Julienne K. Kirk, PharmD, CDE

Evidence summary

A 2012 Cochrane review and meta-analysis of 9 RCTs found that 1261 patients who used SMBG showed a small but statistically significant decrease in HbA1c at 6 months compared with 1063 controls. In 2 other RCTs, patients using SMBG showed a nonsignificant decrease in HbA1c compared with control subjects at 12 months (TABLE 1).¹

TABLE 1
Difference in HbA1c by duration of self-monitoring*

Another meta-analysis reported similar findings. The study grouped 9 RCTs based on the duration of SMBG and examined the change in HbA1c from baseline. In 5 of the trials, SMBG for 6 months was associated with a small decrease in HbA1c, but in the other 4, SMBG for longer than one year didn’t significantly change HbA1c levels.²

Baseline values make a difference
A meta-analysis of 9 RCTs demonstrated that SMBG was marginally superior to non-SMBG for reducing HbA1c when the baseline value was >8%. SMBG didn’t lower HbA1c in patients with a baseline HbA1c <8%. The greatest change in HbA1c occurred in patients with baseline values >10% (TABLE 2).³

In another meta-analysis, 12 of 15 RCTs found SMBG to be better than non-SMBG at reducing HbA1c when the baseline was >8%.⁴

TABLE 2
Patients benefit from self-monitoring when baseline HbA1c is >8%

Limitations of studies
Limitations of the studies (TABLES 1 AND 2) reviewed included methodological quality,^1,3 limited patient compliance reporting,³ heterogeneity,^1,2,4 and small sample size.^2,3

More frequent self-monitoring has no effect
A systematic review of 4 RCTs with a total of 637 patients compared frequent SMBG (4-7 times a week) with less frequent self-monitoring (1-2 times a week) for periods ranging from 3 to 12 months and found no difference in reduction of values (HbA1c reduction difference between the groups = –0.21; 95% confidence interval, –0.57 to 0.15).⁴

Recommendations

The American Diabetes Association advocates SMBG as a guide for patients who use oral or medical nutrition therapies for diabetes. Patients should receive initial instruction in SMBG and routine follow-up evaluation of their technique and ability to use data to adjust therapy.⁵

The American Association of Clinical Endocrinologists (AACE) advises that SMBG can be initiated at the same time as medical therapy, lifestyle modification, specific diabetes education, or dietary consultation. If HbA1c levels are above target, the AACE recommends more frequent SMBG: preprandially, 2 hours postprandially, occasionally between 2 am and 3 am, during illness, or anytime a low glucose level is suspected.⁶

Evidence summary

A 2012 Cochrane review and meta-analysis of 9 RCTs found that 1261 patients who used SMBG showed a small but statistically significant decrease in HbA1c at 6 months compared with 1063 controls. In 2 other RCTs, patients using SMBG showed a nonsignificant decrease in HbA1c compared with control subjects at 12 months (TABLE 1).¹

TABLE 1
Difference in HbA1c by duration of self-monitoring*

Another meta-analysis reported similar findings. The study grouped 9 RCTs based on the duration of SMBG and examined the change in HbA1c from baseline. In 5 of the trials, SMBG for 6 months was associated with a small decrease in HbA1c, but in the other 4, SMBG for longer than one year didn’t significantly change HbA1c levels.²

Baseline values make a difference
A meta-analysis of 9 RCTs demonstrated that SMBG was marginally superior to non-SMBG for reducing HbA1c when the baseline value was >8%. SMBG didn’t lower HbA1c in patients with a baseline HbA1c <8%. The greatest change in HbA1c occurred in patients with baseline values >10% (TABLE 2).³

In another meta-analysis, 12 of 15 RCTs found SMBG to be better than non-SMBG at reducing HbA1c when the baseline was >8%.⁴

TABLE 2
Patients benefit from self-monitoring when baseline HbA1c is >8%

Limitations of studies
Limitations of the studies (TABLES 1 AND 2) reviewed included methodological quality,^1,3 limited patient compliance reporting,³ heterogeneity,^1,2,4 and small sample size.^2,3

More frequent self-monitoring has no effect
A systematic review of 4 RCTs with a total of 637 patients compared frequent SMBG (4-7 times a week) with less frequent self-monitoring (1-2 times a week) for periods ranging from 3 to 12 months and found no difference in reduction of values (HbA1c reduction difference between the groups = –0.21; 95% confidence interval, –0.57 to 0.15).⁴

Recommendations

The American Diabetes Association advocates SMBG as a guide for patients who use oral or medical nutrition therapies for diabetes. Patients should receive initial instruction in SMBG and routine follow-up evaluation of their technique and ability to use data to adjust therapy.⁵

The American Association of Clinical Endocrinologists (AACE) advises that SMBG can be initiated at the same time as medical therapy, lifestyle modification, specific diabetes education, or dietary consultation. If HbA1c levels are above target, the AACE recommends more frequent SMBG: preprandially, 2 hours postprandially, occasionally between 2 am and 3 am, during illness, or anytime a low glucose level is suspected.⁶

Evidence summary

A 2012 Cochrane review and meta-analysis of 9 RCTs found that 1261 patients who used SMBG showed a small but statistically significant decrease in HbA1c at 6 months compared with 1063 controls. In 2 other RCTs, patients using SMBG showed a nonsignificant decrease in HbA1c compared with control subjects at 12 months (TABLE 1).¹

TABLE 1
Difference in HbA1c by duration of self-monitoring*

Another meta-analysis reported similar findings. The study grouped 9 RCTs based on the duration of SMBG and examined the change in HbA1c from baseline. In 5 of the trials, SMBG for 6 months was associated with a small decrease in HbA1c, but in the other 4, SMBG for longer than one year didn’t significantly change HbA1c levels.²

Baseline values make a difference
A meta-analysis of 9 RCTs demonstrated that SMBG was marginally superior to non-SMBG for reducing HbA1c when the baseline value was >8%. SMBG didn’t lower HbA1c in patients with a baseline HbA1c <8%. The greatest change in HbA1c occurred in patients with baseline values >10% (TABLE 2).³

In another meta-analysis, 12 of 15 RCTs found SMBG to be better than non-SMBG at reducing HbA1c when the baseline was >8%.⁴

TABLE 2
Patients benefit from self-monitoring when baseline HbA1c is >8%

Limitations of studies
Limitations of the studies (TABLES 1 AND 2) reviewed included methodological quality,^1,3 limited patient compliance reporting,³ heterogeneity,^1,2,4 and small sample size.^2,3

More frequent self-monitoring has no effect
A systematic review of 4 RCTs with a total of 637 patients compared frequent SMBG (4-7 times a week) with less frequent self-monitoring (1-2 times a week) for periods ranging from 3 to 12 months and found no difference in reduction of values (HbA1c reduction difference between the groups = –0.21; 95% confidence interval, –0.57 to 0.15).⁴

Recommendations

The American Diabetes Association advocates SMBG as a guide for patients who use oral or medical nutrition therapies for diabetes. Patients should receive initial instruction in SMBG and routine follow-up evaluation of their technique and ability to use data to adjust therapy.⁵

The American Association of Clinical Endocrinologists (AACE) advises that SMBG can be initiated at the same time as medical therapy, lifestyle modification, specific diabetes education, or dietary consultation. If HbA1c levels are above target, the AACE recommends more frequent SMBG: preprandially, 2 hours postprandially, occasionally between 2 am and 3 am, during illness, or anytime a low glucose level is suspected.⁶

The youngest Americans—those born in the year 2000 or thereafter—have more than a 1 in 3 lifetime risk of developing diabetes, according to the Centers for Disease Control and Prevention.¹ That estimate, coupled with the fact that more than 2 out of 3 adults and 1 in 6 children between the ages of 2 and 19 years are overweight,² would seem to indicate a need for widespread diabetes screening. But limited health care resources, a lack of evidence that mass screening improves outcomes, and differences among leading medical associations about whom and when to screen argue against it.

At the same time, widespread obesity is making the presentation of hyperglycemia more complex and the forms of diabetes harder to classify. Many cases don’t follow the classic patterns, in which type 1 (formerly called juvenile diabetes) virtually always emerges in childhood and type 2 (previously known as adult-onset diabetes) is strictly an adult disease. Our evolving understanding of diabetes has led researchers to focus on prediabetes (defined as impaired fasting glucose, impaired glucose tolerance, or both) and latent autoimmune diabetes in adults (LADA), a recently reported type 1 variant that some have labeled type 1.5.³

In the face of growing complexity, the US Preventive Services Task Force (USPSTF) last year upgraded its recommendation for screening for type 2 diabetes, and researchers have developed new risk calculation tools. We’ve taken a look at the changing clinical landscape and sorted through the latest evidence to help you make sense of the latest risk and diagnostic developments in diabetes care.

Screening for type 2: A look at guidelines

Type 2 diabetes accounts for approximately 90% of the cases you’ll see.^1,4,5 The American Diabetes Association (ADA) calls for routine screening, starting at 45 years of age and continuing every 3 years thereafter in the absence of risk. But for those who are overweight or obese and have 1 or more additional risk factors, screening is recommended at any age.⁵ In addition to a body mass index (BMI) ≥25, risks include physical inactivity, a first-degree relative with type 2 diabetes, blood pressure >135/80 mm Hg (or controlled with an antihypertensive), high-density lipoproteins (HDL) <35 mg/dL, triglycerides >250 mg/dL, polycystic ovary syndrome, impaired glucose tolerance or impaired fasting glucose, and acanthosis nigricans, a pigmented thickening of the skin folds of the neck (TABLE).^6,7 Patients with the metabolic syndrome—abdominal obesity (defined as a waist circumference of >40” in men and >35” in women) and ≥2 of the following: raised triglyceride levels, elevated blood pressure, elevated fasting plasma glucose, and reduced HDL cholesterol—are at especially high risk of both cardiovascular disease and type 2 diabetes.⁸

The ADA screening recommendations, however, are not based on prospective outcome studies, nor are they widely followed. Until recently, the USPSTF only recommended screening adults with hypertension and hyperlipidemia.

In 2008, after an assessment of new findings and research updates, the USPSTF revised its recommendation: The task force now calls for screening asymptomatic adults with sustained blood pressure >135/80 mm Hg—regardless of lipid profile.⁷ For patients with diabetes and hypertension, the USPSTF concluded, evidence shows that early intervention—including lowering blood pressure below conventional targets—can prevent long-term adverse outcomes of diabetes and reduce the risk of cardiovascular events.

