Abstract
Objectives
Mortality and associated risk factors in young and elderly haemodialysis patients with end-stage kidney disease (ESKD) have not been well examined in China. Therefore, we aimed to assess the all-cause mortality and risk factors associated with all-cause mortality between young and elderly haemodialysis patients in China.
Design
A population-based multicentre retrospective cohort study.
Setting
Using the Dialysis Initiation based on Fuzzy mathematics Equation study data, patients with ESKD undergoing maintenance haemodialysis from 24 centres in China from 1 January 2008 to 30 September 2015.
Participants
1601 enrolled patients with ESKD were categorised into young group (18–44 years old) and elderly (≥60 years old) group.
Outcome measures
The primary outcome was all-cause mortality. We estimated overall survival using a log-rank test. Cox proportional hazard regression analysis was implemented to identify risk factors and HR associated all-cause mortality.
Results
During a mean follow-up of 48.17±25.59 months, of the 1601 subjects, 319 (19.92%) patients death, including 64 (9.97%) in young group and 255 (26.59%) in elderly group, respectively. The cumulative survival in elderly group was lower than young group (Log Rank tests=63.31, p<0.001). Multivariate Cox proportional hazards analysis showed the cardiovascular disease (HR, 2.393; 95% CI 1.532 to 3.735; p<0.001), cerebrovascular disease (HR, 2.542; 95% CI 1.364 to 4.739; p=0.003) and serum albumin<3.5 g/dL (HR, 1.725; 95% CI 1.091 to 2.726; p=0.020) at the haemodialysis initiation were associated with increased risk of all-cause mortality in elderly groups; however, the cardiovascular disease only was associated with increased risk of all-cause mortality in young groups.
Conclusions
The all-cause mortality of elderly haemodialysis patients were higher than young haemodialysis patients in China. Identified risk factors associated all-cause mortality may inform development of age-appropriate treatment, intervention strategies and improve survival prognosis of this unique population.
Keywords: End stage renal failure, Dialysis, Nephrology
Strengths and limitations of this study.
A population-based multicentre retrospective cohort study for assessing the all-cause mortality and associated risk factors between young and elderly haemodialysis patients.
Has a relatively long follow-up period.
Using the Dialysis Initiation based on Fuzzy mathematics Equation study data in China.
Limitations including some important variables were missing from patient’s baseline data, many variables assessed as risk factors were not updated regularly and the selection bias of the included patients were from high-quality haemodialysis units in developed areas mainland China.
Introduction
Haemodialysis is the major modality of renal replacement therapy for patients with end-stage kidney disease (ESKD) worldwide,1 approximately 553 000, accounting for 91.0% of all dialysis patients in China.2 The incidence of patients with ESKD undergoing maintenance haemodialysis with an annual increasing rate of 52.9% in China.3 The numbers of elderly haemodialysis patients has been increasing in recent years; meanwhile, young patients also account for significant proportion in haemodialysis patients.4–8
The 2021 United States Renal Data System (USRDS) Annual Data Report indicated that adjusted incidence of ESKD among individuals aged 65–74 years was 1307 pmp and among individuals aged more than 75 years was 1587 pmp.9 The 2016 annual data report from China Kidney Disease Network (CK-NET) showed that among individuals aged 18–44 years, the incidence of dialysis patients was 66.94 pmp, among individuals aged more than 65 years, the incidence of dialysis patients was 286.31 pmp.8 As expected, elderly haemodialysis patients were associated with higher mortality risk than the young haemodialysis patients,10 The Dialysis Outcomes and Practice Patterns Study showed that elderly haemodialysis patients aged more than 75 years had a threefold to sixfold higher mortality risk than younger patients aged less than 45 years, adjusted for demographics and comorbidities.11 The 2021 USRDS Annual Data Report showed that elderly dialysis patients with more than 65 years old had significant higher mortality rate than younger patients with less than 44 years of age.9 The reason for the difference in mortality of haemodialysis patients may be due to the most elderly haemodialysis patients suffered from more common cardiovascular disease, malignancies, frailty, malnutrition, diabetes and poorer quality of life versus younger haemodialysis patients.10–13 To improve the outcomes of haemodialysis patients, it is critical to identify the contributions of potentially risk factors. However, the mortality rate and associated risk factors between young and elderly undergoing haemodialysis patients have not been well assessed in China. Therefore, the objective of this retrospective cohort study was to evaluate the clinical characteristics, all-cause mortality and associated risk factors between young and elderly patients with ESKD undergoing maintenance haemodialysis in China.
Material and methods
Data sources
We conducted this population-based multicentre retrospective cohort study from the database of Dialysis Initiation based on Fuzzy mathematics Equation study (DIFE),14 which focuses on adults patients with ESKD undergoing maintenance haemodialysis in China and aims to establish a novel equation of timing of haemodialysis initiation based on a fuzzy mathematical method.15 16
Study participants
The enrolled patients with ESKD undergoing maintenance haemodialysis from 24 haemodialysis centres in China from 1 January 2008 to 30 September 2015. Patients who met the inclusion criteria and did not meet the exclusion criteria were included in our study. The inclusion criteria were the following: (1) 18–44 or more than 60 years old, (2) undergoing haemodialysis treatment for more than 3 months and (3) haemodialysis treatment with three times a week. The exclusion criteria were the following: (1) patients undergone peritoneal dialysis or kidney transplantation before or after haemodialysis, (2) patients with the presence of chronic infection defined as lasted more than 3 months, including bacterial, viral, fungal infection and so on, (3) liver cirrhosis and (4) cancer diseases.
