Non-Tuberculous Mycobacteria in Solid Organ Transplant Recipients

Authors: Eric Cober and Daniel R. Kaul


Non-tuberculous mycobacteria (NTM) may cause significant disease in recipients of solid organ transplants. This monograph will provide an overview of the epidemiology, clinical manifestations, diagnosis, and treatment of disease caused by NTM in this population.


The NTM group includes over 125 species of mycobacteria and excludes members of the Mycobacterium tuberculosis complex (M. tuberculosis, M. bovis, M. africanum) and Mycobacterium leprae (the cause of Hansen’s disease). NTM are further divided into rapidly growing mycobacterium or RGM (primarily M. chelonae, M. abscessus, M. fortuitum). Further classification is based on the presence of absence of pigmentation in light or dark conditions. The clinically most important examples include the photochromagens (e.g., M. kansasii, M. marinum) and the non-chromagens (e.g., M. avium complex).


NTM are distributed widely in the environment, and depending on species may be present in treated or untreated water and in soil. NTM are found worldwide although geographic distribution may be species specific (e.g., M. kansasii in the central United States). In the North America, M. avium complex (MAC) is the most common cause of NTM disease followed by the RGM, M. marinum, and M. kansasii (7). Pulmonary cases predominate, followed by skin and soft tissue infections and disseminated disease (7). In most patients with pulmonary NTM infection no source of is identified, although in the case of skin or soft tissue infection a specific source (e.g., fish tank cleaning in patient with M. marinum infection) can be identified. Person to person transmission has not been documented and animals do not appear to be a source of human infection (Griffith, Aksamit et al. 2007).

Traditional risk factors for disease with NTM include underlying lung disease or the “Lady Windemere syndrome” (pulmonary MAC), cystic fibrosis (pulmonary infection), late stage infection with human immunodeficiency virus (disseminated MAC), presence of an indwelling catheter (RGM bacteremia), and immunodeficiency including interferon gamma receptor deficiency. Among SOT recipients the rates of NTM infection are low with the highest rates reported among heart and lung recipients (0.24%-2.3%) with lower rates among recipients of kidney transplants (0.16%-0.38%) (Doucette and Fishman 2004). NTM disease is not a reportable infection in the United States, and rates are likely to vary from center to center.

back to top

Clinical manifestations pertinent to treatment

NTM infections in solid organ transplant recipients generally manifest as skin and soft tissue infections, pleuropulmonary disease, musculoskeletal infections, or disseminated disease. In kidney and heart recipients, the most frequent presentation is skin and soft tissue infection (Doucette and Fishman 2004). The skin lesions characteristically are erythematous or violaceous painful nodules that may ulcerate or form subcutaneous abscesses which spontaneously drain. Skin infections with NTM may be single lesions, or multiple lesions associated with dissemination. Cutaneous lesions can also manifest as non-healing wounds occurring after accidental injury (Jie, Matas et al. 2005). The rapidly growing mycobacteria (M. fortuitum, M. abscessus, M. chelonae) and M. marinum are the NTM species most likely to cause skin and soft tissue infections (Nagy and Rubin 2001), although any NTM species, including M. avium complex and M. kansasii, can be responsible (Patel, Roberts et al. 1994). RGM are a well-described cause of nosocomial wound infections (De Groote and Huitt 2006; Garrison, Morris et al. 2009). Allograft incisional site infections and peritoneal catheter site infections with M. abscessus have recently been described in renal transplant recipients (Garrison, Morris et al. 2009).

Pleuropulmonary disease is the most common presentation of NTM infection reported in lung transplant recipients, and occurs in recipients of other solid organs with less frequency. Clinical symptoms are non-specific and may include cough, dyspnea, sputum production, hemoptysis, and constitutional symptoms such as fevers, weight loss, and malaise. These symptoms may be less prominent in transplant recipients as compared with immunocompentent patients. Radiographic findings for MAC pulmonary disease typically involve one of two patterns: apical fibrocavitary lung disease or bronchiectasis with multiple small pulmonary nodules, often less than 5 mm in size and clustered together. Clinical and radiographic presentations for other NTM depend on the particular species involved. M. kansasii may cause apical cavitary lung disease and constitutional symptoms similar to that of reactivation tuberculosis. M. abscessus is increasingly recognized as a virulent organism in lung transplant recipients (Chernenko, Humar et al. 2006). In addition to lung parenchymal involvement, there have been reports of M. abscessus causing empyema, chest wall and sternal wound infections, and widely disseminated disease following lung transplantation (Fairhurst, Kubak et al. 2002; Taylor and Palmer 2006; Zaidi, Elidemir et al. 2009). Radiographic patterns of M. abscessus are usually similar to the nodular bronchiectatic form of MAC, with less than 15% of cases involving cavitation (Griffith, Girard et al. 1993).

Although less common than skin and soft tissue infection and pleuropulmonary disease, NTM can cause disseminated infection in solid organ transplant recipients. For non-HIV patients disseminated MAC may present as fever of unknown origin, but disseminated disease caused by M. kansasii and the rapidly growing mycobacteria more typically presents as multiple subcutaneous nodules (Griffith, Aksamit et al. 2007). In one review of 82 solid organ transplant recipients with NTM infection, 67% presented with soft tissue and musculoskeletal disease, and over one-half of those patients had multi-focal involvement of non-contiguous areas suggestive of disseminated disease (Patel, Roberts et al. 1994). Although rapidly growing mycobacteria have been reported to cause central venous catheter blood stream infections in hematopoietic stem cell recipients (HSCT), this has not commonly been described in SOT recipients.

