Bioimpedance analysis is often routinely performed in any dialysis unit to guide fluid management but can provide a reproduceable assessment of fat and muscle mass. We wished to determine the clinical significance of low muscle or high fat mass and the determinants that influence their change. We performed retrospective analysis of 824 patients on peritoneal dialysis who underwent routine repeated bioimpedance analysis measurements using the body composition monitor (BCM). Lean tissue index (LTI) was an independent predictor of mortality when sex, age, PD vintage and diabetes status were included in the models (HR 0.93; 95% CI 0.86-1.00, p < 0.05) and when baseline serum albumin was included in a separate model (HR 0.86; 95% CI: 0.79-0.93, p < 0.001). High fat tissue index (FTI) was an independent predictor of mortality when demographic factors were included (HR 0.87; 95% CI: 0.78-0.97, p < 0.02), but not with the addition biochemical parameters. Changes in body composition of 206 patients over a 2-year follow-up period could not be predicted by baseline demographics, functional or biochemical assessments. However, there was a strong inverse relationship between changes in LTI and FTI. There were no associations between changes in body composition with prescribed dialysate glucose. We showed body composition changes are common and complex. LTI was an independent predictor of survival. Changes in LTI and FTI could not be predicted by baseline parameters. BCM may be a sensitive and accurate tool to monitor changes in body composition during dialysis treatment.