can talk about cartilage regeneration.
DR. MOSKOWITZ: Disease retardation or reversal would be the holy grail of OA management: to be able to treat OA with medications—the so-called DMOADS—that not only relieve symptoms but also slow the disease down or actually reverse the disease process. Some of the agents described as possibly having disease-modification potential include glucosamine, chondroitin sulfate, intra-articular hyaluronans, diacerein, avocado/soybean unsaponifiables, and doxycycline among others.17 Further studies will help to define the purported role of such agents in disease modification.
DR. BLOCK: I think it’s fair to say that, as of today, there really are no strategies that have been proven to be effective at retarding the progression and pain of OA over long periods of time. I think that is the gold standard that we’re looking for, and there is suggestive evidence from studies performed on each of the agents that Dr. Moskowitz just mentioned and several others.
I think there is a lot of work being done to study various metalloproteinase inhibitors, especially the collagenase 3 (MMP-13) inhibitors, and the aggrecanase inhibitors. I think that if some of these things are demonstrated to be effective in delaying the progression of OA, they could be extremely exciting, but one needs to keep in mind that whatever we use as a DMOAD has to be extremely safe because it’s going to be used for decades in people who are basically asymptomatic. In order to justify that on a public health basis, it has to be extraordinarily safe. It’s a very high bar.
DR. GALL: Okay. Dr. Block, can you now attempt to discuss cartilage regeneration and the use of the stem cell?
DR. BLOCK: Yes. Mesenchymal stem cell technology is extremely exciting, and it has paid off in some areas of bone and other connective tissues. As everybody knows, there is one U.S. Food and Drug Administration-approved approach for replacing cartilage that has had some success, but it’s used for isolated chondral defects in young and otherwise healthy people—not for OA.
The problem, of course, is that as we understand OA better and better, it’s more than just cartilage degeneration, and more importantly, the cartilage, which is very important, is a very thin and delicate tissue. If we replace the cartilage alone without normalizing the aberrant biomechanics across the joint—without normalizing the subchondral bone that the cartilage sits on—we have very little chance of getting long-term relief.
Having said that, efforts are being carried out in a variety of laboratories around the world to focus on not only mesenchymal cell-directed cartilage formation but also on actual repair of most tissues of the joint. Once we understand how to grow new cartilage and actually have it integrate with the adjacent tissue, which we still are not able to do, and once we’re able to repair some of the subchondral bone, I think that there is a lot of promise for that strategy in the long run. But, in the short-term, again, it’s a very difficult process that is not yet well understood.
DR. MOSKOWITZ: Well said. I agree. I think that we’re on the brink of being able to repair, particularly, very focal defects—maybe 4- or 5-mm defects. But, to repair the joint, as Dr. Block pointed out, you can’t just repair the cartilage because joint changes such as inflammation, meniscal alterations, varus and valgus deformities, and other changes impacting the joint are going to impair cell-based regeneration. Until we get those impacts controlled, I think it’s going to be some time before we can say to someone, “We’re going to replace your cartilage and normalize your joint.” However, I am optimistic and think that down the line, we may be able to do it.
DR. GALL: Dr. Block also mentioned that the underlying bone has changed too, and so, that needs to be attacked as well in these approaches.
DR. BLOCK: My personal view is that in the short-term future—the next 5 to 10 years—big strides will be made in strategies for better pain relief, and there are a variety of biologics that are in phase II and phase III testing that already look very exciting. Again, my own personal research interest is in trying to normalize the mechanical loads across the joints so that we can protect the joints better and so that the OA won’t progress, and I think those kinds of strategies will be available over the next couple of years.
DR. GALL: I’d like to thank you both for a really enlightening and comprehensive discussion of this common disease. Just to review, we’ve discussed what OA is and defined it, talked about its burden and its diagnosis, and mentioned the various risk factors for OA, particularly in light of what we can do to prevent OA or prevent the progression of the disease. We also had a comprehensive talk about the treatment of OA from both the standpoint of pain and the actual disease, and we touched upon the future of this as well as the research that’s being done in these areas.
I really appreciate your wisdom and discussion, and I think we’re providing the primary care physicians and providers with a really comprehensive look at this disease.
FoxP2 Media LLC is the publisher of The Medical Roundtable.
DR. BLOCK: Throughout the 20th century, we always considered OA to be primarily a degenerative disease of cartilage, but during the past decade or more, 2 things have become apparent: First, from a clinical perspective, OA is not just a structural degenerative process, it is primarily a very painful disease with pain being its most prominent feature. Second, the degenerative process involves not only the cartilage but the entire joint and all the involved structures within it.
A reasonable working definition of OA right now would be a painful degenerative process involving all the joint structures and is not primarily inflammatory but involves progressive deterioration of joint structures including articular cartilage. It’s important to keep in mind that pain itself is critical because as individuals age, everybody has some degenerative processes in their joints, and if we look hard enough, we can find the pathological features of OA in all elderly people. However, not all elderly people actually have the clinical disease.
DR. GALL: In a moment, we’ll talk about the causes of this, but, Dr. Moskowitz, would you tell us what the burden of this disease is here in the United States and worldwide?
DR. MOSKOWITZ: Well, it’s the most common form of arthritis and affects nearly 27 million or more Americans.1 It’s the most common disabling disease of the rheumatologic group, and it’s very disabling among all diseases.
DR. GALL: In the United States, you say that there’s about—
DR. MOSKOWITZ: About 27 million people with OA. And, in a decade, it is anticipated that the number will significantly increase.
DR. GALL: How many of those people are affected to the point that they’re unable to do their normal activities of daily living (ADLs)?
DR. MOSKOWITZ: Well, in terms of disability, about 80% of patients with OA have some degree of movement limitation, 25% cannot perform major ADLs, 11% of adults with knee OA need help with personal care, and 14% require help with routine needs.2
DR. GALL: Dr. Block, you talked about pain as a major manifestation of this disease. Does everybody who has OA present with pain?
DR. BLOCK: Again, it’s a definitional question; there is a distinction between clinically evident OA and asymptomatic structural degeneration of the joint. The population that Dr. Moskowitz was just describing, which comprises 20 to 30 million people, has doctor-diagnosed symptomatic OA, which is a painful disease. A much higher number of people who have structural degenerative disease would be diagnosed by radiography and said to fulfill a radiographic definition of OA, but all of the people who Dr. Moskowitz refer to have pain.
DR. MOSKOWITZ: Pain is what brings a patient to the doctor. I’ll often ask my patients, “How are you doing and what can I do for you?” They have limited motion and can’t get around. They say, “Dr. Moskowitz, I just want to get rid of the pain. I’ll worry about the swelling and getting around later.” And, that’s what really motivates them to get medical attention.
DR. GALL: Often, we will get consultations from primary care physicians for a patient who has OA that’s been picked up by radiography, and the patients will come and really not complain of pain in that area or possibly might have another underlying disease that’s causing the pain that’s not related to the OA. Dr. Block, do you want to say a word about that?
DR. BLOCK: Sure. By the age of 80 years, essentially, 100% of the population will have evidence of radiographic OA in at least one of their large joints. If you start with that number and you’re then referred an elderly patient merely for the radiographic appearance of OA, that doesn’t tell you very much about what’s going on clinically, and so, most of us would not even describe that as a clinical disease.
They may have a degenerative process in their knee, hip, etc, but of that group, say at the age of 80 years where almost 100% of people will have radiographic evidence of OA, only about 15% to 20% have symptomatic disease.3 That’s the culprit that I think we’re talking about clinically—people who not only have degenerative structural disease but also have symptoms that are interfering with their quality of life or their daily activities.
DR. MOSKOWITZ: I think that’s an important point, Dr. Block. For example, we all see patients who come in complaining of hip pain and you examine their knees as well and they have terrible genu varum (bowleg) and you say, “Do your knees bother you?” and they say, “Doctor, my knees are fine.” You think, my goodness, how can you walk? It isn’t necessarily true that if you have a deformity or involvement of a joint that you’re going to have pain, but if you have painful joints, that brings you to the doctor.