The enrolled patients were categorised into young group (18–44 years old) and elderly group (≥60 years old) according to their age on the initiation of the haemodialysis. Patients were followed up to 30 September 2016, and the primary outcome was all-cause mortality of haemodialysis patients.
All started haemodialysis and survival prognostic data of patients were obtained and reviewed from electronic medical records of outpatient and inpatient of 24 haemodialysis centres. The demographics data included age, gender, height, weight, body mass index (BMI). Clinical data included date of haemodialysis initiation, type of haemodialysis associated vascular access, causes of ESKD. Nutritional status were assessed using serum albumin and Geriatric Nutritional Risk Index (GNRI).17 Comorbid conditions were assessed using the Charlson comorbidity index (CCI).18 Heart failure defined according to the New York Heart Association functional class. The laboratory tests were performed within 1 month before haemodialysis initiation, and the tested items included haemoglobin, blood urea nitrogen, creatinine, uric acid, serum sodium, serum potassium, serum calcium, serum phosphorus, carbon dioxide combining power, serum albumin, cholesterol, parathyroid hormone. The estimated glomerular filtration rate (eGFR) was calculated based on serum creatinine at the time of haemodialysis initiation using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.19 The survival time of the patients was measured in months and was calculated from the date of haemodialysis initiation to the date of death or to the final follow-up date.
Patient and public involvement
Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.
Statistical analysis
Continuous variables with normal distribution were expressed as means and SD, whereas variables with non-normal distribution were expressed as medians and IQRs, and categorical variables were presented as frequencies with percentages. A student independent t-test or a Mann-Whitney U test was used for analysis in the case of normally or non-normally distributed continuous variables. Categorical variables data were compared using χ2 tests or Fisher’s exact test. A multiple imputation approach was applied to process missing data. Patients were censored because of loss to follow-up, withdraw or survival to end of the study. The survival prognosis was compared using the Kaplan-Meier survival curves and log-rank test. The HR of mortality of elderly patients relative to young patients was assessed using univariate and multivariate Cox proportional hazards regression analysis. The risk factors associated all-cause mortality were assessed using univariate and multivariable Cox proportional hazards regression analysis, adjusting for sex, hypertension, diabetes, heart failure, cardiovascular disease, cerebrovascular disease, eGFR, serum albumin, GNRI and CCI scores. All hypothesis tests were evaluated using two-sided with 95% CIs and p<0.05 indicated statistical significance. All analyses were performed with SPSS software (V.19.0, IBM, Armonk, New York, USA).
Results
Participants
A total of 1680 young and elderly patients with ESKD were screened. A total of 79 patients were excluded as lacking of important demographic, laboratory and survival data. Finally, a total of 1601 patients were subsequently enrolled in our study. The research flowchart is shown in figure 1.
Figure 1.
Flowchart of the study.
Baseline patient characteristics
The baseline demographics and clinical characteristics of the young and elderly undergoing haemodialysis patients are described in table 1. A total of 642 young patients and 959 elderly patients were enrolled in this study. The young and elderly patients had different sex and BMI distribution, 417 males (64.95%) in young group compared with 535 males (55.79%) in elderly group, respectively (X2=13.404, p<0.001). The young patients had a higher average BMI than the elderly patients (p=0.014). There were also obviously differences of causes of ESKD, lower prevalence of diabetes (7.32%), higher glomerulonephritis (44.39%) in young group compared with the higher prevalence of diabetes (29.82%), lower glomerulonephritis (15.53%) in the elderly group (X2=233.6, p<0.001). At the haemodialysis initiation, the elderly patients showed a higher mean eGFR levels than the young patients (6.37±3.30 mL/min/1.73 m2 vs 5.25±3.02 mL/min/1.73 m2, p<0.001). The young patients showed significantly higher levels of diastolic pressure, serum creatinine, serum phosphorus, serum albumin and parathyroid hormone. However, the elderly patients showed higher levels of CCI scores, serum calcium, haemoglobin and serum sodium.
Table 1.
