Biopsy-confirmed Mycobacterium kansasii peritonitis with CT peritoneography in a young PD patient: a case report

Abstract

Background

Mycobacterium kansasii is an exceptionally rare cause of peritoneal dialysis (PD)-related peritonitis and may present with subtle, indolent features that delay recognition. This case illustrates how “silent” atypical peritonitis can be unmasked by pairing contrast-enhanced computed tomography (CT) with CT peritoneography: the former identified nodular biopsy targets, while the latter revealed loculated dialysate flow and failure of free intraperitoneal circulation.

Case presentation

A 20-year-old male had received automated PD for 3 years and was being evaluated for combined heart–kidney transplantation because of severe cardiomyopathy. He presented with vague abdominal discomfort, declining ultrafiltration, mild pedal edema, and weight loss. PD effluent (PDE) showed mononuclear-predominant leukocytosis (49–150 cells/µL), but repeated PDE cultures, AFB staining, and molecular testing were negative. Contrast-enhanced CT revealed multiple peritoneal nodules and omental stranding; CT peritoneography exposed a trapped contrast-containing pocket, providing functional evidence of impaired dialysate circulation. Ultrasound-guided biopsy showed necrotizing granulomatous inflammation, and tissue culture subsequently grew M. kansasii with rifampicin resistance. Because the isolate was also resistant to moxifloxacin and ciprofloxacin, fluoroquinolone-based therapy was avoided. Guided by susceptibility testing, we treated him with isoniazid, azithromycin, ethambutol, and amikacin, and the PD catheter was removed for refractory atypical mycobacterial peritonitis with progressive catheter dysfunction, prompting transition to hemodialysis. Symptoms and inflammatory markers resolved within 6 weeks; however, antimycobacterial therapy was curtailed at 165–179 days when the patient sustained an out-of-hospital sudden cardiac arrest of undetermined cause, in the setting of his severe underlying cardiomyopathy.

Conclusions

This case adds to the limited literature on M. kansasii PD peritonitis and describes an imaging-guided diagnostic pathway in which contrast-enhanced CT localized peritoneal lesions, CT peritoneography demonstrated impaired dialysate circulation, and targeted biopsy enabled tissue diagnosis. Susceptibility-guided therapy and catheter removal were followed by clinical control of the infection, although definitive cure could not be established because follow-up ended with the patient’s death. As a single case, these observations are hypothesis-generating and should not be interpreted as establishing diagnostic criteria or standard management.

Introduction

Peritoneal dialysis (PD)-related peritonitis remains a major challenge in sustaining high-quality, long-term PD therapy. Early diagnosis and prompt treatment are critical to minimizing complications, preserving peritoneal membrane integrity, and avoiding technique failure or transfer to hemodialysis. However, atypical infections can evade early recognition—especially when clinical signs are subtle and standard diagnostic thresholds are not met. Nontuberculous mycobacteria (NTM) are rare causes of PD peritonitis, accounting for less than 1% of cases [1–4], with a population-based incidence of approximately 1 per 1000 PD patient-years [5]. Among these, Mycobacterium kansasii is particularly uncommon, with only a handful of cases reported since the first description in 1992 [6, 7]; in a systematic review of 57 reported NTM PD peritonitis cases, only 3 (5.3%) were attributable to M. kansasii [1]. Unlike bacterial peritonitis, M. kansasii is characterized indolent, diagnostically elusive, and refractory to empirical antibacterial therapy, often requiring prolonged multidrug treatment with catheter removal [8]. Recent systematic and scoping reviews confirm that PD-associated NTM peritonitis remains rare, is dominated by rapidly growing species, and is frequently complicated by diagnostic delay, catheter loss, and transfer to hemodialysis [9, 10].

This case illustrates the diagnostic complexity of M. kansasii peritonitis in a young PD patient with vague abdominal symptoms and low-grade inflammatory findings in the PD effluent (PDE). Although peritoneal biopsy or tissue-based diagnosis has been reported previously in NTM PD peritonitis, this case highlights a distinct imaging-guided diagnostic pathway. Contrast-enhanced computed tomography (CT) identified peritoneal nodules that served as targets for ultrasound-guided biopsy, while CT peritoneography demonstrated loculated dialysate flow and failure of free intraperitoneal circulation. The case is further distinguished by rifampicin resistance reported through phenotypic susceptibility testing. Together, these features highlight important diagnostic and therapeutic considerations in atypical PD peritonitis.

