Peritoneal dialysis-associated peritonitis caused by Enterococcus avium
- 1 Nephrology Department, Centro Hospitalar de Lisboa Ocidental EPE Hospital de Santa Cruz, Carnaxide, Portugal
- 2 Nephrology Department, Hospital Doutor Nélio Mendonça, Funchal, Portugal
- Correspondence to Dr Rita Calça; arrcalca@gmail.com
Abstract
Peritonitis remains a common and serious complication of peritoneal dialysis. Peritonitis caused by gram-positive organisms includes coagulase-negative staphylococci, Streptococcus spp and Enterococcus spp. We present a rare case of peritoneal dialysis-associated peritonitis, where persisting abdominal pain and worsening laboratory findings despite antibiotic therapy led to the identification of Enterococcus avium, requiring Tenckoff catheter removal and temporary transfer to haemodialysis. The available literature reports only few cases where peritonitis is caused by this agent, underlining the need to consider atypical microbial agents when heterogeneous clinical course is presented.
Background
Peritoneal dialysis (PD)-associated peritonitis is a major cause of PD technique failure and conversion to long-term haemodialysis.1
Approximately 45%–65% of peritonitis cases are caused by gram-positive organisms, mostly coagulase-negative Staphylococcus.2 A recent study3 showed that coagulase-negative staphylococci caused 39% of PD-associated peritonitis, Streptococcus spp caused 13% and Enterococcus spp caused 4%. Enterococcus spp are habitual in the human bowel and urinary tract flora, but they can also be major pathogens causing urinary tract infections, bacteraemia and antibiotic-resistant nosocomial infections.4 Specifically, Enterococcus avium is an agent with usually low virulence and rarely associated with severe disease. We present a case with a severe and atypical course of peritonitis, leading to a demanding management of a patient with PD, where E. avium was found to be the causative agent.
Case presentation
A 37-year-old man first presented to our walk-in clinic for noticing turbid peritoneal effluent fluid, which was assumed to be PD-associated peritonitis. He was started on empiric intraperitoneal cefazolin and ceftazidime according to unit protocol (20 mg/kg/day) and was dismissed. Five days later, he was admitted to our ward complaining additionally of abdominal pain, nausea and vomiting. He denied diarrhoea or urinary symptoms. His medical history included chronic glomerulonephritis resulting in end-stage renal disease and hypertension. He was on PD for 5 years. His dialysis strategy consisted of four exchanges of 1800 mL dialysate per day and dialysis adequacy parameters were within the therapeutic target (Kt/V 2.24). He had an exit-site infection due to Proteus mirabilis and Serratia marcescens in May 2016 and tunnel infection without an identified agent in August 2018 (3 months before admission). Home medications included acetylsalicylic acid, alphacalcidol, perindopril, furosemide and sevelamer. He denied recent travels, pets, public bathing or swimming and was not on immunosuppressants.
On admission, his blood pressure was 122/66 mm Hg, heart rate was 103 beats/min, respiratory rate was 19 cycles/min and tympanic temperature was 38.1°C. His abdomen was diffusely tender with normal bowel sounds. Erythema and oedema over the subcutaneous pathway of the PD catheter was detected. Laboratory data showed haemoglobin of 101 g/L, white cell count of 9.3×109 cells/L, C reactive protein of 8.3 mg/dL, platelet count of 296×109/L, sodium of 138 mmol/L, potassium of 4.9 mmol/L and no acidosis. Peritoneal dialysate findings were compatible with peritonitis (white cell count of the effluent was 4.775 x 109/L with 91% predominant neutrophils) and were worse than on the first visit to walk-in clinic (0.302 x 109/L with 51% neutrophils). Tunnel soft tissue ultrasound showed pericatether abscess, while abdominal CT was normal. Due to these findings, antibiotic therapy was switched to intraperitoneal cefazolin (20 mg/kg/day) and gentamicin (0.6 mg/kg/day) plus oral ciprofloxacin (250 mg q12h). Despite widening of antibiotic spectrum, the patient’s clinical condition worsened with persisting abdominal pain and elevation of inflammatory laboratory findings (C reactive protein of 11.7 mg/dL and effluent white cell count 5.302 x 109/L with 96% predominant neutrophils). We ended up removing the PD catheter 4 days after admission. PD was temporarily suspended, switching to haemodialysis. Culture of the peritoneal effluent revealed E. avium, which was susceptible to ampicillin, vancomycin and gentamicin (with an elevated minimum inhibitory concentration (MIC)) and resistant to ciprofloxacin, ceftazidime and other cephalosporins. Targeted antibiotic regimen was started with intravenous ampicillin 1000 mg q6h and gentamicin (0,6 mg/kg q48h) and the patient’s clinical condition improved. He was discharged on intravenous vancomycin (target concentration 10–20 mg/L) and gentamicin q48h and completed a full-antibiotic course of 21 days.
Outcome and follow-up
On follow-up no signs of recurrent infection were identified and he was scheduled for Tenckoff catheter insertion after having completed antibiotics.
PD was fully re-initiated and he remained peritonitis free for 12 months, by which time he underwent a successful kidney transplant.