Muskuloskeletal infection manifesting as septic arthritis or tenosynovitis is relatively common manifestation in solid organ transplant recipients, and often associated with disseminated infection. Joints commonly involved include ankles, knees, elbows, wrists, and interphalangeals (Patel, Roberts et al. 1994). NTM are an unusal cause of prosthetic joint infections, with a small number of cases described in transplant recipients (Gupta and Clauss 2009). In one case series of 8 immunocompetent patients, time to onset range from one to 720 weeks, with median time to onset of 312 weeks (Eid, Berbari et al. 2007). Genitourinary infections have been described, including a M. haemophilum epididymal abscess(Keller, Mak et al. 2008) and in several solid organ transplant recipients NTM urinary tract infections manifested by sterile pyuria.(Patel, Roberts et al. 1994; Pinho, Santos et al. 2009). Mycobacterial growth or AFB positivity of urine should be interpreted cautiously as NTM may be found in the urinary tract of healthy individuals.

back to top

Laboratory Diagnosis

Clinical specimens should be examined for mycobacteria using an appropriate method such as the preferred flourochrome stain or alternatively the Ziehl-Neelsen or Kinyoun stains. Smear microscopy is variably sensitive and is more likely to be positive with a high underlying organism burden (Wright, Wallace et al. 1998). Specimens should be inoculated onto both solid and liquid media. Some fastidious species including M. haemophilum, M. marinum, and M. chelonae grow better at lower temperatures than most other NTM, so all specimens from skin lesions, joint fluid or bones should be incubated at 28o – 30o C and 35 o – 37o C (and supplemented with iron-containing compounds if M. haemophilum is suspected) (Brown-Elliott, Griffith et al. 2002). Time to growth of less than seven days indicates a RGM, although on primary specimens with low inoculation burdens growth of RGM may take longer than 7 days.

Historically, speciation of NTM was a prolonged process and determined by growth patterns and biochemical tests. Currently, commercially available DNA probes (e.g., AccuProbe®, manufacturer Gen-Probe, San Diego, CA) specific for MAC, M. kansasii, and M. gordonae can identify isolates from both solid and liquid culture media within a few hours with a specificity and sensitivity approaching 100% (Chapin-Robertson, Dahlberg et al. 1993). Identification of other NTM species requires more traditional techniques. If a clinical specimen is smear positive, growth on culture media can still take ten days or longer, delaying species identification and treatment decisions. Unfortunately, there is no widely available test for detection and speciation of NTM on direct clinical specimens. High-performance liquid chromatography (HPLC) can identify NTMs via mycolic acid patterns from processed clinical specimens, but given initial equipment expenditures and required expertise is not available in many clinical laboratories (Butler and Guthertz 2001). Polymerase restriction endonuclease analysis (PRA) can also identify many NTM species directly from clinical specimens but has similar limitations for use in most laboratories (Kirschner, Springer et al. 1993).

Clinical Diagnosis

Diagnosis of NTM disease can be challenging as it is difficult to distinguish true infection from colonization or laboratory contamination. Although a positive mycobacterial culture for a NTM species is necessary for diagnosis, it does not necessarily confirm the presence of actual disease. Isolation of a NTM species on culture must always be interpreted in the context of a patient’s clinical symptoms and/or radiographic findings. The ATS/IDSA has published diagnostic criteria for pulmonary NTM disease based on a combination of clinical, radiographic and microbiologic findings (Table 1 ) (Griffith, Aksamit et al. 2007). Microbiologic factors that may increase the clinical significance of a NTM isolate include the burden of organisms recovered (smear positive), repeated isolation in multiple samples, recovery from a normally sterile site, and the specific NTM species isolated. Recovery of a species with high pathogenic potential such as M. kansasii or M. abscessus in a transplant recipient with compatible radiographic findings should prompt careful consideration for treatment. Recovery of a less virulent species such as M. fortuitum or of a species more likely to represent a contaminant such as M. gordonae or M. terrae complex would require a higher threshold to initiate treatment (Griffith, Aksamit et al. 2007). In uncertain cases, a “watchful waiting” approach utilizing radiological, clinical, and microbiological follow-up is appropriate.

For suspected pulmonary infection (clinical symptoms plus compatible radiographic findings), 3 sputum samples or bronchoalveolar lavage (BAL) fluid for AFB staining and mycobacterial cultures with or without a transbronchial biopsy should be obtained. More common infectious etiologies such as fungal, bacterial, and viral infections and non-infectious etiologies such as PTLD or other malignancy must be ruled out. Of particular importance, sensitivity of cultures from BAL fluid for molds is quite low (approximately 40-50%), and the absence of growth of mold in culture should not be relied upon to rule out an invasive fungal infection. Given broad differences in antimicrobial susceptibilities among NTM species and the relative rarity of NTM disease, empiric treatment for mycobacteria pending culture results in smear negative patients is usually not recommended unless tuberculosis is strongly suspected.

