Changes in mortality hazard of the Korean long-term dialysis population: The dependencies of time and modality switch

The effect of dialysis modality on annual mortality: A prospective cohort study

Despite numerous studies on the effect of each dialysis modality on mortality, the issue remains controversial. We investigated the hazard rate of mortality in patients with incident end-stage renal disease (ESRD) concerning initial dialysis modality (hemodialysis vs. peritoneal dialysis). Using a nationwide, multicenter, prospective cohort in South Korea, we studied 2207 patients, of which 1647 (74.6%) underwent hemodialysis. We employed the weighted Fine and Gray model over the follow-up period using inverse probability of treatment and censoring weighting. Landmark analysis was used for identifying the changing effect of dialysis modality on individuals who remained event-free at each landmark point. No significant difference in hazard rate was observed overall. However, the peritoneal dialysis group had a significantly higher hazard rate than the hemodialysis group among patients under 65 years after 4- and 5- year follow-up. A similar pattern was observed among those with diabetes mellitus. Landmark analysis also showed the higher hazard rate for peritoneal dialysis at 2 years for the education-others group and at 3 years for the married group. These findings may inform dialysis modality decisions, suggesting a preference for hemodialysis in young patients with diabetes, especially for follow-ups longer than 3 years.

Rate and reasons for peritoneal dialysis dropout following haemodialysis to peritoneal dialysis switch: a systematic review and meta-analysis

BACKGROUND

Patient experiences and survival outcomes can be influenced by the circumstances related to dialysis initiation and subsequent modality choices. This systematic review and meta-analysis aimed to explore the rate and reasons for peritoneal dialysis (PD) dropout following haemodialysis (HD) to PD switch.

METHOD

This systematic review conducted searches in four databases, including Medline, PubMed, Embase, and Cochrane. The protocol was registered on PROSPERO (study ID: CRD42023405718). Outcomes included factors leading to the switch from HD to PD, the rate and reasons for PD dropout and mortality difference in two groups (PD first group versus HD to PD group). The Critical Appraisal Skills Programme (CASP) checklist and the GRADE tool were used to assess quality.

RESULTS

4971 papers were detected, and 13 studies were included. On meta-analysis, there was no statistically significant difference in PD dropout in the PD first group (OR: 0.81; 95%CI: 0.61, 1.09; I2 = 83%; P = 0.16), however, there was a statistically significant reduction in the rate of mortality (OR: 0.48; 95%CI: 0.25, 0.92; I2 = 73%; P = 0.03) compared to the HD to PD group. The primary reasons for HD to PD switch, included vascular access failure, patient preference, social issues, and cardiovascular disease. Causes for PD dropout differed between the two groups, but inadequate dialysis and peritonitis were the main reasons for PD dropout in both groups.

CONCLUSION

Compared to the PD first group, a previous HD history may not impact PD dropout rates for patients, but it could impact mortality in the HD to PD group. The reasons for PD dropout differed between the two groups, with no statistical differences. Psychosocial reasons for PD dropout are valuable to further research. Additionally, establishing a consensus on the definition of PD dropout is crucial for future studies.

Prediction of premature all-cause mortality in patients receiving peritoneal dialysis using modified artificial neural networks

Premature all-cause mortality is high in patients receiving peritoneal dialysis (PD). The accurate and early prediction of mortality is critical and difficult. Three prediction models, the logistic regression (LR) model, artificial neural network (ANN) classic model and a new structured ANN model (ANN mixed model), were constructed and evaluated using a receiver operating characteristic (ROC) curve analysis. The permutation feature importance was used to interpret the important features in the ANN models. Eight hundred fifty-nine patients were enrolled in the study. The LR model performed slightly better than the other two ANN models on the test dataset; however, in the total dataset, the ANN models fit much better. The ANN mixed model showed the best prediction performance, with area under the ROC curves (AUROCs) of 0.8 and 0.79 for the 6-month and 12-month datasets. Our study showed that age, diastolic blood pressure (DBP), and low-density lipoprotein cholesterol (LDL-c) levels were common risk factors for premature mortality in patients receiving PD. Our ANN mixed model had incomparable advantages in fitting the overall data characteristics, and age is a steady risk factor for premature mortality in patients undergoing PD. Otherwise, DBP and LDL-c levels should receive more attention for all-cause mortality during follow-up.