Case presentation

A 20-year-old Thai male on automated peritoneal dialysis for 3 years due to kidney failure from IgA vasculitis (formerly Henoch–Schönlein purpura) presented with anorexia, vague abdominal discomfort, reduced ultrafiltration (declining to ~ 100–200 mL/day), and mild (1+) bilateral pedal edema, with weight loss from 53 to 49 kg over 2 months. He denied fever, or cloudy PDE. He was unemployed with no occupation-related environmental exposure, had no prior PD-related peritonitis or recent antibiotic use, and was not receiving immunosuppressive therapy; the exit site was clean without discharge, erythema, or tunnel tenderness, so no exit-site culture was obtained. His history was notable for severe nonischemic cardiomyopathy (left ventricular ejection fraction 9%) and WHO Group 2 pulmonary hypertension. He was under evaluation for combined heart–kidney transplantation. His home regimen alternated between nocturnal intermittent PD (10 h, 5 cycles, 1.8-L fill, low-calcium 1.5% and 2.5% dextrose) and CAPD (two, 1.8-L exchanges of 2.5% dextrose).

Routine PDE analysis showed mononuclear-predominant leukocytosis fluctuating around the ISPD threshold of 100 cells/µL (49–150 cells/µL; 65–82% mononuclear cells). Given the indolent presentation and atypical PDE findings of low-grade, mononuclear-predominant leukocytosis without cloudy effluent, empirical intraperitoneal antibiotics were initially withheld while serial PDE samples underwent culture, AFB staining, and molecular testing. Three serial PDE cultures, with acid-fast bacilli (AFB) staining and molecular testing (PCR and GeneXpert), were negative for bacteria, fungi, and mycobacteria. Pre-episode dialysis adequacy was a Kt/V of 1.2 and a weekly creatinine clearance of 33.2 L/week/1.73 m². Peritoneal equilibration testing (PET) classified him as a high-average transporter (baseline D/P creatinine 0.81; 0.71 during the episode, without a clear change in transport category). Serum albumin fell from 4.1 to 3.2 g/dL, and high-sensitivity CRP was elevated at 7.6 mg/dL (reference, 0–5).

Plain abdominal radiography was unremarkable. Approximately 6 weeks after symptom onset, contrast-enhanced abdominal CT revealed multiple enhancing peritoneal nodules —the largest 2.7 × 1.1 × 2.4 cm, one with tiny calcification—along the anterior abdomen, right paracolic gutter, right subdiaphragmatic region, and liver surface. CT also demonstrated diffuse peritoneal thickening up to 4 mm (normal < 1–2 mm) and omental stranding (Fig. 1A–D). These nodules provided the target for ultrasound-guided biopsy 3 days later, using an 18-gauge coaxial needle (4 passes; no complication); histology showed necrotizing granulomatous inflammation with negative AFB, modified AFB, and Grocott methenamine silver stains (Fig. 1G–H). Tissue PCR was not be performed because the limited core biopsy specimen was exhausted by the histochemical stains for the granulomatous inflammation. CT peritoneography—performed after dialysate drainage and intraperitoneal instillation of 60 mL of iobitridol (Xenetix 300; Guerbet; 300 mg iodine/mL) diluted in 2,000 mL of dialysate, with delayed imaging at 1 and 4 h—demonstrated a loculated contrast-opacified pocket, indicating impaired free dialysate circulation (Fig. 1E–F). Four weeks after biopsy, tissue culture grew Mycobacterium kansasii. Phenotypic susceptibility testing by broth microdilution (CLSI M24S, 2nd ed.) reported rifampicin resistance; the isolate was also resistant to moxifloxacin and ciprofloxacin but susceptible to amikacin and clarithromycin; the complete minimum inhibitory concentration profile is presented in Table 1 [11].

Fig. 1.

Radiologic and histopathologic features of Mycobacterium kansasii peritonitis in a PD patient. On contrast-enhanced computed tomography (CT), enhancing peritoneal thickening lines both sides of the peritoneal cavity, measuring up to 4 mm on each side (black arrows, A), together with nodular mesenteric peritoneal thickening showing heterogeneous enhancement (mean attenuation ≈ 13 HU pre-contrast rising to ≈ 36 HU post-contrast; black arrow, B) and additional enhancing parietal peritoneal and omental thickening with fat stranding (white arrows, C–D). CT peritoneography—a modality rarely applied in infectious peritonitis—demonstrates a large loculated intraperitoneal contrast pocket (black stars, E–F), indicating failure of free dialysate circulation and consequent ultrafiltration failure. The nodular lesions seen on contrast-enhanced CT, rather than on the peritoneogram, were the target for guided ultrasound-guided biopsy. Hematoxylin and eosin–stained sections of the peritoneal nodule reveal necrotizing granulomatous inflammation, with early caseous necrosis (white star) and surrounding granulomas (black arrows, G) progressing to full-thickness necrosis (H); acid-fast, modified acid-fast, and Grocott methenamine silver stains were negative

Table 1.