Patients with suspected disease at other sites (e.g., skin or soft tissue infection) usually require biopsy and mycobacterial cultures of affected areas. The finding of granulomatous inflammation is suggestive (but by no means diagnostic) of mycobacterial infection. In difficult to diagnose cases, nucleic acid based testing of fresh frozen specimens or paraffin embedded tissues may be helpful. Mycobacterial blood cultures (using special techniques) may be positive in patients with disseminated disease or catheter related infections

back to top


Choice of antimicrobial therapy for transplant recipients with NTM infection is similar to immunocompetent patients and depends upon the NTM species isolated and the severity and site of infection. In general, treatment requires a combination of antimicrobials over prolonged periods of time, and can be complicated by resistance, additive toxicities with concurrent use of immunosuppressant medications, and drug interactions. The initial treatment regimen is based upon predictable antimicrobial susceptibility patterns for NTM species, and should be modified when results of in vitro susceptibility testing becomes available (Table 2 ). Failure rates remain high and relapses can occur.

Clinicians must be particularly mindful of drug-drug interactions between antimycobacterial agents and immunosuppressant drugs (Table 3 ). The rifamycins, particularly rifampin, can significantly reduce serum levels of calcineurin inhibitors (CNI) and mTOR inhibitors (e.g., sirolimus) via induction of cytochrome p450 isoenzyme CYP3A4. Thus, the use of rifamycins has been associated with acute rejection rates as high as 35 (Aguado, Herrero et al. 1997; Finch, Chrisman et al. 2002). Rifabutin is a less potent inducer of cytochrome p450, but still should be used with caution. If a rifamycin and CNI (or mTOR inhibitor) are used concurrently, the dose of immunosuppressant should be initially increased 3–5 fold and levels closely monitored (Aguado, Torre-Cisneros et al. 2009). Further, rifampin can reduce corticosteroid levels (McAllister, Thompson et al. 1983) and also mycophenolate levels through a novel mechanism unrelated to CYP3A4 induction (Kuypers, Verleden et al. 2005; Naesens, Kuypers et al. 2006). Finally, rifamycins reduce levels of most triazole agents in patients on concurrent antifungal therapy. Clarithromycin (and to a lesser extent azithromycin) can increase serum levels of calcineurin inhibitors (CNI) and sirolimus via inhibition of CYP3A4 (Sadaba, Lopez de Ocariz et al. 1998; Doucette and Fishman 2004). Although this interaction can result in CNI toxicity, it does not usually preclude concurrent use with close monitoring.

Susceptibility testing for most clinically-significant NTM isolates should be performed, although there is not always a correlation between in vitro susceptibility for many antimicrobials and clinical response (Griffith, Aksamit et al. 2007). ATS/IDSA recommends limited initial susceptibility testing for MAC isolates to clarithromycin and for M. kansasii isolates to rifampin. Given the complexity of drug-drug interactions in transplant recipients, it may be prudent to test M. kansasii isolates to secondary agents with the initial evaluation. All clinically-significant RGM isolates should be tested against a broad panel of antimicrobial agents (De Groote and Huitt 2006). When using susceptibility results to guide subsequent therapy for NTM infections, clinicians should be aware that both treatment successes with drugs resistant in vitro and treatment failures with drugs susceptible in vitro can occur.

The ATS/IDSA has published treatment guidelines for the most frequently encountered NTM (Griffith, Aksamit et al. 2007). Recommendations from these guidelines as well as other literature are summarized in Table 2. Traditional anti-tuberculosis agents such as the rifamycins and ethambutol can be effective against many of the NTMs but have little utility for treatment of RGMs. Fortunately, a number of more traditional antibiotics and newer agents possess activity against RGMs. In general, combination regimens should be used given the development of resistance on monotherapy for most isolates. When feasible, the intensity of immunosuppression should be decreased. Specific comments regarding the treatment of several specific NTM species in transplant recipients are discussed below.

The standard regimen for pulmonary MAC infections is a three-drug regimen that includes a macrolide plus two other companion medications (e.g., ethambutol and a rifamycin). Given the interaction with CNI and mTOR inhibitors, rifabutin should probably be the preferred rifamycin in transplant recipients, although it should be used with caution as well (2004). Many clinicians prefer to avoid rifamycins altogether in transplant recipients receiving CNIs or mTOR inhibitors. Alternatives to rifamycins include a two-drug regimen of clarithromycin and ethambutol or substituting a second-line agent such as a fluoroquinolone or aminoglycoside for the rifamycin, although neither approach has been validated in clinical trials (2004). A two-drug regimen should be limited to patients with mild nodular or bronchiectactic pulmonary disease and for disseminated MAC with low organism burden as concern for development of resistance remains (Griffith, Aksamit et al. 2007).

Treatment of M. kansasii can be particularly problematic in transplant recipients because of drug interactions. Outcomes of M. kansasii infections treated medically with a multi-drug rifampin-based regimen are generally quite good, with sputum clearance rates at 6 months of 100% and relapse rates of < 1% in several series (Griffith 2002). This compares favorably with non-rifampin based regimens historically, and is the basis for emphasizing rifampin as the cornerstone of M. kansasii treatment. The ATS/IDSA guidelines suggest for disseminated M. kansasii infection in AIDS patients who cannot receive a rifamycin because of incompatible antiretroviral therapy, a newer macrolide (e.g., clarithromycin or azithromycin) or moxifloxacin could be substituted (Griffith, Aksamit et al. 2007). Floroquinolones and linezolid have in vitro activity but clinical experience is limited (Guna, Munoz et al. 2005). For rifampin-resistant isolates in pulmonary disease, a regimen of intravenous streptomycin or amikacin, high-dose isoniazid (900 mg daily), high-dose ethambutol (25mg/kg/day), and sulfamethazole (1 gm tid) for 12-15 months from culture negativity resulted in sputum conversion in 18 of 20 pts with only one relapse(Wallace, Dunbar et al. 1994). However, overlapping nephrotoxicity between aminoglycosides and CNIs is a concern. An alternative suggested by the ATS/IDSA guidelines for a non-rifampin based regimen includes a macrolide, moxifloxacin and one other agent based on in vitro susceptibilities, but clinical outcomes have not been studied(Griffith, Aksamit et al. 2007).