DR. GALL: Dr. Block, once again, I would like you to summarize the etiology of OA. Obviously, there has been much research and much information that has come to light in recent years regarding the causes of this disease.
DR. BLOCK: As I said, throughout the 20th century, people interested in the pathophysiology of OA were quite convinced that it was a degenerative process of the articular cartilage. As people age, the articular cartilage starts degenerating, and when there is absence of intact cartilage that ought to be providing essentially frictionless articulation of the bones during motion, there is deterioration, and that’s what we always considered to be OA.
Of course, the missing link in that paradigm is that cartilage itself has no nerve or blood supply and is therefore painless, and yet, people have horrible pain with this disease. So, our understanding over the past 10 or 15 years has really dramatically progressed, and we have begun to appreciate both degenerative and reactive processes in the subchondral bone, the ligaments, and the muscle.
The way we look at the pathophysiology of the onset and progression of the disease today is that there is primarily a degenerative process that stimulates reactive processes throughout the joint and even in the immune and inflammatory pathways. So, the reaction to that degenerative process sometimes causes stimulation of nociceptors and a lot of pain because of either local inflammation or mechanical alterations.
The degeneration itself, as I said, is painless. Often, people can have very degenerative joints, as Dr. Moskowitz just mentioned, and have no pain and therefore no clinical symptoms, so the disease is a combination of the degeneration and the secondary stimulated pain.
DR. GALL: So, is this primarily a biochemical or a cellular disease? Is it a genetic disease?
DR. BLOCK: Yes to all.
DR. MOSKOWITZ: There are a number of factors, Dr. Gall, involved in this condition. First, for example, is aging. We see that the frequency of OA increases logarithmically as we age, and there are pathologic changes with age that appear to lead to OA. For example, pigmented collections in the joint, comprising advanced glycation end products, occur that may predispose a patient to OA. You’re going to find more OA in an aging population.
The second factor is trauma. Professional football and basketball players, for example, who have chronic trauma, meniscal injuries, or cruciate ligament injuries are predisposed to OA. Third—the big risk factor, of course—is being overweight or obese. Patients who are overweight have a much higher frequency of OA, particularly in the knees. Last, there are genetic factors. For example, OA in the hands, which is more often seen in women, is genetically oriented so that if your mother had it or your aunt had it, you’re more likely to get it. Accordingly, there are multifactorial etiological factors that we have to try to address.
DR. GALL: Can you say a word about the different areas of the body that are associated with OA and talk a little bit about the difference between OA of the hands that we see frequently in women in relation to OA of the knee, hips, and spine, which can be devastating for individuals?
DR. MOSKOWITZ: Any joint can have it. It can be in the ankle, for example, if there has been trauma, but we’re looking particularly at the knees, the hips, the hands, and the spine. It’s interesting—certain joints in the hands are predisposed to OA including the distal and proximal interphalangeal joints. It tends not to involve the metacarpophalangeal joints, although it can in severe cases. Of course, the first carpometacarpal joint is frequently involved, and OA in this joint can be very disabling.
I have patients, for example, who are pianists or people who use their hands in their daily occupation. Their problem is often unfortunately minimized, with the doctor saying, “Oh, you’ve got a touch of OA, you can live okay with it.” However, this is hard to do because of pain and decreased function—OA can be a very disabling disease. In a number of women, the joints of the hands can be very inflamed. There’s a form of hand OA called erosive inflammatory OA, which can be especially troublesome because it’s very destructive. It’s almost like a low-grade rheumatoid arthritis where you actually get a lot of joint breakdown and deformity—it can be very disabling. OA of the knees or hips obviously can be a problem because of difficulties with ambulation.
DR. BLOCK: Our medical model trains us that, in general, we should look for an abnormality on a laboratory test to make a diagnosis. We do look for radiographic evidence of OA because, as I have said several times already, almost every elderly patient will have radiographic evidence of OA-like degeneration; however, we primarily use radiography adjunctively to make sure nothing else is going on or to assess the severity of the degeneration.
We make a diagnosis purely clinically. We look for someone who has the appropriate degree of pain in specific joints and does not have evidence of a systemic inflammatory source of that pain or a systemic rheumatic disease and in whom the pain appears to be articular rather than extra-articular. Of course, clinically, we feel for crepitus when we move the joint because of the degenerative processes, and we feel for bony enlargements or osteophytes around the joint to assist in making the diagnosis of OA.
Radiography is helpful in the sense that it can tell us about the severity of the degeneration and help us to make sure that we’re not missing something else such as a malignancy; however, radiography itself is not critical for the clinical diagnosis of OA.
DR. GALL: Dr. Block, how can the clinician who first sees a patient with OA differentiate it from a common type of inflammatory arthritis such as rheumatoid arthritis?
DR. MOSKOWITZ: If I may comment on that; you’re looking at 2 things. First, you are looking for symptoms and signs that you expect to be associated with OA, and then, you are looking for signs and symptoms that are not consistent with OA. As we said earlier, you may get some inflammation in the knee. You may get some swelling, and you can have evidence of increased synovial fluid, but you wouldn’t have the diffuse inflammatory reaction that you would have in rheumatoid arthritis, and there are different joints involved. For example, in rheumatoid arthritis, there may be involvement of the proximal interphalangeal and metacarpophalangeal joints of the hands, wrists, elbows, or shoulders. You would have involvement of peripheral joints that are different from those that are characteristically involved with OA.
In OA, the patient doesn’t have systemic findings such as fever, weight loss, or generalized prolonged stiffness. The patient often has localized stiffness in the involved joint but not generalized stiffness like that seen in rheumatoid arthritis. In OA, you can have multiple joints involved, such as 2 knees and 1 hip, but involvement doesn’t come on all at once and with the same severity.
DR. BLOCK: I completely agree with everything Dr. Moskowitz said. Additionally, if one is still confused after all of the clinical signs, this is where radiography may be useful because the radiographic appearance of OA is really different from that of the inflammatory arthritides.
In OA, one expects asymmetric narrowing in the joint itself. Additionally, OA results in subchondral sclerosis instead of periarticular osteopenia, which is seen in inflammatory arthritis. Finally, often, in OA, there is osteophyte formation, which is in contrast to the erosions that may be seen in the inflammatory arthritides. This is one area where radiography can really help.
DR. MOSKOWITZ: You have to be careful with respect to diagnosis in older people because if you do a serum rheumatoid factor study, the rheumatoid factor may be positive, but this may be unrelated to the patient’s symptoms. Studies have shown that a positive rheumatoid factor may be nonspecifically related to aging. So, you have to be careful—this finding can be a red herring in the diagnosis.
DR. BLOCK: I agree completely, and because OA is so common, one needs to bear in mind that a patient might have both inflammatory arthritis and underlying OA at the same time.
DR. MOSKOWITZ: Great point. We all see patients who come in to the office with OA of the hands, and then they develop rheumatoid arthritis or lupus engrafted on that previous involvement.
DR. GALL: I’d like to move on now to the management of the disease and, Dr. Moskowitz, would you talk to us about the first approach, which would be prevention? This is an approach that has caught the attention of the Centers for Disease Control and the Arthritis Foundation, and there is a major ongoing campaign at this time about what we can do to prevent this disease. Could you summarize that for us, Dr. Moskowitz?
DR. MOSKOWITZ: Absolutely. It’s generally thought that there’s nothing you can do to prevent OA. Well, we now think that you can slow it down and perhaps prevent it. Number one, we talked about weight. If a patient were to lose just 10 or 12 pounds, that could have a significant impact on the destructive effect of weight on the progression or the development of the disease. Programmed exercise is also very important. People say, “Well, I have arthritis. I can’t exercise,” and this is a problem because if you’re overweight and you have OA of the knees, joint overuse may temporarily lead to more symptoms. But walking at a reasonable pace and duration can be helpful with respect to symptoms and disability—the body puts out pain-relieving endorphins, muscles are strengthened, and joint range-of-motion is improved. Walking on a treadmill at a reasonable pace helps to achieve weight loss.
The patient can walk on a prescribed basis according to his or her capabilities. It’s important not to ask the patient to do something you know they’re not going to be able to do. We wouldn’t ask them to lose 20 pounds in the next 2 months. We have to make the program practical—these things are not easy to do, so we’ve got to be very supportive.