Baseline demographics and clinical characteristics of the young and elderly haemodialysis patients
| Variables | Young group (n=642） | Elderly group (n=959） | P value |
| Age (year) | 34.31±6.80 | 68.55±6.53 | <0.001 |
| Male (%) | 417 (64.95%) | 535 (55.79%) | <0.001 |
| Body mass index (kg/m2) | 25.34±3.32 | 24.45±3.82 | 0.014 |
| Systolic pressure (mm Hg) | 152.61±24.66 | 153.38±23.05 | 0.093 |
| Diastolic pressure (mm Hg) | 92.94±16.28 | 81.64±13.15 | <0.001 |
| Cause of ESKD (%) | |||
| Diabetic nephropathy | 47 (7.32%) | 286 (29.82%) | <0.001 |
| Hypertensive nephropathy | 56 (8.72%) | 168 (17.51%) | |
| Glomerulonephritis | 285 (44.39%) | 149 (15.53%) | |
| Others | 254 (39.56%) | 356 (37.12%) | |
| Comorbidity (%) | |||
| Diabetes | 84 (13.08%) | 446 (46.51%) | <0.001 |
| Hypertension | 255 (39.72%) | 688 (71.74%) | <0.001 |
| Coronary artery disease | 66 (10.28%) | 303 (31.59%) | <0.001 |
| Heart failure | 183 (28.50%) | 378 (39.42%) | <0.001 |
| Cerebrovascular disease | 15 (2.33%) | 68 (7.09%) | <0.001 |
| CCI Scores | 2.36±0.57 | 5.57±0.75 | <0.001 |
| Types of vascular accesses | |||
| AVF | 136 (21.18%) | 244 (25.44%) | 0.055 |
| TCC | 59 (9.19%) | 103 (10.74%) | |
| NTC | 447 (69.63%) | 612 (63.82%) | |
| Laboratory tests | |||
| Haemoglobin (g/L) | 81.60±22.50 | 85.20±19.50 | 0.001 |
| Blood urea nitrogen (mg/dL) | 98.30±40.97 | 81.44±32.66 | <0.001 |
| Serum creatinine (mg/dL) | 12.44±5.39 | 8.33±3.24 | <0.001 |
| eGFR (mL/min/1.73 m2) | 5.25±3.02 | 6.37±3.30 | <0.001 |
| Serum albumin (g/dL) | 3.51±0.65 | 3.43±0.61 | 0.014 |
| Serum sodium (mmol/L) | 138.65±4.01 | 139.07±4.71 | 0.007 |
| Serum potassium (mmol/L) | 4.79±0.92 | 4.77±0.96 | 0.336 |
| Serum phosphorus (mmol/L) | 2.23±0.79 | 1.84±0.59 | <0.001 |
| Serum calcium (mmol/L) | 1.99±0.40 | 2.01±0.36 | 0.001 |
| Parathyroid hormone (pg/mL) | 352.86±324.60 | 281.88±263.90 | 0.002 |
AVF, arteriovenous fistula; eGFR, estimated glomerular filtration rate; ESKD, end-stage kidney disease; NTC, non-tunnelled catheter; TCC, tunnelled cuffed catheter.
Mortality
Of the 1601 subjects, the mean follow-up duration was 48.17±25.59 months, 319 (19.92%) patients death, including 64 (9.97%) in young group and 255 (26.59%) in elderly group, the HR of mortality of elderly patients relative to young patients were assessed using Cox proportional hazards regression analysis. Multivariate Cox proportional hazards analysis showed that the HR of overall all-cause mortality of elderly patients relative to young patients, 7.129 (95% CI 3.827 to 13.279, p<0.001), the HR of 1-year mortality of elderly patients relative to young patients, 3.922 (95% CI 1.367 to 11.249, p<0.001), the HR of 3-year mortality of elderly patients relative to young patients, 6.543 (95% CI 3.002 to 14.260, p<0.001). The comparison of mortality between young and elderly haemodialysis patients is summarised in table 2.
Table 2.
The comparison of mortality between young and elderly haemodialysis patients
| Variables | Young group | Elderly group | Univariate analysis | Multivariate analysis | ||||||
| Event (%) |
Total | Event (%) | Total | HR | 95% CI | P value | HR | 95% CI | P value | |
| Overall mortality | 64 (9.97) | 642 | 255 (26.59) | 959 | 2.888 | 2.195 to 3.798 | <0.001 | 7.129 | 3.827 to 13.279 | <0.001 |
| 1-year mortality | 17 (2.65) | 642 | 53 (5.53) | 959 | 2.115 | 1.225 to 3.652 | 0.007 | 3.922 | 1.367 to 11.249 | <0.001 |
| 3-year mortality | 48 (10.50) | 457 | 149 (20.84) | 715 | 2.09 | 1.510 to 2.894 | <0.001 | 6.543 | 3.002 to 14.260 | <0.001 |
Cox regression was used to calculate HR, adjustment for sex, hypertension, diabetes, heart failure, cardiovascular disease, cerebrovascular disease, eGFR and serum albumin.
eGFR, estimated glomerular filtration rate.
Survival prognosis
The Kaplan-Meier survival curve of all-cause mortality between young and elderly displayed the overall cumulative survival was lower in elderly group than young group (Log Rank tests=63.31, p<0.001), which are shown in figure 2. The median survival time of young and old patients was 93.59 and 78.63 months, respectively. The HR of mortality of elderly patients relative to young patients, 2.884 (95% CI 2.309 to 3.602, p<0.001). To assess the effects of the timing of haemodialysis initiation on survival of young and elderly patients, patients were divided into the early dialysis starting subgroup (eGFR>5 mL/min.1.73 m2) and the late dialysis starting subgroup (eGFR≤5 mL/min.1.73 m2) according to eGFR at the initiation of haemodialysis. The Kaplan-Meier survival curve showed that overall cumulative survival was lower in the elderly group than the young group no matter whether in the early dialysis starting subgroup (Log Rank tests=36.06, p<0.001) or in the late dialysis starting subgroup (Log Rank tests=23.16, p<0.001), which are shown in figure 3A and figure 3B, respectively. In the early dialysis starting subgroup, the HR of mortality of elderly patients relative to young patients, 3.428, (95% CI 2.525 to 4.653, p<0.001). In the late dialysis starting subgroup, the HR of mortality of elderly patients relative to young patients, 2.399, (95% CI 1.714 to 3.358, p<0.001).
