Peritonitis episodes in peritoneal dialysis. A 5-year retrospective single center cohort study

ACHAIKI IATRIKI | 2020; 39(2): 72–77

Research Article

Adamantia Mpratsiakou1, Marios Papasotiriou1, Ioannis Kehagias2, Evdoxia Avramopoulou1, Theodoros Ntrinias1, Evangelos Papachristou1, Dimitrios S. Goumenos1

1University Hospital of Patras, Department of Nephrology and Kidney Transplantation, Patras, Greece
2University Hospital of Patras, Department of Surgery, Patras, Greece

Received: 25 March 2020; Accepted: 11 May2020

Corresponding author: Dimitrios S. Goumenos, Department of Nephrology and Renal Transplantation, University Hospital of Patras, Patras, Greece, Tel.: +30 2613 603366, Fax: +30 2610 994424, E-mail:

Key words: Peritoneal dialysis, peritonitis, CCPD



Background: Peritonitis is a major complication of peritoneal dialysis (PD) contributing to patient mortality and to treatment modality failure. Many factors have been associated with PD peritonitis, some of which include previous exit site infection, obesity and S. aureus nasal carriage. The aim of this study was to describe the nature of peritonitis, risk factors and outcomes in patients on PD in a large PD providing facility in Greece.
Methods: In this retrospective cohort study we included all patients receiving PD between 1/1/2015 and 31/12/2019. We examined peritonitis rate, causative organisms, clinical outcomes and differences between automated peritoneal dialysis (APD) and continuous ambulatory peritoneal dialysis (CAPD) and patients’ characteristics as risk factors on peritonitis incidence.
Results: The overall peritonitis rate was 0.31 episodes per patient, per year and the median peritonitis free survival was 57 months. The majority of peritonitis cases was caused by Gram-positive organisms. S. aureus was the most common organism; however, no methicillin resistant isolates were found. Continuous cycling peritoneal dialysis modality was independently associated with a higher peritonitis rate (χ2= 8.14, p=0.04), while female gender marginally correlated (χ2=4.66, p=0.05). Finally, the vast majority of our patients were hospitalized with a mean duration of 12.3 ± 8.2 days. No patient died due to peritonitis, while 2 patients were permanently transferred to hemodialysis.
Conclusions: The incidence of peritonitis in our center is low and the most common organism is S. aureus. The absence of methicillin resistance isolates is probably attributed to the absence of vancomycin from the initial antibiotic regimen.


Peritonitis is a serious complication in patients receiving peritoneal dialysis (PD), contributing to patient morbidity and mortality. Moreover, peritonitis is a significant cause of treatment modality failure and PD discontinuation with subsequent initiation of hemodialysis (HD) [1]. There are several predisposing factors related to PD peritonitis which can be classified into infection, dialysis, medical and environmental related risk factors, and are thoroughly addressed in the International Society of PD guidelines for peritonitis prevention and treatment [2]. Among those, the most important include previous exit-site infection and Staphylococcus aureus (S. aureus) nasal carriage, obesity, hypoalbuminemia and invasive interventions without prophylactic antibiotics administration, and finally improper training, the use of bio-incompatible fluids, prior hemodialysis and smoking. Risk factors, such as previous exit site infection, can easily be associated with subsequent peritonitis, nevertheless, other factors that have been repeatedly reported in large patient series, like hypoalbuminemia and depression have not an obvious pathophysiological connection to peritonitis and there are no data to show that treatment of these factors would reduce peritonitis rate [1-9].

Various registries and centers have reported findings regarding the microbiology of organisms causing peritonitis in their patient populations and have evaluated treatment and subsequent outcomes, including catheter removal and death [7, 10]. Apart from the international registries and surveys analyzing overall trends in PD-related peritonitis, knowledge of local peritonitis rates, microbiological profiles and antibiotic resistance patterns are essential to guide optimal local treatment practices and form clinical paradigms. In this retrospective analysis, we have used data collected from patients on PD during the last five years, from a tertiary hospital with the scope of re-assessing the aforementioned parameters.


