Microbiology and Outcome of Peritoneal Dialysis-related Peritonitis in Elderly Patients: a Retrospective Study in Hunan Province, China

The microbiology and outcomes of peritoneal dialysis (PD) related peritonitis in elderly patients have not been thoroughly investigated. We aimed to investigate the microbiological distribution and clinical outcome in elderly patients with PD-associated peritonitis to guide clinical practice. We conducted a prospective, one center cohort study in Hunan province, China from September 1, 2014 to December 31, 2020. Among incident patients (n =279), basic clinical characteristics, pathogen distribution and prognosis of elderly PD patients (up to 65 years, n = 64) were compared with those of PD patients aged under 65 years (n = 215). The survival rate and technical survival rate were analyzed by Kaplan-Meier method. Among the 279 patients there were 394 peritonitis episodes, including 88 in elderly group, and 306 in younger groups. G+ bacteria were the main pathogenic bacteria in both groups (43.2% and 38.0%, respectively). Staphylococcus was the most common G+ bacteria. Peritonitis caused by fungal infection was signicantly higher in elderly patients (P = 0.01), however, there is no signicant difference in the proportions of G + bacteria and G- bacteria between the two groups. The most common G- bacteria was Escherichia coli. Interestingly, we found that Acinetobacter baumannii, polymicrobial infection and culture negative peritonitis in the elderly patients was signicantly higher than that in other patients. Additionally, elderly PD patients had higher peritonitis-related mortality (HR=7.27, P = 0.01). However, the technical survival rate was similar (P = 0.67) in both elderly and other patients. Taken together, this retrospective cohort study found that elderly patients had a higher probability of peritonitis caused by of Acinetobacter baumannii, fungi and polymicrobial infections. In addition, the elderly peritonitis patients had a higher risk of mortality. Understanding the characteristics of microbiology and clinical outcome in elderly patients will help us to take effective measures to reduce the incidence of elderly PD-associated peritonitis.

cohort study found that elderly patients had a higher probability of peritonitis caused by of Acinetobacter baumannii, fungi and polymicrobial infections. In addition, the elderly peritonitis patients had a higher risk of mortality. Understanding the characteristics of microbiology and clinical outcome in elderly patients will help us to take effective measures to reduce the incidence of elderly PD-associated peritonitis.

Background
Chronic kidney disease (CKD) is a serious public health problem, and its incidence is increasing year by year. With the aging of the world's population, the number of elderly patients with end-stage renal disease (ESRD) is increasing. Peritoneal dialysis (PD) is an important kidney replacement therapy. PD has been reported to protect residual renal functions better and other advantages in many studies 1 . Therefore, more and more ESRD patients, especially elderly patients choose to receive PD treatment 2 . However, peritonitis is still a common and serious complication of PD. PD-associated peritonitis is the major cause of death in 15% of patients on PD 3,4 . Moreover, a single episode of severe peritonitis or multiple peritonitis episodes usually leads to peritoneal ultra ltration failure and is the most common cause of switching to hemodialysis 5 . Whether elderly patients on PD have an increased risk of peritonitis? Here, we did a retrospective cohort study based on data from the Second Xiangya Hospital of Central South University. We analyzed 394 incident patients on PD between September 1, 2014 and December 31, 2020. Patients were separated into two age groups: up to 65 and over of 65 years old, in order to understand the characteristics of PD-related peritonitis in the elderly, and provide guidance for clinical prevention and treatment.

Materials And Methods
Study population 394 peritonitis episodes from the Second Xiangya Hospital of Central South University between September 1, 2014 and December 31, 2020 were included. We excluded patients who were under 18 years old and patients with no regular follow-up and incomplete data. Patients were divided into two groups: the elderly group (≥65 years old) and the young and middle-aged group (< 65 years old). Approval of study by research ethics committee of Central South University and informed consents were provided prior to their inclusion in the study.

The diagnosis of peritonitis
The diagnosis of PD-associated peritonitis follows the 2016 Guidelines of the International Peritoneal Dialysis Association: (1) clinical features consistent with peritonitis, i.e., abdominal pain or cloudy dialysis e uent; (2) dialysis e uent white cell count > 0.1 × 10 9 /L (after a dwell time of at least 2 hours), with .50% neutrophils; and (3) positive dialysis e uent culture . The diagnosis of PD-associated peritonitis requires any two of the following features 6 .

Relevant de nitions
Peritonitis related death was de ned as death of the patient due to peritonitis or death within 4 weeks from the onset of peritonitis. Cure: Peritonitis is completely relieved after treatment, and there is no need to continue anti-infection. Extubation: Because of peritonitis, the peritoneal dialysis cannot be continued and the peritoneal dialysis tube is permanently removed and transferred to hemodialysis. Multiple organism infection: refers to two or more pathogenic microorganisms in the peritoneal dialytic e uent.
Statistical analysis SPSS25.0 software was used for statistical analysis, mean ± standard deviation or median and range, as appropriate. Student's t-test was used for comparison between normal distribution data sets, and Wilcoxon rank-sum test was used for comparison between non-normal distribution data sets. Statistical data were presented by composition ratio or rate, Chi-square test or Fisher exact test were used for comparison, and Kaplan-Meier method was used to calculate cumulative survival rate and draw survival curve, and Log-rank method was used to test differences between groups. Hazard ratios (HRs) and 95% con dence intervals (CIs) for the incidence of PD-associated peritonitis were assessed using Cox regression analysis and interaction analysis with the confounding factors of sex, age and primary disease. Values of P < 0.05 was considered statistically signi cant.

