Clinical Characteristics and Outcomes of Paediatric Non-tuberculous Mycobacterial Infection: Single Institution Retrospective Review Over Past 20 Years
Non-tuberculous mycobacterial (NTM) infection is uncommon yet clinically significant as it could cause severe morbidity and mortality. This study reviewed all paediatric patients under 20 years of age diagnosed to have NTM infection in Queen Mary Hospital during 1 January 1999 to 30 June 2018. Total 7 patients were identified with median age of diagnosis at 6 years old and median duration of follow up for 30 months. Majority (86%) have underlying comorbidities such as haemic malignancies or history of stem cell transplantation. Clinical manifestations varied from lymphadenitis, pulmonary infection, osteomyelitis, arthritis to systemic bacteraemia. Infective agents include Mycobacterium fortuitum, M. abscessus and M. chelonae. Mortality rate was high (29%). High index of suspicion and early recognition of NTM infection is important especially in immunocompromised and at-risk individuals. Prompt administration of appropriate therapy improves patient outcomes. Multidisciplinary collaboration is crucial. Treatment could be difficult and prolonged. Drug-related toxicities are common.
Keyword : Diagnosis; Morbidity; Nontuberculous mycobacteria; Therapeutics
Non-tuberculous mycobacteria (NTM) are ubiquitous in the environment such as water, soil, dust, animals, and birds. By far, more than 190 species of NTM have been identified, and approximately 60 of which are suspected or known to be pathogenic in humans. Despite of their low virulence and indolent clinical course of infection, NTM infection could cause severe morbidity and mortality. Clinicians should not overlook this group of relatively less common yet clinically significant infection, especially in oncology and transplant settings. Retrospective review on clinical data helps summarise experience and remind physicians not to overlook this possibility especially in susceptible group of patients.
A single-centre retrospective study was conducted in the University Department of Paediatrics and Adolescent Medicine of Queen Mary Hospital (QMH), a university-affiliated hospital offering tertiary and quaternary clinical services with encounter of highly complex patients in daily practice. All patients under 20 years of age and diagnosed to have NTM infection in QMH during 1st January 1999 to 30th June 2018 were retrieved through Clinical Data Analysis and Reporting System using the diagnostic code (ICD-9) for "non-tuberculous mycobacterial infection" (031.9). Information on patient demographics, presence of underlying co-morbidities, clinical manifestations of infection, duration between symptom onset and confirmation of laboratory diagnosis, infective agent identified, sensitivity pattern, treatment given and duration, presence of complications as well as final outcome were collected and studied from written and electronic medical records through the Clinical Management System. All clinical isolates were sent to the Department of Health Public Health Laboratory Service Branch Centre for Health Protection for identification and susceptibility testing under standard broth dilution method and interpreted according to the Clinical and Laboratory Standards Institute. Patients without definitive diagnosis of NTM infection were excluded from the study. BCGiosis and immune reconstitution inflammatory syndrome (IRIS) due to Bacille Calmette-Guérin (BCG) strain of Mycobacterium bovis were excluded as M. bovis is considered part of M. tuberculosis complex (MTBC). MPT64 antigen detection in culture isolates were performed to exclude MTBC from genuine NTM infections.
This study was primarily descriptive in nature. Continuous variables were expressed as median and range. Duration of follow up was defined from the date of first clinical presentation or diagnosis to date of last follow up (censorship) or death.
This study complies with the Declaration of Helsinki and approval from Institutional Review Board of the University of Hong Kong / Hospital Authority Hong Kong West Cluster had been obtained (UW 18-597 / HKUCTR-2581).
Patient Demographics (see Table 1)
Total 7 patients with genuine NTM infection were identified (3 males, 4 females) with median age at diagnosis of 6 years old (range 4-20 years old) and median duration of follow up of 30 months (range 3 days up to 120 months). Majority (86%, 6 out of 7) have underlying comorbidities, such as haemic malignancies or post-stem cell transplantation (n=4), autoimmune diseases or immunodeficiency (n=1) and chronic respiratory disease (n=1). Clinical manifestations include systemic bacteraemia (n=3), lymphadenitis (n=2), osteomyelitis or arthritis (n=2), and pulmonary infection (n=2). Median time from symptom onset to confirmation of laboratory diagnosis is 1 month.
