After adjusting for eGFR, the association between
sESAM quartile and the composite primary endpoint was attenuated
(Table 4). Results were similar when the creatinine-derived
MDRD equation was used to determine eGFR (adjusted IRR 1.09
(95% CI 0.8e1.48), p ¼ 0.5773). As a continuous variable, each SD
increase in sESAM (18.0 ng/mL) was associated with a 46% increase
in the rate of the composite endpoint (IRR 1.46, 95% CI 1.3e1.65,
p < 0.0001), and the association persisted after adjustment for
demographics and clinical factors (IRR 1.27, 95% CI 1.11e1.46). This
association was no longer apparent after adjusting for eGFR (IRR
1.01, 95% CI 0.85e1.2) (Supplementary Table 1).