Although mass screening remains controversial, regular assessment of risk factors and targeting individuals with established risk is clearly indicated (PATIENT HANDOUT). The importance of early detection was highlighted by the United Kingdom Prospective Diabetes Study, in which approximately half of the patients with newly diagnosed type 2 diabetes already had evidence of complications.⁹

TABLE
Type 1, type 2, and gestational diabetes: Diagnostic clues

Validated risk calculators can boost detection rates

In an attempt to improve detection rates of type 2 diabetes and prediabetes, researchers in both the United States and the United Kingdom recently developed easy-to-use risk calculation tools. The Diabetes Risk Calculator (available at http://www.diabetes.org/food-nutrition-lifestyle/lifestyle-prevention/risk-test.jsp), published in 2008, was validated with findings from the Third National Health and Nutrition Survey.¹⁰ The calculator uses answers to questions about age, waist circumference, history of gestational diabetes mellitus (GDM), height, race/ethnicity, hypertension, family history, and exercise to determine whether an individual is at high risk for undetected diabetes. The tool has a low positive predictive value (14%), but a negative predictive value >99%.¹⁰

The QDScore Diabetes Risk Calculator (www.qdscore.org), another new tool, is designed to estimate an individual’s 10-year risk of developing type 2 diabetes.¹¹ The program, which calculates risk based on answers to questions about family history of diabetes, patient history of cardiovascular disease, smoking, treatment for hypertension, BMI, ethnicity, and steroid use, was validated with data collected from 2.5 million patients in practices throughout England and Wales. The screening tool showed a high degree of discrimination in reflecting differences in disease prevalence related to ethnic and socioeconomic risk factors.¹¹

Pinning down a type 2 (or prediabetes) diagnosis

The ADA, American Association of Clinical Endocrinologists (AACE), USPSTF, and World Health Organization/International Diabetes Federation agree on the diagnostic criteria for type 2 diabetes: a fasting glucose >126 mg/dL, a random plasma glucose ≥200 mg/dL (that must be confirmed on a subsequent day), or both.^5,7,12,13 Patient history, risk factors, and additional laboratory tests can help clinicians distinguish between type 1 and type 2 diabetes.

An oral glucose tolerance test (OGTT) is also an option for diagnosis, but time and scheduling difficulties limit the routine use of this test in primary care. Hemoglobin A1c is not recommended as a diagnostic test because of a lack of standardization.¹

Prediabetes and type 2 risk. One in 4 (25.9%) US adults 20 years of age or older and more than 1 in 3 (35.9%) of those 60 years of age or older have prediabetes,¹⁴ defined as impaired fasting glucose (100-125 mg/dL), impaired glucose tolerance (2-hour glucose test results of 140-199 mg/dL), or both. Prediabetes increases the risk of developing type 2 diabetes by an estimated 30% over a 4-year period,¹⁵ and 70% over 30 years,¹⁶ although lifestyle interventions can substantially lower the risk. In a recently released consensus statement, an AACE task force noted that in addition to the increased risk of type 2 diabetes, patients with prediabetes face a greater risk of macrovascular complications.¹⁷

Type 2 in kids can be mistaken for type 1

As childhood obesity has surged, type 2 diabetes has been diagnosed at an increasingly early age—even in children younger than 10 years.¹⁸ Minority youth, primarily African Americans, Hispanics, and Asians/Pacific Islanders, are at increased risk.¹⁴ Symptoms can be insidious in children and adolescents and easily missed or mistaken for type 1 diabetes, in part because type 2 diabetes is still relatively rare in this age group.¹⁹

Preteens at risk. In a recent study of BMI and metabolic syndrome risk factors in 8- to 14-year-olds, however, researchers concluded that children who are overweight in early adolescence may be at risk for type 2 diabetes as well as cardiovascular disease before they reach their teens.²⁰ There is evidence of a genetic predisposition for type 2 diabetes and defects of β-cell function,^5,21 and family history, in addition to weight, is an important consideration in identifying type 2 diabetes in young patients.

Although young adults with type 1 and type 2 diabetes can present with similar symptoms, there may be certain clues to a type 2 diagnosis. Acanthosis nigricans, which is related to insulin resistance and occurs most frequently in obese adolescents, points to a type 2 diagnosis. Increased insulin and C-peptide levels are indicators of type 2 diabetes. Low levels are not necessarily an indication of type 1, however, because patients with type 2 diabetes may have low levels of insulin and C-peptide because of glucose toxicity and lipotoxicity at the time of diagnosis.²² Treatment with insulin may be necessary until glucose toxicity resolves.

Type 1 diabetes: Beyond childhood

Approximately 5% to 10% of patients with diabetes have type 1, which is defined as idiopathic or cellular immune-mediated autoimmune β-cell destruction.⁵ The rate of destruction is variable—it generally progresses more rapidly in infants and children than in adults. Some people with type 1 diabetes retain residual β-cell function, but have little or no insulin secretion; this manifests as a low or undetectable level of serum C-peptide.

Most cases of type 1 diabetes are diagnosed in patients younger than 18 years. But type 1 diabetes is increasingly recognized as a disorder that also develops in early adulthood, usually before the age of 40.

Arriving at a type 1 diagnosis

Patients with type 1 diabetes often present with modest hyperglycemia, but may rapidly progress to severe hyperglycemia and diabetic ketoacidosis (DKA) when infection or other physical stressors occur.

While screening for autoantibodies in asymptomatic individuals is not recommended,⁵ patients with blood glucose levels ≥200 mg/dL and symptoms of polydipsia, polyuria, and polyphagia who do not meet the profile for type 2 diabetes may be candidates for additional laboratory work. Approximately 85% to 90% of patients with type 1 diabetes will have antibodies to islet cells or glutamic acid decarboxylase (GAD).^5,23

Even without antibody testing, there are distinguishing characteristics that help support a type 1 diagnosis. As a general rule, individuals who develop type 1 diabetes—especially children—are not obese, although patients usually gain weight over time. In addition, many patients with type 1 diabetes have an auto-immune disease, such as celiac or Graves’ disease, hypothyroidism, adrenal anemia, or pernicious anemia; and a first-degree relative with type 1 diabetes. DKA, with acute symptoms of polydipsia and/or polyuria and recent, unintentional weight loss, is suggestive of—but not definitive for—type 1 diabetes.

A recently validated type 1 risk calculator may be particularly useful for screening patients who have a sibling, parent, or child with type 1 diabetes. Using age, BMI, C-peptide concentration, and OGTT results, the algorithm was highly predictive of type 1 diabetes in family members of patients who tested positive for islet cell antibodies.²⁴

Patient doesn’t “fit” type 1 or 2? Consider LADA

LADA, a gradual, progressive form of type 1 diabetes, can be difficult to identify. Circulating GAD or islet cell antibodies are present, but patients don’t have an absolute need for insulin at the time of diagnosis. Thus, they’re often thought to have type 2 diabetes.²⁵ Individuals with LADA show no signs of insulin resistance, however, and over time, β cells decline and insulin usually becomes necessary.

There are no universal recommendations for testing for LADA. Rather, the diagnosis should be considered in those who don’t fit the classic profile for type 1 or type 2 diabetes,²⁶ but have some of the following features:

• age <50 years
• acute symptoms of polydipsia, polyuria, and/or unintentional weight loss
• BMI <25
• a personal history of autoimmune disease
• a family history of autoimmune disease.²⁷

A prospective analysis found that the majority of LADA patients had at least 2 of these distinguishing characteristics.²⁸ Other recent research found heterogeneity among patients with LADA. Noting that not all patients with LADA become insulin-dependent, researchers concluded that the need for insulin is linked to the degree of autoimmunity and β-cell failure.²⁹

When GDM complicates prenatal care

Any degree of carbohydrate intolerance that is first recognized during pregnancy is classified as GDM, whether or not the condition resolves after delivery. A GDM diagnosis does not preclude the possibility of undiagnosed type 2 diabetes or prediabetes, or (rarely) type 1 diabetes.

Approximately 7% of all pregnancies in the United States are complicated by GDM, totaling more than 200,000 cases annually.⁵ The rate of GDM is in direct proportion to the prevalence of type 2 diabetes in the population in question, and ranges from 1% to 14%. GDM is the diagnosis in nearly 90% of pregnancies complicated by diabetes.⁵

The GDM screening controversy

Screening for GDM—whether it should be done universally or selectively on the basis of risk factors—is highly controversial. The USPSTF maintains that there is insufficient evidence to recommend for or against screening women with no history of GDM. The American College of Obstetricians and Gynecologists (ACOG)³⁰ and ADA⁵ recommend selective screening based on patient history, clinical presentation, and, possibly, prior impaired glucose test results or other abnormal laboratory values. AACE calls for universal screening of pregnant women, starting at 20 weeks for high-risk individuals and between 24 and 28 weeks for those at low risk.¹²

Identifying patients at risk. Maternal age (>25 years), obesity (BMI ≥30), prior GDM or delivery of a large-forgestational-age infant, belonging to a high-risk ethnic group, glycosuria, history of glucose resistance or glucose tolerance, and a first-degree relative with diabetes (TABLE) are risk factors for GDM. Women at high risk—those who meet all or most of these criteria—should undergo early screening: at the first prenatal visit, according to ACOG;³⁰ upon confirmation of pregnancy (ADA);⁵ at 20 weeks’ gestation (AACE);¹² or between 24 and 28 weeks’ gestation (USPSTF).⁷ ADA and ACOG recommend a 2-stage approach, starting with a 50-g 1-hour OGTT and following up with a 100-g 3-hour OGTT if the first test results are not definitive.^5,30 Testing for patients at average risk—which includes any pregnant woman with even a single risk factor, such as being older than 25 years—should be done between 24 and 28 weeks’ gestation, according to ACOG and ADA; testing is not required for women who are <25 years, have a normal body weight, and no other risk factors.

GDM screening in primary care. Because most women fit the criteria for average or high risk,³¹ family physicians may find universal screening to be more practical than individual risk assessment. Universal screening is associated with favorable outcomes,³² but screening limited to those at high and average risk also has evidence to support it. In a study of 25,118 deliveries, only 4% of women with GDM were missed by the exclusion of low-risk patients.³³

In the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study, researchers tracked 25,505 women from 9 countries and found a continuous relationship between the risk of macrosomia and the rise in maternal glucose levels.³⁴ The impact on the developing fetus of varying degrees of glucose was studied after a 75-g 2-hour OGTT. The risk of macrosomia increased with fasting blood glucose >75 mg/dL, 1-hour glucose levels >105 mg/dL, and 2-hour glucose concentration >90 mg/dL.³⁵ The most compelling results for adverse effects were associated with fasting glucose levels, rather than glucose tolerance tests.

2 abnormal results needed for a GDM diagnosis

In the absence of unequivocal hyperglycemia, there are 2 diagnostic standards for GDM: The Carpenter-Coustan Conversion and the National Diabetes Data Group Conversion. The Carpenter-Coustan Conversion uses lower glucose values for fasting (≥95 mg/dL) and subsequent 1-, 2-, and 3-hour levels (≥180, 155, and 140 mg/dL, respectively) and is more widely used. But expert opinion also supports the National Diabetes Data Group Conversion criteria (fasting plasma glucose, ≥105 mg/dL; ≥190, 165, and 145 mg/dL for 1-, 2-, and 3-hour OGTT, respectively), and there are no data from clinical trials to prove the superiority of either standard.³⁰

Both sets of standards require 2 or more thresholds to be met or exceeded for a GDM diagnosis. Women with only 1 abnormal value should be monitored carefully, however, as they, too, may be at increased risk for macrosomia and other morbidities.³⁰

Postpartum follow-up. Obtain a fasting glucose reading or perform an OGTT around the time of the postpartum checkup for any patient who was diagnosed with GDM. ACOG recommends using an OGTT to more accurately diagnose type 2 diabetes or prediabetes in these patients, who are at significantly elevated risk.³⁰

Acknowledgement

The authors wish to thank Carol Hildebrandt, a research assistant with no potential conflict of interest, for her help with this manuscript.