Study population

This is a retrospective descriptive cohort study that included all patients who received PD in our center between 1/1/2015 and 31/12/2019. Data collection included demographics, cause of primary renal disease, comorbidities at the start of PD (heart failure, coronary disease, hypertension and diabetes mellitus), microbiology of peritonitis episodes (up to two organisms for polymicrobial episodes), the antibiotic treatment regimen and duration of hospitalization.

All PD-related peritonitis cases were diagnosed according to the International Society of Peritoneal Dialysis (ISPD) guidelines which recommend a diagnosis of peritonitis with at least two of the following: 1. Clinical signs and symptoms such as abdominal pain and/or cloudy effluent, 2. Dialysis effluent white cell count of more than 100 cells/μL after a dwell time of more than 2 hours and more than 50% of cells of polymorphonuclear type, 3. Positive dialysis effluent culture. Recurrent, relapsing and repeat peritonitis episodes were also labeled and accounted for according to the 2016 ISPD guidelines. Recurrent, is an episode that occurs within 4 weeks of completion of therapy of a prior episode but with a different organism; relapsing peritonitis is defined as an episode that occurs within 4 weeks of completion of therapy of a prior episode with the same organism or one sterile episode; whereas repeat peritonitis is defined as an episode that occurs more than 4 weeks after completion of therapy of a prior episode with the same organism. Relapsing episodes were not counted as another episode during the calculation of peritonitis rate [2].

All patients with peritonitis received at presentation a standard empiric double antibiotic regimen consisting of a 2nd generation cephalosporin (cefuroxime axetil) for covering Gram-positive strains and a 3rd generation cephalosporin (ceftazidime) for covering Gram-negative isolates in patients with well-preserved residual urine production or an aminoglycoside (amikacin) in patients with no residual kidney function. Antibiotic regimen was subsequently altered depending on the cultured organism and specific drug susceptibilities. No patient in our center received vancomycin as initial antibiotic treatment while there was no systematic use of topical exit site antibiotic prophylaxis. All patients were properly trained to clean the exit site with antiseptic agents and covered it with sterile dressing. Outcomes examined were catheter removal, transfer to HD and patient death.

Statistical analysis

Results are expressed as means ± standard deviations for continuous variables and frequencies and percentages for categorical variables. The Kolmogorov-Smirnov test was used to examine data distribution normality. Normally distributed continuous data were compared using student’s t-test and skewed continuous data with Mann-Whitney U test. The chi-square test was used in order to examine the relationship between potential predisposing factors and peritonitis incidence. Median peritonitis free survival was calculated using the Mantel-Cox test. For this calculation, the first peritonitis episode was considered as the end-point event, while patients that either did not have any peritonitis at the end of follow up or died or were transferred to hemodialysis or were lost to follow up before having a peritonitis episode, they were censored. All tests were 2-tailed and statistical significance was considered for p-values <0.05. All statistical analyses were performed using SPSS for Windows (version 16.0 SPSS Inc. Chicago II USA) and GraphPad Prism (version 5.00 for Windows, GraphPad Software, San Diego California USA).


Population characteristics and risk factors for peritonitis

During the five-year study period, 70 patients received PD at our treatment center and 45 episodes of peritonitis occurred in 26 patients. Six of these episodes were after tunnel infection. Two out of the 45 episodes were relapsing, whereas four were repeat episodes. No recurrent episodes were observed. The overall peritonitis rate was 0.31 episodes per patient, per year, significantly lower than the one advocated by the international society of peritoneal dialysis guidelines. The median peritonitis free survival was 57 months (range of 1st peritonitis event was between 2 and 68 months after PD initiation) (Figure 1). Patients baseline clinical characteristics and PD modality at the time of data collection are presented in Table 1.

Figure 1. Peritonitis-free survival for incident PD patients was calculated using the Mantel-Cox test. Patients were censored for death, transfer to HD or end of follow-up period. Time is expressed in months since initiation of PD.