While trying to prevent disease progression, you want to have the patient strengthen their joint-related muscles. You want them to do muscle-building exercises because if you have joint stability, you’re likely going to decrease the osteoarthritic process as well. Losing weight is very important. Exercise is important. Using hot and cold applications for local therapy, which is safer than medication, is helpful.
Importantly, you don’t want to limit the patient’s activities any more than necessary. The way that I’ve always practiced medicine is that you don’t want to take away the patient’s ability to live his or her life as normally as possible. You don’t want to prescribe “don’ts,” like “don’t walk” and “don’t climb steps.” You want the patient to walk and do exercises, but the exercises have to be programmed so that the patient is able to do them.
DR. GALL: This is actually a very common question that primary care physicians are faced with when a patient with this disease comes in and is afraid to exercise and has a difficult time losing weight. But, I think it’s important that we understand that the cartilage gets its nourishment by being compressed and released and it has no blood supply of its own, so a lack of exercise is going to further cartilage degeneration.
DR. MOSKOWITZ: There are other non-weight bearing activities such as aquatic exercises that most people can do. You don’t have to be able to swim. You can do water exercises where you have the buoyancy of the water allowing for passive assistive exercise.
DR. BLOCK: Let me just add one other thing to the exercise discussion. In addition to there being a structural benefit to controlled exercise, every single systematic evaluation that has been performed has demonstrated that there is a really substantial and sustained pain relief component to exercise. People who have OA who are able to exercise regularly get substantial pain relief just from the exercise, and this is really important because the pain is what’s slowing them down in the first place.
The problem, of course, is maintaining an exercise regimen for a long term. Remember, this is a disease of decades, and we are not that good at behavioral modification strategies. Patients often give up on the exercise after a while. But, if you can maintain it, there are structural benefits as well as really substantial pain relief benefits.
DR. MOSKOWITZ: Yes, I agree. The other thing is that it’s hard for the busy physician in the office to start teaching the patient how to exercise, so we need to use arthritis health professionals such as occupational therapists and physical therapists to create exercise programs and teach patients how to exercise more effectively.
DR. BLOCK: Can I add one more thing as long as we’re discussing prevention? One of the major societal risk factors right now is recreational trauma. We have a population of young individuals who are being brought up in various organized sports, and there’s a very substantial risk factor for adolescent and young women who are, for example, playing soccer. They are getting knee and anterior cruciate ligament injuries, and over the upcoming 10 to 15 years, these people are at an enormous risk for developing knee OA. So, we need to develop some strategies to reduce the recreational trauma that results in early OA in this population.
DR. GALL: So, now we realize that weight control and exercise are important. Can we now talk about the medical management of both the pain and the arthritis?
DR. MOSKOWITZ: I know we don’t want to take too much time with this, but before we finish, I wanted to mention that I keep getting asked about tai chi or acupuncture for treatment. There are data suggesting that these programs can be helpful, but they should be adjunctive considerations for patients not responding to more routine therapeutic programs.4
DR. GALL: Yes, I would certainly agree with that. These are adjunctive therapies that may be helpful in select patients, but we need to focus on the more proven aspects of the prevention of the disease. Let’s now divide our talk about treatment into pain management and disease management. Can we start with pain management?
DR. BLOCK: I think that, first of all, it’s really important not to neglect the adjunctive measures that we’ve already discussed, and so, pain management is multifactorial, but one needs to include a physical and occupational therapist because although regular exercise dramatically reduces pain, local measures such as heat or ice are often very useful.
When they don’t get sufficient relief from adjunctive measures, symptomatic patients with OA will need to move to the pharmacologic arena sooner or later. From my practice and from the literature, I would say that there are 2 kinds of pain that need to be treated in OA: Pain that occurs during a painful flare that may last a couple of weeks and chronic ongoing pain that may develop as the disease progresses.
During short painful flares, most of the organizations that have published guidelines over the years have suggested that acetaminophen is a good place to start.5,6 I think that the systematic literature at this point suggests that acetaminophen might be very good for short-term pain relief—meaning, a few weeks at most. It’s probably not very effective for long-term pain relief, and it has a fair amount of potential complications, especially in the elderly. I don’t tend to use a lot of acetaminophen in people who are having chronic pain from OA, but during short-term pain flares, it can be very useful.
One can choose various analgesics including topical nonsteroidal agents, which can be very effective for local and monoarticular pain, for example, pain in superficial joints such as the knees or fingers. However, sooner or later, I believe that if patients don’t have a contraindication, they will end up taking either nonsteroidal anti-inflammatory agents (NSAIDs) or COX-2 inhibitors (coxibs), and that’s probably for a good reason. The nonsteroidals and coxibs have been demonstrated repeatedly to retain pain relief during 2-year studies, and they’re almost unique among the pharmacologic agents in that they retain that pain relief over a couple of years.7,8
Of course, if people are undergoing adequate pharmacologic therapy and they have painful flares in single joints, then intra-articular therapies such as glucocorticoids or hyaluronans are often very effective.
DR. GALL: I’d like to just go back to the NSAIDs for a moment. There is increasing emphasis on the risk of NSAIDs, particularly in the elderly. How do we balance the benefits of NSAID use for chronic therapy versus the risks, and how would you monitor the patient?
DR. MOSKOWITZ: Well, I agree with the premises that Dr. Block mentioned. I think acetaminophen has varying efficacy, but I think it’s worth trying. If I am right, I think it was stated in the past that the acetaminophen dose could go up to 4 g/d.9 The current recommendation is that doses of acetaminophen should not exceed 3 g/d so as not to incur hepatic or renal toxicity. The problem is that acetaminophen is often present in other medicines that patients are taking, and so an excess total intake is not uncommon.10
I think a trial of acetaminophen is worthwhile because even if it doesn’t relieve all the pain, it’s a floor of analgesia that you can add to with nonsteroidals.
With regard to the question about whether nonsteroidals are dangerous in older people, I think that they can be. Most things we use are unfortunately associated with some risk, although pain itself is also a risk factor. Pain increases blood pressure and pulse rate. If you relieve the pain, you’re probably relieving more of the stress on the patient overall as opposed to the risk of the NSAID itself. If you use nonselective NSAIDs, I would suggest adding a proton pump inhibitor, especially if the patient has an increased risk of peptic ulcer disease. If you use a COX-2 selective inhibitor, data suggest there is a decreased gastrointestinal (GI) risk. Overall, I think that NSAIDs can be effective and reasonably safe. I’m more careful about using NSAIDs in individuals older than 75 or 80 years; however, many of these patients are physiologically younger than their numerical age, and they can more safely tolerate these agents.
Intra-articular steroids and intra-articular hyaluronans can be very helpful in the management of knee OA, and they have comfortable safety factors.
DR. GALL: I’ll just make a comment about my own approach to that. When I put a patient on an NSAID, and I do use them frequently, I mark it on a problem list. I’m careful to ask the patients about bleeding and to warn them about the signs of GI bleeding, and I do periodic blood counts to look at hemoglobin and creatinine levels. I also do a urinalysis to rule out silent GI or renal effects of these drugs, but I do use NSAIDs frequently.
Additionally, with the recent attention to the pain component of OA over the past decade, there has been a lot of systematic investigation into various neuroactive pain-directed medications. In fact, duloxetine, which is a neuroactive agent, was approved for use a couple of years ago. There are non–anti-inflammatory, non-purely analgesic medicines that can be very helpful for the treatment of the pain component of OA.
DR. GALL: Thank you. Could we just say a word about opioids and tramadol as adjunctive therapies and what the pluses and minuses of using these drugs are?
DR. MOSKOWITZ: I think there is not only a rationale but a place for these agents in treating OA, but we’ve got to be careful. I’m a little less concerned about using tramadol than using the classic opioids, per se, and I think tramadol can be a reasonable bridge when people are not responding to NSAIDS. A number of years ago, when I was writing a piece for the Arthritis Foundation pamphlet, I stated that there is never a place for opioids in the treatment of OA, and I can’t tell you how many of my colleagues wrote and said, “Dr. Moskowitz, there are times when we feel they need to be available.” I agree, but I think my guidance is that, if you use an opioid, use the lowest dose possible and for the shortest time possible.