Correspondence
Julienne K. Kirk, PharmD, CDE, Department of Family and Community Medicine, Wake Forest University School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1084; jkirk@wfubmc.edu

The youngest Americans—those born in the year 2000 or thereafter—have more than a 1 in 3 lifetime risk of developing diabetes, according to the Centers for Disease Control and Prevention.¹ That estimate, coupled with the fact that more than 2 out of 3 adults and 1 in 6 children between the ages of 2 and 19 years are overweight,² would seem to indicate a need for widespread diabetes screening. But limited health care resources, a lack of evidence that mass screening improves outcomes, and differences among leading medical associations about whom and when to screen argue against it.

At the same time, widespread obesity is making the presentation of hyperglycemia more complex and the forms of diabetes harder to classify. Many cases don’t follow the classic patterns, in which type 1 (formerly called juvenile diabetes) virtually always emerges in childhood and type 2 (previously known as adult-onset diabetes) is strictly an adult disease. Our evolving understanding of diabetes has led researchers to focus on prediabetes (defined as impaired fasting glucose, impaired glucose tolerance, or both) and latent autoimmune diabetes in adults (LADA), a recently reported type 1 variant that some have labeled type 1.5.³

In the face of growing complexity, the US Preventive Services Task Force (USPSTF) last year upgraded its recommendation for screening for type 2 diabetes, and researchers have developed new risk calculation tools. We’ve taken a look at the changing clinical landscape and sorted through the latest evidence to help you make sense of the latest risk and diagnostic developments in diabetes care.

Screening for type 2: A look at guidelines

Type 2 diabetes accounts for approximately 90% of the cases you’ll see.^1,4,5 The American Diabetes Association (ADA) calls for routine screening, starting at 45 years of age and continuing every 3 years thereafter in the absence of risk. But for those who are overweight or obese and have 1 or more additional risk factors, screening is recommended at any age.⁵ In addition to a body mass index (BMI) ≥25, risks include physical inactivity, a first-degree relative with type 2 diabetes, blood pressure >135/80 mm Hg (or controlled with an antihypertensive), high-density lipoproteins (HDL) <35 mg/dL, triglycerides >250 mg/dL, polycystic ovary syndrome, impaired glucose tolerance or impaired fasting glucose, and acanthosis nigricans, a pigmented thickening of the skin folds of the neck (TABLE).^6,7 Patients with the metabolic syndrome—abdominal obesity (defined as a waist circumference of >40” in men and >35” in women) and ≥2 of the following: raised triglyceride levels, elevated blood pressure, elevated fasting plasma glucose, and reduced HDL cholesterol—are at especially high risk of both cardiovascular disease and type 2 diabetes.⁸

The ADA screening recommendations, however, are not based on prospective outcome studies, nor are they widely followed. Until recently, the USPSTF only recommended screening adults with hypertension and hyperlipidemia.

In 2008, after an assessment of new findings and research updates, the USPSTF revised its recommendation: The task force now calls for screening asymptomatic adults with sustained blood pressure >135/80 mm Hg—regardless of lipid profile.⁷ For patients with diabetes and hypertension, the USPSTF concluded, evidence shows that early intervention—including lowering blood pressure below conventional targets—can prevent long-term adverse outcomes of diabetes and reduce the risk of cardiovascular events.

Although mass screening remains controversial, regular assessment of risk factors and targeting individuals with established risk is clearly indicated (PATIENT HANDOUT). The importance of early detection was highlighted by the United Kingdom Prospective Diabetes Study, in which approximately half of the patients with newly diagnosed type 2 diabetes already had evidence of complications.⁹

TABLE
Type 1, type 2, and gestational diabetes: Diagnostic clues

Validated risk calculators can boost detection rates

In an attempt to improve detection rates of type 2 diabetes and prediabetes, researchers in both the United States and the United Kingdom recently developed easy-to-use risk calculation tools. The Diabetes Risk Calculator (available at http://www.diabetes.org/food-nutrition-lifestyle/lifestyle-prevention/risk-test.jsp), published in 2008, was validated with findings from the Third National Health and Nutrition Survey.¹⁰ The calculator uses answers to questions about age, waist circumference, history of gestational diabetes mellitus (GDM), height, race/ethnicity, hypertension, family history, and exercise to determine whether an individual is at high risk for undetected diabetes. The tool has a low positive predictive value (14%), but a negative predictive value >99%.¹⁰

The QDScore Diabetes Risk Calculator (www.qdscore.org), another new tool, is designed to estimate an individual’s 10-year risk of developing type 2 diabetes.¹¹ The program, which calculates risk based on answers to questions about family history of diabetes, patient history of cardiovascular disease, smoking, treatment for hypertension, BMI, ethnicity, and steroid use, was validated with data collected from 2.5 million patients in practices throughout England and Wales. The screening tool showed a high degree of discrimination in reflecting differences in disease prevalence related to ethnic and socioeconomic risk factors.¹¹

Pinning down a type 2 (or prediabetes) diagnosis

The ADA, American Association of Clinical Endocrinologists (AACE), USPSTF, and World Health Organization/International Diabetes Federation agree on the diagnostic criteria for type 2 diabetes: a fasting glucose >126 mg/dL, a random plasma glucose ≥200 mg/dL (that must be confirmed on a subsequent day), or both.^5,7,12,13 Patient history, risk factors, and additional laboratory tests can help clinicians distinguish between type 1 and type 2 diabetes.

An oral glucose tolerance test (OGTT) is also an option for diagnosis, but time and scheduling difficulties limit the routine use of this test in primary care. Hemoglobin A1c is not recommended as a diagnostic test because of a lack of standardization.¹

Prediabetes and type 2 risk. One in 4 (25.9%) US adults 20 years of age or older and more than 1 in 3 (35.9%) of those 60 years of age or older have prediabetes,¹⁴ defined as impaired fasting glucose (100-125 mg/dL), impaired glucose tolerance (2-hour glucose test results of 140-199 mg/dL), or both. Prediabetes increases the risk of developing type 2 diabetes by an estimated 30% over a 4-year period,¹⁵ and 70% over 30 years,¹⁶ although lifestyle interventions can substantially lower the risk. In a recently released consensus statement, an AACE task force noted that in addition to the increased risk of type 2 diabetes, patients with prediabetes face a greater risk of macrovascular complications.¹⁷

Type 2 in kids can be mistaken for type 1

As childhood obesity has surged, type 2 diabetes has been diagnosed at an increasingly early age—even in children younger than 10 years.¹⁸ Minority youth, primarily African Americans, Hispanics, and Asians/Pacific Islanders, are at increased risk.¹⁴ Symptoms can be insidious in children and adolescents and easily missed or mistaken for type 1 diabetes, in part because type 2 diabetes is still relatively rare in this age group.¹⁹

Preteens at risk. In a recent study of BMI and metabolic syndrome risk factors in 8- to 14-year-olds, however, researchers concluded that children who are overweight in early adolescence may be at risk for type 2 diabetes as well as cardiovascular disease before they reach their teens.²⁰ There is evidence of a genetic predisposition for type 2 diabetes and defects of β-cell function,^5,21 and family history, in addition to weight, is an important consideration in identifying type 2 diabetes in young patients.

Although young adults with type 1 and type 2 diabetes can present with similar symptoms, there may be certain clues to a type 2 diagnosis. Acanthosis nigricans, which is related to insulin resistance and occurs most frequently in obese adolescents, points to a type 2 diagnosis. Increased insulin and C-peptide levels are indicators of type 2 diabetes. Low levels are not necessarily an indication of type 1, however, because patients with type 2 diabetes may have low levels of insulin and C-peptide because of glucose toxicity and lipotoxicity at the time of diagnosis.²² Treatment with insulin may be necessary until glucose toxicity resolves.

Type 1 diabetes: Beyond childhood

Approximately 5% to 10% of patients with diabetes have type 1, which is defined as idiopathic or cellular immune-mediated autoimmune β-cell destruction.⁵ The rate of destruction is variable—it generally progresses more rapidly in infants and children than in adults. Some people with type 1 diabetes retain residual β-cell function, but have little or no insulin secretion; this manifests as a low or undetectable level of serum C-peptide.

Most cases of type 1 diabetes are diagnosed in patients younger than 18 years. But type 1 diabetes is increasingly recognized as a disorder that also develops in early adulthood, usually before the age of 40.

Arriving at a type 1 diagnosis

Patients with type 1 diabetes often present with modest hyperglycemia, but may rapidly progress to severe hyperglycemia and diabetic ketoacidosis (DKA) when infection or other physical stressors occur.

While screening for autoantibodies in asymptomatic individuals is not recommended,⁵ patients with blood glucose levels ≥200 mg/dL and symptoms of polydipsia, polyuria, and polyphagia who do not meet the profile for type 2 diabetes may be candidates for additional laboratory work. Approximately 85% to 90% of patients with type 1 diabetes will have antibodies to islet cells or glutamic acid decarboxylase (GAD).^5,23

Even without antibody testing, there are distinguishing characteristics that help support a type 1 diagnosis. As a general rule, individuals who develop type 1 diabetes—especially children—are not obese, although patients usually gain weight over time. In addition, many patients with type 1 diabetes have an auto-immune disease, such as celiac or Graves’ disease, hypothyroidism, adrenal anemia, or pernicious anemia; and a first-degree relative with type 1 diabetes. DKA, with acute symptoms of polydipsia and/or polyuria and recent, unintentional weight loss, is suggestive of—but not definitive for—type 1 diabetes.

A recently validated type 1 risk calculator may be particularly useful for screening patients who have a sibling, parent, or child with type 1 diabetes. Using age, BMI, C-peptide concentration, and OGTT results, the algorithm was highly predictive of type 1 diabetes in family members of patients who tested positive for islet cell antibodies.²⁴

Patient doesn’t “fit” type 1 or 2? Consider LADA

LADA, a gradual, progressive form of type 1 diabetes, can be difficult to identify. Circulating GAD or islet cell antibodies are present, but patients don’t have an absolute need for insulin at the time of diagnosis. Thus, they’re often thought to have type 2 diabetes.²⁵ Individuals with LADA show no signs of insulin resistance, however, and over time, β cells decline and insulin usually becomes necessary.

There are no universal recommendations for testing for LADA. Rather, the diagnosis should be considered in those who don’t fit the classic profile for type 1 or type 2 diabetes,²⁶ but have some of the following features:

• age <50 years
• acute symptoms of polydipsia, polyuria, and/or unintentional weight loss
• BMI <25
• a personal history of autoimmune disease
• a family history of autoimmune disease.²⁷

A prospective analysis found that the majority of LADA patients had at least 2 of these distinguishing characteristics.²⁸ Other recent research found heterogeneity among patients with LADA. Noting that not all patients with LADA become insulin-dependent, researchers concluded that the need for insulin is linked to the degree of autoimmunity and β-cell failure.²⁹

When GDM complicates prenatal care

Any degree of carbohydrate intolerance that is first recognized during pregnancy is classified as GDM, whether or not the condition resolves after delivery. A GDM diagnosis does not preclude the possibility of undiagnosed type 2 diabetes or prediabetes, or (rarely) type 1 diabetes.