Microbiology (see Table 2)
Infective agents identified including M. abscessus complex (n=3) (M. abscessus subsp. abscessus =2, M. abscessus subsp. massiliense =1), M. fortuitum (n=2), and M. chelonae (n=2). In general, the mycobacteria identified were sensitive to amikacin, showing intermediate sensitivity to cefoxitin, linezolid and meropenem; variable sensitivity to clarithromycin, levofloxacin, linezolid and cotrimoxazole; and resistant to ciprofloxacin, doxycycline and moxifloxacin. Mortality rate was up to 29% (2 out of 7 died).
Treatment and Related Toxicities (see Table 3)
For the 2 patients with pneumonia and systemic bacteraemia due to mycobacteria, both died before treatments were completed, the one with M. chelonae infection (Case 2) was treated with intravenous (IV) imipenem-cilastatin with oral clarithromycin while the other with M. fortuitum infection (Case 4) was treated with IV imipenem-cilastatin with amikacin for 3 weeks during intensive phase followed by oral levofloxacin as continuation therapy (planned for 6 months but was given for around 5 months then patient died of progressive respiratory failure).
For the 2 patients with lymphadenitis, the one with M. fortuitum lymphadenitis (Case 1) was cured with 6 months of IV meropenem and amikacin while the other patient with M. abscessus lymphadenitis (Case 7) was treated with IV amikacin and cefoxitin/tigecycline for 4 weeks as initial treatment followed by oral clarithromycin for 4 months in total as maintenance therapy.
For the patient with central line infection due to M. chelonae (Case 3), prompt removal of central line followed by treatment with triple agents (IV imipenem-cilastatin, IV amikacin and oral clarithromycin) suffices to eradicate the catheter-related mycobacterial infection.
The patient with cystic fibrosis with pneumonia due to M. massiliense (Case 5) was treated according to United Kingdom Royal Brompton Hospital clinical guideline for the care of children with cystic fibrosis 2011 with 3-week intensive phase of IV amikacin, IV meropenem, IV cefoxitin and oral clarithromycin followed by prolonged continuation phase with nebulised amikacin and oral azithromycin daily for 1 year then continued with oral azithromycin 3 day per week for 2 more years for anti-inflammatory effect.
For the remaining patient with right groin, popliteal, ankle osteomyelitis and subcutaneous abscesses (Case 6), initial treatment was failed with 2-month course of oral clarithromycin, three 2-week courses of IV amikacin, 2 weeks of IV levofloxacin, 1 month of IV imipenem-cilastatin, 2 weeks of IV cefoxitin, and 1-month course of interferon-gamma (3 days per week). The patient was successfully salvaged with 2 re-purposed drugs (oral bedaquiline and clofazimine) together with cotrimoxazole.
Drug-related toxicities were encountered in 3 out of the 7 patients treated (Cases 5-7), and all manifested as drug fever and fixed drug eruptions - 2 due to the beta-lactam cefoxitin (Cases 5, 6) while 1 due to cotrimoxazole (Case 7). One also developed hepatitis (Case 6) with alanine transaminase and aspartate transaminase elevated above 1,000 requiring discontinuation of all anti-mycobacterial agents for 1 week before resuming antimicrobials when liver function normalised.
Classification of Mycobacteria
Mycobacteria are aerobic acid-fast bacilli under genus Actinobacteria. Over 190 species had been identified so far, and approximately 60 of which are suspected or known to be pathogenic in humans. NTM refer to mycobacteria apart from MTBC, M. leprae and M. ulcerans. Distinguishing NTM from MTBC is of clinical significance as rapidly-growing NTM are in general not susceptible in vitro to anti-TB drugs but susceptible to traditional bactericidal agents. Besides, TB is a notifiable disease requiring public health tracking while NTM is not on the contrary. NTM comprises mycobacteria in environment which are non-pathogenic and also NTM which may cause diseases in humans.1,2 NTM could be categorised by Runyon's classification into photochromogens, scotochromogens, non-chromogens (such as Mycobacterium avium complex MAC) and rapid growers known as rapidly growing mycobacteria (RGM)3 (such as M. abscessus, M. fortuitum and M. chelonae). RGM are environmental organisms found worldwide that typically grow within 1 week in suitable culture medium. Potential diagnostic limitations in distinguishing different species occur and different susceptibility patterns for different species.4
Clinical Manifestations of NTM Infections
Clinical syndromes caused by NTM include pulmonary disease5,6 (Cases 2, 4 and 5), lymphadenitis7,8 (either localised cervical lymphadenitis or complicated disseminated lymphadenitis) (Cases 1 and 7), skin and soft tissue infections9 (either superficial chronic cutaneous lesions or deep-seated infections involving tendons, synovium, bones and joints) (Case 6), intra-vascular catheter related infections10-13 and/or systemic bacteraemia (Cases 2, 3 and 4) as well as continuous ambulatory peritoneal dialysis-related peritonitis.14 For local case cohort, we encountered almost all different clinical manifestations as stated above.