I think if someone is in a lot of pain and they’re having a flare, particularly an older individual, it may be safer to use the opioids than to put them on NSAIDs, particularly if they have had past trouble with NSAIDS. I don’t think that we should never use opioids when treating OA. What are your thoughts on that, Dr. Block?
DR. BLOCK: I agree. I think that there is good evidence that the opiates provide really substantial pain relief, so they’re effective in OA. The problem, of course, is that they have very high side effect profiles, and the people that they’re the most dangerous for with regard to falling and injuring oneself are actually the same people who are at high risk with nonsteroidals—the very elderly with congestive heart failure. It really does put us in a bind.
I think that I agree with what you just said, which is that they are effective, and at times, I think that there is clearly a place for opiates in OA, but one needs to be judicious and careful when using them. Tramadol is, I think, much safer. It’s a weak opiate. On the other hand, it also has much less pain-relieving potential, so this is the bind we face.
Well, I think what we haven’t mentioned yet are the surgical approaches, which are critically important.
DR. MOSKOWITZ: Should we discuss glucosamine and chondroitin sulfate?
DR. BLOCK: So, more than half of patients who have symptomatic OA use various complementary medicine approaches. They are widely used because they’re widely believed to be beneficial, and I think that there has been a fair amount of systematic evaluation, specifically of glucosamine and chondroitin sulfate, over the past several years.
When we look at the independently sponsored studies, the evidence suggests that the side effect profile for both agents is very good. As long as they were manufactured in a safe manner, they are very safe and one shouldn’t worry about using them.
On the other hand, the efficacy above placebo is very low, but that doesn’t mean that they’re not effective in individual patients. The thing that’s really important to note here is that whenever there is a placebo-controlled study with pain as an outcome in OA, the placebo response is huge. More than half of the people who get placebo get very substantial clinically significant pain relief, and more importantly, it’s durable. In all of these placebo-controlled studies that go on for 2 years, the placebo group with pain relief had sustainability over 2 years,11,12 and so, if one gets more than a placebo response from glucosamine, and if it’s providing pain relief, I’m happy for them to use it.
The Glucosamine/Chondroitin Arthritis Intervention Trial (GAIT),7 the large National Institutes of Health multicenter study, which was a null study, showed that fundamentally, there was no systematic advantage of a combination glucosamine plus chondroitin sulfate. Even when that came out as a null study and was publicized as such, the market for glucosamine and chondroitin sulfate in the following year didn’t decrease in the United States.13,14
People who are getting a benefit, whether it’s due to an endorphin effect from placebo or the agent, are going to continue using the agent. As long as the agents are safe, I’m comfortable with my patients using them if they really are feeling like they are getting relief.
DR. MOSKOWITZ: Yes, I think that there are conflicting data, but there are a number of studies that seem to show that these agents do help some people. For example, this question about the GAIT study7 not showing an effect, if you’re using glucosamine and chondroitin sulfate in individuals who have high pain to start with—if they had a score of more than 300 mm on the Western Ontario and McMaster Universities Arthritis Index (WOMAC)—they appeared to do better, but there wasn’t a predefined endpoint, so the result has to be substantiated in a repeat study. However, I see enough people—like you do, Dr. Block—who seem to respond. They’re safe agents, and I think that they do have merit in patients who are not getting a benefit from other agents.
DR. GALL: Before we move on to surgery and the future of therapy, I just want to discuss intra-articular steroids. What are your thoughts on these?
DR. MOSKOWITZ: I would be lost without them. I remember when they were first used by Dr. Joseph Hollander in Philadelphia, and they failed to help because he used cortisone acetate instead of the hydrocortisone derivative, and cortisone has to be converted to hydrocortisone before it works.15 I use intra-articular corticosteroids a lot. The general recommendation is that you shouldn’t use it more than 4 times a year in the knee. I think maybe if somebody is older—for example, 88 years—and they cannot tolerate other therapy, you might use them that often. Studies have shown that an intra-articular corticosteroid injection every 3 months did not lead to deleterious anatomic effects.16 I think using them 2 or 3 times a year, if indicated, can be symptomatically effective and safe, and I don’t think that will lead to joint breakdown if the patient is careful about doing their exercises and other management.
DR. GALL: Let’s move on to the use of surgery, and without going into all the surgical procedures, when is it appropriate for the clinician and the patient to consider surgical intervention in OA?
DR. BLOCK: It is entirely patient-oriented. You can’t tell by structure or by X-ray. When the patient’s OA is painful and severe enough, when it cannot be controlled with medication and medical regimens, or when it’s interfering with their lifestyle and they can’t live with it, they’ll tell you they need surgery.
DR. MOSKOWITZ: I completely agree. I think you can’t tell them. They’ll let you know. They’ll say, “I just have so much pain and disability. I need something done.”
DR. GALL: I think it’s important also to bring the surgeon in, not as a plumber to fix something, but as a partner in making the decision. Certainly the primary care physician, often the rheumatologist, can also provide guidance and work with the patients and the surgeons in making that decision.
DR. MOSKOWITZ: Yes. Let me just add one thing: I don’t think the patient should be in an extreme state of pain and disability in order to have surgery. If the condition interferes with their ADLs and the pain is really encumbering their daily activities, then it’s not unreasonable to consider surgery. I don’t think they have to be so bad that they’re limping and in terrible distress.
DR. BLOCK: Having said that, we should mention that joint replacement, at least in the knees and the hips, is extraordinarily effective in relieving pain.
DR. GALL: Yes. I would agree and these advances have continued to increase over the years, and these replacements last for a longer period of time, so the amount of satisfaction has increased, and the number of complications in experienced centers has certainly decreased.
Let’s end by talking about the future of OA. There are 2 areas that I’d like to touch on: One is the disease-modifying OA drugs (DMOADs) and the other is cartilage regeneration with stem cells and other techniques. Dr. Moskowitz, maybe you can talk about the DMOADs, and Dr. Block, you can talk about cartilage regeneration.
DR. MOSKOWITZ: Disease retardation or reversal would be the holy grail of OA management: to be able to treat OA with medications—the so-called DMOADS—that not only relieve symptoms but also slow the disease down or actually reverse the disease process. Some of the agents described as possibly having disease-modification potential include glucosamine, chondroitin sulfate, intra-articular hyaluronans, diacerein, avocado/soybean unsaponifiables, and doxycycline among others.17 Further studies will help to define the purported role of such agents in disease modification.
DR. BLOCK: I think it’s fair to say that, as of today, there really are no strategies that have been proven to be effective at retarding the progression and pain of OA over long periods of time. I think that is the gold standard that we’re looking for, and there is suggestive evidence from studies performed on each of the agents that Dr. Moskowitz just mentioned and several others.
I think there is a lot of work being done to study various metalloproteinase inhibitors, especially the collagenase 3 (MMP-13) inhibitors, and the aggrecanase inhibitors. I think that if some of these things are demonstrated to be effective in delaying the progression of OA, they could be extremely exciting, but one needs to keep in mind that whatever we use as a DMOAD has to be extremely safe because it’s going to be used for decades in people who are basically asymptomatic. In order to justify that on a public health basis, it has to be extraordinarily safe. It’s a very high bar.
DR. GALL: Okay. Dr. Block, can you now attempt to discuss cartilage regeneration and the use of the stem cell?
DR. BLOCK: Yes. Mesenchymal stem cell technology is extremely exciting, and it has paid off in some areas of bone and other connective tissues. As everybody knows, there is one U.S. Food and Drug Administration-approved approach for replacing cartilage that has had some success, but it’s used for isolated chondral defects in young and otherwise healthy people—not for OA.
The problem, of course, is that as we understand OA better and better, it’s more than just cartilage degeneration, and more importantly, the cartilage, which is very important, is a very thin and delicate tissue. If we replace the cartilage alone without normalizing the aberrant biomechanics across the joint—without normalizing the subchondral bone that the cartilage sits on—we have very little chance of getting long-term relief.
