Important differences in patient survival exist between peritoneal dialysis (PD) and haemodialysis (HD). Several different studies have shown that PD yields a better survival rate than HD in the first and second years of treatment, especially in younger patients and non-diabetic patients with low comorbidity, whereas HD produces better results in diabetic patients, elderly patients, and in patients with greater comorbidity. In recent years, interesting changes have occurred in PD units in the Canary Islands, such as the introduction of peritoneal dialysis solutions with bicarbonate dialysate and low content of glucose degradation products, extended use of automated dialysis, and continuity of physicians and nurses in PD units, in addition to enhancing visits for advanced chronic kidney disease (ACKD).
This situation led us to perform our study with the primary objective of comparing medium-term survival among incident dialysis patients on HD versus PD in recent years in the Canary Islands, and as a secondary objective, to compare survival between these two types of dialysis by subgroups as defined by age, sex and diabetes.
This was a retrospective cohort study comparing survival between HD and PD patients starting dialysis in the Canary Islands between 01/01/2006 and 31/12/2009, with adjustment based on the propensity score analysis. We analysed data from the RERCAN database, which collects data on demographic variables, changes in type of dialysis, province and hospital of the patient, and mortality and its causes. We calculated Kaplan-Meier estimates of survival based on the overall population and stratified by age, sex and diabetes. We applied a Cox proportional hazards model for survival to estimate the relative mortality risk of PD compared with HD, using as independent variables: age, sex, quartiles of propensity score, the province of the patient, and diabetes. Finally, we applied a Cox model with time-dependent effects, using as a fixed risk factor the initial type of dialysis in order to assess the effect of PD versus HD on short and medium-term survival.
The cohort included 1469 patients (173 PD and 1296 HD), with a mean age of 62.5 years, 65% male. Mean follow-up was 16.2±12.4 months. Factors associated with greater probability of choosing PD were younger age and living in the province of Las Palmas. The cumulative mortality in the intention to treat (ITT) analysis was 27.1% in the HD group and 8.7% in the PD group, P<.0001. The cumulative probability of survival by ITT using PD vs HD was 96.6% versus 89% at 6 months (P<.001), 96% versus 80% at 12 months (P<.001), 90% versus 65% at 24 months (P<.001), 82% versus 58% at 36 months (P<.001) and 73% versus 45% at 46 months (P<.001). In the subgroup analysis, survival was also higher in PD patients compared to HD patients both over and under 65 years old, in both diabetic and non-diabetic patients, and in both genders. The same analysis from the 90th day onward produced similar results. In the ITT analysis, the Cox-adjusted mortality risk for PD was 61% lower than for HD (RR: 0.398, 95% CI 0.237-0.669, P=.001), adjusted for age, diabetes, sex, patient's province and propensity score. Broken down by years of survival on the technique used, the relative risk of death for PD compared with HD in the first year was also significantly lower (RR 0.509, 95% CI: 0.259-0.999, P=.049). From year 2 onwards, only age was a risk factor for mortality (RR: 2.785, 95% CI: 1.525-5.086, P=.001) and no differences were shown between the two dialysis techniques.
In the Canary Islands, PD has demonstrated survival advantages over HD in the short and medium term. It is remarkable that this benefit was found in young and old patients, men and women, and diabetic and non-diabetic patients, and that this advantage was maintained even after years of being on dialysis.