Patients characteristics
Of the 279 patients included in the study, 64 were older than 65 years. The median ages at dialysis initiation were respectively 72 and 48 years for the elderly group and the other patients. Detailed information about patient characteristics is provided in Table 1. Among the 279 patients there were 394 peritonitis episodes, including 88 in the elderly group, and 306 in other groups. Until the occurrence of peritonitis, the average dialysis age of the elderly group was 36.98 ± 3.51 months, and the other patients were 30.29 ± 1.60 months. Chronic glomerulonephritis was the main primary disease in PD-peritonitis patients, followed by hypertension and diabetic nephropathy. Table 1 shows the different demographic and baseline clinical parameter between the two groups across age, that elderly patients had lower lower serum albumin, serum creatinine, uric acid, and serum phosphorus than the younger PD patients ( Table 1). caused by fungal infection was signi cantly higher in the elderly group than other patients (X2 = 6.55, P = 0.01), but there is no signi cant difference in the proportions of G + bacteria and G-bacteria between the two groups (P = 0.38, 0.18).
We further analyzed G + bacteria, and the results showed that staphylococcus was the most common G + organism in both groups, the proportions of staphylococcus, streptococcus and enterococcus in the two groups were not statistically signi cant (P = 0.12,0.71, 0.50). The most common G-bacteria was Escherichia coli, but the difference between the two groups was not statistically signi cant (χ 2 = 0.08, P = 0.77), while the proportion of Acinetobacter baumannii in the elderly group was signi cantly higher than that in other patients (χ 2 = 9.25, P = 0.002), as shown in Table 2. According to the results of pathogen culture, peritonitis episodes were divided into single G + bacteria, single G-bacteria, single fungi, mixed pathogenic infection and culture negative peritonitis. The proportion of single gram-positive bacteria and gram-negative bacteria in the two groups was similar. The infection rate of single fungus in the elderly group was higher than that of other patients, but the difference was not statistically signi cant(P = 0.14). The incidence of polymicrobial infection infection and culture negative peritonitis in the elderly group was signi cantly higher than that in other patients (χ 2 = 6.41, P = 0.01; χ 2 = 4.69, P = 0.03), as shown in Table 3. In terms of treatment outcome, the cure rates of the two groups were 72.70% and 80.10%, the extubation rates were 19.30% and 17.30%, and the mortality rates were 6.80% and 0.70%, respectively. The cure rate of peritonitis in elderly group was lower than that in other patients ,but the difference was not statistically signi cant (χ 2 = 2.18, P = 0.14), while the mortality of peritonitis in the elderly group was signi cantly higher than that in other patients(χ 2 = 12.521, P = 0.000402). There was no signi cant difference in extubation rate between the two groups, as shown in Table 4.

Survival time and technical survival time of peritonitis patients
Taking all-cause death of the patients as the end point, the survival time of the two groups were analyzed. Figure 1 showed the 1-year, 3-year, and 5-year survival rates of the elderly group were 92%, 89%, and 82%, respectively, while younger patients were 99%, 98%, and 94% respectively. We also performed COX regression analysis on survival time, further con rming that age is an independent factor that affects the survival outcome of patients with peritoneal dialysis-related peritonitis (p = 0.02). Compared with other patients, the risk of death after peritonitis in the elderly group is signi cantly higher, HR = 7.27 (95%CI: 1.45-36.45). With "removal of the peritoneal dialysis tube" as the end point, the technical survival rate of the two groups of patients was analyzed, and there was no signi cant difference between the two groups of patients (P = 0.67), as shown in Fig. 2.