For RGMs, in vitro susceptibility testing (once available) should guide therapy. M. fortuitum isolates are usually more susceptible to antimicrobials (including multiple oral agents) than other RGMs (De Groote and Huitt 2006; Griffith, Aksamit et al. 2007). Recently M. fortuitum isolates have been demonstrated to contain an inducible erythromycin methylase (erm) gene (Nash, Zhang et al. 2005), and caution should be exercised with macrolides despite a susceptible MIC, especially as monotherapy (Griffith, Aksamit et al. 2007) Isolates of M. chelonae or M. abscessus are usually reliably susceptible only to the macrolides and parenteral agents. M. abscessus pulmonary infections are particularly difficult to treat medically and the goal of therapy is often control rather than cure (Griffith, Aksamit et al. 2007). Some authors have advocated periodic intravenous therapy interspersed with combination oral therapy (macrolide plus a second-line agent such as a floroquinolone) (De Groote and Huitt 2006; Griffith, Aksamit et al. 2007). Several newer agents including linezolid and tigecycline possess in vitro activities against M. abscessus and M. chelonae and may be options in cases of resistance or interolerance to first line agents (Wallace, Brown-Elliott et al. 2001; Wallace, Brown-Elliott et al. 2002; Rodriguez Diaz, Lopez et al. 2003; Garrison, Morris et al. 2009). Clinical experience remains limited, although there have been case reports of successful therapy with linezolid in a patient with extensive soft tissue infection with M. chelonae and successful therapy with tigecycline in a HSCT recipient with necrotizing M. chelonae pneumonia (Brown-Elliott, Wallace et al. 2001; Peres, Khaled et al. 2009). In both cases, macrolide resistance developed on treatment, and the newer agents were used as salvage therapy. Specific recommendations are included in Table 2.

Optimal duration of therapy for NTM infections in transplant recipients remains unclear. The ATS/IDSA guidelines for duration of therapy should be considered as minimums, and extended therapy should be considered depending on the clinical, radiologic and microbiologic response of the patient and the net state of immunosuppression (Doucette and Fishman 2004). For pulmonary NTM disease, generally at least 18 months of treatment is considered appropriate and with at least 12 months of therapy occurring after respiratory cultures have been clear of mycobacteria. Skin and soft tissue infections can often be treated for shorter periods of time (e.g., 4 months and clinical resolution). The ATS/IDSA guidelines recommend lifelong therapy for disseminated MAC infections in AIDS patients unless immune reconstitution is achieved (Griffith, Aksamit et al. 2007). Duration of treatment in transplant patients with disseminated disease may depend on the net state of immunosuppression (e.g., waning of T-cell depletion caused by cell-depleting therapies) and, while indefinite secondary prophylaxis in transplant recipients can be considered, this approach is often limited by toxicities, drug interactions, and concerns for resistance (Doucette and Fishman 2004). Catheter-related NTM blood stream infections in HSCT patients have generally responded favorably to catheter removal and much shorter courses of antimicrobials. Given low incidence of NTM disease in organ recipients, primary prophylaxis is not routinely recommended.

The optimal approach for patients with NTM colonization or infection prior to transplantation remains unclear, particularly for lung transplant recipients in whom the distinction between NTM colonization and disease is difficult. In one study, the incidence of post-transplant invasive disease with M. abscessus was predicted by pre-transplant isolation of that species of NTM (OR 6.13, CI 3.2-11.4), although the survival of patients with post-transplant disease was not significantly different than those without (Chalermskulrat, Sood et al. 2006). Nonetheless, many centers consider pulmonary infection with M. abscessus a contraindication to lung transplantation. Patients with active NTM pulmonary disease at time of transplantation should receive treatment post-transplantation. Some authors advocate achieving smear negativity prior to transplantation (Zaidi, Elidemir et al. 2009). The approach to lung transplant recipients believed to be only colonized with NTM prior to transplantation remains uncertain, but many centers will treat with antimycobacterial drugs for a period of time after transplantation. For potential non-lung recipients with NTM disease pre-transplant, clinical response, radiological response (if lung disease), and demonstrated ability to tolerate treatment are reasonable prerequisites to organ transplantation and the required additional immunosuppression.

