Automated peritoneal dialysis in Hong Kong: there are two distinct groups of patients

To compare the clinical outcome between continuous ambulatory peritoneal dialysis (CAPD) and automated peritoneal dialysis (APD) in specific subgroups of patients.

We reviewed the clinical outcome of 90 consecutive incident APD patients and 180 CAPD patients in our centre.

The median follow up was 21.9 months (inter-quartile range, 9.5 to 46.5 months). The APD group was younger and had a lower Charlson's score than the CAPD group. Furthermore, the APD group had a highly skewed distribution of the Charlson's score, indicating the possibility of two different groups of patients. Multivariate analysis showed that in addition to the treatment mode (APD vs CAPD) and Charlson's score, there was a significant interaction between the two (P = 0.043) on patient survival. For patients with Charlson's score ≤6, the APD group had a significantly better patient survival than the CAPD group (78.3% vs. 65.4% at 5 years, P = 0.039), while for patients with Charlson's score ≥7, the APD group had a worse patient survival than the CAPD group (16.3% vs. 48.4% at 5 years, P = 0.028). Similarly, Charlson's score and its interaction with treatment mode, but not the APD group per se, were independent predictors of technique survival (P = 0.013). For patients with Charlson's score ≥7, the APD group had a significantly lower technique survival than the CAPD group (8.8% vs. 34.3%, P = 0.001), while for patients with Charlson's score ≤6, the technique survival was similar (44.4% vs. 42.5%, P = 0.15). Peritonitis-free survival was 35.2% and 32.2% for APD and CAPD groups, respectively (P = 0.021), and the difference was not affected by Charlson's score.

Comorbid diseases had a significant interaction with the mode of PD on patient and technique survival of incident PD patients. Our result suggests that APD may offer benefit in, and only in, young patients with minimal comorbid diseases.