Approximately 7% of all pregnancies in the United States are complicated by GDM, totaling more than 200,000 cases annually.⁵ The rate of GDM is in direct proportion to the prevalence of type 2 diabetes in the population in question, and ranges from 1% to 14%. GDM is the diagnosis in nearly 90% of pregnancies complicated by diabetes.⁵

The GDM screening controversy

Screening for GDM—whether it should be done universally or selectively on the basis of risk factors—is highly controversial. The USPSTF maintains that there is insufficient evidence to recommend for or against screening women with no history of GDM. The American College of Obstetricians and Gynecologists (ACOG)³⁰ and ADA⁵ recommend selective screening based on patient history, clinical presentation, and, possibly, prior impaired glucose test results or other abnormal laboratory values. AACE calls for universal screening of pregnant women, starting at 20 weeks for high-risk individuals and between 24 and 28 weeks for those at low risk.¹²

Identifying patients at risk. Maternal age (>25 years), obesity (BMI ≥30), prior GDM or delivery of a large-forgestational-age infant, belonging to a high-risk ethnic group, glycosuria, history of glucose resistance or glucose tolerance, and a first-degree relative with diabetes (TABLE) are risk factors for GDM. Women at high risk—those who meet all or most of these criteria—should undergo early screening: at the first prenatal visit, according to ACOG;³⁰ upon confirmation of pregnancy (ADA);⁵ at 20 weeks’ gestation (AACE);¹² or between 24 and 28 weeks’ gestation (USPSTF).⁷ ADA and ACOG recommend a 2-stage approach, starting with a 50-g 1-hour OGTT and following up with a 100-g 3-hour OGTT if the first test results are not definitive.^5,30 Testing for patients at average risk—which includes any pregnant woman with even a single risk factor, such as being older than 25 years—should be done between 24 and 28 weeks’ gestation, according to ACOG and ADA; testing is not required for women who are <25 years, have a normal body weight, and no other risk factors.

GDM screening in primary care. Because most women fit the criteria for average or high risk,³¹ family physicians may find universal screening to be more practical than individual risk assessment. Universal screening is associated with favorable outcomes,³² but screening limited to those at high and average risk also has evidence to support it. In a study of 25,118 deliveries, only 4% of women with GDM were missed by the exclusion of low-risk patients.³³

In the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study, researchers tracked 25,505 women from 9 countries and found a continuous relationship between the risk of macrosomia and the rise in maternal glucose levels.³⁴ The impact on the developing fetus of varying degrees of glucose was studied after a 75-g 2-hour OGTT. The risk of macrosomia increased with fasting blood glucose >75 mg/dL, 1-hour glucose levels >105 mg/dL, and 2-hour glucose concentration >90 mg/dL.³⁵ The most compelling results for adverse effects were associated with fasting glucose levels, rather than glucose tolerance tests.

2 abnormal results needed for a GDM diagnosis

In the absence of unequivocal hyperglycemia, there are 2 diagnostic standards for GDM: The Carpenter-Coustan Conversion and the National Diabetes Data Group Conversion. The Carpenter-Coustan Conversion uses lower glucose values for fasting (≥95 mg/dL) and subsequent 1-, 2-, and 3-hour levels (≥180, 155, and 140 mg/dL, respectively) and is more widely used. But expert opinion also supports the National Diabetes Data Group Conversion criteria (fasting plasma glucose, ≥105 mg/dL; ≥190, 165, and 145 mg/dL for 1-, 2-, and 3-hour OGTT, respectively), and there are no data from clinical trials to prove the superiority of either standard.³⁰

Both sets of standards require 2 or more thresholds to be met or exceeded for a GDM diagnosis. Women with only 1 abnormal value should be monitored carefully, however, as they, too, may be at increased risk for macrosomia and other morbidities.³⁰

Postpartum follow-up. Obtain a fasting glucose reading or perform an OGTT around the time of the postpartum checkup for any patient who was diagnosed with GDM. ACOG recommends using an OGTT to more accurately diagnose type 2 diabetes or prediabetes in these patients, who are at significantly elevated risk.³⁰

Acknowledgement

The authors wish to thank Carol Hildebrandt, a research assistant with no potential conflict of interest, for her help with this manuscript.

Correspondence
Julienne K. Kirk, PharmD, CDE, Department of Family and Community Medicine, Wake Forest University School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1084; jkirk@wfubmc.edu

The youngest Americans—those born in the year 2000 or thereafter—have more than a 1 in 3 lifetime risk of developing diabetes, according to the Centers for Disease Control and Prevention.¹ That estimate, coupled with the fact that more than 2 out of 3 adults and 1 in 6 children between the ages of 2 and 19 years are overweight,² would seem to indicate a need for widespread diabetes screening. But limited health care resources, a lack of evidence that mass screening improves outcomes, and differences among leading medical associations about whom and when to screen argue against it.

At the same time, widespread obesity is making the presentation of hyperglycemia more complex and the forms of diabetes harder to classify. Many cases don’t follow the classic patterns, in which type 1 (formerly called juvenile diabetes) virtually always emerges in childhood and type 2 (previously known as adult-onset diabetes) is strictly an adult disease. Our evolving understanding of diabetes has led researchers to focus on prediabetes (defined as impaired fasting glucose, impaired glucose tolerance, or both) and latent autoimmune diabetes in adults (LADA), a recently reported type 1 variant that some have labeled type 1.5.³

In the face of growing complexity, the US Preventive Services Task Force (USPSTF) last year upgraded its recommendation for screening for type 2 diabetes, and researchers have developed new risk calculation tools. We’ve taken a look at the changing clinical landscape and sorted through the latest evidence to help you make sense of the latest risk and diagnostic developments in diabetes care.

Screening for type 2: A look at guidelines

Type 2 diabetes accounts for approximately 90% of the cases you’ll see.^1,4,5 The American Diabetes Association (ADA) calls for routine screening, starting at 45 years of age and continuing every 3 years thereafter in the absence of risk. But for those who are overweight or obese and have 1 or more additional risk factors, screening is recommended at any age.⁵ In addition to a body mass index (BMI) ≥25, risks include physical inactivity, a first-degree relative with type 2 diabetes, blood pressure >135/80 mm Hg (or controlled with an antihypertensive), high-density lipoproteins (HDL) <35 mg/dL, triglycerides >250 mg/dL, polycystic ovary syndrome, impaired glucose tolerance or impaired fasting glucose, and acanthosis nigricans, a pigmented thickening of the skin folds of the neck (TABLE).^6,7 Patients with the metabolic syndrome—abdominal obesity (defined as a waist circumference of >40” in men and >35” in women) and ≥2 of the following: raised triglyceride levels, elevated blood pressure, elevated fasting plasma glucose, and reduced HDL cholesterol—are at especially high risk of both cardiovascular disease and type 2 diabetes.⁸

The ADA screening recommendations, however, are not based on prospective outcome studies, nor are they widely followed. Until recently, the USPSTF only recommended screening adults with hypertension and hyperlipidemia.

In 2008, after an assessment of new findings and research updates, the USPSTF revised its recommendation: The task force now calls for screening asymptomatic adults with sustained blood pressure >135/80 mm Hg—regardless of lipid profile.⁷ For patients with diabetes and hypertension, the USPSTF concluded, evidence shows that early intervention—including lowering blood pressure below conventional targets—can prevent long-term adverse outcomes of diabetes and reduce the risk of cardiovascular events.

Although mass screening remains controversial, regular assessment of risk factors and targeting individuals with established risk is clearly indicated (PATIENT HANDOUT). The importance of early detection was highlighted by the United Kingdom Prospective Diabetes Study, in which approximately half of the patients with newly diagnosed type 2 diabetes already had evidence of complications.⁹

TABLE
Type 1, type 2, and gestational diabetes: Diagnostic clues

Validated risk calculators can boost detection rates

In an attempt to improve detection rates of type 2 diabetes and prediabetes, researchers in both the United States and the United Kingdom recently developed easy-to-use risk calculation tools. The Diabetes Risk Calculator (available at http://www.diabetes.org/food-nutrition-lifestyle/lifestyle-prevention/risk-test.jsp), published in 2008, was validated with findings from the Third National Health and Nutrition Survey.¹⁰ The calculator uses answers to questions about age, waist circumference, history of gestational diabetes mellitus (GDM), height, race/ethnicity, hypertension, family history, and exercise to determine whether an individual is at high risk for undetected diabetes. The tool has a low positive predictive value (14%), but a negative predictive value >99%.¹⁰

The QDScore Diabetes Risk Calculator (www.qdscore.org), another new tool, is designed to estimate an individual’s 10-year risk of developing type 2 diabetes.¹¹ The program, which calculates risk based on answers to questions about family history of diabetes, patient history of cardiovascular disease, smoking, treatment for hypertension, BMI, ethnicity, and steroid use, was validated with data collected from 2.5 million patients in practices throughout England and Wales. The screening tool showed a high degree of discrimination in reflecting differences in disease prevalence related to ethnic and socioeconomic risk factors.¹¹

Pinning down a type 2 (or prediabetes) diagnosis

The ADA, American Association of Clinical Endocrinologists (AACE), USPSTF, and World Health Organization/International Diabetes Federation agree on the diagnostic criteria for type 2 diabetes: a fasting glucose >126 mg/dL, a random plasma glucose ≥200 mg/dL (that must be confirmed on a subsequent day), or both.^5,7,12,13 Patient history, risk factors, and additional laboratory tests can help clinicians distinguish between type 1 and type 2 diabetes.

An oral glucose tolerance test (OGTT) is also an option for diagnosis, but time and scheduling difficulties limit the routine use of this test in primary care. Hemoglobin A1c is not recommended as a diagnostic test because of a lack of standardization.¹

Prediabetes and type 2 risk. One in 4 (25.9%) US adults 20 years of age or older and more than 1 in 3 (35.9%) of those 60 years of age or older have prediabetes,¹⁴ defined as impaired fasting glucose (100-125 mg/dL), impaired glucose tolerance (2-hour glucose test results of 140-199 mg/dL), or both. Prediabetes increases the risk of developing type 2 diabetes by an estimated 30% over a 4-year period,¹⁵ and 70% over 30 years,¹⁶ although lifestyle interventions can substantially lower the risk. In a recently released consensus statement, an AACE task force noted that in addition to the increased risk of type 2 diabetes, patients with prediabetes face a greater risk of macrovascular complications.¹⁷

Type 2 in kids can be mistaken for type 1

As childhood obesity has surged, type 2 diabetes has been diagnosed at an increasingly early age—even in children younger than 10 years.¹⁸ Minority youth, primarily African Americans, Hispanics, and Asians/Pacific Islanders, are at increased risk.¹⁴ Symptoms can be insidious in children and adolescents and easily missed or mistaken for type 1 diabetes, in part because type 2 diabetes is still relatively rare in this age group.¹⁹

Preteens at risk. In a recent study of BMI and metabolic syndrome risk factors in 8- to 14-year-olds, however, researchers concluded that children who are overweight in early adolescence may be at risk for type 2 diabetes as well as cardiovascular disease before they reach their teens.²⁰ There is evidence of a genetic predisposition for type 2 diabetes and defects of β-cell function,^5,21 and family history, in addition to weight, is an important consideration in identifying type 2 diabetes in young patients.

