Effect of the centre volume on time on peritoneal dialysis therapy: A cause-specific analysis with the retrospective...

BackgroundThis study was carried out to evaluate the association between the peritoneal dialysis (PD) centre volume and the risk of transfer to haemodialysis (HD).MethodsThis retrospective analysis used the French Language Peritoneal Dialysis Registry data of incident adult patients between 2002 and 2018. Centre volume was defined by the number of patients starting PD by centre and by year; PD centres were separated in 2 groups with a cut-off value of 10 new patients by year and by centre based on the graphical representation of the regression spline. A mixed Cox model was used to estimate the role of the centre volume on the centre effect. A Cox model was performed to assess the association between the centre volume and the cause-specific risk of transfer to HD.ResultsOf the 15653 patients included, 4293 (27%) started PD in a centre with a volume greater than 10. The transfer to HD heterogeneity between centre was reduced by the centre volume (variance reduction of 13%). In the multivariable analysis, the centre volume was associated with lower risk of transfer to HD (cause-specific hazard ratio (cs-HR): 0.86, 95% confidence interval (CI): [0.80-0.92]), but not with the risk of death (cs-HR: 1.06 [95% CI: 1.00-1.13]) or kidney transplantation (cs-HR: 0.97 [95% CI: 0.90-1.06]). Centre volume was associated with lower risk of transfer to HD due to small solute clearance or inadequate ultrafiltration (cs-HR = 0.80 [95% CI: 0.72-0.89]), but was not associated with transfer to HD by peritonitis (cs-HR: 0.97 [95% CI: 0.82-1.16]), or catheter-related issues (cs-HR: 0.85 [95% CI: 0.67-1.09]).ConclusionsThe centre volume reduced the risk of transfer to HD due to small solute clearance or inadequate ultrafiltration. At the centre level, quality improvement programme should focus on PD prescription. Networking between centres may improve the outcome on PD.