Peritoneal dialysis (PD)-related infections due to nontuberculous Mycobacteria (NTM) constitute a rare complication of PD, which is even rarer in children and adolescent, accounts less than 1% of peritonitis cases.1,2 Noteworthy is the fact that the incidence and prevalence rate of PD-related infections caused by NTM seems to have been increasing.3 The diagnostic problem for these specific infections based on the fact that can be misidentified as other pathogens including diphtheroids or Corynebacterium species, leading to delayed diagnosis.1 When suspected, there is a need for further evaluation requiring acid-fast bacilli stain and culture on specific media or automated system analysis even whole-genome sequencing (WGS). Species including Mycobacterium fortuitum, followed by Mycobacterium abscessus, Mycobacterium chelonae and Mycobacterium avium are the most common NTM isolates known to cause PD-related infections, both PD-related peritonitis or PD catheter-related infection such as exit-site infection (ESI) and tunnel infection.1,2,4 Problematic is also the appropriate management and treatment of these infections, especially with respect to the relocation of the exit site and catheter removal, which is remain unclear. However, in general two agents with in vitro activity against the clinical isolate of NTM infection for a minimum duration of 4 months, seems to be the appropriated treatment based on the ISPD recommendations.2

Data about the incidence, diagnostic methods, management, treatment and prognosis of NTM infections in children undergoing PD are scarce. It is noteworthy that there are no retrospective studies that include the full number of pediatric cases found in the literature, but usually include a smaller number of cases. Based on that, we conducted a thorough systematic review of literature focusing in PD-related infections caused by NTM in children and adolescents less than 18 years-old, because there are very limited data available. Accordingly, this literature review aimed to investigate in depth these rare and difficult to cure infections.

The search strategy yielded 910 scientific articles which were reviewed. After selection using PRISMA methodology, a total of 25 cases of PD-related infections caused by NTM in children and adolescents were identified in 19 articles and presented in Fig. 1.7–25 The total number of cases involved 23 patients. The first report of NTM peritonitis in a PD child appeared in 1983.7 Their mean (±SE) age was 7.35±5.63 years (range, 3 months–18 years), while, the most cases occurred in males (15/25, 60%) (Table 1). Twelve (48%) of the cases were reported from Japan, followed by 7 (28%) which were from USA and 1 case each was from Qatar, Singapore, Spain, India, Canada and Saudi Arabia, respectively.

Fig. 1.

PRISMA flow diagram of literature search, eligibility and inclusion process.

Diseases leading to ESRD were described in 23 of the 25 reviewed cases, in which was found that was secondary to congenital anomalies of the kidney and urinary tract (CAKUT) in 9 cases (39.1%), especially hypo- and or dysplastic kidney in 7 out of them. Twelve (52%) of 25 reported cases experienced peritonitis, 9 (36%) cases experienced ESI and the rest 3 (12%) cases suffered by both peritonitis/ESI.

Mean of vintage on PD in 19 cases with reported data was 16.89 months (range, 1 month–72 months), while in 11/19 (57.9%) vintage duration was equal or less than 12 months. Continuous ambulatory peritoneal dialysis (CAPD) was the commonest PD method used in 52% of cases. The clinical profile of PD-associated with NTM infections was non-specific, specifically, fever (52%), abdominal pain (44%), cloudy effluent (40%) were the most common manifestations. However, in 12 cases with PD-associated with NTM infections including ESI, either alone or combined with peritonitis, 91.7% of cases showed inflammatory skin lesions or were purulent.

Microbiological findings and management of 25 cases of PD–related infections caused by NTM presented in Table 2. The most prevalent organisms responsible for PD-associated with NTM infections was M.abscessus and M. fortuitum (28%), each, followed by M.chelonae (24%) and M. avium (16%) (Table 3). M. chelonae PD-associated infections occurred in children with a mean age of 12 months (range, 3 months–26 months), constituting the NTM pathogen which affects the youngest age group among patients reviewed. Only cases with confirmed peritoneal effluent or exit site tissue culture were included in this review. The pathogens were further verified by molecular methods (PCR) in 3 of the 25 cases, 2 cases by MALDI–TOF analysis and 1 case by WGS method.

In 23 (95.8%) of 24 cases with reported data, the PD catheter was removed and patients were given a novel PD catheter (25%) continuing PD methodology or most commonly switched to hemodialysis (75%). In 3 cases, a transition from PD to hemodialysis was necessitated and this was subsequently followed by successful renal transplantation, highlighting a favorable clinical course despite the complexity of their prior management.

All patient initially received empirical antibacterial treatment, which was subsequently modified in the majority of cases to ensure adequate coverage against NTM, following mycobacterial identification and susceptibility testing. The treatment duration ranged from 1 month to 12 months with a mean and median of the treatment, in cases with reported data, being 5.6 and 6 months, respectively. The anti-NTM regimen was selected according to specific isolates and/or in vitro anti-mycobacterial susceptibility. The most commonly used agents for the treatment of PD-associated with NTM infections included macrolides, aminoglycosides, fluoroquinolones, tetracyclines, carbapenems, oxazolidinones, rifampin, ethambutol and rifabutin. The outcome was favorable in all cases, with the sole exception of one patient who unfortunately died due to the infection. Notably, this particular case represented the first ever published report of PD-associated with NTM infections more than four decades ago.7

Our systematic review revealed 25 pediatric cases of PD-associated with NTM infections published to date, most of them reported as isolated case reports. M. abscessus and M. fortuitum were the predominant species in this cohort mirrors trends in both adult and pediatric populations.26,27 However, there appears to be a shift in recent years toward broader recognition of species such as M. chelonae and M. avium, partly due to improved identification techniques including MALDI–TOF and PCR.4 Diagnostic delays remain a major issue, with initial cultures often misidentified as diphtheroids or contaminated samples.1 Furthermore, the lack of clinical suspicion, especially in patients presenting with mild or atypical symptoms, such as localized exit-site inflammation without systemic signs, contributes to diagnostic inertia.