Having said that, efforts are being carried out in a variety of laboratories around the world to focus on not only mesenchymal cell-directed cartilage formation but also on actual repair of most tissues of the joint. Once we understand how to grow new cartilage and actually have it integrate with the adjacent tissue, which we still are not able to do, and once we’re able to repair some of the subchondral bone, I think that there is a lot of promise for that strategy in the long run. But, in the short-term, again, it’s a very difficult process that is not yet well understood.
DR. MOSKOWITZ: Well said. I agree. I think that we’re on the brink of being able to repair, particularly, very focal defects—maybe 4- or 5-mm defects. But, to repair the joint, as Dr. Block pointed out, you can’t just repair the cartilage because joint changes such as inflammation, meniscal alterations, varus and valgus deformities, and other changes impacting the joint are going to impair cell-based regeneration. Until we get those impacts controlled, I think it’s going to be some time before we can say to someone, “We’re going to replace your cartilage and normalize your joint.” However, I am optimistic and think that down the line, we may be able to do it.
DR. GALL: Dr. Block also mentioned that the underlying bone has changed too, and so, that needs to be attacked as well in these approaches.
DR. BLOCK: My personal view is that in the short-term future—the next 5 to 10 years—big strides will be made in strategies for better pain relief, and there are a variety of biologics that are in phase II and phase III testing that already look very exciting. Again, my own personal research interest is in trying to normalize the mechanical loads across the joints so that we can protect the joints better and so that the OA won’t progress, and I think those kinds of strategies will be available over the next couple of years.
DR. GALL: I’d like to thank you both for a really enlightening and comprehensive discussion of this common disease. Just to review, we’ve discussed what OA is and defined it, talked about its burden and its diagnosis, and mentioned the various risk factors for OA, particularly in light of what we can do to prevent OA or prevent the progression of the disease. We also had a comprehensive talk about the treatment of OA from both the standpoint of pain and the actual disease, and we touched upon the future of this as well as the research that’s being done in these areas.
I really appreciate your wisdom and discussion, and I think we’re providing the primary care physicians and providers with a really comprehensive look at this disease.
FoxP2 Media LLC is the publisher of The Medical Roundtable.
DR. BLOCK: Throughout the 20th century, we always considered OA to be primarily a degenerative disease of cartilage, but during the past decade or more, 2 things have become apparent: First, from a clinical perspective, OA is not just a structural degenerative process, it is primarily a very painful disease with pain being its most prominent feature. Second, the degenerative process involves not only the cartilage but the entire joint and all the involved structures within it.
A reasonable working definition of OA right now would be a painful degenerative process involving all the joint structures and is not primarily inflammatory but involves progressive deterioration of joint structures including articular cartilage. It’s important to keep in mind that pain itself is critical because as individuals age, everybody has some degenerative processes in their joints, and if we look hard enough, we can find the pathological features of OA in all elderly people. However, not all elderly people actually have the clinical disease.
DR. GALL: In a moment, we’ll talk about the causes of this, but, Dr. Moskowitz, would you tell us what the burden of this disease is here in the United States and worldwide?
DR. MOSKOWITZ: Well, it’s the most common form of arthritis and affects nearly 27 million or more Americans.1 It’s the most common disabling disease of the rheumatologic group, and it’s very disabling among all diseases.
DR. GALL: In the United States, you say that there’s about—
DR. MOSKOWITZ: About 27 million people with OA. And, in a decade, it is anticipated that the number will significantly increase.
DR. GALL: How many of those people are affected to the point that they’re unable to do their normal activities of daily living (ADLs)?
DR. MOSKOWITZ: Well, in terms of disability, about 80% of patients with OA have some degree of movement limitation, 25% cannot perform major ADLs, 11% of adults with knee OA need help with personal care, and 14% require help with routine needs.2
DR. GALL: Dr. Block, you talked about pain as a major manifestation of this disease. Does everybody who has OA present with pain?
DR. BLOCK: Again, it’s a definitional question; there is a distinction between clinically evident OA and asymptomatic structural degeneration of the joint. The population that Dr. Moskowitz was just describing, which comprises 20 to 30 million people, has doctor-diagnosed symptomatic OA, which is a painful disease. A much higher number of people who have structural degenerative disease would be diagnosed by radiography and said to fulfill a radiographic definition of OA, but all of the people who Dr. Moskowitz refer to have pain.
DR. MOSKOWITZ: Pain is what brings a patient to the doctor. I’ll often ask my patients, “How are you doing and what can I do for you?” They have limited motion and can’t get around. They say, “Dr. Moskowitz, I just want to get rid of the pain. I’ll worry about the swelling and getting around later.” And, that’s what really motivates them to get medical attention.
DR. GALL: Often, we will get consultations from primary care physicians for a patient who has OA that’s been picked up by radiography, and the patients will come and really not complain of pain in that area or possibly might have another underlying disease that’s causing the pain that’s not related to the OA. Dr. Block, do you want to say a word about that?
DR. BLOCK: Sure. By the age of 80 years, essentially, 100% of the population will have evidence of radiographic OA in at least one of their large joints. If you start with that number and you’re then referred an elderly patient merely for the radiographic appearance of OA, that doesn’t tell you very much about what’s going on clinically, and so, most of us would not even describe that as a clinical disease.
They may have a degenerative process in their knee, hip, etc, but of that group, say at the age of 80 years where almost 100% of people will have radiographic evidence of OA, only about 15% to 20% have symptomatic disease.3 That’s the culprit that I think we’re talking about clinically—people who not only have degenerative structural disease but also have symptoms that are interfering with their quality of life or their daily activities.
DR. MOSKOWITZ: I think that’s an important point, Dr. Block. For example, we all see patients who come in complaining of hip pain and you examine their knees as well and they have terrible genu varum (bowleg) and you say, “Do your knees bother you?” and they say, “Doctor, my knees are fine.” You think, my goodness, how can you walk? It isn’t necessarily true that if you have a deformity or involvement of a joint that you’re going to have pain, but if you have painful joints, that brings you to the doctor.
DR. GALL: Dr. Block, once again, I would like you to summarize the etiology of OA. Obviously, there has been much research and much information that has come to light in recent years regarding the causes of this disease.
DR. BLOCK: As I said, throughout the 20th century, people interested in the pathophysiology of OA were quite convinced that it was a degenerative process of the articular cartilage. As people age, the articular cartilage starts degenerating, and when there is absence of intact cartilage that ought to be providing essentially frictionless articulation of the bones during motion, there is deterioration, and that’s what we always considered to be OA.
Of course, the missing link in that paradigm is that cartilage itself has no nerve or blood supply and is therefore painless, and yet, people have horrible pain with this disease. So, our understanding over the past 10 or 15 years has really dramatically progressed, and we have begun to appreciate both degenerative and reactive processes in the subchondral bone, the ligaments, and the muscle.
The way we look at the pathophysiology of the onset and progression of the disease today is that there is primarily a degenerative process that stimulates reactive processes throughout the joint and even in the immune and inflammatory pathways. So, the reaction to that degenerative process sometimes causes stimulation of nociceptors and a lot of pain because of either local inflammation or mechanical alterations.
The degeneration itself, as I said, is painless. Often, people can have very degenerative joints, as Dr. Moskowitz just mentioned, and have no pain and therefore no clinical symptoms, so the disease is a combination of the degeneration and the secondary stimulated pain.
DR. GALL: So, is this primarily a biochemical or a cellular disease? Is it a genetic disease?
DR. BLOCK: Yes to all.
DR. MOSKOWITZ: There are a number of factors, Dr. Gall, involved in this condition. First, for example, is aging. We see that the frequency of OA increases logarithmically as we age, and there are pathologic changes with age that appear to lead to OA. For example, pigmented collections in the joint, comprising advanced glycation end products, occur that may predispose a patient to OA. You’re going to find more OA in an aging population.
The second factor is trauma. Professional football and basketball players, for example, who have chronic trauma, meniscal injuries, or cruciate ligament injuries are predisposed to OA. Third—the big risk factor, of course—is being overweight or obese. Patients who are overweight have a much higher frequency of OA, particularly in the knees. Last, there are genetic factors. For example, OA in the hands, which is more often seen in women, is genetically oriented so that if your mother had it or your aunt had it, you’re more likely to get it. Accordingly, there are multifactorial etiological factors that we have to try to address.