Surgery can serve an important role for treatment of some NTM infections. Infected foreign bodies should be removed and abscesses drained. The surgical approach to NTM pulmonary disease is determined by the species isolated and the extent of the infection. Surgical excision of a solitary pulmonary nodule incidentally found to be MAC infection can be considered curative without need for antimicrobial therapy. Lung resection surgery can be considered in patients with more extensive MAC disease if confined to one lung, although there are no universally accepted criteria for selection of patients most likely to benefit. M. kansasii generally responds well to medical management and surgical management is usually not required. In M. abscessus pulmonary infections, cure with medical therapy is unlikely and surgical resection of limited disease in conjunction with medical management is the only reliably curative treatment.

back to top

Infection Control Measures

As person to person transmission of NTM is not a major route of transmission, isolation of hospitalized patients with NTM is not required. Nosocomial outbreaks may occur related to contamination of medical devices or medications (Saiman and Siegel 2003). Recommended sterilization and decontamination techniques of medical equipment (particularly whirlpool baths and respiratory devices) should be followed to reduce the risk of hospital-acquired NTM infection.


Clinically significant infection with NTM in recipients of solid organs is relatively uncommon, occurring in <3% of patients. Lung transplant recipients are at highest risk, with pulmonary disease predominating. Skin and soft tissue infection, occasionally with dissemination, is more commonly seen in recipients of renal or heart transplants. Disseminated MAC, as observed in AIDS patients, is very rare in SOT recipients. Distinguishing pulmonary colonization with NTM from disease is challenging, but ATS/IDSA guidelines provide some guidance. Given that treatment of NTM generally requires long durations of treatment with poorly tolerated drugs that often interact with immunosuppressive medications, in questionable cases a “watchful waiting” approach possibly combined with reduction in immunosuppression may be appropriate. Rifamycins in particular should be used with great caution in SOT recipients as interactions with immunosuppressant medications are difficult to manage and rejection or drug toxicity may occur.


1. (2004). "Nontuberculous mycobacteria." Am J Transplant 4 Suppl 10: 42-6. [PubMed]

2. Aguado, J. M., J. A. Herrero, et al. (1997). "Clinical presentation and outcome of tuberculosis in kidney, liver, and heart transplant recipients in Spain. Spanish Transplantation Infection Study Group, GESITRA." Transplantation 63(9): 1278-86. [PubMed]

3. Aguado, J. M., J. Torre-Cisneros, et al. (2009). "Tuberculosis in solid-organ transplant recipients: consensus statement of the group for the study of infection in transplant recipients (GESITRA) of the Spanish Society of Infectious Diseases and Clinical Microbiology." Clin Infect Dis 48(9): 1276-84. [PubMed]

4. Brown-Elliott, B. A., D. E. Griffith, et al. (2002). "Diagnosis of nontuberculous mycobacterial infections." Clin Lab Med 22(4): 911-25, vi. [PubMed]

5. Brown-Elliott, B. A., R. J. Wallace, Jr., et al. (2001). "Successful treatment of disseminated Mycobacterium chelonae infection with linezolid." Clin Infect Dis 33(8): 1433-4. [PubMed]

6. Butler, W. R. and L. S. Guthertz (2001). "Mycolic acid analysis by high-performance liquid chromatography for identification of Mycobacterium species." Clin Microbiol Rev 14(4): 704-26, table of contents. [PubMed]

7. Cassidy, P. M., K. Hedberg, et al. (2009). "Nontuberculous mycobacterial disease prevalence and risk factors: a changing epidemiology." Clin Infect Dis 49(12): e124-9. [PubMed]

8. Chalermskulrat, W., N. Sood, et al. (2006). "Non-tuberculous mycobacteria in end stage cystic fibrosis: implications for lung transplantation." Thorax 61(6): 507-13. [PubMed]

9. Chapin-Robertson, K., S. Dahlberg, et al. (1993). "Detection and identification of Mycobacterium directly from BACTEC bottles by using a DNA-rRNA probe." Diagn Microbiol Infect Dis 17(3): 203-7. [PubMed]

10. Chernenko, S. M., A. Humar, et al. (2006). "Mycobacterium abscessus infections in lung transplant recipients: the international experience." J Heart Lung Transplant 25(12): 1447-55. [PubMed]

11. De Groote, M. A. and G. Huitt (2006). "Infections due to rapidly growing mycobacteria." Clin Infect Dis 42(12): 1756-63. [PubMed]

12. Doucette, K. and J. A. Fishman (2004). "Nontuberculous mycobacterial infection in hematopoietic stem cell and solid organ transplant recipients." Clin Infect Dis 38(10): 1428-39. [PubMed]

13. Eid, A. J., E. F. Berbari, et al. (2007). "Prosthetic joint infection due to rapidly growing mycobacteria: report of 8 cases and review of the literature." Clin Infect Dis 45(6): 687-94. [PubMed]

14. Fairhurst, R. M., B. M. Kubak, et al. (2002). "Mycobacterium abscessus empyema in a lung transplant recipient." J Heart Lung Transplant 21(3): 391-4. [PubMed]

15. Finch, C. K., C. R. Chrisman, et al. (2002). "Rifampin and rifabutin drug interactions: an update." Arch Intern Med 162(9): 985-92. [PubMed]

16. Garrison, A. P., M. I. Morris, et al. (2009). "Mycobacterium abscessus infection in solid organ transplant recipients: report of three cases and review of the literature." Transpl Infect Dis 11(6): 541-8. [PubMed]

17. Garrison, A. P., M. I. Morris, et al. (2009). "Mycobacterium abscessus infection in solid organ transplant recipients: report of three cases and review of the literature." Transpl Infect Dis. [PubMed]

18. Griffith, D. E. (2002). "Management of disease due to Mycobacterium kansasii." Clin Chest Med 23(3): 613-21, vi.  [PubMed]