The variable that was independently associated with peritonitis was CCPD modality (χ2= 8.14, p=0.04), while female gender marginally correlated (χ2=4.66, p=0.05). Other variables such as diabetes mellitus, hypertension, coronary disease, age over 65 years and continuous ambulatory PD (CAPD), nocturnal intermittent PD (NIPD) and intermittent PD (IPD) methods showed no significant correlation.

Microbiology of peritonitis

Microbiology results were available for all cases. Overall, 29 peritonitis episodes were culture positive (67.4%) and 14 culture negative (32,6%). Gram-positive organisms were the most common cause with 24 cases (55.8%). S. aureus was the most frequent isolated Gram-positive organism (8 cases) followed by Streptococcus spp. No methicillin resistant S. aureus isolates were found. Gram-negative organisms were isolated in 5 cases (11.6%), with Pseudomonas aeruginosa being the most isolated organism (2 cases). Fungal and mycobacterial peritonitis infections did not occur. Isolated microorganisms and resistance to antibiotics are presented in Tables 2, 3.

Hospitalization and outcomes

The vast majority of patients were hospitalized (93.3% rate of hospitalization). Mean duration of hospital stay was 12.3 ± 8.2 days. Patients with peritonitis from Gram-negative organisms had a marginally longer in-hospital stay (17.5 ± 5 vs 13.8 ± 8 days, p=ns). The number of cases were not sufficient to extract safe results concerning the correlation of microorganism’s Gram stain and catheter removal. Overall, in 8 cases the catheter was removed (17.8%), in 2 cases peritonitis was due to Gram-negative and in 6 cases due to Gram-positive organism. No patient died due to peritonitis while 2 patients were permanently transferred to hemodialysis.


In this study, we present data concerning peritonitis from a contemporary cohort of patients on PD. The main findings of this study are the low incidence rate of PD peritonitis, the high rate of S. aureus isolates that surpass that of coagulase negative isolates and the long duration of in-hospital stay of patients.

The incidence of peritonitis in our center was 0.31 episodes per patient-years  which is lower than the advocated rate from the ISPD, that is 0.5 episodes per patient-years [2]. This low incidence is probably due to patient education. In Greece, patient selection for PD is based on nephrologist and patient decision, as there is no national protocol. This selection is based on a combination of patient factors and local unit practices, while patient education is not protocol based either. As far as it concerns our center, a team of physicians and nurses specialized in PD, train and supervise each patient entering PD in a five-day program. According to Mujais et al, peritonitis rate in the USA was 0.367 per patient, per year and in Canada was 0.434 per patient, per year [10]. In our center, the vast majority of peritonitis cases were caused by Gram-positive organisms, with S. aureus being the most common. According to Kavanagh et al, who studied the peritonitis episodes in Scotland, CoN Staphylococcus was the most common organism, whereas S. aureus was the second one [7]. In our study no methicillin resistant isolates were found, whereas in Scotland 13.5% of S. aureus isolates were methicillin resistant. The results of Kan et al, who studied the peritonitis episodes in Australia, were similar to our center’s as S. aureus was the most common Gram-positive organism, followed by streptococcus spp. [11]. As for Gram-negative organisms, pseudomonas spp was the most common isolate, whereas in Mujais’s et al study E. coli, pseudomonas spp. and klebsiella spp. contributed equally in the US, whereas E. coli was the most common Gram-negative organism in Canada and Scotland [7, 10]. In our center, fungal and mycobacterial peritonitis infections did not occur, whereas fungal infections were observed in 3.92% of episodes in the US and in 3.66% of episodes in Canada [7, 10].

Failure to identify an organism is common and two case series reported an overall incidence reaching 20 to 40% of cases that were culture negative [12]. Culture-negative peritonitis is treated with empiric antibiotics and has a clinical course that is similar to that of coagulase-negative Staphylococcus infection. In our study, 32,6% of peritonitis episodes were culture negative, whereas in North America this rate was about 17% [10]. Finally, in our cohort, in 8 cases the catheter was removed (17.8%), in 2 cases peritonitis was due to Gram negative and in 6 cases due to Gram-positive organism. No patient died due to peritonitis, while two patients were permanently transferred to hemodialysis.