Antimycobacterial susceptibility profile of the Mycobacterium kansasii isolate

Antimicrobial agent	MIC (µg/mL)	Interpretation
Amikacin	4	S
Clarithromycin	0.5	S
Rifampicin	> 8	R
Linezolid	16	I
Trimethoprim–sulfamethoxazole	> 8	R
Moxifloxacin	> 8	R
Ciprofloxacin	4	R
Streptomycin	64	—
Rifabutin	> 8	R
Isoniazid	8	—
Ethambutol	> 16	—
Doxycycline	> 16	R
Ethionamide	> 20	—

MIC, minimum inhibitory concentration; S, susceptible; I, intermediate; R, resistant; —, no CLSI interpretive criterion available. MICs were determined by broth microdilution; categorical interpretations were assigned, where applicable, according to CLSI M24S, 2nd edition (2023) [11]

Empirical isoniazid, rifampicin, and ethambutol were started while culture results were pending. After susceptibility testing showed resistance to rifampicin and fluoroquinolones, the regimen was changed to isoniazid 300 mg daily, azithromycin 500 mg daily, ethambutol 1000 mg 3 times weekly, and amikacin 500 mg in 100 mL normal saline infused intravenously over 1 h after each thrice-weekly hemodialysis session through the venous return line before circuit disconnection. Amikacin was the only reliably active parenteral agent; post-infusion serum concentrations of 35–70 mg/L guided dose adjustment. Azithromycin was selected based on the clarithromycin MIC as the CLSI macrolide class surrogate, whereas isoniazid was retained empirically because CLSI provides no breakpoint for M. kansasii [11, 12]. Because of refractory infection, worsening symptoms, and progressive inflow–outflow dysfunction, the PD catheter was removed and the patient transitioned to hemodialysis. Symptoms and inflammatory markers resolved within 6 weeks. Isoniazid and ethambutol were continued for 179 days and azithromycin and amikacin for 165 days, until the patient died following an out-of-hospital sudden cardiac arrest of undetermined cause in the setting of severe cardiomyopathy. At death, the infection remained clinically controlled without documented recurrence, although definitive cure could not be established.

Discussion and conclusions

This case represents a rare instance of PD-associated peritonitis due to M. kansasii, a species seldom implicated in PD infections [6, 8]. Among NTM, M. kansasii is more commonly associated with pulmonary disease, and its manifestation as PD peritonitis is exceptional. Although tissue-based diagnosis has been reported in NTM PD peritonitis [1, 2], this case is distinguished by an imaging-guided pathway in which contrast-enhanced CT localized nodular biopsy targets and CT peritoneography exposed loculated dialysate flow; to our knowledge, it is also the first documented rifampicin-resistant M. kansasii peritonitis in a PD patient.

The patient’s indolent symptoms and fluctuating, low-grade PDE leukocyte counts (49–150 cells/µL), frequently near the 2022 ISPD threshold of 100 cells/µL with mononuclear predominance, posed a diagnostic challenge [13]. Similar low-grade inflammation has been noted in other NTM infections [14], underscoring the importance of clinical judgment when laboratory thresholds are inconclusive. Serial cultures, stains, and molecular testing on PDE remained negative, and the organism was ultimately identified from peritoneal tissue, reinforcing that tissue sampling may be decisive when non-invasive diagnostics fail.

The 2022 ISPD criteria are PDE-based, requiring 2 of: compatible clinical features; PDE WBC > 100 cells/µL with > 50% PMN; and positive PDE culture [13]. In this case, clinical features were present, but the PDE leukocyte count fluctuated around the threshold with mononuclear predominance, and the diagnosis was secured by tissue—rather than PDE—culture. Although the guidelines do not list tissue culture among the criteria, they do not exclude it; this case therefore illustrates a rare, tissue-confirmed diagnostic pathway when conventional PDE studies remain unrevealing, rather than a deficiency of the guidelines.

The discordance between negative PDE PCR/GeneXpert and positive tissue culture likely reflected a low organism burden, effluent dilution, and infection compartmentalized within granulomatous nodules rather than freely circulating in dialysate. Conventional PCR has limited sensitivity in paucibacillary PD-associated mycobacterial peritonitis, whereas enriched culture can improve yield [15]; emerging metagenomic next-generation sequencing may assist culture-negative or atypical cases, although culture remains indispensable for susceptibility testing [16, 17].

