SO048CREATININE INDEX AND IT'S ASSOCIATION WITH MORTALITY AND HOSPITALISATION AMONG INCIDENT HAEMODIALYSIS PATIENTS

Abstract

Abstract Background and Aims Studies have suggested that higher lean body mass (LBM) is associated with increased longevity among haemodialysis (HD) patients. It has been shown that creatinine index is a surrogate marker of lean body mass and can be estimated using a simplified equation based on age, gender, serum creatinine and KT/V(sp). The present study was conducted to investigate the association between estimated creatinine index and all-cause mortality in a large racially diverse cohort of incident HD patients. We also examined the relationship between creatinine index and hospitalisation incidence in this cohort. Method Incident patients (aged≥18 years) between January, 2010 and December, 2018 who survived six months of HD were included in this retrospective observational study. Baseline demographic data was collected at study commencement, followed by clinical and laboratory data during the 6 month exposure period. Patients were followed from the index date (last day of exposure period) until first of the following – death, withdrawal or end of study (June 30, 2019). The exposure of interest was Creatinine Index (CI) estimated using a simplified formula (Canaud et al 2014) based on patient demographics (age and gender), pre-dialysis serum creatinine and KT/V(sp). Patients were stratified into 4 groups based on CI quartiles - Q1(<=17.48), Q2(17.49-18.91), Q3(18.92-20.54) and Q4(20.55+). Primary outcome was all-cause mortality during the overall follow-up. Adjusted hazard ratios were estimated using a Cox regression model for association between CI quartiles and mortality. Negative binomial regression models were used to assess the relationship between CI quartiles and hospitalisation. Results A total of 3172 incident HD patients (mean age 60.4±11.5 years, 56.2% male, 57.4% Chinese and 70.4% with diabetes as cause of ESRD) were included in the study. During the median follow-up of 2.8 (1.4-4.7) years, 755 (23.8%) patients died. The mean creatinine index for entire cohort was 19.1±2.3 mg/kg/day. The patients in the lowest CI quartile (Q1) were older, females, had higher incidence of diabetes and comorbidity but lower levels of albumin, haemoglobin, BMI and SGA scores compared to higher quartiles (p<0.001). Following multivariate analysis the adjusted hazard of death were as follows - Q4 (aHR-1, reference), Q1 (aHR-2.14, 95%CI 1.56-2.94), Q2 (aHR-1.58, 95%CI 1.21-2.05), Q3 (aHR-1.33, 95%CI 1.05-1.68), p<0.001. Patients in the lowest quartile of CI (Q1) had 64% higher incidence of hospitalisation (IRR 1.64, 95%CI 1.36–1.98, p<0.001) compared with patients in Q4 (IRR-1, reference). Conclusion Creatinine index derived from a simplified equation is useful in predicting mortality and hospitalisation among Southeast Asian incident HD patients. Low CI was associated with increased risk of all-cause mortality and hospitalisation in our cohort. The present study is limited by the lack of data on residual renal function which might confound the results.