Although young adults with type 1 and type 2 diabetes can present with similar symptoms, there may be certain clues to a type 2 diagnosis. Acanthosis nigricans, which is related to insulin resistance and occurs most frequently in obese adolescents, points to a type 2 diagnosis. Increased insulin and C-peptide levels are indicators of type 2 diabetes. Low levels are not necessarily an indication of type 1, however, because patients with type 2 diabetes may have low levels of insulin and C-peptide because of glucose toxicity and lipotoxicity at the time of diagnosis.²² Treatment with insulin may be necessary until glucose toxicity resolves.

Type 1 diabetes: Beyond childhood

Approximately 5% to 10% of patients with diabetes have type 1, which is defined as idiopathic or cellular immune-mediated autoimmune β-cell destruction.⁵ The rate of destruction is variable—it generally progresses more rapidly in infants and children than in adults. Some people with type 1 diabetes retain residual β-cell function, but have little or no insulin secretion; this manifests as a low or undetectable level of serum C-peptide.

Most cases of type 1 diabetes are diagnosed in patients younger than 18 years. But type 1 diabetes is increasingly recognized as a disorder that also develops in early adulthood, usually before the age of 40.

Arriving at a type 1 diagnosis

Patients with type 1 diabetes often present with modest hyperglycemia, but may rapidly progress to severe hyperglycemia and diabetic ketoacidosis (DKA) when infection or other physical stressors occur.

While screening for autoantibodies in asymptomatic individuals is not recommended,⁵ patients with blood glucose levels ≥200 mg/dL and symptoms of polydipsia, polyuria, and polyphagia who do not meet the profile for type 2 diabetes may be candidates for additional laboratory work. Approximately 85% to 90% of patients with type 1 diabetes will have antibodies to islet cells or glutamic acid decarboxylase (GAD).^5,23

Even without antibody testing, there are distinguishing characteristics that help support a type 1 diagnosis. As a general rule, individuals who develop type 1 diabetes—especially children—are not obese, although patients usually gain weight over time. In addition, many patients with type 1 diabetes have an auto-immune disease, such as celiac or Graves’ disease, hypothyroidism, adrenal anemia, or pernicious anemia; and a first-degree relative with type 1 diabetes. DKA, with acute symptoms of polydipsia and/or polyuria and recent, unintentional weight loss, is suggestive of—but not definitive for—type 1 diabetes.

A recently validated type 1 risk calculator may be particularly useful for screening patients who have a sibling, parent, or child with type 1 diabetes. Using age, BMI, C-peptide concentration, and OGTT results, the algorithm was highly predictive of type 1 diabetes in family members of patients who tested positive for islet cell antibodies.²⁴

Patient doesn’t “fit” type 1 or 2? Consider LADA

LADA, a gradual, progressive form of type 1 diabetes, can be difficult to identify. Circulating GAD or islet cell antibodies are present, but patients don’t have an absolute need for insulin at the time of diagnosis. Thus, they’re often thought to have type 2 diabetes.²⁵ Individuals with LADA show no signs of insulin resistance, however, and over time, β cells decline and insulin usually becomes necessary.

There are no universal recommendations for testing for LADA. Rather, the diagnosis should be considered in those who don’t fit the classic profile for type 1 or type 2 diabetes,²⁶ but have some of the following features:

• age <50 years
• acute symptoms of polydipsia, polyuria, and/or unintentional weight loss
• BMI <25
• a personal history of autoimmune disease
• a family history of autoimmune disease.²⁷

A prospective analysis found that the majority of LADA patients had at least 2 of these distinguishing characteristics.²⁸ Other recent research found heterogeneity among patients with LADA. Noting that not all patients with LADA become insulin-dependent, researchers concluded that the need for insulin is linked to the degree of autoimmunity and β-cell failure.²⁹

When GDM complicates prenatal care

Any degree of carbohydrate intolerance that is first recognized during pregnancy is classified as GDM, whether or not the condition resolves after delivery. A GDM diagnosis does not preclude the possibility of undiagnosed type 2 diabetes or prediabetes, or (rarely) type 1 diabetes.

Approximately 7% of all pregnancies in the United States are complicated by GDM, totaling more than 200,000 cases annually.⁵ The rate of GDM is in direct proportion to the prevalence of type 2 diabetes in the population in question, and ranges from 1% to 14%. GDM is the diagnosis in nearly 90% of pregnancies complicated by diabetes.⁵

The GDM screening controversy

Screening for GDM—whether it should be done universally or selectively on the basis of risk factors—is highly controversial. The USPSTF maintains that there is insufficient evidence to recommend for or against screening women with no history of GDM. The American College of Obstetricians and Gynecologists (ACOG)³⁰ and ADA⁵ recommend selective screening based on patient history, clinical presentation, and, possibly, prior impaired glucose test results or other abnormal laboratory values. AACE calls for universal screening of pregnant women, starting at 20 weeks for high-risk individuals and between 24 and 28 weeks for those at low risk.¹²

Identifying patients at risk. Maternal age (>25 years), obesity (BMI ≥30), prior GDM or delivery of a large-forgestational-age infant, belonging to a high-risk ethnic group, glycosuria, history of glucose resistance or glucose tolerance, and a first-degree relative with diabetes (TABLE) are risk factors for GDM. Women at high risk—those who meet all or most of these criteria—should undergo early screening: at the first prenatal visit, according to ACOG;³⁰ upon confirmation of pregnancy (ADA);⁵ at 20 weeks’ gestation (AACE);¹² or between 24 and 28 weeks’ gestation (USPSTF).⁷ ADA and ACOG recommend a 2-stage approach, starting with a 50-g 1-hour OGTT and following up with a 100-g 3-hour OGTT if the first test results are not definitive.^5,30 Testing for patients at average risk—which includes any pregnant woman with even a single risk factor, such as being older than 25 years—should be done between 24 and 28 weeks’ gestation, according to ACOG and ADA; testing is not required for women who are <25 years, have a normal body weight, and no other risk factors.

GDM screening in primary care. Because most women fit the criteria for average or high risk,³¹ family physicians may find universal screening to be more practical than individual risk assessment. Universal screening is associated with favorable outcomes,³² but screening limited to those at high and average risk also has evidence to support it. In a study of 25,118 deliveries, only 4% of women with GDM were missed by the exclusion of low-risk patients.³³

In the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study, researchers tracked 25,505 women from 9 countries and found a continuous relationship between the risk of macrosomia and the rise in maternal glucose levels.³⁴ The impact on the developing fetus of varying degrees of glucose was studied after a 75-g 2-hour OGTT. The risk of macrosomia increased with fasting blood glucose >75 mg/dL, 1-hour glucose levels >105 mg/dL, and 2-hour glucose concentration >90 mg/dL.³⁵ The most compelling results for adverse effects were associated with fasting glucose levels, rather than glucose tolerance tests.

2 abnormal results needed for a GDM diagnosis

In the absence of unequivocal hyperglycemia, there are 2 diagnostic standards for GDM: The Carpenter-Coustan Conversion and the National Diabetes Data Group Conversion. The Carpenter-Coustan Conversion uses lower glucose values for fasting (≥95 mg/dL) and subsequent 1-, 2-, and 3-hour levels (≥180, 155, and 140 mg/dL, respectively) and is more widely used. But expert opinion also supports the National Diabetes Data Group Conversion criteria (fasting plasma glucose, ≥105 mg/dL; ≥190, 165, and 145 mg/dL for 1-, 2-, and 3-hour OGTT, respectively), and there are no data from clinical trials to prove the superiority of either standard.³⁰

Both sets of standards require 2 or more thresholds to be met or exceeded for a GDM diagnosis. Women with only 1 abnormal value should be monitored carefully, however, as they, too, may be at increased risk for macrosomia and other morbidities.³⁰

Postpartum follow-up. Obtain a fasting glucose reading or perform an OGTT around the time of the postpartum checkup for any patient who was diagnosed with GDM. ACOG recommends using an OGTT to more accurately diagnose type 2 diabetes or prediabetes in these patients, who are at significantly elevated risk.³⁰

Acknowledgement

The authors wish to thank Carol Hildebrandt, a research assistant with no potential conflict of interest, for her help with this manuscript.

Correspondence
Julienne K. Kirk, PharmD, CDE, Department of Family and Community Medicine, Wake Forest University School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1084; jkirk@wfubmc.edu

In 1998 an estimated 9.5 million people had a diagnosis of type 2 diabetes mellitus in the United States.¹ Tight control of blood glucose concentrations to near-normal levels has been shown to reduce the microvascular complications of type 2 diabetes without increasing major macrovascular complications.^2,3 However, control of blood glucose is complex and involves multiple organ systems. Patients with type 2 diabetes often do not achieve desirable glucose control despite the use of oral hypoglycemic agents or insulin. Insulin resistance—the diminished ability of insulin to exert its biological activity over a broad range of glucose levels—may contribute to the difficulty in controlling type 2 diabetes.^4,5

Thiazolidinediones represent a newer class of drugs that affect insulin resistance.^6,7 Troglitazone (Rezulin) is the first drug in this class to be approved by the US Food and Drug Administration (FDA) for the treatment of type 2 diabetes. Troglitazone can be used in combination with a sulfonylurea or insulin to improve glycemic control.^8-10 Troglitazone increases the responsiveness of insulin-dependent tissues through a mechanism thought to involve receptors that regulate the transcription of a number of insulin-responsive genes.^11,12 It increases insulin-dependent glucose disposal in skeletal muscle, enhancing the effects of circulating insulin.

Metformin (glucophage) lowers plasma glucose by decreasing hepatic glucose output through the inhibition of gluconeogenesis and by increasing peripheral glucose use by skeletal muscle. It is a biguanide that was introduced in Europe in 1957 and has been available in the United States since 1995.^13,14 Metformin is indicated in patients with type 2 diabetes as a monotherapy along with diet; it can also be used concomitantly with a sulfonylurea or insulin.^15,16 The efficacy of metformin on glycemic control has been demonstrated as a monotherapy and in combination with a sulfonylurea.^16,17

Although there are data to support the use of metformin or troglitazone in combination with a sulfonylurea,^8-10,14-16 randomized comparisons of the relative effects of these combinations on glycemic control are lacking. Information about the efficacy, safety, tolerability, and cost of these combination therapies may help in pharmacotherapy decision making. Our primary goal was to compare the effects of troglitazone with those of metformin on the Hb A1c levels of patients with type 2 diabetes who were already receiving sulfonylurea therapy. Secondary outcomes included the comparative effects of these combinations on fasting plasma glucose (FPG) and C-peptide levels. We also compared safety, tolerability, and cost of the 2 drugs.

