Joint associations of obesity and estimated GFR with clinical outcomes: a population-based cohort study

BACKGROUND: Despite the interrelationships between obesity, eGFR and albuminuria, few large studies examine how obesity modifies the association between these markers of kidney function and adverse clinical outcomes. METHODS: We examined the joint associations between obesity, eGFR and albuminuria on four clinical outcomes (death, end-stage renal disease [ESRD], myocardial infarction [MI], and placement in a long-term care facility) using a population-based cohort with procedures from Alberta. Obesity was defined by body mass index ≥35 kg/m(2) as defined by a fee modifier applied to an eligible procedure. RESULTS: We studied 1,293,362 participants, of whom 171,650 (13.3%) had documented obesity (BMI ≥ 35 kg/m(2) based on claims data) and 1,121,712 (86.7%) did not. The association between eGFR and death was J-shaped for participants with and without documented obesity. After full adjustment, obesity tended to be associated with slightly lower odds of mortality (OR range 0.71–1.02; p for interaction between obesity and eGFR 0.008). For participants with and without obesity, the adjusted odds of ESRD were lowest for participants with eGFR > 90 mL/min*1.73m(2) and increased with lower eGFR, with no evidence of an interaction with obesity (p = 0.37). Although albuminuria and obesity were both associated with higher odds of ESRD, the excess risk associated with obesity was substantially attenuated at higher levels of albuminuria (p for interaction 0.0006). The excess risk of MI associated with obesity was observed at eGFR > 60 mL/min*1.73m(2) but not at lower eGFR (p for interaction < 0.0001). Participants with obesity had a higher adjusted likelihood of placement in long-term care than those without, and the likelihood of such placement was higher at lower eGFR for those with and without obesity (p for interaction = 0.57). CONCLUSIONS: We found significant interactions between obesity and eGFR and/or albuminuria on the likelihood of adverse outcomes including death and ESRD. Since obesity is common, risk prediction tools for people with CKD might be improved by adding information on BMI or other proxies for body size in addition to eGFR and albuminuria. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1186/s12882-019-1351-9) contains supplementary material, which is available to authorized users.