The renal effects of nonsteroidal anti-inflammatory drugs in older people:Findings from the established populations for epidemiologic studies of theelderly

OBJECTIVE: To determine whether older people who use nonsteroidal anti-infl ammatory agents (NSAIDs) have increased levels of blood urea nitrogen (BUN) , serum creatinine, and BUN:serum creatinine ratio. DESIGN: Cross-sectional, secondary data analysis. SETTING: Older people living in the communities of East Boston, MA, New Hav en, CT, and Washington and Iowa Counties, Iowa. PARTICIPANTS: A total of 4099 people aged 70 years or older who were partic ipants in the National Institute on Aging's Established Populations for Epi demiologic Studies of the Elderly project, had survived to the 6-year follo w-up interview and had consented to the blood drawing. MEASUREMENTS: We assessed use of the NSAIDs at the 3- and h-year interviews through a drug inventory and visual review of medication containers. Marke rs of renal function assessed through analysis of blood samples drawn at th e time of the interview included BUN and creatinine. RESULTS: Fifteen percent of the cohort reported use of NSAIDs during the 2 weeks preceding the 6-year interview. Controlling for age, sex, and a range of potential confounding variables, NSAID users had significant prevalence odds ratios of 1.9 (95% confidence interval (CI), 1.5-2.3) for being in th e highest quartile of BUN (>23), 1.3 (CI 1.1-1.7) for the highest quartile of serum creatinine (greater than or equal to 1.4), and 1.7 (CI 1.4-2.1) fo r the highest quartile of the BUN:creatinine ratio (greater than or equal t o 19.4). Chronic NSAID users (those who reported NSAID use at both the 3-ye ar and 6-year interviews) accounted for the increased risk of high serum cr eatinine levels. CONCLUSION: Community-dwelling older people who use NSAIDs tend to have hig her levels of common laboratory markers of renal dysfunction. This hypothes is requires further testing in prospective cohort studies designed a priori to evaluate these issues.