DR. GALL: Can you say a word about the different areas of the body that are associated with OA and talk a little bit about the difference between OA of the hands that we see frequently in women in relation to OA of the knee, hips, and spine, which can be devastating for individuals?
DR. MOSKOWITZ: Any joint can have it. It can be in the ankle, for example, if there has been trauma, but we’re looking particularly at the knees, the hips, the hands, and the spine. It’s interesting—certain joints in the hands are predisposed to OA including the distal and proximal interphalangeal joints. It tends not to involve the metacarpophalangeal joints, although it can in severe cases. Of course, the first carpometacarpal joint is frequently involved, and OA in this joint can be very disabling.
I have patients, for example, who are pianists or people who use their hands in their daily occupation. Their problem is often unfortunately minimized, with the doctor saying, “Oh, you’ve got a touch of OA, you can live okay with it.” However, this is hard to do because of pain and decreased function—OA can be a very disabling disease. In a number of women, the joints of the hands can be very inflamed. There’s a form of hand OA called erosive inflammatory OA, which can be especially troublesome because it’s very destructive. It’s almost like a low-grade rheumatoid arthritis where you actually get a lot of joint breakdown and deformity—it can be very disabling. OA of the knees or hips obviously can be a problem because of difficulties with ambulation.
DR. BLOCK: Our medical model trains us that, in general, we should look for an abnormality on a laboratory test to make a diagnosis. We do look for radiographic evidence of OA because, as I have said several times already, almost every elderly patient will have radiographic evidence of OA-like degeneration; however, we primarily use radiography adjunctively to make sure nothing else is going on or to assess the severity of the degeneration.
We make a diagnosis purely clinically. We look for someone who has the appropriate degree of pain in specific joints and does not have evidence of a systemic inflammatory source of that pain or a systemic rheumatic disease and in whom the pain appears to be articular rather than extra-articular. Of course, clinically, we feel for crepitus when we move the joint because of the degenerative processes, and we feel for bony enlargements or osteophytes around the joint to assist in making the diagnosis of OA.
Radiography is helpful in the sense that it can tell us about the severity of the degeneration and help us to make sure that we’re not missing something else such as a malignancy; however, radiography itself is not critical for the clinical diagnosis of OA.
DR. GALL: Dr. Block, how can the clinician who first sees a patient with OA differentiate it from a common type of inflammatory arthritis such as rheumatoid arthritis?
DR. MOSKOWITZ: If I may comment on that; you’re looking at 2 things. First, you are looking for symptoms and signs that you expect to be associated with OA, and then, you are looking for signs and symptoms that are not consistent with OA. As we said earlier, you may get some inflammation in the knee. You may get some swelling, and you can have evidence of increased synovial fluid, but you wouldn’t have the diffuse inflammatory reaction that you would have in rheumatoid arthritis, and there are different joints involved. For example, in rheumatoid arthritis, there may be involvement of the proximal interphalangeal and metacarpophalangeal joints of the hands, wrists, elbows, or shoulders. You would have involvement of peripheral joints that are different from those that are characteristically involved with OA.
In OA, the patient doesn’t have systemic findings such as fever, weight loss, or generalized prolonged stiffness. The patient often has localized stiffness in the involved joint but not generalized stiffness like that seen in rheumatoid arthritis. In OA, you can have multiple joints involved, such as 2 knees and 1 hip, but involvement doesn’t come on all at once and with the same severity.
DR. BLOCK: I completely agree with everything Dr. Moskowitz said. Additionally, if one is still confused after all of the clinical signs, this is where radiography may be useful because the radiographic appearance of OA is really different from that of the inflammatory arthritides.
In OA, one expects asymmetric narrowing in the joint itself. Additionally, OA results in subchondral sclerosis instead of periarticular osteopenia, which is seen in inflammatory arthritis. Finally, often, in OA, there is osteophyte formation, which is in contrast to the erosions that may be seen in the inflammatory arthritides. This is one area where radiography can really help.
DR. MOSKOWITZ: You have to be careful with respect to diagnosis in older people because if you do a serum rheumatoid factor study, the rheumatoid factor may be positive, but this may be unrelated to the patient’s symptoms. Studies have shown that a positive rheumatoid factor may be nonspecifically related to aging. So, you have to be careful—this finding can be a red herring in the diagnosis.
DR. BLOCK: I agree completely, and because OA is so common, one needs to bear in mind that a patient might have both inflammatory arthritis and underlying OA at the same time.
DR. MOSKOWITZ: Great point. We all see patients who come in to the office with OA of the hands, and then they develop rheumatoid arthritis or lupus engrafted on that previous involvement.
DR. GALL: I’d like to move on now to the management of the disease and, Dr. Moskowitz, would you talk to us about the first approach, which would be prevention? This is an approach that has caught the attention of the Centers for Disease Control and the Arthritis Foundation, and there is a major ongoing campaign at this time about what we can do to prevent this disease. Could you summarize that for us, Dr. Moskowitz?
DR. MOSKOWITZ: Absolutely. It’s generally thought that there’s nothing you can do to prevent OA. Well, we now think that you can slow it down and perhaps prevent it. Number one, we talked about weight. If a patient were to lose just 10 or 12 pounds, that could have a significant impact on the destructive effect of weight on the progression or the development of the disease. Programmed exercise is also very important. People say, “Well, I have arthritis. I can’t exercise,” and this is a problem because if you’re overweight and you have OA of the knees, joint overuse may temporarily lead to more symptoms. But walking at a reasonable pace and duration can be helpful with respect to symptoms and disability—the body puts out pain-relieving endorphins, muscles are strengthened, and joint range-of-motion is improved. Walking on a treadmill at a reasonable pace helps to achieve weight loss.
The patient can walk on a prescribed basis according to his or her capabilities. It’s important not to ask the patient to do something you know they’re not going to be able to do. We wouldn’t ask them to lose 20 pounds in the next 2 months. We have to make the program practical—these things are not easy to do, so we’ve got to be very supportive.
While trying to prevent disease progression, you want to have the patient strengthen their joint-related muscles. You want them to do muscle-building exercises because if you have joint stability, you’re likely going to decrease the osteoarthritic process as well. Losing weight is very important. Exercise is important. Using hot and cold applications for local therapy, which is safer than medication, is helpful.
Importantly, you don’t want to limit the patient’s activities any more than necessary. The way that I’ve always practiced medicine is that you don’t want to take away the patient’s ability to live his or her life as normally as possible. You don’t want to prescribe “don’ts,” like “don’t walk” and “don’t climb steps.” You want the patient to walk and do exercises, but the exercises have to be programmed so that the patient is able to do them.
DR. GALL: This is actually a very common question that primary care physicians are faced with when a patient with this disease comes in and is afraid to exercise and has a difficult time losing weight. But, I think it’s important that we understand that the cartilage gets its nourishment by being compressed and released and it has no blood supply of its own, so a lack of exercise is going to further cartilage degeneration.
DR. MOSKOWITZ: There are other non-weight bearing activities such as aquatic exercises that most people can do. You don’t have to be able to swim. You can do water exercises where you have the buoyancy of the water allowing for passive assistive exercise.
DR. BLOCK: Let me just add one other thing to the exercise discussion. In addition to there being a structural benefit to controlled exercise, every single systematic evaluation that has been performed has demonstrated that there is a really substantial and sustained pain relief component to exercise. People who have OA who are able to exercise regularly get substantial pain relief just from the exercise, and this is really important because the pain is what’s slowing them down in the first place.
The problem, of course, is maintaining an exercise regimen for a long term. Remember, this is a disease of decades, and we are not that good at behavioral modification strategies. Patients often give up on the exercise after a while. But, if you can maintain it, there are structural benefits as well as really substantial pain relief benefits.
DR. MOSKOWITZ: Yes, I agree. The other thing is that it’s hard for the busy physician in the office to start teaching the patient how to exercise, so we need to use arthritis health professionals such as occupational therapists and physical therapists to create exercise programs and teach patients how to exercise more effectively.