Immunity Against Mycobacterial Infection
Cell-mediated immunity is the major protective immune response against intracellular bacteria, with non-tuberculous mycobacteria being one of them.15,16 Th1 lymphocytes can produce interferon-gamma (IFN-γ) which activates macrophage to kill phagocytised intracellular microbe. In local cohort, one of the cases (Case 6) had incorporated IFN-g as part of anti-mycobacterial treatment.
Risk Factors for Development of NTM Infections
Steroid and cytotoxic therapy predispose to a wide combination of phagocytic, cell-mediated and even humoral defects.9,17 From local case series, 6 out of 7 have underlying acquired or congenital immunodeficiency render them predispose to non-tuberculous mycobacterial infection, with underlying haemic malignancies or history of stem cell transplantation being the commonest group (n=4), reflecting usage of steroid and/or cytotoxic therapy being an important risk factor for development of NTM infections.
Consideration of Primary Immunodeficiency in Patients with NTM Infections
Among the 7 cases in local cohort, one (Case 7) of them was with good past health and no underlying comorbidities, it is postulated that subtle immune defect could exist and investigations on Mendelian susceptibility to mycobacterial diseases (MSMD) are under way though not yet revealing at the time of publication.
Diagnosis and Treatment
Despite knowing the importance in differentiating TB and NTM, as mentioned they are in practical not easily distinguishable from clinical history, Mantoux test result, radiological pattern and initial laboratory reports. They also share overlapping clinical features such as pulmonary disease and lymphadenopathies. Delayed in diagnosis is common. The disease burden of NTM is unknown in region such as Hong Kong where TB is endemic. As no mandatory reporting is required, epidemiology and clinical studies are scarce. NTM are often assumed to be TB and treated as such, but susceptibility to antimicrobials are in fact very different. Classical anti-TB drugs (streptomycin, isoniazid, rifampicin, pyrazinamide and ethambutol) (S, H, R, E, Z) are generally not useful because of intrinsic resistance.18,19
Treatment against NTM differs in different patients according to the clinical manifestations, type of mycobacteria identified, sensitivity pattern, underlying comorbidities and the presence of drug-related toxicities.20-22 Surgical debridement23 (Case 6) or removal of foreign body (such as central venous catheter) (Case 3) are indicated in selected cases.20-22 Anti-mycobacterial treatment is usually prolonged in terms of months or years with employment of multiple antimicrobial agents in an attempt to avoid development of drug resistance.20-22 Amikacin, imipenem, cefoxitin, clarithromycin and azithromycin are the key drugs commonly used to treat NTM infections.20-22,24 From the sensitivity patterns of local cohort, NTM isolated are generally sensitive to amikacin; demonstrated intermediate sensitivity to cefoxitin, linezolid and meropenem; variable sensitivity to clarithromycin, levofloxacin, linezolid and cotrimoxazole; and resistant to ciprofloxacin, doxycycline and moxifloxacin (see Table 3). Development of multiple-drug resistant (MDR) strains of mycobacteria as well as compliance and drug-related toxicities related to prolonged duration of treatment are the 2 major challenges faced by physicians in combating NTM infections.
Macrolide resistance for NTM, in particular M. abscessus spp. abscessus and bolletii exists due to erm gene. Macrolide resistance is associated with delayed treatment response and possible treatment failure in patient with lung disease on macrolide-containing regimens.25 However, including a macrolide (clarithromycin or azithromycin) in the multidrug regimen may still be considered in such situations as the choice of oral alternatives is limited. Thus, macrolide was incorporated as part of treatment regime for the 3 local cases with M. abscessus infection (Cases 5-7) (including Case 5 which later identified to be M. massiliense).
Drug-related Toxicities (see Supplementary Table 3)
From literature review, around 7% patients would develop adverse drug reactions towards anti-TB/NTM treatment, with transaminitis/ hepatotoxicity being commonest (87%), followed by rash (8%), angioedema (5%) and gastrointestinal intolerance.26 Combination therapy is associated with increased toxicity.26 Treatment time is on average being prolonged by a median of 1 month due to ADR.26 In our study, drug-related toxicities were encountered in 3 out of the 7 patients treated (43%) (Cases 5-7), and all manifested as drug fever and fixed drug eruptions. Hepatotoxicity is encountered in 1 patient only (Case 6) with anti-NTM treatment withheld for 1 week.