Differentiating PD-related infections caused by NTM from those due to conventional bacterial pathogens remains a significant clinical challenge but is essential for timely diagnosis and effective management. While both entities can present with non-specific symptoms such as fever, abdominal pain or cloudy effluent, a key distinguishing feature of NTM infections is their poor or absent response to empirical broad-spectrum antibiotics, which should raise early clinical suspicion in refractory cases.1,2 Diagnostic delays are common, as standard microbiological methods may fail to detect NTM or misidentify them as contaminants, underscoring the need for acid-fast staining and specialized culture techniques, or advanced modalities such as MALDI–TOF, PCR or WGS.4,27 Although epidemiological differences in age or PD vintage between NTM and common bacterial infections are not consistently pronounced, our review shows that NTM infections more frequently require PD catheter removal and a temporary shift to hemodialysis, suggesting a more complex clinical course.3,25,26 Taken together, these factors emphasize the importance of maintaining a high index of suspicion for NTM, particularly in cases unresponsive to standard treatment protocols, to ensure prompt diagnostic confirmation and implementation of appropriate therapy.

The reviewed pediatric cases revealed substantial variation in clinical presentation and therapeutic approaches. Peritonitis was the most frequent manifestation, although ESI were also common. In our study, the mean duration on PD prior to the onset of NTM infection was approximately 17 months, with a substantial proportion of cases occurring within the first year following the initiation of PD as renal replacement therapy. This suggests a potential role of early technique training, catheter care protocols and environmental exposure in disease pathogenesis.2,26 Fever, abdominal pain and cloudy effluent were common but not universal findings. Interestingly, most cases involving ESI were accompanied by skin lesions or purulent discharge, highlighting the importance of local signs in early suspicion of NTM.26

From a microbiological standpoint, culture positivity from peritoneal fluid or exit-site samples remains the cornerstone of diagnosis. However, newer techniques such as MALDI–TOF and PCR provide more rapid and accurate identification, especially in slow-growing or atypical isolates like mycobacteria.4 The use of WGS remains limited in routine clinical practice, but may become more accessible in tertiary centers in the near future and it will probably be the method of choice for the detection of these pathogens.

Treatment strategies varied considerably, both in terms of antibiotic selection and therapy duration. According to the ISPD guidelines, at least two antimycobacterial agents with in vitro activity against the isolate should be used, ideally for a minimum of four months.2 However, in practice as was found in our study, therapy duration ranged from a few weeks to several months and a wide variety of drug combinations, which were often empirical due to limited susceptibility data, were administered. In general, macrolides, fluoroquinolones, amikacin and linezolid were the most frequently used agents in literature, either alone or in combination, which are in accordance with our findings.25,26

Catheter removal was performed in the majority of cases, often after failure of initial empirical therapy. Early removal may be advisable, especially in tunnel or ESI, to prevent progression to peritonitis or systemic dissemination. However, the optimal timing of catheter removal and subsequent reinsertion remains poorly defined in the pediatric setting and should be individualized based on clinical response and microbial eradication. While catheter salvage is possible in select cases of localized ESI, relapsing infections and persistent inflammation frequently necessitate catheter replacement or a temporary switch to hemodialysis.1,4,27

The long-term prognosis of pediatric patients with NTM-related PD infections is generally favorable when appropriately treated, although some cases are associated with PD method failure and transition to hemodialysis. There are no clear recommendations on the optimal timing for reinitiating PD after catheter removal due to NTM infection and decisions are often individualized. Given the chronic nature of the underlying kidney disease in most of these children, preserving peritoneal membrane function and minimizing disruptions in dialysis modality are critical considerations.1,26

This study is limited by the small sample size, reflecting the rarity of PD-related infections due to NTM. The retrospective nature of case reports and the heterogeneity of data present significant challenges. This includes the infrequent reporting of specific peritoneal dialysis solution types in the included case reports, which prevented a comprehensive analysis of their potential influence on infection risk. Furthermore, incomplete reporting on treatment regimens, duration, and outcomes in some cases limits our ability to draw firm conclusions about optimal management strategies. Another emerging issue is the potential underreporting of cases. Given the difficulty in diagnosis and frequent misidentification, the true incidence of NTM-related PD infections may be underestimated, especially in resource-limited settings. Moreover, the inclusion of cases only with confirmed positive cultures may lead to selection bias in existing literature. However, despite these limitations, our study contributes valuable insights into an underreported condition and supports the need for heightened clinical suspicion and tailored microbiological evaluation in pediatric PD patients.

Notwithstanding the slowly growing number of case reports, there remains a lack of high-quality evidence and consensus on optimal management. The heterogeneity in treatment approaches highlights the need for standardized treatment protocols based on susceptibility data and pediatric pharmacokinetics. Future research should aim to establish prospective registries and multicenter collaborations to gather more robust data on incidence, outcomes and antimicrobial susceptibility. Until then, clinical suspicion, early diagnosis, aggressive therapy with appropriate agents and timely catheter management remain the cornerstones of successful outcomes.

The authors declare that they have no potential conflict of interest related to the contents of this article.