Figure 2.
Kaplan-Meier survival curve between the young and elderly groups. The elderly group survival was significantly lower than young group (Log Rank tests=63.31, p<0.001).
Figure 3.
Kaplan-Meier survival curve between young and elderly patients groups in the early and late dialysis starting subgroups. (A) The elderly patients survival was significantly lower than young patients in early dialysis starting subgroup with eGFR>5.0 mL/min 1.73 m2 (Log Rank tests=36.06, p<0.001). (B) The elderly group survival was significantly lower than young group in late dialysis starting subgroup with eGFR≤5.0 mL/min 1.73 m2 (Log Rank tests=23.16, p<0.001). eGFR, estimated glomerular filtration rate.
Risk factors of all-cause mortality
The risk factors associated all-cause mortality of young haemodialysis patients were assessed using Cox proportional hazards regression analysis. Univariate Cox proportional hazards analysis showed that cardiovascular disease (HR, 3.046; 95% CI 1.178 to 7.875; p=0.022) was the risk factor associated all-cause mortality in young haemodialysis patients. Multivariate Cox proportional hazards analysis with a forward stepwise regression method (Forward LR) showed that only cardiovascular disease (HR, 3.349; 95% CI 1.059 to 10.587; p=0.040) was the risk factor associated all-cause mortality in young haemodialysis patients, which are shown in table 3.
Table 3.
Univariate and multivariate Cox proportional hazards regression analysis of risk factors associated with all-cause mortality of young patients
| Variables | Univariate analysis | Multivariate analysis | ||||
| HR | 95% CI | P value | HR | 95% CI | P value | |
| Sex | ||||||
| Female | Reference | |||||
| Male | 1.58 | 0.907 to 2.752 | 0.106 | |||
| Hypertension | 1.558 | 0.502 to 4.836 | 0.443 | |||
| Diabetes | 0.772 | 0.333 to 1.791 | 0.547 | |||
| Heart failure | 0.574 | 0.306 to 1.074 | 0.083 | |||
| Cardiovascular disease | 3.046 | 1.178 to 7.875 | 0.022 | 3.349 | 1.059 to 10.587 | 0.040 |
| Cerebrovascular disease | 3.024 | 0.397 to 23.052 | 0.286 | |||
| eGFR level | ||||||
| eGFR>5 mL/min 1.73 m2 | Reference | |||||
| eGFR≤5 mL/min 1.73 m2 | 1.178 | 0.710 to 1.955 | 0.525 | |||
| Serum albumin | ||||||
| ≥3.5 g/dL | Reference | |||||
| <3.5 g/dL | 1.07 | 0.654 to 1.751 | 0.788 | |||
| CCI scores | ||||||
| CCI (1 to 3) | Reference | |||||
| CCI (4 to 5) | 0.7 | 0.171 to 2.865 | 0.62 | |||
Cox regression was used to calculate HR, adjustment for sex, hypertension, diabetes, heart failure, cardiovascular disease, cerebrovascular disease, eGFR, serum albumin and CCI scores.
CCI, Charlson comorbidity index; eGFR, estimated glomerular filtration rate.
The risk factors associated all-cause mortality of elderly haemodialysis patients were assessed using Cox proportional hazards regression analysis. Univariate Cox proportional hazards analysis showed that cardiovascular disease, cerebrovascular disease, serum albumin<3.5 g/dL, 82≤GNRI<92, GNRI<82 and CCI≥6 were the risk factors associated all-cause mortality in elderly haemodialysis patients. However, multivariate Cox proportional hazards analysis with a forward stepwise method (Forward LR) showed that cardiovascular disease (HR, 2.393; 95% CI 1.532 to 3.735; p<0.001), cerebrovascular disease (HR, 2.542; 95% CI 1.364 to 4.739; p=0.003) and serum albumin<3.5 g/dL (HR, 1.725; 95% CI 1.091 to 2.726; p=0.020) were the risk factors associated all-cause mortality in elderly haemodialysis patients, which are shown in table 4.
Table 4.
Univariate and multivariate Cox proportional hazards analysis of risk factors associated with all-cause mortality of elderly patients
| Variables | Univariate analysis | Multivariate analysis | ||||
| HR | 95% CI | P value | HR | 95% CI | P value | |
| Sex | ||||||
| Female | Reference | |||||
| Male | 1.15 | 0.089 to 1.476 | 0.273 | |||
| Comorbidity | ||||||
| Hypertension | 0.981 | 0.672 to 1.432 | 0.922 | |||
| Diabetes | 1.104 | 0.863 to 1.432 | 0.43 | |||
| Heart failure | 1.245 | 0.969 to 1.600 | 0.086 | |||
| Cardiovascular disease | 1.479 | 1.069 to 2.046 | 0.018 | 2.392 | 1.532 to 3.735 | <0.001 |
| Cerebrovascular disease | 2.901 | 1.896 to 4.438 | <0.001 | 2.542 | 1.364 to 4.739 | 0.003 |
| eGFR level | ||||||
| eGFR>5 mL/min 1.73 m2 | Reference | |||||
| eGFR≤5 mL/min.1.73 m2 | 0.812 | 0.630 to 1.046 | 0.107 | |||
| Serum albumin | ||||||
| ≥3.5 g/dL | Reference | Reference | ||||
| < 3.5 g/dL | 1.584 | 1.235 to 2.031 | <0.001 | 1.725 | 1.091 to 2.726 | 0.020 |
| GNRI | ||||||
| GNRI>98 | Reference | |||||
| 92≤GNRI ≤ 98 | 1.232 | 0.800 to 1.898 | 0.343 | |||
| 82≤GNRI < 92 | 1.625 | 1.094 to 2.414 | 0.016 | |||
| GNRI<82 | 2.063 | 1.316 to 3.234 | 0.002 | |||
| CCI scores | ||||||
| CCI (1 to 3) | Reference | |||||
| CCI (4 to 5) | 1.339 | 0.646 to 2.775 | 0.432 | |||
| CCI (≥6) | 3.038 | 1.494 to 6.177 | 0.002 | |||
Cox regression was used to calculate HR, adjustment for sex, hypertension, diabetes, heart failure, cardiovascular disease, cerebrovascular disease, eGFR, serum albumin, GNRI and CCI scores.
