Study design and participants
We conducted a retrospective cohort study that included 5,128 incident Chinese CAPD patients from seven PD centers in China between January 1, 2005, and May 31, 2023. Patients aged < 18 years or those with a follow-up time of < 3 months were excluded. Patients with malignant tumors or severe liver disease were excluded. Patients with missing data or abnormal values were excluded from this study.
Data collection and follow-up
We constructed the CUS scores to evaluate the prognosis of patients with CAPD. Each condition was assigned a score of 1, 2, 3, or 4 depending on the Charlson Comorbidity Index (CCI)14. Based on our previous studies7,8,9,15,16,17,18,19, the CUS scores comprised nine conditions with relative mortality risks ≥ 1.2 or greater, and these conditions were also weighted based on their relative risks. Clinical conditions (common comorbidities and complications in patients with CAPD) and associated scores were as follows (1 point each): cerebrovascular disease, cardiovascular disease (myocardial infarction or congestive heart failure), peripheral vascular disease, diabetes mellitus, hypertension, hyperlipidemia, malnutrition (serum albumin < 3.8 g/dL), and anemia (hemoglobin < 11.0 g/dL). Patients aged 50 years or older received additional points: 18–49 years, + 0; 50–59 years old, + 1 point; 60–69 years old, + 2 points; 70–79 years old, + 3 points; and 80 years or older, + 4 points (Figure S1). The scores were summed to obtain the total score (CUS score), which was used to assess mortality in patients with CAPD.
Two well-trained nurses collected demographic data, comorbidities, and laboratory data one week (5.4 ± 1.1 days) before the start of PD in each facility, including age at study entry, sex, body mass index (BMI), current smoker, current alcohol use, comorbidities (DM, hypertension, a history of cerebrovascular disease, cardiovascular disease, or peripheral vascular disease, and hyperlipidemia), medication use (calcium channel blockers [CCB], β-blockers, angiotensin-converting enzyme inhibitors/angiotensin II receptor blockers [ACEI/ARBs]), and laboratory measurements (serum albumin and hemoglobin). Weight was measured in the absence of peritoneal dialysis.
The primary outcome was all-cause mortality. Cerebrovascular and cardiovascular (CVD) deaths due to cerebrovascular disease, myocardial infarction, congestive heart failure, and peripheral vascular disease were recorded. The details of the CAPD follow-up have been previously described20. The follow-up period was from the start of PD to the date of death, transfer to hemodialysis, renal transplantation, transfer to another dialysis center, loss to follow-up, or May 31, 2023. Patients who were lost to follow-up were censored on the date of the last examination.
Ethical approval
This study was conducted in accordance with the Declaration of Helsinki and received approval from the ethical and scientific review boards of Zhejiang Provincial People’s Hospital, approval number [QT2023233]. The ethical and scientific review boards of Zhejiang Provincial People’s Hospital have granted an exemption from requiring written informed consent.
Statistical analysis
Continuous variables are presented as means with standard deviations (SDs) for normally distributed data or medians with interquartile ranges (IQR) for skewed data. The normality of the parameters was examined using the Shapiro-Wilk test. Categorical variables are expressed as the number of patients. We first used restricted cubic spline plots to explore the nonlinearity assumptions between the CUS scores and the risk of all-cause mortality, fitting a restricted cubic spline function with four knots (at the 25th, 50th, 75th, and 95th percentiles)21. All patients were categorized based on the threshold value of the CUS scores (hazard ratio [HR] = 1.0) using restricted cubic spline plots for the primary analysis.
Survival was calculated using the Kaplan-Meier method, and differences in survival distributions were assessed using a log-rank test. We primarily used cause-specific hazard models to explore the association between the CUS scores and mortality risk. Patients who experienced events such as transfer to hemodialysis, renal transplantation, transfer to other centers, or loss to follow-up were censored, which impeded the occurrence of death. Thus, Transfer to hemodialysis, renal transplantation, transfer to other centers, and loss to follow-up before death were considered competing risks. We constructed subdistribution hazard models to confirm the associations observed in the primary analysis. The main difference between the two hazard models is that subjects experiencing a competing risk event remain in the risk set in the sub-distribution hazard model but are removed from the cause-specific hazard model22,23. These models were constructed after adjusting for sex, BMI, current smoking status, current alcohol use, medication use, and centers. The results from the multivariable hazard models were presented as HRs and 95% confidence intervals (95% CIs). Stratified analyses were performed to assess the potential effects of sex modification.
To minimize the potential for reverse causation, we conducted analyses that excluded deaths in the first two years of follow-up. In addition, for patients with a short-term follow-up period, interesting outcomes may only be partially observed, with underreporting of mortality incidence. We further analyzed the association in patients with at least 24 months of follow-up to fully observe the outcomes. All analyses were performed using Stata version 15.1. (StataCorp, College Station, TX, USA).