Novel Treatment Agents
For Case 6 who was a 4-year old girl with translocation t (8;21) acute myeloid leukaemia who developed right groin, popliteal, ankle osteomyelitis and subcutaneous abscesses due to M. abscessus after 2 courses of induction chemotherapy, 2 experimental drugs (bedaquiline and clofazimine) were used to salvage this patient as off-label use in view of treatment failure against traditional anti-NTM agents. Bedaquiline and clofazimine are agents to treat MDR-TB and leprosy in adults, with uncertain efficacy in paediatric population. Both are lipophilic agents and able to achieve high bactericidal concentrations in soft tissues. Disease control had been achieved in this patient.28
Besides antimicrobial agents, 2 agents were also considered in the same patient to enhance patient's own cellular immunity against NTM infection, namely pegylated granulocyte colony-stimulating factor (pegfilgrastim) and IFN-g. Pegfilgrastim administration was associated with a significant increase of the inducible IL-12p40 subunit in patient serum.27 Whereas in patients given filgrastim with a much shorter half-life of around 3.5-3.8 hours as compared to 42 hours for pegfilgrastim, IL-12p40 slightly declined and returned to baseline values by day +11 from the commencement of cytokine treatment. It is tempting to speculate that immunoreactive IL-12 in patients given filgrastim may have been degraded as a result of sharp increases in circulating PMN capable of releasing proteolytic enzymes. Side effect profile of pegfilgrastim is similar to filgrastim.
Strengths and Limitations of Study
This is the first study reporting patient characteristics, microbiology, treatment and outcomes of paediatric NTM infections in QMH over past 20 years. Despite the very small number of cases extracted from local cohort of a single paediatric department which makes it difficult to draw definitive conclusions or generalise the findings, this paper crystalises experience and serves as a timely reminder to paediatric oncologists and transplant physicians not to overlook possibility of NTM infections in their specific clinical settings with common encounter of immunocompromised children facing the conglomeration of all five local paediatric oncology centres into single paediatric oncology and transplant centre at Hong Kong Children's Hospital since 2019. In addition, it also enlightens general paediatricians, immunologists and infectious disease specialists to look for concealed immunodeficiencies for children presented with NTM infections without know underlying susceptibility conditions. In the future, a territory-wide study or international collaboration with inclusion of adult cases would be recommended.
To conclude, anticipation and early recognition of NTM infections especially in immunocompromised patients is important to facilitate prompt provision of appropriate therapy to improve clinical outcomes. Concealed immunodeficiencies should be actively sought for in children presented with NTM infections without known underlying susceptibility conditions. Multidisciplinary collaboration is crucial. Treatment could be difficult and prolonged and drug-related toxicities are common.
Declaration of Interest
1. Pozniak A, Bull T. Recently recognized mycobacteria of clinical significance. J Infect 1999;38:157-61.
2. Jarzembowski JA, Young MB. Nontuberculous mycobacterial infections. Arch Pathol Lab Med 2008;132:1333-41.
3. Griffith DE, Brown-Elliott BA, Benwill JL, Wallace Jr RJ. Mycobacterium abscessus. "Pleased to meet you, hope you guess my name...". Ann Am Thorac Soc 2015;12:436-9.
4. Tortoli E, Kohl TA, Brown-Elliott BA, et al. Emended description of Mycobacterium abscessus, Mycobacterium abscessus subsp. abscessus and Mycobacteriumabscessus subsp. bolletii and designation of Mycobacteriumabscessus subsp. massiliense comb. nov. Int J Syst Evol Microbiol 2016;66:4471-9.
5. Wolfe J, Turenne C, Alfa M, Harding G, Thibert L, Kabani A. Mycobacterium branderi from both a hand infection and a case of pulmonary disease. J Clin Microbiol 2000;38:3896-9.
6. Martinez S, McAdams HP, Batchu CS. The many faces of pulmonary nontuberculous mycobacterial infection. AJR Am J Roentgenol 2007;189:177-86.
7. Chetchotisakd P, Mootsikapun P, Anunnatsiri S, et al. Disseminated infection due to rapidly growing mycobacteria in immunocompetent hosts presenting with chronic lymphadenopathy: a previously unrecognized clinical entity. Clin Infect Dis 2000;30:29-34.
8. Springer B, Kirschner P, Rost-Meyer G, Schroder KH, Kroppenstedt RM, Botther EC. Mycobacterium interjectum, a new species isolated from a patient with chronic lymphadenitis. J Clin Microbiol 1993;31:3083-9.