Discussion
The present single-center retrospective cohort study investigated the microbiology and clinical outcomes of elderly PD patients, as compared with younger PD.The results showed that the infection rate of Acinetobacter baumannii in elderly PD patients was higher as compared to younger PD patients. And elderly patients were more prone to fungal infection or polymicrobial infection. In addition, the survival rate of elderly PD patients was inferior to those of younger patients. However, the technical survival rate was similar between the elderly and younger PD groups.
Peritonitis is a major complication of PD, and often resulting in the termination of peritoneal dialysis 7 . In this study, the main pathogenic microorganism of peritonitis in both elderly and younger patients were G + bacteria, accounting for 42.7% and 38.5% respectively. Staphylococcus predominated for PD-related peritonitis in both groups. This bacterial ora distribution and high incidence of Staphylococcus were similar to previous reports 8- 11 . And such results of the predominance of gram-positive bacteria were similar to the studies in America, Canada, Scotland, and Hong Kong [12][13][14][15] . The type of pathogenic microorganism usually indicates a possible cause of infection. Coagulase-negative staphylococcal peritonitis, especially those caused by Staphylococcus epidermidis, is mainly caused by contact contamination. Staphylococcus aureus infection is related to contact contamination or catheter infection. This suggests that PD patients should further strengthen aseptic operation concept, avoid contact infection.
G-bacteria are mostly related to intestinal infection and constipation 16 . In this study, the most common Gbacteria was Escherichia coli. The proportion of G-bacteria in the elderly group(28%) was higher than that in the younger group (22.2%), but the difference was not statistically signi cant. Nevertheless, we still need to take positive measures, including strengthening dietary guidance for elderly PD patients, paying attention to regulating intestinal ora, avoiding hypokalemia, improving constipation, and timely treating intestinal infections, so as to reduce the risk of G-bacteria infections in PD patients. Interestingly, we found that the proportion of Acinetobacter baumannii in elderly group was signi cantly higher than that in younger group. Several researchers proposed that Acinetobacter peritonitis often occurred with an immunocompromised status 17,18 . Since elderly patients often have weakened immune systems, this explains why they are more susceptible to Acinetobacter infection. Acinetobacter baumannii infections are reportedly associated with 10-20% higher attributable mortality and longer length of hospital stay 19,20 ,owing to their high antibiotic resistance rates 21 . This also explains the higher mortality rate among older patients in the study. Therefore, clinicians should pay more attention to the nutrition of elderly patients.
Compared with bacterial peritonitis, fungal peritonitis is rare in PD related peritonitis. representing only 1-12% of overall peritonitis in PD patients 22,23 , yet bringing terrible damage by higher rates of catheter loss, longer hospital stay, permanent transfer to HD, and mortality 6, 24 . For most fungal peritonitis cases, Candida species are the commonest pathogen, accountable for 70-90% of fungal peritonitis in adults 22,24 , where Candida albicans has been classically considered predominant. Our ndings were similar to those of these studies, and fungal infections were higher in elderly patients. Previous studies have shown that malnutrition, diabetes mellitus, prior use of antibiotics, immunosuppressed state, and prolonged time on PD are the main risk factors for fungal peritonitis 23,25 . The elderly often suffer from malnutrition, poor immune function and diabetes mellitus and so on. Therefore, the nutrition management of the elderly peritoneal dialysis patients should be strengthened and the complications should be treated actively.
The incidences of culture-negative peritonitis were 22.7% (20/88) in elderly patients, 35.0% (107/306) in younger patients. The relatively high culture-negative proportion in both groups may be primarily related to the early use of antibiotics prior to admission to our center. In addition, a small part of the reasons may be limited e uent culture technique, although the microbial culture technical in our hospital has been greatly improved in recent years. Therefore, we should follow the recommendations of the ISPD guidelines ,such as centrifugation of PD e uent, incubation in aerobic, microaerophilic and anaerobic environments, using antibiotic neutralisation bottle to further improve the positive rate of e uent microbial culture 6, 26 .
In terms of the clinical outcome of this study, the results showed that the death rates of elderly PD patients was higher than that of younger patients, this may have to do with elderly patients having more comorbidity, poorer performance, and were more frequently malnourished. However, the decannulation rate has no statistical signi cance between the two groups. Furthermore, the survival analysis of the two groups was carried out in this study, and the results showed that the overall survival rate of the elderly group was worse than that of the younger group; but the technical survival rate was similar between elderly and younger PD patients. Regardless of age, peritonitis was the leading cause of technical failure in PD patients. The incidence of peritonitis of elderly PD patients was higher than that of younger PD patients.
De Vecchi et al. 27 also reported that patient survival rate was poorer and the incidence of peritonitis was higher in the elderly PD patients than younger PD patients, whereas the technical survival was similar between the two groups. Lim WH et al. 28 reported that the hazard ratios for technical failure were similar across the age groups despite higher risk of peritonitis-related mortality. Hyunsuk K et al. 29 likewise reported inferior patient survival in elderly PD than in younger PD patients, but similar technical survival; and additional studies similarly showed no differences in the technical survival rates between elderly and younger PD patients [30][31][32] .
The main limitation of our study was single-center, retrospective and relatively small sample size of elderly PD patients. Moreover, the risk factors of peritoneal dialysis-related peritonitis in elderly patients were not further analyzed in this study. Therefore, a multi-center, large sample size, long-term, prospective study is needed in the future to con rm our ndings.
To sum up, this retrospective cohort study found that elderly patients had a higher probability of peritonitis caused by of baumannii, fungi and multiple microorganisms mixed infections. In addition, the elderly peritonitis patients had a higher risk of death. Understanding the characteristics of microbiology and clinical outcome in elderly patients will help to take effective measures to reduce the incidence of PDassociated peritonitis. Figure 1 Patient survival according to elderly patients and younger patients with peritoneal dialysis-associated peritonitis. Log-rank test χ2 =7.867, p=0.005. And COX regression analysis further con rmed that age is an independent factor that affects the survival outcome of patients with peritoneal dialysis-related peritonitis HR=7.27 (95%CI: 1.45-36.45), p=0.02.

Figure 2
Technique survival according to elderly patients and younger patients with peritoneal dialysis-associated peritonitis. Log-rank test χ2 =0.036, p=0.849.