#2879 EFFECT OF THE RATE OF ALBUMIN DECREASE ON LONG-TERM PROGNOSIS IN DIALYSIS PATIENTS

Abstract

Abstract Background and Aims The malnutrition is often prevalent in hemodialysis (HD) patients, and the risk of mortality is strongly correlated with malnutrition. We assume the dynamic change and trend of albumin in HD patients are mortality essential, however there is not enough information on these data. The aim of this study was to investigate the association of long-term prognosis and decrease rate of albumin for one-year. Method We enrolled HD patients in six centers to a retrospective follow-up study. All patients had received HD from January 2014 to December 2014 and serum albumin data were collected every month in this period. We defined decrease rate in serum albumin as (average of albumin value for the year - decrease value for the year) / average of albumin value for the year. We categorized these patients into two groups according to the median value of the decrease rate. Then these patients were observed for three years. The primary and secondary outcome of this study was all-cause mortality and the combined endpoint of mortality and major cardio-cerebrovascular adverse events (MACCE). To evaluate cut-off value in decrease rate of serum albumin, Receiver Operating Characteristic (ROC) curve analysis was performed. Moreover, restricted cubic splines were used to detect the possible nonlinear dependency of the relationship between decrease rate in serum albumin and mortality. Results Six-hundred twenty-one HD patients were enrolled into this study. The median rate of decrease in albumin in the enrolled patients was 7.5%. Compared to the low decrease fluctuation group(n = 321), the high decrease fluctuation group (n = 300) was significantly older (67 vs 66 years; p = 0.03), more frequently had diabetes mellitus (51.7 vs 39.3%; p = 0.02), had lower serum creatinine (9.34 vs 10.69 mg/dl; p <0.001) and lower body mass index (20.3 vs 21.3; p = 0.01). The mean follow-up period was 31.0±10.1 months. During the follow-up period, there were 121 cases (19.5%) in all cause death and 191 cases (30.5%) in combined endpoint, respectively. Kaplan-Meier analysis showed that high decrease fluctuation in serum albumin group had a significantly worse prognosis in both all-cause mortality (Log-rank, p = 0.01) and combined endpoint (Log-rank, p = 0.002) than low decrease fluctuation group (Figure 1). Multi-variate Cox proportional hazard model revealed that high decrease fluctuation in serum albumin was a significantly associated with a higher risk of all-cause mortality and combined endpoint (hazard ratio 1.47 (95%CI: 1.02-2.13), p = 0.03, hazard ratio 1.49 (95%CI:1.11-1.99), p = 0.007). The cut-off value for all-cause mortality in decrease rate of serum albumin was 8.4% by ROC curve analysis. Moreover, through multivariable restricted cubic spline regression, continuous variation in decrease rate in serum albumin was found to be related to all-cause mortality in a nonlinear manner (Figure 2). Conclusion In HD patients, high decrease fluctuation in serum albumin was significantly associated with an increased risk of all-cause mortality compared low decrease fluctuation in serum albumin. Our findings suggest that regarding serum albumin, we might need to pay attention not only to its monthly value, but also to its decreasing fluctuation.