CCI, Charlson comorbidity index; eGFR, estimated glomerular filtration rate; eGFR, estimated glomerular filtration rate; GNRI, Geriatric Nutritional Risk Index.
Discussion
The longitudinal nationwide cohort study was the first to assess that the clinical characteristics, mortality rate and risk factors associated all-cause mortality between young and elderly patients undergoing maintenance haemodialysis in China. Our study demonstrated that the overall mortality rate of elderly haemodialysis patients was higher than that of young haemodialysis patients (26.59% vs 9.97%), the 1-year mortality rate of young and elderly patients was 2.65% and 5.53%, and the 3-year mortality rates of young and elderly patients were 10.50% and 20.84%, respectively. Our findings were consistent with previous studies that the elderly patients undergoing haemodialysis had a higher mortality risk.20 21 Based on the USRDS registry, the 1-year mortality rate after dialysis initiation for older adults patients was currently approximately 30%.22 A recent study showed that the mortality rate among patients more than 65 years old was 54.5% at the first year.23 The 1-year mortality rate of the elderly aged 65–74 years old was 28%, more than 75 years old was 41%.24 Obviously, the 1-year mortality of haemodialysis patients in our study was significantly lower than that in USA, it may be associated with the majority of enrolled patients were from higher quality haemodialysis centres with higher survival quality.
Our study showed the eGFR of initiation of haemodialysis was 5.25±3.02 mL/min/1.73 m2 in the young group compared with 6.37±3.30 mL/min/1.73 m2 in the elderly group, which indicated that the timing of haemodialysis initiation of elderly patients was earlier than young patients. Our findings were consistent with previous studies which timing of haemodialysis initiation of elderly patients was earlier than young patients.25 The possible reasons may be associated with the elderly patients with ESKD may be more likely to start earlier haemodialysis due to more ESKD complications, comorbidity, poor nutritional status and worsen tolerance to uraemia symptoms than young patients with ESKD.4 26–28 Although our study demonstrated that the timing of haemodialysis initiation of elderly patients was earlier than young patients; however, the overall mortality was higher in elderly patients compared with young patients, and the findings were similar to previous studies that early initiation of haemodialysis was not associated with an improvement in survival or clinical outcomes.20 29–31
The timing of haemodialysis initiation is one of the important factors influencing survival prognosis of patients with ESKD.32 33 Previous studies did not assess the impaction of timing of dialysis on survival prognosis in different age patients especially between young and elderly patients together. The Kaplan-Meier survival curve showed that overall survival was lower in elderly than young patients no matter whether in the early dialysis starting subgroup or in the late dialysis starting subgroup. Multivariate Cox regression indicated that the eGFR was not the risk factor associated all-cause mortality in young or elderly haemodialysis patients. Therefore, the timing of dialysis initiation based on eGFR did not affect survival outcomes in elderly and young haemodialysis patients.
A multivariate Cox regression model of 2920 patients with uraemia demonstrated that the risk of death increased 1.25 times for each additional 10 years of age of dialysis patients.34 The poor prognosis of elderly patients may be associated with primary diseases, higher prevalence of cardiovascular disease, diabetes, frailty, cognition disorder, higher levels of CCI scores, lower levels of serum albumin, serious dialysis complications and reduced life expectancy together, which may lead to increased mortality risk compared with younger dialysis patients.10–13 To assess the risk factors associated all-cause mortality between young and elderly undergoing haemodialysis patients in China, we calculated HR using univariate and multivariable Cox proportional hazards regression analysis, adjusting of the demographics and clinical characteristics of the patients, including sex, hypertension, diabetes, heart failure, cardiovascular disease, cerebrovascular disease, eGFR, serum albumin, GNRI and CCI scores. Our study indicated young and elderly haemodialysis patients with cardiovascular disease on the haemodialysis initiation had a significantly increased risk of all-cause mortality than those without such disease. As we know, cardiovascular disease was a leading cause of death among patients with ESKD, accounting for nearly 50% of deaths in haemodialysis population,35 36 and accounted for nearly 40% of deaths in young adults with incident ESKD based on from USRDS data.37 The 2021 USRDS Annual Data Report showed that arrhythmia or cardiac arrest accounted for 34.7% of deaths, and cardiovascular disease, collectively, was the cause of death in 42.5%.9
A systematic review and meta-analysis showed that lower serum albumin was the risk of mortality and cardiac death in patients undergoing haemodialysis.38 Our study also demonstrated that the significantly increased risk of all-cause mortality in elderly haemodialysis patients with serum albumin less than 3.5 g/dL than those with serum albumin ≥3.5 g/dL using multivariable Cox proportional hazards regression analysis. However, serum albumin was not associated with increased risk of all-cause mortality in young haemodialysis patients.
