Comparison of cystatin C and creatinine to determine the incidence of composite adverse outcomes in HIV-infected individuals

Background Cystatin C is an overall biomarker of pathophysiologic abnormalities that accompany chronic kidney disease (CKD). The utility of cystatin C is not fully understood in an HIV-infected population. Methods This prospective study investigated 661 HIV-infected individuals for 4 years to determine the incidence of adverse outcomes, including all-cause mortality, cardiovascular disease, and renal dysfunction. The risk of developing the outcomes was discriminated with a 4 color-coded classification in a 3 × 6 contingency table, that combined 3 grades of dipstick proteinuria with 6 grades of estimated glomerular filtration rate (eGFR) calculated using either serum creatinine (eGFRcr) or cystatin C (eGFRcy): green, low risk; yellow, moderately increased risk; orange, high risk; and red, very high risk. The cumulative incidence of the outcomes was assessed by the Kaplan–Meier method, and the association between color-coded risk and the time to outcome was evaluated using multivariate proportional hazards analysis. Results Compared with eGFRcr, the use of eGFRcy reduced the prevalence of risk ≥orange by 0.8%. The adverse outcomes were significantly more likely to occur to the patients with baseline risk category ≥orange than those with ≤yellow, independent of risk categories based on eGFRcr or eGFRcy. However, in multivariate analysis, risk category ≥orange with eGFRcy-based classification was significantly associated with adverse outcomes, but not the one with eGFRcr. Conclusions Replacing creatinine by cystatin C in the CKD color-coded risk classification may be appropriate to discriminate HIV-infected patients at increased risk of a poor prognosis.