9. Wallace Jr RJ, Brown BA, Onyi GO. Skin, soft tissue, and bone infections due to Mycobacterium chelonae chelonae: importance of prior corticosteroid therapy, frequency of disseminated infections, and resistance to oral antimicrobials other than clarithromycin. J Infect Dis 1992;166:405-12.
10. Woo PC, Li JH, Tang W, Yuen K. Acupuncture mycobacteriosis. N Engl J Med 2001;345:842-3.
11. Raad II, Vartivarian S, Khan A, Bodey GP. Catheter-related infections caused by the Mycobacterium fortuitum complex: 15 cases and review. Rev Infect Dis 1991;13:1120-5.
12. Schinsky MF, McNeil MM, Whitney AM, et al. Mycobacterium septicum sp. nov., a new rapidly growing species associated with catheter-related bacteraemia. Int J Syst Evol Microbiol 2000;50:575-81.
13. Kiska DL, Turenne CY, Dubansky AS, Domachowske JB. First case report of catheter-related bacteremia due to "Mycobacterium lacticola". J Clin Microbiol 2004;42:2855-7.
14. Vera G, Lew SQ. Mycobacterium fortuitum peritonitis in two patients receiving continuous ambulatory peritoneal dialysis. Am J Nephrology 1999;19:586-9.
15. van Kooyk Y, Geijtenbeek TBH. DC-SIGN: escape mechanism for pathogens. Nat Rev Immunol 2003;3:697-709.
16. Medzhitov R. Toll-like receptors and innate immunity. Nat Rev Immunol 2001;1:135-45.
17. Redelman-Sidi G, Sepkowitz KA. Rapidly growing mycobacteria infection in patients with cancer. Clin Infect Dis 2010;51:422-34.
18. Wallace Jr RJ, Brown-Elliott BA, Ward SC, Crist CJ, Mann LB, Wilson RW. Activities of linezolid against rapidly growing mycobacteria. Antimicrob Agents Chemother 2001;45:764-7.
19. Rodriguez Diaz JC, Lopez M, Ruiz M, Royo G. In vitro activity of new fluoroquinolones and linezolid against non-tuberculous mycobacteria. Int J Antimicrob Agents 2003;21:585-8.
20. Griffith DE, Aksamit T, Brown-Elliott BA, et al. An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 2007;175:367-416.
21. Campbell I, Drobniewski F, Novelli V, et al. Management of opportunist mycobacterial infections: Joint Tuberculosis Committee Guidelines 1999. Subcommittee of the Joint Tuberculosis Committee of the British Thoracic Society. Thorax 2000;55:210-8.
22. Sizaire V, Nackers F, Comte E, Portaels F. Mycobacterium ulcerans infection: control, diagnosis, and treatment. Lancet Infect Dis 2006;6:288-96.
23. Lindeboom JA, Kuijper EJ, Bruijnesteijn van Coppenraet ES, Lindeboom R, Prins JM. Surgical excision versus antibiotic treatment for nontuberculous mycobacterial cervicofacial lymphadenitis in children: a multicenter, randomized, controlled trial. Clin Infect Dis 2007;44:1057-64.
24. Bermudez LE, Kolonoski P, Wu M, Aralar PA, Inderlied CB, Young LS. Mefloquine is active in vitro and in vivo against Mycobacterium avium complex. Antimicrob Agents Chemother 1999;43:1870-4.
25. Bermudez LE, Kolonoski P, Petrofsky M, Wu M, Inderlied CB, Young LS. Mefloquine, moxifloxacin, and ethambutol are a triple-drug alternative to macrolide-containing regimens for treatment of Mycobacterium avium disease. J Infect Dis 2003;187:1977-80.
26. Murray MP, Laurenson IF, Hill AT. Outcomes of a standardized triple-drug regimen for the treatment of nontuberculous mycobacterial pulmonary infection. Clin Infect Dis 2008;47:222-4.
27. Bonanno G, Procoli A, Mariotti A, et al. Effects of pegylated G-CSF on immune cell number and function in patients with gynecological malignancies. J Transl Med 2010;8:114.
28. Chan WYK, Ho PL, To KKW, et al. A child with acute myeloid leukemia complicated by calcaneal osteomyelitis due to Mycobacterium abscessus infection after induction chemotherapy successfully salvaged with bedaquiline and clofazimine. Int J Infect Dis 2021;103:9-12.
This web site is sponsored by Johnson & Johnson (HK) Ltd.