19. Griffith, D. E., T. Aksamit, et al. (2007). "An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases." Am J Respir Crit Care Med 175(4): 367-416. [PubMed]

20. Griffith, D. E., W. M. Girard, et al. (1993). "Clinical features of pulmonary disease caused by rapidly growing mycobacteria. An analysis of 154 patients." Am Rev Respir Dis 147(5): 1271-8. [PubMed]

21. Guna, R., C. Munoz, et al. (2005). "In vitro activity of linezolid, clarithromycin and moxifloxacin against clinical isolates of Mycobacterium kansasii." J Antimicrob Chemother 55(6): 950-3. [PubMed]

22. Gupta, A. and H. Clauss (2009). "Prosthetic joint infection with Mycobacterium avium complex in a solid organ transplant recipient." Transpl Infect Dis. [PubMed]

23. Jie, T., A. J. Matas, et al. (2005). "Mycobacterial infections after kidney transplant." Transplant Proc 37(2): 937-9. [PubMed]

24. Keller, M., A. Mak, et al. (2008). "Mycobacterium haemophilum epididymal abscess in a renal transplant patient." J Clin Microbiol 46(7): 2459-60. [PubMed]

25. Kirschner, P., B. Springer, et al. (1993). "Genotypic identification of mycobacteria by nucleic acid sequence determination: report of a 2-year experience in a clinical laboratory." J Clin Microbiol 31(11): 2882-9. [PubMed]

26. Kuypers, D. R., G. Verleden, et al. (2005). "Drug interaction between mycophenolate mofetil and rifampin: possible induction of uridine diphosphate-glucuronosyltransferase." Clin Pharmacol Ther 78(1): 81-8. [PubMed]

27. McAllister, W. A., P. J. Thompson, et al. (1983). "Rifampicin reduces effectiveness and bioavailability of prednisolone." Br Med J (Clin Res Ed) 286(6369): 923-5. [PubMed]

28. Naesens, M., D. R. Kuypers, et al. (2006). "Rifampin induces alterations in mycophenolic acid glucuronidation and elimination: implications for drug exposure in renal allograft recipients." Clin Pharmacol Ther 80(5): 509-21. [PubMed]

29. Nagy, G. S. and R. H. Rubin (2001). "Disseminated Mycobacterium avium-intracellulare in a kidney transplant recipient." Transpl Infect Dis 3(4): 220-30. [PubMed]

30. Nash, K. A., Y. Zhang, et al. (2005). "Molecular basis of intrinsic macrolide resistance in clinical isolates of Mycobacterium fortuitum." J Antimicrob Chemother 55(2): 170-7. [PubMed]

31. Patel, R., G. D. Roberts, et al. (1994). "Infections due to nontuberculous mycobacteria in kidney, heart, and liver transplant recipients." Clin Infect Dis 19(2): 263-73. [PubMed]

32. Peres, E., Y. Khaled, et al. (2009). "Mycobacterium chelonae necrotizing pneumonia after allogeneic hematopoietic stem cell transplant: report of clinical response to treatment with tigecycline." Transpl Infect Dis 11(1): 57-63. [PubMed]

33. Pinho, L., J. Santos, et al. (2009). "Mycobacterium gordonae urinary infection in a renal transplant recipient." Transpl Infect Dis 11(3): 253-6. [PubMed]

34. Rodriguez Diaz, J. C., M. Lopez, et al. (2003). "In vitro activity of new fluoroquinolones and linezolid against non-tuberculous mycobacteria." Int J Antimicrob Agents 21(6): 585-8. [PubMed]

35. Sadaba, B., A. Lopez de Ocariz, et al. (1998). "Concurrent clarithromycin and cyclosporin A treatment." J Antimicrob Chemother 42(3): 393-5. [PubMed]

36. Saiman, L. and J. Siegel (2003). "Infection control recommendations for patients with cystic fibrosis: microbiology, important pathogens, and infection control practices to prevent patient-to-patient transmission." Infect Control Hosp Epidemiol 24(5 Suppl): S6-52. [PubMed]

37. Taylor, J. L. and S. M. Palmer (2006). "Mycobacterium abscessus chest wall and pulmonary infection in a cystic fibrosis lung transplant recipient." J Heart Lung Transplant 25(8): 985-8. [PubMed]

38. Wallace, R. J., Jr., B. A. Brown-Elliott, et al. (2002). "Comparison of the in vitro activity of the glycylcycline tigecycline (formerly GAR-936) with those of tetracycline, minocycline, and doxycycline against isolates of nontuberculous mycobacteria." Antimicrob Agents Chemother 46(10): 3164-7. [PubMed]

39. Wallace, R. J., Jr., B. A. Brown-Elliott, et al. (2001). "Activities of linezolid against rapidly growing mycobacteria." Antimicrob Agents Chemother 45(3): 764-7. [PubMed]

40. Wallace, R. J., Jr., D. Dunbar, et al. (1994). "Rifampin-resistant Mycobacterium kansasii." Clin Infect Dis 18(5): 736-43. [PubMed]

41. Wright, P. W., R. J. Wallace, Jr., et al. (1998). "Sensitivity of fluorochrome microscopy for detection of Mycobacterium tuberculosis versus nontuberculous mycobacteria." J Clin Microbiol 36(4): 1046-9. [PubMed]