The present study has the following several aspects limitation. First, some important variables were missing from patient’s baseline data, such as left ventricular ejection fraction, left ventricular mass index, brain natriuretic peptide and many variables assessed as risk factors were not updated regularly, and the data were collected only at the baseline and at the end of the study. Additionally, the selection bias of the included patients mainly were from high-quality haemodialysis units in developed areas, no coverage of community haemodialysis units with poor medical and economic conditions. Moreover, not all potential risk factors were available for analysis, including residual renal function, haemodialysis prescription, medications. Therefore, in the future, we will conduct a multicentre, large-sample prospective randomised controlled study including comprehensive and detailed variables, haemodialysis centres covering different regions, medical and economic levels in mainland China,
Conclusion
The all-cause mortality of elderly haemodialysis patients were higher than young haemodialysis patients in China. Cardiovascular disease was the risk factors associated all-cause mortality of young and elderly haemodialysis patients, cerebrovascular disease and lower serum albumin were the risk factor associated all-cause mortality of elderly haemodialysis patients. Identified risk factors associated all-cause mortality could be helpful to inform development of age-appropriate treatment and intervention strategies for undergoing haemodialysis patient to improve survival prognosis of this unique population.
Supplementary Material
Acknowledgments
The authors are grateful to all of the staff who conducted the baseline and follow-up surveys, all patients who participated in this study and DIFE group members.
Footnotes
Collaborators: Investigators of the Dialysis Initiation based on Fuzzy mathematics Equation (DIFE) study group: Hongli Lin, MD, PhD, Ying Liu, MD, the First Affiliated Hospital of Dalian Medical University, Dalian, China. Xuefeng Sun, MD, PhD, Yong Wang MD, Chinese PLA General Hospital, Beijing, China. Hongli Jiang, MD, PhD, Meng Wang, MM, The First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, China. Wei Shi, MD, PhD, Xinling Liang, MD, PhD, Guangdong General Hospital, Guangdong Academy of Medical Sciences, Guangzhou, China. Wenhu Liu, MD, PhD, Beijing Friendship Hospital, Capital Medical University. Ping Fu, MD, PhD, Li Zhou, MD, PhD, West China Hospital of Sichuan University, Chengdu, China. Xiaoqiang Ding, MD, PhD, Jianzhou Zou, MD, Zhongshan Hospital, Fudan University, Shanghai, China. Ming Chang, MM, Dalian Municipal Central Hospital, Dalian, China. Xueqing Yu, MD, PhD, Zhihua Zheng, MD, PhD, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou, China. Ning Cao, PhD, Xiaoling Zhang, MM, General Hospital of Shenyang Military Area Command, Shenyang, China. Menghua Chen, PhD, Zhihong Feng, MD, General Hospital of Ningxia Medial University, Yinchuan, China. Zhaohui Ni, MD, PhD, Mingli Zhu, MD, Renji Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai, China. Jing Chen, MD, PhD, Minmin Zhang, MD, PhD, Huashan Hospital, Fudan University, Shanghai, China. Shiren Sun, PhD, Yan Yu, MM, Xijing Hospital, The Fourth Military Medical University, Xi’an, China. Huimin Wang, MM, Wei Wei, MM, General Hospital of Benxi Iron and Steel Co., Ltd, Benxi, China. Yunyan Wang, BSM, Jie Yang, BSM, Daping Hospital and Surgery Institute, The Third Military Medical University, Chongqing, China. Bihu Gao, PhD, Hua Wu, BSM, Affiliated Zhong Shan Hospital of Dalian University, Dalian, China. Jianqin Wang, PhD, Wenge Wang, BSM, Lanzhou University Second Hospital, Lanzhou, China. Lirong Hao, PhD, Yang Zhou, MD, the First Affiliated Hospital of Harbin Medical University, Harbin, China. Suhua Li, MD, Yingping Sun, MM, The First Affiliated Hospital of Xinjiang Medical University, Urumchi, 21 China. Qiang He, PhD, Xiaogang Shen, MD, Zhejiang Provincial People’s Hospital, Hangzhou, China. Hongmei Liu, BSM, Xianghui Kong, MM, An Steel Group Hospital, Anshan, China. Fengmin Shao, PhD, Xiaogang Shen, MD, Henan Provincial People’s Hospital and The People’s Hospital of Zhengzhou University, Zhengzhou, China. Wei Li, PhD, Yang Wang, MPH, Fuwai Hospital, National Center for Cardiovascular Disease, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing, China. Ximing Sun, PhD, Degang Wang, PhD, School of Control Science and Engineering, Dalian University of Technology, Dalian, China.