Although there are limited reports on peritonitis rate in PD from other centers in Greece, these are in abstract form and cover limited time period. Published data mostly concern interesting cases due to infrequent isolated strains and do not give a full perspective of the most common and important microbiology patterns that Greek nephrologist have to face. In a published large cohort from our dialysis center, that included 168 patients and 121 recorded peritonitis cases over a period of 12 years until 2004, the most common isolates were Gram positive organisms and coagulase negative staphylococci, while S. aureus was only detected once [13]. In comparison to these results, the microbiology pattern of peritonitis isolates of the current cohort has similarities as well as important differences. First of all, Gram-positive organisms are in both cohorts the primary causes of peritonitis, nevertheless, while in the late 90’s and early years after 00’s, coagulase negative staphylococci were the most common cause of peritonitis, 10 years later this pattern has changed in favor of S. aureus isolates [13]. This difference, although striking, is difficult to apprehend as PD patient training and all procedures have been kept the same throughout the years. An important measure for the prevention of S. aureus related peritonitis, which was previously not followed but has now already been implemented in our center, is the routine screening and eradication of S. aureus nasal carriage by intranasal mupirocin, early after catheter insertion, and the daily application of mupirocin cream to the catheter exit site. A common finding between the two different cohorts from our center is the high rate of culture negative peritonitis which reached 27% of cases in the late 90’s while currently there is a small increase to more than 30% of cases [13]. This extremely high rate of culture negative peritonitis, although not affecting curing rates, constitutes a problem of quality of care and its solution requires review and revision of effluent sampling and culturing methods in the microbiology department.

In our center the antibiotic regimen consists of a 2nd generation cephalosporin for Gram-positive strains and a 3rd generation cephalosporin or an aminoglycoside for Gram-negative strains, with no empiric coverage with vancomycin, which resulted in no methicillin resistant S. aureus (MRSA) isolates. However, the duration of hospital stay was long with an average duration of 12.3 ± 8.2 days. According to Kan et al, in Australia there is a high incidence of CoN Staphylococcus methicillin resistant and rising rates of MRSA PD peritonitis are likely because of high rates of MRSA carriage [11]. Moreover, in Scotland, 13.5% of the S. aureus peritonitis episodes were due to MRSA [7]. Absence of MRSA isolates in our cohort could be attributed to the standard antibiotic regimen used with no addition of vancomycin which is instead commonly used as a primary option in other cohorts and could contribute to the emergence of MRSA isolates. This treatment strategy thus, confers low rates of treatment failures with the advantage of long-term sustainability of no treatment resistant strains. The drawbacks of this therapeutic choice include both a long in-hospital stay and the subsequent increase of nursing workload and overall treatment costs. A solution to this would be the incorporation in our patient’s training module of the vancomycin enrichment protocol, that would enable them to stay at home, a policy that is already followed in large PD centers locally and abroad [2]. Nevertheless, we consider that this treatment strategy is the cause of the emergence of resistant strains which in turn lead to treatment failures, higher morbidity and in the end higher overall costs. In conclusion, it should be emphasized that empiric therapy should be guided by regional antibiotic resistance patterns but at the same time the appropriate use of antibiotics should aim at keeping the emergence of resistant microorganisms as low as possible [7, 11].

Finally, as far the risks factors for PD related peritonitis are concerned, in our cohort, CCPD as PD modality and probably female gender were associated with more episodes of peritonitis. On the other hand, diabetes was not shown to be a predisposing factor for peritonitis. Although different PD modalities have not been associated systematically with peritonitis, nevertheless, there are several risk factors that have been repeatedly related to peritonitis but are not mechanistically supported nor explained by either basic or clinical evidence. Thus, whether CCPD modality is a risk factor for peritonitis must be evaluated in larger series. In other cohorts, risk factors which predispose to increased peritonitis incidence include previous exit site infection, invasive interventions, diabetes, older age and BMI above 30 kg/m2 which in our cohort were not confirmed [2, 14].

