The Problem
A 97 year-old independently living female presents with her family because she is having increasing difficulty getting around at home. She has a history of a permanent pacemaker for “cardiac conduction disease,” atrial fibrillation, coronary artery disease, and hypothyroidism. She describes bilateral knee pain moderately severe in intensity, in her left leg greater than her right. She has nocturnal pain that wakes her when she is rolling over in bed. The pain is worse with standing and better with sitting. She denies swelling, redness, or a history of gout. On examination, she has no swelling redness or warmth on either knee, but she has joint line tenderness bilaterally. You review an old knee x-ray that demonstrates medial compartment narrowing bilaterally. You make the diagnosis of osteoarthritis and proceed with conservative measures including topical diclofenac gel, lidocaine patches, stretching exercises, and a prescription for walking as tolerated. She returns several weeks later complaining of ongoing pain somewhat addressed with conservative measures. You elect to do a steroid injection of her left knee, which improves her pain. She calls back 6 weeks later and requests pain medication. She tells you that she had shoulder surgery in the 1990s at which time she received propoxyphene/acetaminophen that has since been removed from the market. She requests a substitute. You consider a narcotic. You decide to review the evidence.
The Question
Compared with NSAIDs, are narcotics safer for older patients and associated with fewer side effects?
The Search
You open PubMed, and enter the terms “analgesics AND arthritis.”
Our Critique
This pharmacoepidemiologic study provides new information for the care of our elderly patients. Limitations of this study include no information on quality of life and no information on BMI, alcohol or tobacco use, or use of over-the-counter medications.
Clinical Decision
You prescribe a coxib (a cyclo-oxygenase-2 inhibitor) and recommend a proton pump inhibitor. In the meantime, the family has moved her to assisted living.
The Evidence
“The Comparative Safety of Analgesics in Older Adults With Arthritis” (Arch. Intern. Med. 2010;170:1968-76).
P Study Cohort: The eligible cohort consisted of Medicare beneficiaries in Pennsylvania and New Jersey who qualified for pharmaceutical assistance programs based on income. Insurance coverage was provided for all medications without restriction. The study cohort consisted of the eligible adults who had recorded diagnoses for osteoarthritis or rheumatoid arthritis on two separate occasions. At the time of the second diagnosis, their first new analgesic prescription was recorded. New use of a nonselective NSAID, coxib, or an opioid was determined by excluding persons who received these drugs in the 180 days before the index date. Eligible subjects also were excluded if they had cancer, used hospice in the proceeding year, or received analgesics from two categories at the same time, either as combination products or as two separate medications. Propensity score matching was used and attempts were made to balance exposure groups upon potential confounders. Potential confounders included a history of medication use related to the diagnoses of cardiovascular disease, osteoporosis, bone fracture, and gastrointestinal, liver, and renal disease.
P Outcome Safety Events: Safety events were defined as all clinically significant on intent health effects related to analgesics. Cardiovascular events included MI, stroke, heart failure, revascularization, and out-of-hospital cardiac death. Gastrointestinal events included upper and lower GI tract bleeding and bowel obstruction. Acute kidney injury included hospitalizations for acute renal failure requiring dialysis. Liver toxic effects included inpatient and outpatient events. Fractures included hip, pelvis, wrist, and humerus but not spine. Trauma diagnoses denoting a fall also were identified.
P Analgesic Exposure: Exposure was categorized based on analgesic class: coxib, opioid, or nonselective NSAID. Medication use came from pharmacy dispensing claims records. Subjects were allowed to enter the analysis only once.
P Results: The final cohort consisted of 12,840 subjects. Almost 85% of subjects were women with a mean age of 80.0 years. Most (80%) had osteoarthritis. Comorbidities were similar across groups. Mean follow-up time was 117 days for NSAIDs, 202 days for coxibs, and 137 days for opioids. Compared with nonselective NSAIDs, more CV events were observed with coxibs (hazard ratio 1.28; 95% confidence interval, 1.01-1.62) and opioids (HR, 1.77; CI, 1.39-2.24). Compared with nonselective NSAID users, a similar risk for GI bleeding was observed for opioid users (HR, 1.07; CI, 0.65-1.76), but a lower risk was observed for coxib users (HR, 0.60; 95% CI, 0.35-1.00). Coxib and nonselective NSAID users had a similar risk for fracture. Fracture risk was elevated with opioid use (HR, 4.47; 95% CI, 3.12-6.41), compared with nonselective NSAID use. Compared with use of nonselective NSAIDs, an increased risk of safety events requiring hospitalization was observed with opioid use (HR, 1.68; 95% CI, 1.37-2.07), but not coxibs (HR, 1.12; CI, 0.91-1.38). Compared with nonselective NSAID users, an increased risk of all-cause mortality was observed among opioid users (HR, 1.87; 95 CI, 1.39-2.53), but not coxib users (HR, 1.16; CI, 0.85-1.57).