42. Zaidi, S., O. Elidemir, et al. (2009). "Mycobacterium abscessus in cystic fibrosis lung transplant recipients: report of 2 cases and risk for recurrence." Transpl Infect Dis 11(3): 243-8. [PubMed]

back to top


Table 1 – ATS/IDSA Criteria for Diagnosis of Nontuberculous Pulmonary Disease ADDIN EN.CITE  ADDIN (Griffith, Aksamit et al. 2007)

Clinical Criteria

Pulmonary symptoms
Nodular or cavitary opacities on CXR
Thoracic CT scan with multifocal bronchiectasis with multiple small nodules


Exclusion of other diagnoses

  Microbiologic Criteria

Positive culture results from at least two expectorated sputum samples
Positive culture result from at least one bronchial wash or lavage in patient with classic symptoms and radiographic findings for nodular bronchiectatic NTM lung disease unable to produce sputum
Lung biopsy with evidence of granulomatous inflammation or AFB on histopathology and either a tissue culture or sputum specimen or BAL washing with growth of an NTM

Diagnosis requires both clinical and microbiological criteria.


Table 2: Treatment regimens for NTM

NTM species Preferred regimen, based upon ATS/IDSA guidelines Alternative regimens and/or antibiotics Duration Adjuvant treatment Comments
Mycobacterium Avium complex For cavitary disease:
Clarithromycin 500-1000 mg po daily or azithromycin  250-500 mg po daily


Rifampin 600 mg po daily (450 mg po daily if wt < 50 kg) (a)


Ethambutol 15 mg/kg/day
For severe disease or previously treated, consider adding amikacin or streptomycin IV to the above three-drug regimen.
Alternatives to rifampin, as part of a multi-drug regimen: • Levofloxacin • Moxifloxacin • Amikacin • Streptomycin • Rifabutin
In patients with mild disease and only nodular-bronchiectatic lung disease without cavitation, can consider  a two-drug regimen of a macrolide and ethambutol (Griffith, Askamit 2007))
Skin, soft tissue, and bone infections – 6-12 months
Pulmonary infections -  at least 12 months after negative sputum cultures
Surgical management if : • Lymphadenitis • Abscess formation • Intolerance of drug therapy • Retained foreign body
Surgical resection of focal pulmonary disease can be considered in centers with extensive experience and in conjunction with medical treatment.  Clinical evidence is limited to small series.
Macrolides are the cornerstone of MAC therapy, although  they should always be used with companion medications to prevent development of resistance.
Consider substitution of rifabutin (or alternative agent) for rifampin given potent drug interaction

 Mycobacterium kansasii

Rifampin 10 mg/kg/day (max 600 mg/day) (b,c)
Ethambutol 15mg/kg/day
 INH 5 mg/kg/day (max 300 mg/day) + pyridoxine 50 mg/day (d)
Rifampin 600 mg and ethambutol 25mg/kg and clarithromycin 500-1000 mg, all dosed three times weekly
Rifampin-resistant isolates:
3 of the following agents based on in vitro susceptibilities: • Macrolide • Moxifloxacin (e) • Ethambutol • ulfamethoxazole • Streptomycin
IV streptomycin or amikacin, INH (900 mg daily), ethambutol (25mg/kg/day), and sulfamethazole (1 gm po tid) (Wallace, Dunbar 1994)
Agents with in vitro activity, but little clinical experience: • Linezolid
Pulmonary infections – 12 months of negative sputum cultures. Given the excellent response to medical treatment with rifamycin-based regimens, surgical management is generally not required.




Consider substitution of rifabutin, a macrolide, or moxifloxacin for rifampin as less potent drug interaction with immunosuppressants.


Mycobacterium fortuitum

2 of the following agents with in vitro activity based on susceptibility testing: (percentage of untreated isolates usually susceptible):
• Imipenem (100%)
• Amikacin (100%)
• Fluoroquinolone  (100%)
• ulfonamides(100%) (sulfamethoxazole or TMP/SMX)
• Clarithromycin (80%) (f)
• Doxycycline 100 mg po daily (50%)
• Cefoxitin (50%)
As part of multi-drug regimen:

•  Linezolid (g)

Skin and soft tissue infections – 4 months
Bone infections – 6 months
Pulmonary infections – at least 12 months of negative sputum cultures
Surgical management if :
• Extensive disease
• Abscess formation
• Intolerance of drug therapy
• Retained foreign body




Mycobacterium chelonae

A macrolide (e.g., Clarithromycin 1000 mg po daily)
One other active agent based on in vitro susceptibility testing (percentage of untreated isolates usually susceptible):
• Tobramycin 5mg/kg per day in 2-3 divided doses (100%)
• Imipenem 1 g IV q 6 hrs (60%)[d]
• Amikacin (50%)
• Doxycycline (25%)
• Ciprofloxacin (20%)
As part of multi-drug regimen  (percentage of untreated isolates usually susceptible):
• Linezolid (90%)
• Tigecycline
Duration of treatment same as for mycobacterium fortuitum. Surgical management same as for mycobacterium fortuitum.


