Effect of Hemodiafiltration and Hemodialysis on Postoperative Mortality in Patients With Kidney Failure

BACKGROUND AND HYPOTHESIS: Patients receiving chronic dialysis have increased risks of postoperative mortality and morbidity. Studies have suggested that patients receving hemodiafiltration (HDF) compared to hemodialysis (HD) may benefit from improved cardiovascular health, reduced infections and better overall survival. This study aimed to determine if patients receving HDF had improved postoperative outcomes after elective surgery compared to patients receiving HD. METHODS: Bi-national population cohort study evaluated all incident and prevalent patients receiving chronic hemodialysis in Australia and New Zealand. Surgeries included upper gastrointestinal, anorectal, hernia surgery, cholecystectomy and appendectomy. The primary outcome was 30-day all-cause mortality. Secondary outcomes included cardiovascular and infective outcomes. As there was potential for data to be correlated with multiple surgeries clustered at the patient and centre level, a multi-level mixed effects logistic regression model was used, with patient level covariates as fixed effects and dialysis centre as random intercepts, whereby patients were nested within centres. RESULTS: During the study period, there were 1750 surgeries in 1669 patients, of which 416 (23.8%) were receiving HDF. The median age was similar in both groups, as was the prevalence of comorbidities. Patients receiving HDF had received dialysis for a shorter period (0.9 vs. 2.3 years, p < 0.001). The 30-day mortality rate was lower in patients receiving HDF than in those receiving HD (1.9 vs. 3.3%, p < 0.005). Using multi-level mixed effects logistic regression, HDF was not associated with reduced mortality odds (OR 0.84, 95% CI 0.37-1.93) compared to patients receiving HD. Patients on HDF experienced a lower rate of a composite of infections (5.5 vs. 9.8%, p = 0.012), and required less blood transfusions (7.7 vs. 12.4%, p = 0.009). The occurrence of cardiovascular complications was not significantly different between the two groups (1.4 vs. 2.3%, p = 0.280). CONCLUSIONS: The use of HDF was not associated with lower postoperative mortality odds but lower rates of infective complications compared to HD were observed.