Peritoneal dialysis-associated infection caused by Mycobacterium abscessus in a pediatric patient on continuous peritoneal dialysis without switching to hemodialysis

Abstract

Most peritoneal dialysis (PD)-associated infections caused by Mycobacterium abscessus (M. abscessus) require a transfer from PD to hemodialysis (HD). Here, we report a pediatric case of exit-site and tunnel infections caused by M. abscessus, for whom PD was continued with catheter replacement, debridement of the infected site, and the administration of multiple antibacterial agents. A 10-year-old boy with end-stage kidney disease secondary to juvenile nephronophthisis with NPHP1 deletion, for whom PD was initiated at the age of 9 years, was admitted to the hospital with complaints of fever, pus at the exit-site of the PD catheter, and poor PD drainage. The dialysis effluent culture results were negative; however, M. abscessus was detected in the pus at the exit-site of the PD catheter. The management of HD was expected to be challenging owing to the presence of developmental disorders. Therefore, PD was continued with the simultaneous removal of the PD catheter, reinsertion of a new catheter at a new site, and debridement of the infected site. Multiple antibacterial therapies were administered for 2 months, and the patient was eventually discharged without switching to HD. To the best of our knowledge, this is the first pediatric case of a PD-associated infection caused by M. abscessus, for whom PD was continued without switching to HD. This treatment strategy is not generally recommended but may be an option for patients without peritonitis who have difficulty switching to HD.

Introduction

Mycobacterium abscessus (M. abscessus) belongs to a rapidly growing group of non-tuberculous mycobacteria (NTM) transmitted via water and soil. Peritoneal dialysis (PD)-associated infections caused by M. abscessus are rare and difficult to treat owing to the high drug resistance and lack of clear treatment strategies [1, 2]. Therefore, the removal of PD catheter and conversion to hemodialysis (HD) are often necessary [3]. Herein, we describe a case of M. abscessus exit-site infection (ESI) and tunnel infection (TI) in a pediatric patient who was successfully treated without conversion to HD.

Case report

A 10-year-old boy with developmental disorders visited our hospital complaining of pus discharge, redness, and swelling around the exit-site of a PD catheter for 2 days. The patient was diagnosed with end-stage kidney disease secondary to juvenile nephronophthisis with NPHP1 deletion and started PD at 9 years of age. The chlorhexidine gluconate was used for exit-site care. The patient had no fever, abdominal pain, or cloudy effluent in the PD; therefore, we suspected an ESI and administered an oral antibiotic (cefaclor; 30 mg/kg body weight per day). However, the patient developed fever and poor PD drainage on the same day and was admitted to our hospital on the following day. Upon admission, the vital signs of the patient were as follows: blood pressure, 130/68 mmHg; pulse, 120 bpm; and temperature 38.1 ℃. Physical examination revealed redness around the exit-site of the PD, pus discharge, and tenderness in the subcutaneous tunnel. Rebound tenderness or muscle guarding was not observed in the abdomen. Blood tests revealed a white blood cell count of 5900/µL and C-reactive protein level of 0.33 mg/dL. The effluent in the PD was not cloudy, and the white blood cell count in the effluent was 36/µL (mononuclear cells: 13/µL, polymorphonuclear cells: 23/µL). Radiographic examinations revealed no abnormal positioning of the PD catheter.

Cefazolin (CEZ; 25 mg/kg body weight per day) was administered as the first-line treatment for the ESI. On the fourth day of hospitalization, the patient became afebrile; however, pus discharge from the exit-site persisted. M. abscessus was detected in the pus collected during the first visit, and the culture of the dialysis effluent was negative. Based on the results of drug susceptibility testing for M. abscessus (Table 1), on the eighth day, we changed the antibiotics to intravenous amikacin (AMK; 5 mg/kg body weight every 3 days), imipenem/cilastatin (IPM/CS; 8 mg/kg body weight per day), and oral clarithromycin (CAM; 4 mg/kg body weight per day). The AMK dosage was adjusted using therapeutic drug monitoring. As poor PD drainage persisted after admission, we suspected obstruction of the PD catheter by the greater omentum. Nevertheless, the management of HD was anticipated to be challenging, because the patient had developmental disorders. Thus, we planned to remove the PD catheter, perform debridement of the infected site, and reinsertion a new catheter at a new site to continue the PD. On day 14 of hospitalization, the PD catheter was simultaneously removed and reinserted, and the greater omentum was resected. A large amount of pus was drained from the exit-site, and debridement was performed (Fig. 1). M. abscessus was detected on the subcutaneous cuff of the catheter and surrounding tissue; however, it was not detected in the deep cuff, catheter tip, or intraperitoneal fluid. Therefore, PD was resumed on postoperative day 3. Antibacterial treatment was continued for 2 months, and the patient was discharged on day 78 without any recurrence. The patient did not experience recurrence of M. abscessus infection for approximately 5 years (Fig. 2). After treatment, the patient’s residual renal function (RRF) decreased, possibly because of long-term AMK administration. No other side effects, such as ototoxicity, were observed.

Table 1.

Antibiotic susceptibility of Mycobacterium abscessus isolates

Antibiotics	MIC (µg/mL)	Interpretation
Amikacin	16	S
Cefoxitin	≥ 256	R
Ciprofloxacin	≥ 4	R
Clarithromycin	1	S
Doxycycline	≥ 256	R
Imipenem/cilastatin	8	I
Linezolid	≥ 256	R
Meropenem	≥ 16	I
Moxifloxacin	≥ 32	R
Trimethoprim/sulfamethoxazole	≥ 80	R
Tobramycin	≥ 16	R

I intermediate, MIC minimum inhibitory concentration, R resistant, S susceptible

Fig. 1.