DR. BLOCK: Can I add one more thing as long as we’re discussing prevention? One of the major societal risk factors right now is recreational trauma. We have a population of young individuals who are being brought up in various organized sports, and there’s a very substantial risk factor for adolescent and young women who are, for example, playing soccer. They are getting knee and anterior cruciate ligament injuries, and over the upcoming 10 to 15 years, these people are at an enormous risk for developing knee OA. So, we need to develop some strategies to reduce the recreational trauma that results in early OA in this population.
DR. GALL: So, now we realize that weight control and exercise are important. Can we now talk about the medical management of both the pain and the arthritis?
DR. MOSKOWITZ: I know we don’t want to take too much time with this, but before we finish, I wanted to mention that I keep getting asked about tai chi or acupuncture for treatment. There are data suggesting that these programs can be helpful, but they should be adjunctive considerations for patients not responding to more routine therapeutic programs.4
DR. GALL: Yes, I would certainly agree with that. These are adjunctive therapies that may be helpful in select patients, but we need to focus on the more proven aspects of the prevention of the disease. Let’s now divide our talk about treatment into pain management and disease management. Can we start with pain management?
DR. BLOCK: I think that, first of all, it’s really important not to neglect the adjunctive measures that we’ve already discussed, and so, pain management is multifactorial, but one needs to include a physical and occupational therapist because although regular exercise dramatically reduces pain, local measures such as heat or ice are often very useful.
When they don’t get sufficient relief from adjunctive measures, symptomatic patients with OA will need to move to the pharmacologic arena sooner or later. From my practice and from the literature, I would say that there are 2 kinds of pain that need to be treated in OA: Pain that occurs during a painful flare that may last a couple of weeks and chronic ongoing pain that may develop as the disease progresses.
During short painful flares, most of the organizations that have published guidelines over the years have suggested that acetaminophen is a good place to start.5,6 I think that the systematic literature at this point suggests that acetaminophen might be very good for short-term pain relief—meaning, a few weeks at most. It’s probably not very effective for long-term pain relief, and it has a fair amount of potential complications, especially in the elderly. I don’t tend to use a lot of acetaminophen in people who are having chronic pain from OA, but during short-term pain flares, it can be very useful.
One can choose various analgesics including topical nonsteroidal agents, which can be very effective for local and monoarticular pain, for example, pain in superficial joints such as the knees or fingers. However, sooner or later, I believe that if patients don’t have a contraindication, they will end up taking either nonsteroidal anti-inflammatory agents (NSAIDs) or COX-2 inhibitors (coxibs), and that’s probably for a good reason. The nonsteroidals and coxibs have been demonstrated repeatedly to retain pain relief during 2-year studies, and they’re almost unique among the pharmacologic agents in that they retain that pain relief over a couple of years.7,8
Of course, if people are undergoing adequate pharmacologic therapy and they have painful flares in single joints, then intra-articular therapies such as glucocorticoids or hyaluronans are often very effective.
DR. GALL: I’d like to just go back to the NSAIDs for a moment. There is increasing emphasis on the risk of NSAIDs, particularly in the elderly. How do we balance the benefits of NSAID use for chronic therapy versus the risks, and how would you monitor the patient?
DR. MOSKOWITZ: Well, I agree with the premises that Dr. Block mentioned. I think acetaminophen has varying efficacy, but I think it’s worth trying. If I am right, I think it was stated in the past that the acetaminophen dose could go up to 4 g/d.9 The current recommendation is that doses of acetaminophen should not exceed 3 g/d so as not to incur hepatic or renal toxicity. The problem is that acetaminophen is often present in other medicines that patients are taking, and so an excess total intake is not uncommon.10
I think a trial of acetaminophen is worthwhile because even if it doesn’t relieve all the pain, it’s a floor of analgesia that you can add to with nonsteroidals.
With regard to the question about whether nonsteroidals are dangerous in older people, I think that they can be. Most things we use are unfortunately associated with some risk, although pain itself is also a risk factor. Pain increases blood pressure and pulse rate. If you relieve the pain, you’re probably relieving more of the stress on the patient overall as opposed to the risk of the NSAID itself. If you use nonselective NSAIDs, I would suggest adding a proton pump inhibitor, especially if the patient has an increased risk of peptic ulcer disease. If you use a COX-2 selective inhibitor, data suggest there is a decreased gastrointestinal (GI) risk. Overall, I think that NSAIDs can be effective and reasonably safe. I’m more careful about using NSAIDs in individuals older than 75 or 80 years; however, many of these patients are physiologically younger than their numerical age, and they can more safely tolerate these agents.
Intra-articular steroids and intra-articular hyaluronans can be very helpful in the management of knee OA, and they have comfortable safety factors.
DR. GALL: I’ll just make a comment about my own approach to that. When I put a patient on an NSAID, and I do use them frequently, I mark it on a problem list. I’m careful to ask the patients about bleeding and to warn them about the signs of GI bleeding, and I do periodic blood counts to look at hemoglobin and creatinine levels. I also do a urinalysis to rule out silent GI or renal effects of these drugs, but I do use NSAIDs frequently.
Additionally, with the recent attention to the pain component of OA over the past decade, there has been a lot of systematic investigation into various neuroactive pain-directed medications. In fact, duloxetine, which is a neuroactive agent, was approved for use a couple of years ago. There are non–anti-inflammatory, non-purely analgesic medicines that can be very helpful for the treatment of the pain component of OA.
DR. GALL: Thank you. Could we just say a word about opioids and tramadol as adjunctive therapies and what the pluses and minuses of using these drugs are?
DR. MOSKOWITZ: I think there is not only a rationale but a place for these agents in treating OA, but we’ve got to be careful. I’m a little less concerned about using tramadol than using the classic opioids, per se, and I think tramadol can be a reasonable bridge when people are not responding to NSAIDS. A number of years ago, when I was writing a piece for the Arthritis Foundation pamphlet, I stated that there is never a place for opioids in the treatment of OA, and I can’t tell you how many of my colleagues wrote and said, “Dr. Moskowitz, there are times when we feel they need to be available.” I agree, but I think my guidance is that, if you use an opioid, use the lowest dose possible and for the shortest time possible.
I think if someone is in a lot of pain and they’re having a flare, particularly an older individual, it may be safer to use the opioids than to put them on NSAIDs, particularly if they have had past trouble with NSAIDS. I don’t think that we should never use opioids when treating OA. What are your thoughts on that, Dr. Block?
DR. BLOCK: I agree. I think that there is good evidence that the opiates provide really substantial pain relief, so they’re effective in OA. The problem, of course, is that they have very high side effect profiles, and the people that they’re the most dangerous for with regard to falling and injuring oneself are actually the same people who are at high risk with nonsteroidals—the very elderly with congestive heart failure. It really does put us in a bind.
I think that I agree with what you just said, which is that they are effective, and at times, I think that there is clearly a place for opiates in OA, but one needs to be judicious and careful when using them. Tramadol is, I think, much safer. It’s a weak opiate. On the other hand, it also has much less pain-relieving potential, so this is the bind we face.
Well, I think what we haven’t mentioned yet are the surgical approaches, which are critically important.
DR. MOSKOWITZ: Should we discuss glucosamine and chondroitin sulfate?
DR. BLOCK: So, more than half of patients who have symptomatic OA use various complementary medicine approaches. They are widely used because they’re widely believed to be beneficial, and I think that there has been a fair amount of systematic evaluation, specifically of glucosamine and chondroitin sulfate, over the past several years.
When we look at the independently sponsored studies, the evidence suggests that the side effect profile for both agents is very good. As long as they were manufactured in a safe manner, they are very safe and one shouldn’t worry about using them.
On the other hand, the efficacy above placebo is very low, but that doesn’t mean that they’re not effective in individual patients. The thing that’s really important to note here is that whenever there is a placebo-controlled study with pain as an outcome in OA, the placebo response is huge. More than half of the people who get placebo get very substantial clinically significant pain relief, and more importantly, it’s durable. In all of these placebo-controlled studies that go on for 2 years, the placebo group with pain relief had sustainability over 2 years,11,12 and so, if one gets more than a placebo response from glucosamine, and if it’s providing pain relief, I’m happy for them to use it.
