In this study comparing HD and PD, we found that the overall survival of PD was better than that of HD for people who were younger than 60 years old, while HD was associated with lower risk of death in the patients who older than 70 years of age. This difference was still significant after adjustment for a variety of confounding factors.
We compared 501 HD patients and 436 PD patient and found that PD patients had better survival rate than HD patients (The 1-, 3- and 5-year survival rates were 98.1% vs 96.3%, 86.3% vs 82.7%, and 72.5% vs 71.1%), even though the difference was not statistically significant. Our result concurred with the observational study in Beijing, China 12. For subgroup analyses by age, we found that PD was a better choice for younger patients while HD was superior to PD for the elderly.
Key findings from our research are that the survival rates of both dialysis methods were higher than those reported in previous studies. PD has advantages in young patients, while HD has advantages in older patients. The advantage of peritoneal dialysis in younger ESRD patients was first discovered in our study.
A study with 871 ESRD patients in Singapore showed patients who initiated dialysis with HD experienced better survival outcomes than those who initiated dialysis with PD in Singapore 7. This may be relating to the older age of PD in Singapore study (51 years old for HD vs 64 years old for PD). In contrast, in our cohort, patients with PD were younger (56 years old for HD vs 51 years old for PD). It can be seen that the age at which dialysis is started may be the key to patient survival. A 2008 study in Taiwan suggested that PD has a better prognosis in non-diabetic patients younger than 55 years of age, in other subgroups, the prognosis is similar 11. Previous study in South Korea has found that the survival outcomes of these two dialysis modalities were similar for patients younger than 55 years old, whereas hemodialysis is superior to peritoneal dialysis for the elderly 6. Previous studies in Taiwan suggest that the overall survival time of PD patients has improved in recent years 11. This may be related to the continuous improvement of peritoneal dialysis training methods and new dialysate applications 14. Therefore, we have reason to believe that the survival time of PD patients will be prolonged through better chronic disease management and peritoneal dialysis technology.
As a retrospective cohort study, our study had unavoidable retrospective bias and selective bias. The huge demand for kidney replacement therapy and the shortage of HD equipment have made clinicians and patients prefer PD to HD in recent years, which would lead to significant patient selection bias. And may be the reason for the significantly higher over all survival rate of patients in this study. Our study record only baseline laboratory test only. However, the results of these tests may change during dialysis treatment. Therefore, hemoglobin, plasma albumin, calcium, phosphorus and other items could not reflect the situation of patients in the treatment process. In addition, no specific cause of death was investigated in this study, therefore, no further analysis of the cause of death.
Our research was a regional retrospective study, the sample size was relatively large, and the follow-up duration was longer than previous studies. This study could be representative of the actual situation in the region. The finding that ESRD patients prefer different dialysis modality at different ages would provide evidence for clinicians and patients when they choose dialysis method. We plan to continue the current cohort and observe the survival differences between the two dialysis methods over a longer period of time.
This study found that 1-, 3-, and 5-year survival rates in ESRD patients were significantly higher than in previous studies. We thought the reason may be the screening of clinicians when patients enter renal replacement therapy. Compared with less developed regions, the source of patients in large medical centers may be better. In addition, in recent years, HD and PD technology have continued to be improved, management levels have been continuously improved, and medical insurance reimbursement coverage has been continuously expanded. These were important reasons for the prolonged survival time of these patients.
With regard to the risk factors associated with survival, the risk factors reported by previous studies were not totally consistent, including age 12,15−17, diabetes 12,15−17, and serum albumin 15, BMI 12, cardiovascular disease 17. As most studies found that age is one of the most important factors affecting survival in dialysis patients, which is consistent with our findings. Therefore, we performed subgroups analysis by age. We found that the risk factors were not completely similar in different age subgroup (< 60y and > 70y). In the younger subgroup diabetes, CCI, cardiovascular disease, total cholesterol, and hemodialysis were risk factors of all-cause mortality. Peritoneal dialysis fluids containing glucose have harmful effect on glycaemia control of PD patients and influence the long-term survival. Additionally, prior history of cardiovascular disease and diabetes mellitus were found to be associated with left ventricular ejection fraction for PD patients 18. In the older subgroup, while PD and chronic pulmonary disease were risk factors. Hypercholesterolemia was associated with increased mortality in HD patients without myocardial infarction/ cardiovascular disease 18. Higher pulmonary artery systolic pressure predicted development of right ventricular dysfunction, which portends a poor prognosis 20.
