β2-microglobulin, not muscle mass, accounts for the association between eGFR discordance and mortality in CKD.

Introduction The discordance between cystatin C- and creatinine-based eGFR (eGFRdiff) predicts mortality in CKD, but what this index actually reflects remains unclear. We examined the relative contributions of several biomarkers, representing muscle mass, middle-molecule retention, cardiac injury, inflammation, and physical function, to the eGFRdiff-mortality association in a nationally representative US population. Materials and Methods We analyzed 1503 adults with CKD stages 1-4 from NHANES 1999-2004 with mortality follow-up through 2019. The exposure was percentage eGFRdiff. Survey-weighted Cox models estimated HRs for mortality. Attenuation analyses quantified the contribution of each candidate biomarker. Residualization analyses decomposed the B2M signal by adjusting for creatinine-based, cystatin C-based, and combined eGFR estimates separately. Results 911 all-cause and 313 cardiovascular deaths occurred. Each 1-SD more negative eGFRdiff was associated with higher all-cause (HR 1.29; 95%CI 1.15-1.46) and cardiovascular mortality (HR 1.46; 95%CI 1.23-1.72). DXA-derived muscle mass yielded negligible attenuation (1.3%), and physical limitation produced modest attenuation (11.2%). In contrast, B2M produced the largest attenuation (73.4%), followed by hs-cTnT (32.0%) and NT-proBNP (23.0%). Residualization on eGFRcr preserved the B2M attenuation (80.1%), whereas residualization on eGFRcys abolished it (-7.9%). B2M residualized on eGFRcr-cys retained 21.9% attenuation. Conclusions Middle-molecule retention (represented by B2M), not muscle mass, is the primary contributor to the prognostic value of eGFRdiff, with cardiac biomarkers contributing secondarily. Residualization patterns support a biological signal consistent with shrunken pore syndrome rather than a mathematical artifact of how eGFR is estimated. Calculable from routine assays, eGFRdiff may offer a practical indicator of middle-molecule burden and cardiac stress for CKD risk stratification.