Imaging was pivotal. CT demonstrated enhancing peritoneal nodules and omental thickening, while CT peritoneography revealed loculation and a trapped dialysate pocket—a pattern overlapping with tuberculous peritonitis [18] but not previously described in M. kansasii PD peritonitis. Imaging enabled targeted biopsy showing necrotizing granulomatous inflammation, with the diagnosis confirmed from tissue culture while PDE cultures remained negative [1, 2]. The differential diagnosis included malignancy, tuberculous or fungal peritonitis, and encapsulating peritoneal sclerosis. There were considered unlikely given negative tuberculosis and fungal studies, absence of malignant cells, granulomatous histology and subsequent growth of M. kansasii. Rifampicin resistance in M. kansasii, reported by phenotypic susceptibility testing, was another notable feature. Though uncommon, resistance has been reported in pulmonary isolates, typically linked to rpoB gene mutations [19]. Our patient had no prior rifampicin exposure, suggesting de novo resistance. A rifampicin-sparing, susceptibility-guided regimen led to resolution, reinforcing the importance of early susceptibility testing and individualized therapy [12].

Catheter removal aligned with ISPD principles for refractory or atypical mycobacterial peritonitis and proved essential for clinical resolution [13]. Observational studies show poor outcomes when catheters are retained: in one Hong Kong series, 74% required removal and only 7% resumed PD [8]; a multicenter Israeli study reported a 63% transition to permanent hemodialysis [2]. Consistent with recent reviews documenting frequent catheter loss and hemodialysis transfer in NTM PD peritonitis [9, 10], our patient required catheter removal and permanent hemodialysis, whereas early, susceptibility-guided therapy was associated with prompt clinical improvement. Although the peritoneal infection was controlled, the patient subsequently died of an out-of-hospital sudden cardiac arrest of undetermined cause, unrelated to the peritonitis and reflecting the burden of his severe underlying cardiomyopathy.

This single case is hypothesis-generating and cannot establish standard practice. Tissue PCR and metagenomic sequencing were not performed, and management was individualized because evidence for rifampicin-resistant M. kansasii PD peritonitis is limited. No exit-site culture was obtained because the site was clinically normal and lacked purulent discharge; therefore, asymptomatic exit-site colonization as a potential source cannot be excluded.

In summary, this case adds to the limited literature on M. kansasii PD peritonitis and describes an imaging-guided diagnostic pathway using contrast-enhanced CT, CT peritoneography, and targeted biopsy, followed by susceptibility-guided therapy and catheter removal. As a single case, these observations are hypothesis-generating and should not be interpreted as establishing diagnostic criteria or standard management.

Abbreviations

AFB

Acid–fast bacilli

CAPD

Continuous ambulatory peritoneal dialysis

CLSI

Clinical and Laboratory Standards Institute

CRP

C–reactive protein

CT

Computed tomography

D/P

Dialysate–to–plasma creatinine ratio

GeneXpert

GeneXpert MTB/RIF molecular assay

hs

CRP–High–sensitivity C–reactive protein

IgA

Immunoglobulin A

ISPD

International Society for Peritoneal Dialysis

Kt/V

Dialysis adequacy index based on urea clearance

MIC

Minimum inhibitory concentration

NTM

Nontuberculous mycobacteria

PCR

Polymerase chain reaction

PD

Peritoneal dialysis

PDE

Peritoneal dialysis effluent

PET

Peritoneal equilibrium test

PMN

Polymorphonuclear leukocytes

rpoB

RNA polymeraseβ–subunit gene

WBC

White blood cell

WHO

World Health Organization

Author contributions

TL: Data curation, Investigation, Writing – original draft. KS: Investigation, Visualization. AS: Investigation, Visualization. NU: Investigation, Methodology. DLW: Formal analysis, Methodology, Writing – review & editing. TK: Conceptualization, Funding acquisition, Project administration, Supervision, Writing – review & editing. All authors read and approved the final manuscript.

Funding

This Research is funded by Thailand Science Research and Innovation Fund Chulalongkorn University [Grant No. HEA_FF_69_123_3000_012]. The funding bodies played no role in the design of the study, collection, analysis, or interpretation of data, or in writing the manuscript.

Data availability

No datasets were generated or analyzed during the current study.

Declarations

Ethics approval and consent to participate

Ethical approval was waived by the Institutional Review Board (IRB), Faculty of Medicine, Chulalongkorn University (IRB exemption for individual case reports), as individual case reports do not require formal ethics review under Thai regulations. Written informed consent to participate was obtained from the patient in accordance with the Declaration of Helsinki.

Consent for publication

Written informed consent was obtained from the patient for the publication of anonymized clinical details and images.

Competing interests

TK has received consultancy fees from Visterra (as a country investigator), personal funding from the National Research Council of Thailand, and speaking honoraria from AstraZeneca, Baxter Healthcare, and Fresenius Medical Care, all outside the submitted work. The remaining authors declare that they have no competing interests.