Mycobacterium abscessus

A macrolide (clarithromycin 1000 mg po daily or azithromycin 250-500 mg po daily)
the following IV agents:
• Amikacin
• Cefoxitin 12g IV/day in divided doses or imipenem IV if cefoxitin unavailable.  In combination with amikacin until clinical stability or at least 2 wks of treatment.
For macrolide-resistant isolates, a combination of parenteral agents based on in vitro susceptibility testing.
Alternative second-line agents as part of multi-drug regimen:
• Linezolid
• Tigecycline
• Moxifloxacin
• Telithromycin
Efficacy of inhaled amikacin unknown, but could be considered in patients at high risk of renal failure on intravenous amikacin.
Skin and soft tissue infections – 4 months
Bone infections – 6 months
Pulmonary infections – no antibiotic regimens have been shown to reliably produce long-term sputum conversion. Several approaches reasonable.  Some advocate periodic antibiotics (macrolides and/or parenteral agents) either to control symptomatic disease flares or at regular intervals for disease suppression.
Surgical management for non-pulmonary infection if:
• Extensive disease
• Abscess formation
• Intolerance of drug therapy
• Retained foreign body
Surgical management of limited, focal pulmonary disease is the only predictably curative therapy.


M. marinum



Multidrug regimen with at least two active agents:
• Rifampin (a)
• Rifabutin (a)
• Ethambutol
• Clarithromycin
• Sulfonamides
• Doxycycline  (intermediate susc)
> • Minocycline  (intermediate susc)



1 – 2 months after resolution of symptoms, typically 3-4 months total duration

Surgical management if:
• Infection involving closed spaces of the hand
• Failure to respond to medical management
Rifampin or rifabutin should be avoided if other alternatives available
Mycobacterium haemophilum Optimal treatment regimens unkown, but should involve a multidrug regimen
Agents considered active in vitro:
• Amikacin
• Clarithromycin
• Ciprofloxacin
• Rifampin
• Rifabutin
Agents considered variably active:
• Doxycycline
• Sulfonamides


Optimal duration is unknown, but soft tissue infection should be treated at least a few months beyond resolution of symptoms.
Treatment duration for disseminated disease should generally be longer.



MAC=mycobacterium avium complex, INH=isoniazid, IV=intravenous
a – Rifabutin is a less potent inducer of p450, and can be substituted for rifampin.  Should still be used with caution in patients on concurrent CNIs or sirolimus.
b – Isolates susceptible to rifampin can also be considered susceptible to rifabutin.
c – Consider substitution of rifabutin, a newer macrolide, or moxifloxacin for the rifampin.
d – M. kansasii may be reported as resistant to INH with MICs of 0.2 or 1 ug/mL, the standard concentrations tested for M. tuberculosis.  In patients with no prior exposure to INH, the drug may still be useful in the treatment of M kansasii infection, because isolates are often susceptible to slightly higher drug concentrations.
e – Test susceptibilities specifically to moxifloxacin.  Ciprofloxacin susceptibilities not representative of moxifloxacin.
f – M. fortuitum has inducible erm gene, theoretical concern for treatment failure on macrolides, especially when used as monotherapy. (Nash, Zhang et al. 2005)
g – Linezolid less active against M. fortuitum than M. abscessusmm. chelonae. (Rodriguez Diaz, Lopez et al. 2003)


Table 3 - Antimycobacterial Medications for Treatment of NTM Infections

Antibiotic Selected drug interactions with immunosuppressive medications Anticipated effect on selected agents Selected toxicities and side effects Monitoring parameters
• Azithromycin
• Clarithromycin
CNI (cyclosporine and tacrolimus)
Increased levels of immunosuppressive agent (clarithromycin with more pronounced effect) GI (nausea,vomiting, diarrhea)
Decreased hearing
Clinical symptoms
Periodic LFTs during first 2-3 months
CNI and sirolimus levels
• Rifampin
• Rifabutin
Decreased levels of immunosuppressive agent Hepatitis
Orange discoloration of urine
GI effects
Hypersensitivity(rash, fever)
Flu-like syndrome
Hematologic (leucopenia, granulocytopenia)
Clinical symptoms
Periodic CBC, LFTs
CNI and sirolimus levels
Ethambutol None   Optic neuritis Periodic testing for red/green color discrimination and visual acuity
Clinical symptoms
• Ciprofloxacin
• Levofloxacin
• Moxifloxacin
Increased level of immunosuppressive medication (mild) GI effects
Tendonitis and tendon rupture
CNS (headache, insomnia, agitation)
QT prolongation
Clinical symptoms
CNI and sirolimus levels
• Minocycline
• Doxycycline



GI effects
Vertigo, dizziness
Increased LFTs
Clinical symptoms
Tigecycline None   GI effects
Vertigo, dizziness
Increased LFTs
Clinical symptoms
CNIs  Increased risk of nephrotoxicity Nephrotoxicity
Vestibular and auditory toxicity (dizziness, vertigo, ataxia, tinnitus, hearing loss)
Clinical symptoms Periodic serum creatinine measurements Baseline and periodic hearing tests in high risk individuals

None   Hepatitis
Hypersensitivity (rash, fever)
Periodic LFTs
Clinical symptoms
Linezolid No interactions with immunosuppressants
Interaction with serotonergic agents (many antidepressants) can result in serotonin syndrome
  Myelosuppressive (leukopenia, anemia, thrombocytopenia)
Optic neuropathy
Peripheral neuropathy
Periodic CBC.
Clinical symptoms

CNI=calcineurin inhibitor, GI=gastrointestinal, LFT=liver function tests, CBC=complete blood count


Non-Tuberculous Mycobacteria in Solid Organ Transplant Recipients