Renal disease from habitual antipyretic analgesic consumption: an assessment of the epidemiologic evidence.

The quantitative evidence relating habitual antipyretic analgesic consumption to the development of renal disease is reviewed. For purposes of analysis, "abuse" of analgesics is defined as regular, usually daily consumption. The prevalence of habitual consumption demonstrates marked geographic variation for poorly defined reasons. Prevalence tended to be highest in those populations in which phenacetin was available and popular. The prevalence of nephropathy among habitual consumers also shows marked interpopulation variability. A highly significant linear relationship exists between the prevalence of habitual analgesic consumption in a given population and the prevalence of nephropathy in the subset of that population that habitually consumes analgesics. This relationship may represent a dose-response curve relating amount of analgesic intake to prevalence of nephropathy. Four cross-sectional studies, 1 longitudinal, and 1 case-control study have shown significant differences in prevalence of nephropathy between habitual users of phenacetin-containing analgesics and control populations. Conversely, 2 cross-sectional studies and 1 case control study showed no difference between habitual users of analgesics and control populations. However, those studies showing no difference were performed in populations with low prevalence of habitual consumption of compound analgesics and a lower prevalence of nephropathy than was found in those populations where differences were observed. The relative risk for developing various manifestations of analgesic nephropathy has been estimated in 4 studies and varies depending on which abnormality is considered. Data are available only for subjects consuming phenacetin-containing compound analgesics. The relative risk for elevated serum creatinine is in the range of 8 to 11, about twice that for an abnormal renal concentrating test (approximately 4). The relative risk for clinical papillary necrosis is approximately 18, and for cause-specific death (and by inference for end-stage renal disease) about 4. The difference in risk between most clinical manifestations of analgesic nephropathy and death (or end-stage renal disease) may be due to the fact that most patients with nephropathy do not progress to terminal renal failure. Rheumatology clinic studies indicate that the prevalence of nephropathy in habitual consumers of phenacetin-containing compounds is higher than that for habitual consumers of aspirin alone. Removal of phenacetin from compound analgesics in Scandinavian countries appears to have reduced the prevalence of papillary necrosis and death from interstitial nephritis.(ABSTRACT TRUNCATED AT 400 WORDS)