Discussion
Enterococcus spp have emerged as one of the major pathogens causing nosocomial and community acquired infections, including urinary tract infections, intra-abdominal infections, endocarditis, meningitis and bloodstream infections.4 5 From the spectrum of known Enterococci, there have been described 23 species capable of being pathogenic to humans, being E. faecalis and E. faecium the most common.5
E. avium, formerly known as group Q Streptococcus due to the presence of group Q antigen, was first identified in 1955 in human faeces samples4–7 and is rarely reported as pathogenic, being responsible for approximately 1% of infections in humans.3 8
PD-associated peritonitis caused by E. avium is a rare event. To our best knowledge, only four other cases were reported in the available literature.4 7 9 10 Furthermore, in a study with 1421 cases of peritonitis from a single-centre in Hong Kong, 29 (2%) were attributable to Enterococcus and only one of the cases was caused by E. avium.10
The most common pathogenic mechanism for PD-associated peritonitis is intraluminal contamination, which occurs if a suboptimal sterile technique is used when handling the system to perform exchanges. Other mechanisms include intra-abdominal bacterial translocation and haematogenous dissemination.2 The most probable mechanism for E. avium peritonitis might be from bowel translocation or urogenital tract contamination, as E. avium usually colonises these systems.1 7 Strengthening this assumption is the fact that most of the reported cases of E. avium PD-associated peritonitis had previous gastrointestinal disease.4 10 Ugur et al 9 reported a E. avium peritonitis in a 10-year-old girl with end-stage renal disease due to neurogenic bladder and hydrocephalus treated with a ventriculoperitoneal shunt. Remarkably, our patient reported no history of gastrointestinal or urinary disease. No other risk factors for E. avium peritonitis are described in the literature.
Enterococcal species are intrinsically resistant to cephalosporins and present high MIC levels of aminoglycosides and macrolides. Therefore, strains of enterococci that are resistant to ampicillin, vancomycin and carbapenems are increasingly common and, for this reason, the correct identification of the exact sensitivity to antibiotics is important.1 10
According to the protocol of our department, empirical treatment of PD-associated peritonitis consists in intraperitoneal ceftazidine plus cefazolin. The International Society for Peritoneal Dialysis (ISPD) guidelines for peritonitis prevention and treatment recommend that enterococcal peritonitis should be treated for 3 weeks with intraperitoneal vancomycin and adding intraperitoneal aminoglycoside in case of severe disease. For peritonitis due to vancomycin-resistant Enterococcus, guidelines suggest intraperitoneal ampicillin or other alternative antibiotics (such as linezolid, quinupristin/dalfopristin, daptomycin or teicoplanin) if the organism is ampicillin-resistant.1 In our patient, E. avium was susceptible to ampicillin, vancomycin and had high MIC for gentamicin. Our review of the literature showed similar susceptibilities among E. avium cases and no resistance to vancomycin has been reported.4 7 The aforementioned susceptibility patterns justify our empirical antibiotic choice failure, which demonstrate the importance of correct microbial identification and, such failure, should raise the suspicion for the presence of an atypical agent.
In our patient, although the sensitivity of gentamicin has a high MIC, we choose to use this antibiotic as it is the only aminoglycoside available in our centre to be administered intraperitoneal. We chose the strategy of using doses adjusted to the patient’s weight and measuring frequent serum gentamicin levels. Nevertheless, clinical improvement was only possible after Tenckoff catheter removal combined with directed antibiotic therapy. Adapa et al 7 also described a case of E. avium peritonitis that required catheter removal in addition to antibiotic therapy. Enterococcus peritonitis was found to be associated with an increased risk of catheter removal, permanent haemodialysis transfer and death in Australia and New Zealand Dialysis and Transplant (ANZDATA) Registry.11 They found also that clinical outcomes Enterococcus peritonitis were worse than outcomes in other gram-positive peritonitis and comparable to outcomes in gram-negative peritonitis.11 Yip et al showed that outcomes in Enterococcus peritonitis were similar to those in coagulase-negative staphylococci peritonitis.10
ISPD guidelines for peritonitis recommend that the PD catheter be removed promptly in episodes of refractory peritonitis, or earlier, if the patient’s clinical condition is deteriorating1 as we proceed on our patient. We think that this measure was crucial for the positive outcome of our case and to preserve the peritoneum for future PD.
Our patient received 21 days of antibiotics as recommended for enterococcal peritonitis, using vancomycin and gentamicin q48h (within haemodialysis sessions) after discharge to ensure antibiotic compliance.
It is recommended that the re-insertion of a new catheter be performed at least 2 weeks after catheter removal and complete resolution of peritoneal symptoms.1 In our particular case, we wait 3 weeks after complete resolution of the acute symptoms given the rarity of the causative agent.
As in all infections, prompt identification and timely initiation of antibiotics will prevent mortality, but in cases where such identification is delayed, widening antibiotic spectrum should prevent clinical deterioration and might be aided by removal of the source of infection, in our case the Tenckoff catheter.
Learning points
-
Peritoneal dialysis (PD)-associated peritonitis remains a major cause for technique failure.
-
Suspect rare causes of peritonitis or antibiotic resistance in patients with PD when the clinical improvement is not obtained after empirical therapy.
-
Delayed microbial identification might lead to difficult clinical course, culminating on removal of Tenckoff catheter.
-
Guideline-suggested empirical antibiotic therapy might not be adequate in cases of Enterococci infections.
-
Multiple drug-resistant Enterococci strains are emerging, making it difficult to initially manage such patient.
Footnotes
-
Contributors RC, FGdS contributed to data acquisition; RC contributed to paper writing; FGdS, ARM, PQB contributed to revision of manuscript; PQB contributed to supervision or mentorship.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Competing interests None declared.
-
Patient consent for publication Obtained.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2021. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer