#2969 One-year Longitudinal Changes in Nutritional Parameters Associated with Low-Protein Diets in Patients Starting with Incremental PD

Abstract Background and Aims Starting peritoneal dialysis (PD) with an incremental approach, that is, a low dose of PD (1-2 dwells a day in continuous ambulatory PD and until 5 sessions a week in automated PD) based on residual kidney function, is a common strategy to increase the acceptance of dialysis for patients affected by advanced chronic kidney disease (CKD) requiring renal replacement therapy. However, no study has provided evidence on the optimal protein intake in patients treated with a low dose of PD. Indeed, patients starting dialysis require an increased protein intake to supply protein loss by dialysis. On the other hand, patients beginning with a low dose of dialysis and a high protein intake may lead to toxins accumulation and metabolic acidosis. Therefore, the maintenance of protein restriction could be indicated as long as malnutrition does not occur. Method The I-COPE study was designed to assess in 24 PD units in Italy the time to full-dose PD in patients undergoing integrated conservative therapy with low-dose PD. Patients who reached full-dose, switched to hemodialysis, received transplant, or died were censored. No specific recommendation on protein intake was included in the study protocol due to the lack of guidelines on this issue, so the prescription was left to decision of each participating center. This is an ad-interim analysis of the I-COPE study aimed at evaluating the longitudinal changes at one year of BMI and serum albumin. Two groups of nutritional interventions were identified a posteriori: non-low protein diet (NLPD, dietary protein intake > 0.6 g/Kg/day) vs low protein diet (LPD, dietary protein intake ≤ 0.6 g/Kg/day) on two visits after baseline (months 6 and 12). We used a mixed linear regression model to evaluate changes in variables over time in the two diet groups, assuming an unstructured covariance matrix; this was done to consider the correlation between repeated measures and missing points for patients not completing the analysis. Results The I-COPE study included 222 patients (age 63.6 ± 14.2 ys, males 65.3%, diabetes 29.2%). LPD was prescribed in 46.3%, while NLPD in 53.7%. At baseline, no intergroup difference was reported for age (P = 0.51), gender (P = 0.94), and diabetes (P = 0.11). 150 patients had at least an evaluation after baseline. The mixed regression model showed no difference in BMI (P = 0.668) and serum albumin (P = 0.297) adjusted for age, gender, diabetes, and GFR at the start of dialysis. As depicted in Fig. 1, serum albumin level was maintained at 12 months in either group. Conclusion We provide first-time observation that LPD may be safely associated with stable serum albumin and BMI over one year of incremental PD treatment. Therefore, in patients with advanced CKD requiring renal replacement therapy, LPD combined with a low dose of PD may provide a suitable nutritional approach to reduce waste metabolic products without increasing the risk of malnutrition.