PD might confer a survival advantage to young and healthier patients due to better preservation of residual renal function compared to those undergoing HD 21 Moreover, young PD patients have better self-management ability to follow the doctors’ advice, guarantee enough nutritional intakes and perform standard operation to avoid infection. Flexible and diverse educational activities may further enhance the survival time of patients with peritoneal dialysis. In addition, home dialysis offers the opportunity to thrive; improves freedom, flexibility and well-being; and strengthens relationships, which could help young patients carry on perform better in PD 22,23. The PD treatment mode provides patients with greater freedom. Compared with HD patients, they could arrange their life more flexibly. Patients could get better financial support by returning to society. The economic burden of PD was relatively low at the same time. The better financial support might lead to longer survival.
Most patients with severe cerebrovascular disease were unable to complete PD alone. These patients with poor prognosis are more likely to choose hemodialysis treatment. Baseline data from our younger subgroup showed a significantly lower proportion of cerebrovascular disease in the PD group (4.69% versus 13.00%, p < 0.01). These patients may affect the overall survival outcomes of HD. Therefore, we performed a sensitivity analysis to exclude patients with cerebrovascular disease. The Cox regression model analysis of age grouping was performed again. The advantages of PD in young patients remained. However, the advantages of PD were narrower than before. This showed that the lower proportion of comorbid cerebrovascular diseases is one possible reason for the better prognosis of younger PD patients.
The CCI in the PD group was lower than that in the HD group (4.09 ± 1.69 versus 4.38 ± 1.94, p = 0.06). This suggests that patients in the PD group are younger and have less comorbidity. CCI has been shown to predict the risk of all-cause death in patients 24. Differences in CCI between PD and HD may lead to different prognosis in the two groups. Therefore, we adjusted the baseline CCI levels of the two groups in the propensity score matching model so that the baseline CCI levels of the two groups of patients were as close as possible. We then performed a multivariate Cox regression analysis again, and the survival advantage of the PD group remained. This suggests that comorbidities might not be the reason for better survival in young PD patients.
However, in the elderly subgroup, there were no significant differences in CCI and prevalence of baseline cerebrovascular disease, diabetes, and other comorbidities. PD has become a risk factor for patient prognosis. Older patients in China generally had poor self-care ability and needed to rely on the assistance of family members and medical institutions, and in-center dialysis provides clinical assessment twice or thrice per week by clinician and nurses. Therefore, it may be more beneficial to elderly ESRD patients. PD can cause significant protein energy loss. Decreased nutritional intake of elderly patients may be the cause of shortened life span. This hypothesis needed to be confirmed in subsequent studies by comparing the nutritional indicators of the two groups during follow-up.
We believe that the factors affecting the survival of dialysis patients are diverse. Economic development, climates, eating habits, medical insurance payment systems, and medical service forms in different regions may be factors that affect the life of patients. Our hospital is located in southern China, and the source of patients is mainly in Guangdong Province. Guangdong's GDP in 2018 exceeded RMB 970 billion, surpassing Australia and slightly less than South Korea. The population is over 100 million. It is one of the fastest growing regions in China. Previous observational studies from other regions and countries had suggested that there may be regional and ethnic differences in the survival of ESRD patients undergoing PD or HD. Our research could at least represent the situation of ESRD patients in Guangdong Province. Taking into account the similarities of race and economic development level, we though our result could be partially extended to Southeast Asian countries or Asian populations.
For younger ESRD patients (< 60 years old), the 1, 3, and 5-year survival rates of PD were higher than those of HD. For older ESRD patients (> 70 years old), the 1, 3, and 5-year survival rates of PD were lower than those of HD. Therefore. The results suggest that younger ESRD patients may be more suitable for PD, and the elderly for HD.