BACKGROUND Although chronic peritoneal dialysis (CPD) is considered the replacement therapy of choice for infants with end-stage renal failure, many questions persist about treatment risks and outcomes. METHODS We present data on 84 infants who started CPD at <1 year of age; these patients represent 12% of the total population of the Italian Registry of Paediatric Chronic Dialysis. We analysed patient records from all children consecutively treated with CPD between 1995 and 2007 in Italy. Growth data analysis was performed only in infants with complete auxological parameters at 0, 6 and 12 months of follow-up. RESULTS Median age at the start of CPD was 6.9 months, weight was 6.1 kg and length 63.6 cm. In one-half of the study population diagnosis leading to renal failure was congenital nephrouropathy. Twenty-eight per cent of the children had at least one pre-existing comorbidity. The mean height standard deviation score was -1.65 at the start of CPD, -1.82 after 12 months and -1.53 after 24 months. Catch-up growth was documented in 50% of patients during dialysis. A positive correlation was observed between longitudinal growth and both exchange volume (R(2) = 0.36) and dialysis session length (R(2) = 0.35), while a negative association was found with the number of peritonitis cases (P = 0.003). Peritonitis incidence was 1:20.7 episode:CPD-months (1:28.3 in the older children from the same registry) and was significantly higher in children with oligoanuria (1:15.5 episode:CPD-months) compared to infants with residual renal function (1:37.4 episode:CPD-months). Catheter survival rate was 70% at 12 months and 51% at 24 months. Catheter-related complications were similar in infants and older children (1:20.5 versus 1:19.8 episode:CPD-months), while clinical complications were more frequent in children under 1 year of age (1:18.3 versus 1:25.2 episode:CPD-months; P < 0.05). During the follow-up period, 33 patients were transplanted (39.3%), 18 were shifted to haemodialysis (21.4%) and 8 died (9.5%). The mortality rate was 4-fold greater than in older children (2.3%). CONCLUSIONS Our data confirm that infants on CPD represent a high-risk group; however, our experience demonstrated that growth was acceptable and a large portion was successfully transplanted. Increased efforts should be aimed at optimizing dialysis efficiency and preventing peritonitis. The higher mortality rate in infants was largely caused by comorbidities.