Contributors: HL, JC and YL conceived and designed the study. JC, JW, YL, FG, GZ and WW were in charge of management of screened, collected, reviewed and recorded the data. JC, JW, YL and MH performed statistical analyses. JC, JW and YL drafted the manuscript. HL was the study guarantor. All authors provided critical revisions to the manuscript text. All authors reviewed and approved the manuscript of final version.
Funding: This work was supported by the special fund for National Health and Family Planning Commission (NHFPC) Scientific Research in the Public Welfare (201502023) and Dalian High-level Talents Innovation Support Program (2021RD01).
Competing interests: None declared.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review: Not commissioned; externally peer reviewed.
Data availability statement
Data are available on reasonable request.
Ethics statements
Patient consent for publication
Not applicable.
Ethics approval
This study involves human participants and was approved by The Ethics Committee of the First Affiliated Hospital of Dalian Medical University China, approval number (LUCK 2014-25). Participants gave informed consent to participate in the study before taking part.
References
- 1.Liyanage T, Ninomiya T, Jha V, et al. Worldwide access to treatment for end-stage kidney disease: a systematic review. Lancet 2015;385:1975–82.:S0140-6736(14)61601-9. 10.1016/S0140-6736(14)61601-9 [DOI] [PubMed] [Google Scholar]
- 2.Wang F, Yang C, Long J, et al. Executive summary for the 2015 annual data report of the china kidney disease network (CK-NET). Kidney Int 2019;95:501–5.:S0085-2538(18)30814-7. 10.1016/j.kint.2018.11.011 [DOI] [PubMed] [Google Scholar]
- 3.Zuo L, Wang M, Chinese Association of Blood Purification Management of Chinese Hospital Association . Current burden and probable increasing incidence of ESRD in china. Clin Nephrol 2010;74 Suppl 1:S20–2. [PubMed] [Google Scholar]
- 4.O’Hare AM, Choi AI, Boscardin WJ, et al. Trends in timing of initiation of chronic dialysis in the united states. Arch Intern Med 2011;171:1663–9. 10.1001/archinternmed.2011.436 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kessler M, Ayav C, Erpelding M-L, et al. Trends in the characteristics of ESRD patients at the initiation of dialysis. Nephrol Ther 2012;8:521–6.:S1769-7255(12)00030-2. 10.1016/j.nephro.2012.02.002 [DOI] [PubMed] [Google Scholar]
- 6.Cheng X, Nayyar S, Wang M, et al. Mortality rates among prevalent hemodialysis patients in beijing: a comparison with USRDS data. Nephrol Dial Transplant 2013;28:724–32. 10.1093/ndt/gfs326 [DOI] [PubMed] [Google Scholar]
- 7.Saran R, Robinson B, Abbott KC, et al. US renal data system 2019 annual data report: epidemiology of kidney disease in the united states. Am J Kidney Dis 2020;75(1 Suppl 1):A6–7.:S0272-6386(19)31009-1. 10.1053/j.ajkd.2019.09.003 [DOI] [PubMed] [Google Scholar]
- 8.Zhang L, Zhao M-H, Zuo L, et al. China kidney disease network (CK-NET) 2016 annual data report. Kidney Int Suppl (2011) 2020;10:e97–185. 10.1016/j.kisu.2020.09.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.SYSTEM U S R D . USRDS annual data report: epidemiology of kidney disease in the united states [R]. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 2021. [Google Scholar]
- 10.Berger JR, Jaikaransingh V, Hedayati SS. End-stage kidney disease in the elderly: approach to dialysis initiation, choosing modality, and predicting outcomes. Adv Chronic Kidney Dis 2016;23:36–43.:S1548-5595(15)00113-5. 10.1053/j.ackd.2015.08.005 [DOI] [PubMed] [Google Scholar]
- 11.Canaud B, Tong L, Tentori F, et al. Clinical practices and outcomes in elderly hemodialysis patients: results from the dialysis outcomes and practice patterns study (DOPPS). Clin J Am Soc Nephrol 2011;6:1651–62. 10.2215/CJN.03530410 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Saeed F, Arrigain S, Schold JD, et al. What are the risk factors for one-year mortality in older patients with chronic kidney disease? an analysis of the cleveland clinic CKD registry. Nephron 2019;141:98–104. 10.1159/000494298 [DOI] [PubMed] [Google Scholar]
- 13.Soler MJ, Montero N, Pascual MJ, et al. Age may explain the association of an early dialysis initiation with poor survival. QJM 2014;107:879–86. 10.1093/qjmed/hcu105 [DOI] [PubMed] [Google Scholar]
- 14.Liu Y, Wang L, Han X, et al. The profile of timing dialysis initiation in patients with end-stage renal disease in china: a cohort study. Kidney Blood Press Res 2020;45:180–93. 10.1159/000504671 [DOI] [PubMed] [Google Scholar]
- 15.Liu Y, Wang D, Chen X, et al. An equation based on fuzzy mathematics to assess the timing of haemodialysis initiation. Sci Rep 2019;9:5871.:5871. 10.1038/s41598-018-37762-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chen J, Liu Y, Chen X, et al. Assessment of dialysis initiation by a fuzzy mathematics equation (ADIFE): a study protocol for a randomised controlled trial. BMJ Open 2019;9:e023162. 10.1136/bmjopen-2018-023162 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Bouillanne O, Morineau G, Dupont C, et al. Geriatric nutritional risk index: a new index for evaluating at-risk elderly medical patients. Am J Clin Nutr 2005;82:777–83. 10.1093/ajcn/82.4.777 [DOI] [PubMed] [Google Scholar]