In conclusion, in this regional cohort we have described the microbiology, treatment patterns and outcomes of peritonitis in PD patients of one of the largest PD centers in Greece. Overall, peritonitis in our center occurs at lower rates than those reported from countries such Australia and are similar to those observed in Western Europe and North America. Moreover, the chosen treatment options in our center has proven its efficacy with the advantage of very low emergence of resistant bacterial isolates but at the cost of longer in-hospital stay. Information regarding the outcomes of peritonitis based on causative organism and collection of further information on risk factors and interventions may help to further guide clinical practice.

Conflict of interest disclosure

All authors have no conflicts of interest regarding this work.

Declaration of funding sources

All authors received no financial support for this study.

Author contributions

AM, EA and TN gathered the clinical data; AM, MP and IK wrote the manuscript; IK inserted the peritoneal catheters; EP and D.S.G. organized and planned the study protocol and edited all the final versions and revisions of the manuscript; all authors provided final approval for the version to be submitted.


1. Cho Y, Johnson DW. Peritoneal dialysis-related peritonitis: towards improving evidence, practices, and outcomes. Am J Kidney Dis. 2014;64(2):278-89.
2. Li PK, Szeto CC, Piraino B, de Arteaga J, Fan S, Figueiredo AE, et al. ISPD Peritonitis Recommendations: 2016 Update on Prevention and Treatment. Perit Dial Int. 2016;36(5):481-508.
3. Pulliam J, Li NC, Maddux F, Hakim R, Finkelstein FO, Lacson E. First-year outcomes of incident peritoneal dialysis patients in the United States. Am J Kidney Dis. 2014;64(5):761-9.
4. Prasad N, Gupta A, Sharma RK, Sinha A, Kumar R. Impact of nutritional status on peritonitis in CAPD patients. Perit Dial Int. 2007;27(1):42-7.
5. Piraino B, Bernardini J, Brown E, Figueiredo A, Johnson DW, Lye WC, et al. ISPD position statement on reducing the risks of peritoneal dialysis-related infections. Perit Dial Int. 201;31(6):614-30.
6. McQuillan RF, Chiu E, Nessim S, Lok CE, Roscoe JM, Tam P, et al. A randomized controlled trial comparing mupirocin and polysporin triple ointments in peritoneal dialysis patients: the MP3 Study. Clin J Am Soc Nephrol. 2012;7(2):297-303.
7. Kavanagh D, Prescott GJ, Mactier RA. Peritoneal dialysis-associated peritonitis in Scotland (1999-2002). Nephrology, dialysis, transplantation: official publication of the European Dialysis and Transplant Association – European Renal Association. 2004;19(10):2584-91.
8. Cnossen TT, Usvyat L, Kotanko P, van der Sande FM, Kooman JP, Carter M, et al. Comparison of outcomes on continuous ambulatory peritoneal dialysis versus automated peritoneal dialysis: results from a USA database. Perit Dial Int. 2011 Nov-Dec;31(6):679-84.
9. Chaudhary K. Peritoneal Dialysis Drop-out: Causes and Prevention Strategies. Int J Nephrol. 2011;2011:434608.
10. Mujais S. Microbiology and outcomes of peritonitis in North America. Kidney Int Suppl. 2006;(103):S55-62.
11. Kan GW, Thomas MA, Heath CH. A 12-month review of peritoneal dialysis-related peritonitis in Western Australia: is empiric vancomycin still indicated for some patients? Perit Dial Int. 2003;23(5):465-8.
12. Bunke M, Brier ME, Golper TA. Culture-negative CAPD peritonitis: the Network 9 Study. Adv Perit Dial. 1994;10:174-8.
13. Fourtounas C, Savidaki E, Dousdabanis P, Hardalias A, Kalliakmani P, Papachristou E, et al. Peritonitis during the first year after commencement of peritoneal dialysis has an impact on technique survival and patient morbidity. Adv Perit Dial. 2006;22:50-4.
14. Ghali JR, Bannister KM, Brown FG, Rosman JB, Wiggins KJ, Johnson DW, et al. Microbiology and outcomes of peritonitis in Australian peritoneal dialysis patients. Perit Dial Int. 2011;31(6):651-62.