The Glucosamine/Chondroitin Arthritis Intervention Trial (GAIT),7 the large National Institutes of Health multicenter study, which was a null study, showed that fundamentally, there was no systematic advantage of a combination glucosamine plus chondroitin sulfate. Even when that came out as a null study and was publicized as such, the market for glucosamine and chondroitin sulfate in the following year didn’t decrease in the United States.13,14
People who are getting a benefit, whether it’s due to an endorphin effect from placebo or the agent, are going to continue using the agent. As long as the agents are safe, I’m comfortable with my patients using them if they really are feeling like they are getting relief.
DR. MOSKOWITZ: Yes, I think that there are conflicting data, but there are a number of studies that seem to show that these agents do help some people. For example, this question about the GAIT study7 not showing an effect, if you’re using glucosamine and chondroitin sulfate in individuals who have high pain to start with—if they had a score of more than 300 mm on the Western Ontario and McMaster Universities Arthritis Index (WOMAC)—they appeared to do better, but there wasn’t a predefined endpoint, so the result has to be substantiated in a repeat study. However, I see enough people—like you do, Dr. Block—who seem to respond. They’re safe agents, and I think that they do have merit in patients who are not getting a benefit from other agents.
DR. GALL: Before we move on to surgery and the future of therapy, I just want to discuss intra-articular steroids. What are your thoughts on these?
DR. MOSKOWITZ: I would be lost without them. I remember when they were first used by Dr. Joseph Hollander in Philadelphia, and they failed to help because he used cortisone acetate instead of the hydrocortisone derivative, and cortisone has to be converted to hydrocortisone before it works.15 I use intra-articular corticosteroids a lot. The general recommendation is that you shouldn’t use it more than 4 times a year in the knee. I think maybe if somebody is older—for example, 88 years—and they cannot tolerate other therapy, you might use them that often. Studies have shown that an intra-articular corticosteroid injection every 3 months did not lead to deleterious anatomic effects.16 I think using them 2 or 3 times a year, if indicated, can be symptomatically effective and safe, and I don’t think that will lead to joint breakdown if the patient is careful about doing their exercises and other management.
DR. GALL: Let’s move on to the use of surgery, and without going into all the surgical procedures, when is it appropriate for the clinician and the patient to consider surgical intervention in OA?
DR. BLOCK: It is entirely patient-oriented. You can’t tell by structure or by X-ray. When the patient’s OA is painful and severe enough, when it cannot be controlled with medication and medical regimens, or when it’s interfering with their lifestyle and they can’t live with it, they’ll tell you they need surgery.
DR. MOSKOWITZ: I completely agree. I think you can’t tell them. They’ll let you know. They’ll say, “I just have so much pain and disability. I need something done.”
DR. GALL: I think it’s important also to bring the surgeon in, not as a plumber to fix something, but as a partner in making the decision. Certainly the primary care physician, often the rheumatologist, can also provide guidance and work with the patients and the surgeons in making that decision.
DR. MOSKOWITZ: Yes. Let me just add one thing: I don’t think the patient should be in an extreme state of pain and disability in order to have surgery. If the condition interferes with their ADLs and the pain is really encumbering their daily activities, then it’s not unreasonable to consider surgery. I don’t think they have to be so bad that they’re limping and in terrible distress.
DR. BLOCK: Having said that, we should mention that joint replacement, at least in the knees and the hips, is extraordinarily effective in relieving pain.
DR. GALL: Yes. I would agree and these advances have continued to increase over the years, and these replacements last for a longer period of time, so the amount of satisfaction has increased, and the number of complications in experienced centers has certainly decreased.
Let’s end by talking about the future of OA. There are 2 areas that I’d like to touch on: One is the disease-modifying OA drugs (DMOADs) and the other is cartilage regeneration with stem cells and other techniques. Dr. Moskowitz, maybe you can talk about the DMOADs, and Dr. Block, you can talk about cartilage regeneration.
DR. MOSKOWITZ: Disease retardation or reversal would be the holy grail of OA management: to be able to treat OA with medications—the so-called DMOADS—that not only relieve symptoms but also slow the disease down or actually reverse the disease process. Some of the agents described as possibly having disease-modification potential include glucosamine, chondroitin sulfate, intra-articular hyaluronans, diacerein, avocado/soybean unsaponifiables, and doxycycline among others.17 Further studies will help to define the purported role of such agents in disease modification.
DR. BLOCK: I think it’s fair to say that, as of today, there really are no strategies that have been proven to be effective at retarding the progression and pain of OA over long periods of time. I think that is the gold standard that we’re looking for, and there is suggestive evidence from studies performed on each of the agents that Dr. Moskowitz just mentioned and several others.
I think there is a lot of work being done to study various metalloproteinase inhibitors, especially the collagenase 3 (MMP-13) inhibitors, and the aggrecanase inhibitors. I think that if some of these things are demonstrated to be effective in delaying the progression of OA, they could be extremely exciting, but one needs to keep in mind that whatever we use as a DMOAD has to be extremely safe because it’s going to be used for decades in people who are basically asymptomatic. In order to justify that on a public health basis, it has to be extraordinarily safe. It’s a very high bar.
DR. GALL: Okay. Dr. Block, can you now attempt to discuss cartilage regeneration and the use of the stem cell?
DR. BLOCK: Yes. Mesenchymal stem cell technology is extremely exciting, and it has paid off in some areas of bone and other connective tissues. As everybody knows, there is one U.S. Food and Drug Administration-approved approach for replacing cartilage that has had some success, but it’s used for isolated chondral defects in young and otherwise healthy people—not for OA.
The problem, of course, is that as we understand OA better and better, it’s more than just cartilage degeneration, and more importantly, the cartilage, which is very important, is a very thin and delicate tissue. If we replace the cartilage alone without normalizing the aberrant biomechanics across the joint—without normalizing the subchondral bone that the cartilage sits on—we have very little chance of getting long-term relief.
Having said that, efforts are being carried out in a variety of laboratories around the world to focus on not only mesenchymal cell-directed cartilage formation but also on actual repair of most tissues of the joint. Once we understand how to grow new cartilage and actually have it integrate with the adjacent tissue, which we still are not able to do, and once we’re able to repair some of the subchondral bone, I think that there is a lot of promise for that strategy in the long run. But, in the short-term, again, it’s a very difficult process that is not yet well understood.
DR. MOSKOWITZ: Well said. I agree. I think that we’re on the brink of being able to repair, particularly, very focal defects—maybe 4- or 5-mm defects. But, to repair the joint, as Dr. Block pointed out, you can’t just repair the cartilage because joint changes such as inflammation, meniscal alterations, varus and valgus deformities, and other changes impacting the joint are going to impair cell-based regeneration. Until we get those impacts controlled, I think it’s going to be some time before we can say to someone, “We’re going to replace your cartilage and normalize your joint.” However, I am optimistic and think that down the line, we may be able to do it.
DR. GALL: Dr. Block also mentioned that the underlying bone has changed too, and so, that needs to be attacked as well in these approaches.
DR. BLOCK: My personal view is that in the short-term future—the next 5 to 10 years—big strides will be made in strategies for better pain relief, and there are a variety of biologics that are in phase II and phase III testing that already look very exciting. Again, my own personal research interest is in trying to normalize the mechanical loads across the joints so that we can protect the joints better and so that the OA won’t progress, and I think those kinds of strategies will be available over the next couple of years.
DR. GALL: I’d like to thank you both for a really enlightening and comprehensive discussion of this common disease. Just to review, we’ve discussed what OA is and defined it, talked about its burden and its diagnosis, and mentioned the various risk factors for OA, particularly in light of what we can do to prevent OA or prevent the progression of the disease. We also had a comprehensive talk about the treatment of OA from both the standpoint of pain and the actual disease, and we touched upon the future of this as well as the research that’s being done in these areas.
I really appreciate your wisdom and discussion, and I think we’re providing the primary care physicians and providers with a really comprehensive look at this disease.
FoxP2 Media LLC is the publisher of The Medical Roundtable.