- 18.Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373–83. 10.1016/0021-9681(87)90171-8 [DOI] [PubMed] [Google Scholar]
- 19.Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604–12. 10.7326/0003-4819-150-9-200905050-00006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hwang S-J, Lin M-Y, Chen H-C, et al. Increased risk of mortality in the elderly population with late-stage chronic kidney disease: a cohort study in taiwan. Nephrol Dial Transplant 2008;23:3192–8. 10.1093/ndt/gfn222 [DOI] [PubMed] [Google Scholar]
- 21.Foley RN. Epidemiology and risk factors for early mortality after dialysis initiation. Semin Nephrol 2017;37:114–9.:S0270-9295(16)30131-0. 10.1016/j.semnephrol.2016.12.001 [DOI] [PubMed] [Google Scholar]
- 22.Saran R, Robinson B, Abbott KC, et al. US renal data system 2017 annual data report: epidemiology of kidney disease in the united states. Am J Kidney Dis 2018;71(3 Suppl 1):S0272-6386(18)30042-8. 10.1053/j.ajkd.2018.01.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wachterman MW, O’Hare AM, Rahman O-K, et al. One-year mortality after dialysis initiation among older adults. JAMA Intern Med 2019;179:987–90. 10.1001/jamainternmed.2019.0125 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Vandecasteele SJ, Kurella Tamura M. A patient-centered vision of care for ESRD: dialysis as a bridging treatment or as a final destination? J Am Soc Nephrol 2014;25:1647–51. 10.1681/ASN.2013101082 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Feng L, Jin AZ, Allen JC, et al. Timing of commencement of maintenance dialysis and mortality in young and older adults in singapore. BMC Nephrol 2017;18:176.:176. 10.1186/s12882-017-0590-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Crews DC, Scialla JJ, Liu J, et al. Predialysis health, dialysis timing, and outcomes among older united states adults. J Am Soc Nephrol 2014;25:370–9. 10.1681/ASN.2013050567 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lassalle M, Labeeuw M, Frimat L, et al. Age and comorbidity may explain the paradoxical association of an early dialysis start with poor survival. Kidney Int 2010;77:700–7. 10.1038/ki.2010.14 [DOI] [PubMed] [Google Scholar]
- 28.Park JY, Yoo KD, Kim YC, et al. Early dialysis initiation does not improve clinical outcomes in elderly end-stage renal disease patients: a multicenter prospective cohort study. PLoS One 2017;12:e0175830. 10.1371/journal.pone.0175830 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Zhao Y, Pei X, Zhao W. Timing of dialysis initiation and mortality risk in chronic kidney disease: a meta-analysis. Ther Apher Dial 2018;22:600–8. 10.1111/1744-9987.12721 [DOI] [PubMed] [Google Scholar]
- 30.Wright S, Klausner D, Baird B, et al. Timing of dialysis initiation and survival in ESRD. Clin J Am Soc Nephrol 2010;5:1828–35. 10.2215/CJN.06230909 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Chang JH, Rim MY, Sung J, et al. Early start of dialysis has no survival benefit in end-stage renal disease patients. J Korean Med Sci 2012;27:1177–81. 10.3346/jkms.2012.27.10.1177 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Zhang Y, Hu C, Bian Z, et al. Impact of timing of initiation of dialysis on long-term prognosis of patients undergoing hemodialysis. Exp Ther Med 2018;16:1209–15. 10.3892/etm.2018.6281 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.T CC, J BP, M DL, et al. Dialysis initiation, modality choice, access, and prescription: conclusions from a kidney disease improving global outcomes (KDIGO) controversies conference [J]. Kidney Int 2019. [DOI] [PubMed] [Google Scholar]
- 34.Beddhu S, Samore MH, Roberts MS, et al. Impact of timing of initiation of dialysis on mortality. J Am Soc Nephrol 2003;14:2305–12. 10.1097/01.asn.0000080184.67406.11 [DOI] [PubMed] [Google Scholar]
- 35.Locatelli F, Pozzoni P, Tentori F, et al. Epidemiology of cardiovascular risk in patients with chronic kidney disease. Nephrol Dial Transplant 2003;18 Suppl 7:vii2–9. 10.1093/ndt/gfg1072 [DOI] [PubMed] [Google Scholar]
- 36.Foley RN, Murray AM, Li S, et al. Chronic kidney disease and the risk for cardiovascular disease, renal replacement, and death in the united states medicare population, 1998 to 1999 [J]. J Am Soc Nephrol 2005;16:489–95. 10.1681/ASN.2004030203 [DOI] [PubMed] [Google Scholar]
- 37.Modi ZJ, Lu Y, Ji N, et al. Risk of cardiovascular disease and mortality in young adults with end-stage renal disease: an analysis of the US renal data system. JAMA Cardiol 2019;4:353–62. 10.1001/jamacardio.2019.0375 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Ma L, Zhao S. Risk factors for mortality in patients undergoing hemodialysis: a systematic review and meta-analysis. Int J Cardiol 2017;238:151–8.:S0167-5273(17)31125-7. 10.1016/j.ijcard.2017.02.095 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available on reasonable request.