Hospital-Volume Effects on Perioperative Outcomes in Peritoneal Dialysis Catheter Implantation: Analysis of 2,505 Cases.

BACKGROUND

Evidence regarding volume-outcome effects on peritoneal dialysis (PD) catheter implantation is limited. This study aimed to investigate associations between hospital volume (annual caseload of catheter implantation) and perioperative outcomes.

METHODS

Clinical data for patients who underwent PD catheter implantation from 2007 to 2012 were extracted from the Japanese nationwide Diagnosis Procedure Combination database. Hospital volume was divided into tertiles: low-volume (1 - 6 cases/year), medium-volume (7 - 13 cases/year), and high-volume (≥ 14 cases/year). Multivariate logistic regression analysis for the occurrence of any adverse events and blood transfusion, and gamma-distributed log-linked linear regression analysis for postoperative length of stay were conducted with explanatory variables of hospital volume, age, sex, Charlson comorbidity index, history of hemodialysis, type of anesthesia, and type of hospital.

RESULTS

Among 906, 855, and 744 cases in the low-volume, medium-volume, and high-volume groups, overall adverse events were 10.0%, 7.6%, and 6.0%, transfusion rates were 1.3%, 1.1%, and 0.9%, and median postoperative stays were 12, 10, and 9 days, respectively. In multivariate analyses, compared with the low-volume group, medium-volume and high-volume groups were associated with a lower incidence of overall adverse events (odds ratio [OR] = 0.71, p = 0.058, and OR = 0.59, p = 0.013, respectively) and shorter postoperative stay (% difference = -10.5%, p = 0.023, and % difference = -18.5%, p = 0.001, respectively), while no significant association was detected for transfusion.

CONCLUSIONS

Less frequent adverse events and shorter stays were observed in higher-volume centers. Inverse volume-outcome relationships in PD catheter implantation were confirmed.