Methods

Study Subjects

We studied 32 patients (20 men, 12 women) with type 2 diabetes who were already taking a sulfonylurea. We randomly screened individuals found in a database of family medicine patients who had been given a diagnosis of type 2 diabetes mellitus. Patients were eligible if they were aged 30 to 75 years, had poorly controlled diabetes defined by an Hb A1c level between 8.5% and 16% at the screening visit, and were able to give informed consent. We excluded women of childbearing potential. The other exclusion criteria were: a history or laboratory evidence of renal or hepatic insufficiency; a history of alcohol abuse (including binge drinking within the past year); concomitant treatment with insulin, cholestyramine, potentially nephrotoxic drugs, or glucocorticoids (except topical or inhaled glucocorticoids); plans for radiographic studies involving the use of intravenous iodinated contrast during the course of our study; and known intolerance or sensitivity to a biguanide or troglitazone. The protocol was approved by the institutional review board at Wake Forest University Baptist Medical Center.

Study Design

At baseline we randomized the patients to receive either metformin or troglitazone for a 14-week period. The study was divided into 2 phases: a 2-week dose-titration period and a 12-week open-label comparison of metformin and troglitazone. If randomized to metformin, the patient took 500 mg with the evening meal for 2 days, then 500 mg twice daily with the morning and evening meals for 5 days. During the second week of the study, the patient took 500 mg with the morning meal and 1000 mg with the evening meal. After week 2, all patients randomized to metformin therapy were taking 1000 mg with the morning and evening meals. Patients randomized to troglitazone took 200 mg daily with the evening meal for 2 weeks and then 400 mg daily for the remaining 12 weeks of the study. If at any time during the study a patient experienced a FPG of less than 80 mg/dL, the oral sulfonylurea was decreased by one half of the original dose and then discontinued if further blood glucose readings were less than 80 mg/dL on more than one occasion.

Patients were required to make a screening visit at least 1 week before entry into the trial. We obtained a past medical history, body weight and height, and blood tests (serum creatinine, serum bicarbonate, liver enzymes, Hb A1c, FPG, and C-peptide levels). We recorded body weight and repeated blood tests 6 to 8 weeks after randomization and at the end of the study period (14 weeks). In addition, participants randomized to troglitazone had monthly liver enzyme tests. We also instructed patients to perform home blood glucose monitoring twice daily, in the morning before breakfast and at bedtime. We validated blood glucose monitors for accuracy by checking control solutions and performing check strip tests at study visits. Patients received standardized information about diabetes from a certified diabetes educator consisting of a general overview of diabetes mellitus, a medication review, instructions for blood glucose monitoring, a review of complications associated with diabetes, and nutritional advice. Each patient was given the same written information about diabetes and counseled on the signs and symptoms of high and low blood glucose. No patient enrolled in this trial reported problems reading or understanding written instructions. We assessed compliance using pill counts during each scheduled follow-up visit. We asked patients about adverse events at each visit after beginning drug therapy; any reported events were recorded.

Statistical Analysis

We performed an analysis of efficacy by intention to treat. We included all patients who received at least one dose of troglitazone or metformin, and selected a sample size adequate for detecting clinically meaningful differences in treatment effects. A sample of 16 patients in each group allowed detection of a mean absolute difference in Hb A1c level reduction between groups from a baseline of 1.2% (±0.2%) with a power greater than 0.80 (a = 0.05, 2-tailed test). We evaluated baseline differences between treatment groups using analysis of variance and chi-square procedures. Paired t tests were used to examine changes in variables over time. Analyses were performed using the Statistical Package for the Social Sciences for Personal Computers (Version 8.0, SPSS, Inc, Chicago, Ill).

Results

The baseline demographic and disease-related characteristics of the participants are outlined in Table 1. There were no significant differences at baseline between treatment groups with respect to age; body mass index (BMI); Hb A1c, FPG, or C-peptide levels; or the duration of diabetes. Ninety-seven percent of the patients took their assigned medication for the 14 weeks of the study. All patients were receiving an oral sulfonylurea, with 85% taking glipizide (Glucotrol XL) 10 mg to 20 mg per day, 9% taking glimepiride (Amaryl) 4 to 8 mg per day, and 6% taking glyburide (generic or DiaBeta) 10 to 20 mg per day.

Table 2 contains the changes in glycemic control parameters observed in each treatment group. At the end of a 3-month treatment period, Hb A1c values decreased significantly for each group when compared with the values obtained at baseline. The mean Hb A1c level among those receiving metformin fell from 9.9% ±1.6 to 7.8% ±1.3 (P <.001). Among patients in the troglitazone treatment group, the mean Hb A1c level fell from 10% ±1.6 to 7.4% ±1.7 (P <.001). The mean FPG level fell from 229 mg/dL ±75 to 138 mg/dL ±36 (P <.001) in the patients receiving metformin and from 210 mg/dL ±79 to 127 mg/dL ±33 (P <.001), in those receiving troglitazone. For each treatment group this amounts to a 60% reduction in FPG levels. For those patients receiving metformin, fasting C-peptide levels fell from 6.9 ng/mL ±2.3 to 4.7 ng/mL ±1.6 (P <.001), and in troglitazone-treated patients, it fell from 6.5 ng/mL ±3.9 to 4.5 ng/mL ±2.3 (P = .004). Although both troglitazone and metformin significantly improved glycemic control, there was no significant difference between the 2 groups in any treatment effect measured. In addition, BMI did not significantly change from baseline to the end of the study in either treatment group.

Safety was assessed on the basis of the results of serum FPG levels, creatinine and liver function tests, and home glucose monitoring. There were no elevations in liver enzymes or serum creatinine in those individuals receiving either troglitazone or metformin. No patient reported hypoglycemic symptoms or blood glucose values of less than 70 mg/dL on more than one occasion. Tolerability was evaluated using a questionnaire of potential side effects that was answered during each study visit. There was one dropout from the trial in the metformin treatment arm, which was attributed to moderate nausea and diarrhea after 1 month of treatment. Six additional participants in the metformin group reported mild nausea and bloating in the first 2 weeks of treatment with metformin; however, no other adverse effects were ascribed to the study medications.

Discussion

There has been much interest in combined pharmacologic therapy for type 2 diabetes, especially when target Hb A1c levels are not achieved with monotherapy. The American Diabetes Association recommends that the goal of treatment in type 2 diabetes is an Hb A1c level of less than 7%, with additional action suggested at values greater than 8%.¹⁸ This small study addressed patients with type 2 diabetes who were already receiving treatment with moderate to maximum doses of a sulfonylurea. The patients in this study were obese (mean BMI = 33 kg/m²) and had uncontrolled diabetes as evidenced by baseline Hb A1c levels. Our results show that metformin and troglitazone had very similar efficacy for those patients in terms of reductions in Hb A1c, FPG, and C-peptide levels when used in combination with a sulfonylurea. Furthermore, safety and tolerability in terms of symptomatic adverse events, hypoglycemia, changes in serum creatinine, and changes in liver enzymes were good for both combinations during the 14 weeks of this clinical trial. Previous studies^9,10,14,16 have shown these combinations effective, but those studies did not directly compare them in a controlled fashion. Troglitazone and metformin are both approved by the FDA for combination therapy with a sulfonylurea. Although metformin is also FDA approved as a monotherapy, it is common practice to start with a sulfonylurea when type 2 diabetes is diagnosed.

The United Kingdom Prospective Diabetes Study (UKPDS) showed metformin to be beneficial as a monotherapy for obese patients with type 2 diabetes, but raised concern about combination sulfonylurea/metformin therapy.¹⁹ In the UKPDS, metformin was shown to reduce overall mortality in obese patients with serum creatinine levels less than 1.5 mg/dL, but increased mortality was associated with the addition of metformin to sulfonylurea therapy. The baseline differences between patients treated with metformin alone and those for whom metformin was added to a sulfonyurea (plus the small number of deaths overall) led the UKPDS investigators to question the validity of this observation. Special precautions are recommended when prescribing metformin,²⁰ mainly because of potential problems with severe lactic acidosis observed in the past with another biguanide (phenformin). Accordingly, metformin is contraindicated in congestive heart failure, in the presence of renal or hepatic insufficiency, during periods of hypoxemia or dehydration, and for heavy alcohol drinkers. It should also be withheld before, during, and after the administration of iodinated intravenous contrast. Tolerability of metformin can be problematic during the dose-titration phase for some patients because of gastrointestinal side effects, but adherence to treatment can be optimized by educating patients that this is usually a transient side effect.

Troglitazone has been associated with elevated hepatic enzymes in approximately 2% of patients in clinical trials; very rare severe or fatal hepatic dysfunction has also been reported.²¹ Accordingly, the manufacturer recommends periodic assay of serum alanine aminotransferase levels (baseline and monthly for the first year of therapy, then quarterly).²² Similar laboratory monitoring is not required for metformin.

Simplicity of a prescribed drug regimen is a consideration for patients, especially with regard to compliance. Metformin requires at least twice-daily administration; troglitazone can be taken once per day. Comparative drug costs are also important to consider. As of 1999, the average wholesale cost in the United States for a 1-month supply of the doses in our study is approximately $142 for troglitazone and $75 for metformin.²³ Newer thiazolidinediones, such as rosiglitazone and pioglitazone, will add competition and continued scrutiny to the efficacy, safety, and costs of this drug class.

Conclusions

Metformin and troglitazone improved glycemic profiles and C-peptide levels with equal success with no significant differences in safety or tolerability. Although our study had sufficient power to detect a clinically meaningful difference in Hb A1c reduction, it was based on a small number of patients and a short study duration. The true test of the effectiveness of these combination therapies must come from large clinical trials of sufficient duration that assess their effects on diabetes-related morbidity and mortality.

In 1998 an estimated 9.5 million people had a diagnosis of type 2 diabetes mellitus in the United States.¹ Tight control of blood glucose concentrations to near-normal levels has been shown to reduce the microvascular complications of type 2 diabetes without increasing major macrovascular complications.^2,3 However, control of blood glucose is complex and involves multiple organ systems. Patients with type 2 diabetes often do not achieve desirable glucose control despite the use of oral hypoglycemic agents or insulin. Insulin resistance—the diminished ability of insulin to exert its biological activity over a broad range of glucose levels—may contribute to the difficulty in controlling type 2 diabetes.^4,5

Thiazolidinediones represent a newer class of drugs that affect insulin resistance.^6,7 Troglitazone (Rezulin) is the first drug in this class to be approved by the US Food and Drug Administration (FDA) for the treatment of type 2 diabetes. Troglitazone can be used in combination with a sulfonylurea or insulin to improve glycemic control.^8-10 Troglitazone increases the responsiveness of insulin-dependent tissues through a mechanism thought to involve receptors that regulate the transcription of a number of insulin-responsive genes.^11,12 It increases insulin-dependent glucose disposal in skeletal muscle, enhancing the effects of circulating insulin.