Photographs of the exit-site of peritoneal dialysis catheter. a Before surgery. b During the surgery, a large amount of pus was discharged from the peritoneal dialysis catheter insertion site

Fig. 2.

Clinical course of the patient. AMK amikacin, CAM clarithromycin, CCL cefaclor, CEZ cefazolin, IPM/CS imipenem/cilastatin, M. abscessus Mycobacterium abscessus

Discussion

PD-associated infections caused by M. abscessus often cause difficulties for patients to continue PD. The guidelines of the International Society for Peritoneal Dialysis (ISPD) recommend the removal of the PD catheter in NTM peritonitis [1], and Chamarthi et al. recommended catheter removal and temporary conversion to HD for ESI caused by M. abscessus [4]. Although there are only a few case reports on pediatric cases of PD-associated infections due to M. abscessus, all patients underwent PD catheter removal, with or without peritonitis, and conversion to HD [5–8]. In a study of 28 patients with PD-associated infections caused by M. abscessus [2], 14 patients with peritonitis underwent catheter removal and conversion to HD. In other reports, catheter removal and insertion were performed simultaneously owing to ESI/TI in three patients [3, 4, 9]; however, all patients developed peritonitis and subsequently underwent HD. In contrast, another article reported that PD was not interrupted in two adult patients [10]; one patient with ESI was treated with multiple antibacterial treatments, and the other patient with ESI/TI was managed with multiple antibacterial treatments and debridement of the infected site. Therefore, in general, the catheter should be removed, along with temporary conversion to HD, in patients with ESI caused by M. abscessus; however, PD may be continued if appropriate treatment is initiated before the progression to peritonitis. Song et al. reported that approximately 30% of patients with PD with peritonitis due to NTM required more than 1 month for treatment initiation [11], suggesting that the delay in identifying the causative organism may contribute to the progression to peritonitis and refractoriness of the treatment. In our patient, prompt treatment was initiated, which may have prevented progression to peritonitis. The ISPD guidelines recommend the daily application of antibacterial ointments to the exit-site [12]; however, some reports suggest that exit-site care with gentamicin may cause NTM infection [10, 13]. In the present case, only chlorhexidine gluconate was used for exit-site care. In addition, depending on the severity and infection site, debridement or incisional drainage may help prevent progression to peritonitis [14]. Consistent with this recommendation, we performed debridement of the infected site at the time of PD catheter replacement in the present case.

Currently, there is no consensus on the optimal antibacterial agents, their combination, and the duration of treatment for M. abscessus infections [3]. The most commonly used antibacterial agents are AMK, IPM, cefoxitin, and macrolides, which are often susceptible to these agents; however, their susceptibility patterns vary depending on the infection site and other factors. Therefore, multidrug therapy based on the results of susceptibility tests is the mainstay of treatment [3, 15]. The duration of antibacterial treatment varies according to the infection site and typically requires prolonged administration for at least several months [2, 3, 11, 14]. In our case, the patient became afebrile after using CEZ, which was ineffective for M. abscessus; however, a large amount of pus was drained at the time of catheter replacement. Although clearly identifying the cause of fever was difficult, the patient had a mild cough on admission; therefore, the cause might have been an upper respiratory tract infection rather than ESI. After the results of drug susceptibility testing for M. abscessus were confirmed, we administered the patient with AMK, IPM/CS, and CAM for 2 months based on the previous reports [10, 11].

In our case, the RRF decreased after the treatment course, possibly owing to prolonged AMK administration. AMK-induced nephrotoxicity is often transient, whereas a decrease in RRF may result from the presence of risk factors for AMK-induced nephrotoxicity, such as renal dysfunction or long-term administration [14]. Several other cases of PD-associated peritonitis caused by M. abscessus have been reported; one case was treated by changing AMK to clofazimine [6], and the other case was treated with antibacterial therapy for only 3 weeks [7]. Further studies are needed on the selection of appropriate antibiotics and duration of administration, considering the side effects of treatment.

To the best of our knowledge, this is the first reported case of ESI/TI caused by M. abscessus in a pediatric patient who was able to continue PD without conversion to HD. Our patient had a developmental disorder, which was expected to be challenging to manage with HD. Considering that peritonitis had not yet developed, we decided to remove and reinsert the PD catheter simultaneously with surgical debridement to continue PD. Multiple antibacterial agents were used based on the results of the drug susceptibility testing. Our report suggested that PD catheter replacement, debridement of the infected site, and long-term administration of multiple antibacterial agents based on drug sensitivity testing before the development of peritonitis are the main factors that enabled PD continuation without conversion to HD. Notably, the fact that a pediatric patient who had difficulty managing HD was treated successfully while continuing with PD was a significant finding.

In conclusion, we report the first pediatric case of PD-associated ESI/TI caused by M. abscessus, for whom PD was continued without conversion to HD. M. abscessus infection is difficult to treat, and the ISPD guidelines or previous reports recommend catheter removal and transition to HD [1, 12]. The treatment strategy described in our report is not generally recommended but may an option for patients without peritonitis who have difficulty in switching to HD.

Funding

The authors received no financial support for this study, authorship, or publication.

Declarations

Conflicts of interest

The authors declare no conflicts of interest.

Research involving human participants

All procedures performed in studies involving human participants were in accordance with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Informed consent

Written informed consent was obtained from the legal guardians for the publication of the case report with anonymized patient information.