Abstract
Listeria monocytogenes is a rare cause of peritoneal dialysis-related peritonitis. Only a handful of cases have been reported, and the optimal management is still uncertain. We present a case of Listeria monocytogenes peritonitis and perform a review of the literature to elucidate optimal antibiotic therapy.
1. Introduction
Listeria monocytogenes is a Gram-positive bacillus that causes self-limiting disease in healthy subjects [1]. Risk factors for invasive infection include immunocompromised states, old age, malignancy, liver cirrhosis, pregnancy, diabetes mellitus, and alcoholism [2]. Rarely, Listeria monocytogenes can cause peritoneal dialysis- (PD-) related peritonitis. Optimal management of Listeria peritonitis in PD patients is uncertain.
2. Case Report
We present a case of a 49-year-old male patient with end-stage renal disease due to lupus nephritis who had undergone continuous ambulatory PD for 1 year, without previous episodes of peritonitis. He had concomitant ischemic heart disease and was taking prednisolone 10 mg daily and aspirin 80 mg daily. His dialysis prescription was three 2 L exchanges of 2.5% dextrose bags per day with a calcium concentration of 3.5 mEq/L (Ultrabag, Baxter International Inc.). He achieved an average weekly KT/V of over 1.7, and peritoneal equilibration testing showed that he was a “high” transporter. He produced 500 ml of urine a day. He presented to us in August 2015 with fever, abdominal pain, and turbid dialysis effluent. Physical examination revealed a blood pressure of 140/80 mmHg and a pulse rate of 90 beats/minute. His body surface area was 1.61 square meters. The abdomen was diffusely tender. The exit site was clean and healthy-looking.
Blood work on admission was unremarkable for liver function, electrolytes, and blood counts. Microscopic examination of the effluent revealed more than 1000 leucocytes/mm3 (our laboratory did not give absolute values for counts over 1000), predominantly neutrophils (90%). Gram staining of the effluent did not reveal organisms. Blood culture was negative.
Empirical cefazolin (1 g/2 L dialysate loading and then 250 mg/2 L dialysate three times per day) and gentamicin (40 mg/2 L dialysate daily) were given via the intraperitoneal (IP) route. Owing to a lack of clinical response on as late as the third day of admission, empirical IP vancomycin (2 g/2 L dialysate, or 35 mg/kg body weight/2 L dialysate, once weekly), IP ceftazidime (1 g/2 L dialysate loading and then 250 mg/2 L dialysate three times per day), and IP amikacin (120 mg/2 L dialysate, or 2 mg/kg/2 L dialysate, once daily) were given to achieve broader antimicrobial coverage. In addition, fungal prophylaxis was given in the form of oral nystatin syrup at a dose of 500,000 units four times a day. Vancomycin levels were not checked. Clinical remission was achieved two days later as evidenced by clearing of the effluent and the disappearance of leucocytes on microscopy. Later, effluent bacterial culture came back with Listeria monocytogenes, for which the minimum inhibitory concentration of penicillin was 0.5 μg/ml. Our microbiology laboratory reported susceptibility to only penicillin for Listeria strains.
Accordingly, the antibiotic regime was changed to IP ampicillin (250 mg/2 L dialysate three times per day) and amikacin for 4 weeks despite clinical resolution at the time of starting IP ampicillin.
3. Discussion
Peritoneal dialysis-related peritonitis caused by Listeria monocytogenes is rare. To date, only 19 cases have been reported (Table 1) [3–20]. A high index of suspicion is required for timely diagnosis, especially in patients who are immunocompromised or have other risk factors [2]. As regards to treatment for Listeria infections in general, ampicillin is considered first-line therapy [21]. Trimethoprim/sulphamethoxazole can be considered if contraindications to ampicillin exist, such as allergies [3, 22]. Adding an aminoglycoside can enhance bacterial clearance because ampicillin alone is bacteriostatic [23]. Cephalosporins are generally ineffective [24]. While vancomycin is effective against Gram-positive bacteria, its limited ability to penetrate the eukaryote cell membrane makes it ineffectual against intracellular bacteria, such as Listeria species. Indeed, vancomycin failure in the context of Listeria monocytogenes PD-related peritonitis has been widely reported: In 6 of the reported cases where vancomycin was part of the primary regime, the only case that ended in resolution used amoxicillin as well. Vancomycin failed to clear the infection in the remaining 5 cases, even when used in combination with an aminoglycoside, a cephalosporin, or trimethoprim-sulphamethozazole (Table 1). On the other hand, the reported cases are strongly in favour of ampicillin/amoxicillin as either primary or salvage therapy: intraperitoneal ampicillin/amoxicillin with or without an aminoglycoside resulted in cure in 7 of the reported cases. Oral or intravenous ampicillin or amoxicillin resulted in cure in 5 cases. It is for the abovementioned reasons that we substituted ceftazidime and vancomycin with ampicillin while keeping the amikacin despite clinical resolution of the peritonitis following therapy with vancomycin, ceftazidime, and amikacin.
There is some concern regarding the use of ampicillin through the intraperitoneal route, as some authors have shown that ampicillin has little in vitro antibacterial activity against Enterococci when added to common peritoneal dialysates [25]. In another study, Szeto et al. showed that oral amoxicillin could effectively treat Enterococcal PD-related peritonitis [26]. Whether this holds true for Listeria monocytogenes peritonitis remains to be proven.
In conclusion, Listeria monocytogenes PD-related peritonitis should be treated with an ampicillin- or amoxicillin-based regime with or without the addition of an aminoglycoside. While the majority of reported cases used the intraperitoneal route of administration, it is as yet uncertain whether oral or intravenous administration of ampicillin or amoxicillin might be an equal or better alternative.
Data Availability
Underlying data supporting the results of our study can be obtained by emailing the corresponding author of this article.
Conflicts of Interest
All the authors declare no conflicts of interest.