Metformin (glucophage) lowers plasma glucose by decreasing hepatic glucose output through the inhibition of gluconeogenesis and by increasing peripheral glucose use by skeletal muscle. It is a biguanide that was introduced in Europe in 1957 and has been available in the United States since 1995.^13,14 Metformin is indicated in patients with type 2 diabetes as a monotherapy along with diet; it can also be used concomitantly with a sulfonylurea or insulin.^15,16 The efficacy of metformin on glycemic control has been demonstrated as a monotherapy and in combination with a sulfonylurea.^16,17

Although there are data to support the use of metformin or troglitazone in combination with a sulfonylurea,^8-10,14-16 randomized comparisons of the relative effects of these combinations on glycemic control are lacking. Information about the efficacy, safety, tolerability, and cost of these combination therapies may help in pharmacotherapy decision making. Our primary goal was to compare the effects of troglitazone with those of metformin on the Hb A1c levels of patients with type 2 diabetes who were already receiving sulfonylurea therapy. Secondary outcomes included the comparative effects of these combinations on fasting plasma glucose (FPG) and C-peptide levels. We also compared safety, tolerability, and cost of the 2 drugs.

Methods

Study Subjects

We studied 32 patients (20 men, 12 women) with type 2 diabetes who were already taking a sulfonylurea. We randomly screened individuals found in a database of family medicine patients who had been given a diagnosis of type 2 diabetes mellitus. Patients were eligible if they were aged 30 to 75 years, had poorly controlled diabetes defined by an Hb A1c level between 8.5% and 16% at the screening visit, and were able to give informed consent. We excluded women of childbearing potential. The other exclusion criteria were: a history or laboratory evidence of renal or hepatic insufficiency; a history of alcohol abuse (including binge drinking within the past year); concomitant treatment with insulin, cholestyramine, potentially nephrotoxic drugs, or glucocorticoids (except topical or inhaled glucocorticoids); plans for radiographic studies involving the use of intravenous iodinated contrast during the course of our study; and known intolerance or sensitivity to a biguanide or troglitazone. The protocol was approved by the institutional review board at Wake Forest University Baptist Medical Center.

Study Design

At baseline we randomized the patients to receive either metformin or troglitazone for a 14-week period. The study was divided into 2 phases: a 2-week dose-titration period and a 12-week open-label comparison of metformin and troglitazone. If randomized to metformin, the patient took 500 mg with the evening meal for 2 days, then 500 mg twice daily with the morning and evening meals for 5 days. During the second week of the study, the patient took 500 mg with the morning meal and 1000 mg with the evening meal. After week 2, all patients randomized to metformin therapy were taking 1000 mg with the morning and evening meals. Patients randomized to troglitazone took 200 mg daily with the evening meal for 2 weeks and then 400 mg daily for the remaining 12 weeks of the study. If at any time during the study a patient experienced a FPG of less than 80 mg/dL, the oral sulfonylurea was decreased by one half of the original dose and then discontinued if further blood glucose readings were less than 80 mg/dL on more than one occasion.

Patients were required to make a screening visit at least 1 week before entry into the trial. We obtained a past medical history, body weight and height, and blood tests (serum creatinine, serum bicarbonate, liver enzymes, Hb A1c, FPG, and C-peptide levels). We recorded body weight and repeated blood tests 6 to 8 weeks after randomization and at the end of the study period (14 weeks). In addition, participants randomized to troglitazone had monthly liver enzyme tests. We also instructed patients to perform home blood glucose monitoring twice daily, in the morning before breakfast and at bedtime. We validated blood glucose monitors for accuracy by checking control solutions and performing check strip tests at study visits. Patients received standardized information about diabetes from a certified diabetes educator consisting of a general overview of diabetes mellitus, a medication review, instructions for blood glucose monitoring, a review of complications associated with diabetes, and nutritional advice. Each patient was given the same written information about diabetes and counseled on the signs and symptoms of high and low blood glucose. No patient enrolled in this trial reported problems reading or understanding written instructions. We assessed compliance using pill counts during each scheduled follow-up visit. We asked patients about adverse events at each visit after beginning drug therapy; any reported events were recorded.

Statistical Analysis

We performed an analysis of efficacy by intention to treat. We included all patients who received at least one dose of troglitazone or metformin, and selected a sample size adequate for detecting clinically meaningful differences in treatment effects. A sample of 16 patients in each group allowed detection of a mean absolute difference in Hb A1c level reduction between groups from a baseline of 1.2% (±0.2%) with a power greater than 0.80 (a = 0.05, 2-tailed test). We evaluated baseline differences between treatment groups using analysis of variance and chi-square procedures. Paired t tests were used to examine changes in variables over time. Analyses were performed using the Statistical Package for the Social Sciences for Personal Computers (Version 8.0, SPSS, Inc, Chicago, Ill).

Results

The baseline demographic and disease-related characteristics of the participants are outlined in Table 1. There were no significant differences at baseline between treatment groups with respect to age; body mass index (BMI); Hb A1c, FPG, or C-peptide levels; or the duration of diabetes. Ninety-seven percent of the patients took their assigned medication for the 14 weeks of the study. All patients were receiving an oral sulfonylurea, with 85% taking glipizide (Glucotrol XL) 10 mg to 20 mg per day, 9% taking glimepiride (Amaryl) 4 to 8 mg per day, and 6% taking glyburide (generic or DiaBeta) 10 to 20 mg per day.

Table 2 contains the changes in glycemic control parameters observed in each treatment group. At the end of a 3-month treatment period, Hb A1c values decreased significantly for each group when compared with the values obtained at baseline. The mean Hb A1c level among those receiving metformin fell from 9.9% ±1.6 to 7.8% ±1.3 (P <.001). Among patients in the troglitazone treatment group, the mean Hb A1c level fell from 10% ±1.6 to 7.4% ±1.7 (P <.001). The mean FPG level fell from 229 mg/dL ±75 to 138 mg/dL ±36 (P <.001) in the patients receiving metformin and from 210 mg/dL ±79 to 127 mg/dL ±33 (P <.001), in those receiving troglitazone. For each treatment group this amounts to a 60% reduction in FPG levels. For those patients receiving metformin, fasting C-peptide levels fell from 6.9 ng/mL ±2.3 to 4.7 ng/mL ±1.6 (P <.001), and in troglitazone-treated patients, it fell from 6.5 ng/mL ±3.9 to 4.5 ng/mL ±2.3 (P = .004). Although both troglitazone and metformin significantly improved glycemic control, there was no significant difference between the 2 groups in any treatment effect measured. In addition, BMI did not significantly change from baseline to the end of the study in either treatment group.

Safety was assessed on the basis of the results of serum FPG levels, creatinine and liver function tests, and home glucose monitoring. There were no elevations in liver enzymes or serum creatinine in those individuals receiving either troglitazone or metformin. No patient reported hypoglycemic symptoms or blood glucose values of less than 70 mg/dL on more than one occasion. Tolerability was evaluated using a questionnaire of potential side effects that was answered during each study visit. There was one dropout from the trial in the metformin treatment arm, which was attributed to moderate nausea and diarrhea after 1 month of treatment. Six additional participants in the metformin group reported mild nausea and bloating in the first 2 weeks of treatment with metformin; however, no other adverse effects were ascribed to the study medications.

Discussion

There has been much interest in combined pharmacologic therapy for type 2 diabetes, especially when target Hb A1c levels are not achieved with monotherapy. The American Diabetes Association recommends that the goal of treatment in type 2 diabetes is an Hb A1c level of less than 7%, with additional action suggested at values greater than 8%.¹⁸ This small study addressed patients with type 2 diabetes who were already receiving treatment with moderate to maximum doses of a sulfonylurea. The patients in this study were obese (mean BMI = 33 kg/m²) and had uncontrolled diabetes as evidenced by baseline Hb A1c levels. Our results show that metformin and troglitazone had very similar efficacy for those patients in terms of reductions in Hb A1c, FPG, and C-peptide levels when used in combination with a sulfonylurea. Furthermore, safety and tolerability in terms of symptomatic adverse events, hypoglycemia, changes in serum creatinine, and changes in liver enzymes were good for both combinations during the 14 weeks of this clinical trial. Previous studies^9,10,14,16 have shown these combinations effective, but those studies did not directly compare them in a controlled fashion. Troglitazone and metformin are both approved by the FDA for combination therapy with a sulfonylurea. Although metformin is also FDA approved as a monotherapy, it is common practice to start with a sulfonylurea when type 2 diabetes is diagnosed.

The United Kingdom Prospective Diabetes Study (UKPDS) showed metformin to be beneficial as a monotherapy for obese patients with type 2 diabetes, but raised concern about combination sulfonylurea/metformin therapy.¹⁹ In the UKPDS, metformin was shown to reduce overall mortality in obese patients with serum creatinine levels less than 1.5 mg/dL, but increased mortality was associated with the addition of metformin to sulfonylurea therapy. The baseline differences between patients treated with metformin alone and those for whom metformin was added to a sulfonyurea (plus the small number of deaths overall) led the UKPDS investigators to question the validity of this observation. Special precautions are recommended when prescribing metformin,²⁰ mainly because of potential problems with severe lactic acidosis observed in the past with another biguanide (phenformin). Accordingly, metformin is contraindicated in congestive heart failure, in the presence of renal or hepatic insufficiency, during periods of hypoxemia or dehydration, and for heavy alcohol drinkers. It should also be withheld before, during, and after the administration of iodinated intravenous contrast. Tolerability of metformin can be problematic during the dose-titration phase for some patients because of gastrointestinal side effects, but adherence to treatment can be optimized by educating patients that this is usually a transient side effect.

Troglitazone has been associated with elevated hepatic enzymes in approximately 2% of patients in clinical trials; very rare severe or fatal hepatic dysfunction has also been reported.²¹ Accordingly, the manufacturer recommends periodic assay of serum alanine aminotransferase levels (baseline and monthly for the first year of therapy, then quarterly).²² Similar laboratory monitoring is not required for metformin.

Simplicity of a prescribed drug regimen is a consideration for patients, especially with regard to compliance. Metformin requires at least twice-daily administration; troglitazone can be taken once per day. Comparative drug costs are also important to consider. As of 1999, the average wholesale cost in the United States for a 1-month supply of the doses in our study is approximately $142 for troglitazone and $75 for metformin.²³ Newer thiazolidinediones, such as rosiglitazone and pioglitazone, will add competition and continued scrutiny to the efficacy, safety, and costs of this drug class.

Conclusions

Metformin and troglitazone improved glycemic profiles and C-peptide levels with equal success with no significant differences in safety or tolerability. Although our study had sufficient power to detect a clinically meaningful difference in Hb A1c reduction, it was based on a small number of patients and a short study duration. The true test of the effectiveness of these combination therapies must come from large clinical trials of sufficient duration that assess their effects on diabetes-related morbidity and mortality.

In 1998 an estimated 9.5 million people had a diagnosis of type 2 diabetes mellitus in the United States.¹ Tight control of blood glucose concentrations to near-normal levels has been shown to reduce the microvascular complications of type 2 diabetes without increasing major macrovascular complications.^2,3 However, control of blood glucose is complex and involves multiple organ systems. Patients with type 2 diabetes often do not achieve desirable glucose control despite the use of oral hypoglycemic agents or insulin. Insulin resistance—the diminished ability of insulin to exert its biological activity over a broad range of glucose levels—may contribute to the difficulty in controlling type 2 diabetes.^4,5

Thiazolidinediones represent a newer class of drugs that affect insulin resistance.^6,7 Troglitazone (Rezulin) is the first drug in this class to be approved by the US Food and Drug Administration (FDA) for the treatment of type 2 diabetes. Troglitazone can be used in combination with a sulfonylurea or insulin to improve glycemic control.^8-10 Troglitazone increases the responsiveness of insulin-dependent tissues through a mechanism thought to involve receptors that regulate the transcription of a number of insulin-responsive genes.^11,12 It increases insulin-dependent glucose disposal in skeletal muscle, enhancing the effects of circulating insulin.

Metformin (glucophage) lowers plasma glucose by decreasing hepatic glucose output through the inhibition of gluconeogenesis and by increasing peripheral glucose use by skeletal muscle. It is a biguanide that was introduced in Europe in 1957 and has been available in the United States since 1995.^13,14 Metformin is indicated in patients with type 2 diabetes as a monotherapy along with diet; it can also be used concomitantly with a sulfonylurea or insulin.^15,16 The efficacy of metformin on glycemic control has been demonstrated as a monotherapy and in combination with a sulfonylurea.^16,17

Although there are data to support the use of metformin or troglitazone in combination with a sulfonylurea,^8-10,14-16 randomized comparisons of the relative effects of these combinations on glycemic control are lacking. Information about the efficacy, safety, tolerability, and cost of these combination therapies may help in pharmacotherapy decision making. Our primary goal was to compare the effects of troglitazone with those of metformin on the Hb A1c levels of patients with type 2 diabetes who were already receiving sulfonylurea therapy. Secondary outcomes included the comparative effects of these combinations on fasting plasma glucose (FPG) and C-peptide levels. We also compared safety, tolerability, and cost of the 2 drugs.

Methods

Study Subjects

We studied 32 patients (20 men, 12 women) with type 2 diabetes who were already taking a sulfonylurea. We randomly screened individuals found in a database of family medicine patients who had been given a diagnosis of type 2 diabetes mellitus. Patients were eligible if they were aged 30 to 75 years, had poorly controlled diabetes defined by an Hb A1c level between 8.5% and 16% at the screening visit, and were able to give informed consent. We excluded women of childbearing potential. The other exclusion criteria were: a history or laboratory evidence of renal or hepatic insufficiency; a history of alcohol abuse (including binge drinking within the past year); concomitant treatment with insulin, cholestyramine, potentially nephrotoxic drugs, or glucocorticoids (except topical or inhaled glucocorticoids); plans for radiographic studies involving the use of intravenous iodinated contrast during the course of our study; and known intolerance or sensitivity to a biguanide or troglitazone. The protocol was approved by the institutional review board at Wake Forest University Baptist Medical Center.

Study Design

At baseline we randomized the patients to receive either metformin or troglitazone for a 14-week period. The study was divided into 2 phases: a 2-week dose-titration period and a 12-week open-label comparison of metformin and troglitazone. If randomized to metformin, the patient took 500 mg with the evening meal for 2 days, then 500 mg twice daily with the morning and evening meals for 5 days. During the second week of the study, the patient took 500 mg with the morning meal and 1000 mg with the evening meal. After week 2, all patients randomized to metformin therapy were taking 1000 mg with the morning and evening meals. Patients randomized to troglitazone took 200 mg daily with the evening meal for 2 weeks and then 400 mg daily for the remaining 12 weeks of the study. If at any time during the study a patient experienced a FPG of less than 80 mg/dL, the oral sulfonylurea was decreased by one half of the original dose and then discontinued if further blood glucose readings were less than 80 mg/dL on more than one occasion.

Patients were required to make a screening visit at least 1 week before entry into the trial. We obtained a past medical history, body weight and height, and blood tests (serum creatinine, serum bicarbonate, liver enzymes, Hb A1c, FPG, and C-peptide levels). We recorded body weight and repeated blood tests 6 to 8 weeks after randomization and at the end of the study period (14 weeks). In addition, participants randomized to troglitazone had monthly liver enzyme tests. We also instructed patients to perform home blood glucose monitoring twice daily, in the morning before breakfast and at bedtime. We validated blood glucose monitors for accuracy by checking control solutions and performing check strip tests at study visits. Patients received standardized information about diabetes from a certified diabetes educator consisting of a general overview of diabetes mellitus, a medication review, instructions for blood glucose monitoring, a review of complications associated with diabetes, and nutritional advice. Each patient was given the same written information about diabetes and counseled on the signs and symptoms of high and low blood glucose. No patient enrolled in this trial reported problems reading or understanding written instructions. We assessed compliance using pill counts during each scheduled follow-up visit. We asked patients about adverse events at each visit after beginning drug therapy; any reported events were recorded.

Statistical Analysis

We performed an analysis of efficacy by intention to treat. We included all patients who received at least one dose of troglitazone or metformin, and selected a sample size adequate for detecting clinically meaningful differences in treatment effects. A sample of 16 patients in each group allowed detection of a mean absolute difference in Hb A1c level reduction between groups from a baseline of 1.2% (±0.2%) with a power greater than 0.80 (a = 0.05, 2-tailed test). We evaluated baseline differences between treatment groups using analysis of variance and chi-square procedures. Paired t tests were used to examine changes in variables over time. Analyses were performed using the Statistical Package for the Social Sciences for Personal Computers (Version 8.0, SPSS, Inc, Chicago, Ill).

Results

The baseline demographic and disease-related characteristics of the participants are outlined in Table 1. There were no significant differences at baseline between treatment groups with respect to age; body mass index (BMI); Hb A1c, FPG, or C-peptide levels; or the duration of diabetes. Ninety-seven percent of the patients took their assigned medication for the 14 weeks of the study. All patients were receiving an oral sulfonylurea, with 85% taking glipizide (Glucotrol XL) 10 mg to 20 mg per day, 9% taking glimepiride (Amaryl) 4 to 8 mg per day, and 6% taking glyburide (generic or DiaBeta) 10 to 20 mg per day.

Table 2 contains the changes in glycemic control parameters observed in each treatment group. At the end of a 3-month treatment period, Hb A1c values decreased significantly for each group when compared with the values obtained at baseline. The mean Hb A1c level among those receiving metformin fell from 9.9% ±1.6 to 7.8% ±1.3 (P <.001). Among patients in the troglitazone treatment group, the mean Hb A1c level fell from 10% ±1.6 to 7.4% ±1.7 (P <.001). The mean FPG level fell from 229 mg/dL ±75 to 138 mg/dL ±36 (P <.001) in the patients receiving metformin and from 210 mg/dL ±79 to 127 mg/dL ±33 (P <.001), in those receiving troglitazone. For each treatment group this amounts to a 60% reduction in FPG levels. For those patients receiving metformin, fasting C-peptide levels fell from 6.9 ng/mL ±2.3 to 4.7 ng/mL ±1.6 (P <.001), and in troglitazone-treated patients, it fell from 6.5 ng/mL ±3.9 to 4.5 ng/mL ±2.3 (P = .004). Although both troglitazone and metformin significantly improved glycemic control, there was no significant difference between the 2 groups in any treatment effect measured. In addition, BMI did not significantly change from baseline to the end of the study in either treatment group.

Safety was assessed on the basis of the results of serum FPG levels, creatinine and liver function tests, and home glucose monitoring. There were no elevations in liver enzymes or serum creatinine in those individuals receiving either troglitazone or metformin. No patient reported hypoglycemic symptoms or blood glucose values of less than 70 mg/dL on more than one occasion. Tolerability was evaluated using a questionnaire of potential side effects that was answered during each study visit. There was one dropout from the trial in the metformin treatment arm, which was attributed to moderate nausea and diarrhea after 1 month of treatment. Six additional participants in the metformin group reported mild nausea and bloating in the first 2 weeks of treatment with metformin; however, no other adverse effects were ascribed to the study medications.

Discussion

There has been much interest in combined pharmacologic therapy for type 2 diabetes, especially when target Hb A1c levels are not achieved with monotherapy. The American Diabetes Association recommends that the goal of treatment in type 2 diabetes is an Hb A1c level of less than 7%, with additional action suggested at values greater than 8%.¹⁸ This small study addressed patients with type 2 diabetes who were already receiving treatment with moderate to maximum doses of a sulfonylurea. The patients in this study were obese (mean BMI = 33 kg/m²) and had uncontrolled diabetes as evidenced by baseline Hb A1c levels. Our results show that metformin and troglitazone had very similar efficacy for those patients in terms of reductions in Hb A1c, FPG, and C-peptide levels when used in combination with a sulfonylurea. Furthermore, safety and tolerability in terms of symptomatic adverse events, hypoglycemia, changes in serum creatinine, and changes in liver enzymes were good for both combinations during the 14 weeks of this clinical trial. Previous studies^9,10,14,16 have shown these combinations effective, but those studies did not directly compare them in a controlled fashion. Troglitazone and metformin are both approved by the FDA for combination therapy with a sulfonylurea. Although metformin is also FDA approved as a monotherapy, it is common practice to start with a sulfonylurea when type 2 diabetes is diagnosed.

The United Kingdom Prospective Diabetes Study (UKPDS) showed metformin to be beneficial as a monotherapy for obese patients with type 2 diabetes, but raised concern about combination sulfonylurea/metformin therapy.¹⁹ In the UKPDS, metformin was shown to reduce overall mortality in obese patients with serum creatinine levels less than 1.5 mg/dL, but increased mortality was associated with the addition of metformin to sulfonylurea therapy. The baseline differences between patients treated with metformin alone and those for whom metformin was added to a sulfonyurea (plus the small number of deaths overall) led the UKPDS investigators to question the validity of this observation. Special precautions are recommended when prescribing metformin,²⁰ mainly because of potential problems with severe lactic acidosis observed in the past with another biguanide (phenformin). Accordingly, metformin is contraindicated in congestive heart failure, in the presence of renal or hepatic insufficiency, during periods of hypoxemia or dehydration, and for heavy alcohol drinkers. It should also be withheld before, during, and after the administration of iodinated intravenous contrast. Tolerability of metformin can be problematic during the dose-titration phase for some patients because of gastrointestinal side effects, but adherence to treatment can be optimized by educating patients that this is usually a transient side effect.

Troglitazone has been associated with elevated hepatic enzymes in approximately 2% of patients in clinical trials; very rare severe or fatal hepatic dysfunction has also been reported.²¹ Accordingly, the manufacturer recommends periodic assay of serum alanine aminotransferase levels (baseline and monthly for the first year of therapy, then quarterly).²² Similar laboratory monitoring is not required for metformin.

Simplicity of a prescribed drug regimen is a consideration for patients, especially with regard to compliance. Metformin requires at least twice-daily administration; troglitazone can be taken once per day. Comparative drug costs are also important to consider. As of 1999, the average wholesale cost in the United States for a 1-month supply of the doses in our study is approximately $142 for troglitazone and $75 for metformin.²³ Newer thiazolidinediones, such as rosiglitazone and pioglitazone, will add competition and continued scrutiny to the efficacy, safety, and costs of this drug class.

Conclusions

Metformin and troglitazone improved glycemic profiles and C-peptide levels with equal success with no significant differences in safety or tolerability. Although our study had sufficient power to detect a clinically meaningful difference in Hb A1c reduction, it was based on a small number of patients and a short study duration. The true test of the effectiveness of these combination therapies must come from large clinical trials of sufficient duration that assess their effects on diabetes-related morbidity and mortality.