Antibiotic treatment for non-tuberculous mycobacteria lung infection in people with cystic fibrosis.

Rationale: Cystic fibrosis (CF) is a common genetic condition in which progressive lung disease leads to morbidity and mortality. Non-tuberculous mycobacteria (NTM) are mycobacteria, other than those in the Mycobacterium tuberculosis complex, and are commonly found in the environment. NTM pulmonary infections affect a significant proportion of people with CF worldwide, which may be associated with a more rapid decline in lung function and even death in certain circumstances. Although there are guidelines for the antimicrobial treatment of NTM lung disease, there is no specific evidence from studies of people with CF to inform recommendations for their treatment. It is not clear which antibiotic regimen may be the most effective in the treatment of people with CF. This is an update of a previous review.

Objectives: To compare antibiotic treatment to no antibiotic treatment, or to compare different combinations of antibiotic treatment, for suppressing or eradicating non-tuberculous mycobacteria (NTM) lung infections in people with cystic fibrosis (CF).

Search methods: We searched Cochrane's Cystic Fibrosis Trials Register, online databases (MEDLINE, Embase and PubMed) and online trials registries (www.

Clinicaltrials: gov and the World Health Organization International Clinical Trials Registry). We also searched the reference lists of included studies and relevant reviews. The date of the last search was 14 October 2024.

Eligibility criteria: Randomised controlled trials (RCTs) or quasi-RCTs with a parallel design; non-randomised studies of interventions (NRSIs) with the following designs: instrumental variables; regression discontinuity; interrupted time series; difference-in-differences and fixed-effect designs. These should have compared antibiotic treatment to no antibiotic treatment, or different combinations of antibiotic treatment, in people with CF of any age with NTM pulmonary infection.

Outcomes: We aimed to assess the critical outcomes of microbiological clearance of NTM in sputum, quality of life, adverse events, lung function and pulmonary exacerbations. Further, we planned to assess important outcomes of mortality, nutritional parameters, hospitalisations and use of additional oral antibiotics.

Risk of bias: We planned to use the recommended Cochrane tools for RCTs or NRSIs. These were not suitable for the included study, so we assessed the risk of bias using a tool for case series developed by the Joanna Briggs Institute.

Synthesis methods: We were only able to report the limited results from the single included study narratively. We assessed the certainty of the results using GRADE.

Included studies: Due to a lack of studies of the types planned, we were only able to include a single retrospective case review, which presented data as the change from baseline for some outcomes. It was conducted in Sweden in 2003 and included 11 participants with CF and NTM infection (three males) aged between 10 and 36 years. The study identified the specific cystic fibrosis transmembrane conductance regulator (CFTR) mutation for 10 participants. All participants were chronically colonised with Pseudomonas aeruginosa; 10 participants had been vaccinated with the Bacillus Calmette-Guérin vaccine. Antibiotic selection differed amongst participants and was determined according to in vitro susceptibility testing. Antibiotics included isoniazid, ethambutol, rifampicin (or rifabutin), amikacin, clarithromycin, ciprofloxacin, streptomycin and clofazimine. Of note, at the start of the study, isoniazid was the standard treatment for NTM, and three participants received this drug; however, investigators stated that following severe adverse effects, the drug was excluded in the latter part of the 1980s. Investigators reported data for lung function, weight and adverse events one year before NTM diagnosis, at baseline, at completion of therapy and at the latest follow-up (ranging from one to 14 years). Treatment was considered effective if NTM was cleared and cultures remained negative throughout treatment; it was considered to have failed if there were continued or sporadic positive cultures.

Synthesis of results: We graded all the evidence as very low and are very uncertain of the effects of the different antibiotic regimens on any of the outcomes reported. The study reported that in 10/11 participants, microbiological cultures turned negative. They also stated that five participants reported adverse events; three reported photosensitivity to ciprofloxacin, while each of the following events was reported by one of the five participants: impaired hearing, convulsions, neuropathy and lupus erythematous. There was no consistent effect on lung function. Investigators reported that forced expiratory volume in one second increased by between 1% predicted and 46% predicted in six participants, decreased between 2% predicted and 31% predicted in four participants and remained the same in one participant. They also reported that forced vital capacity increased in eight participants by between 3% predicted and 53% predicted, and decreased in three participants by between 4% predicted and 21% predicted. Two participants died as a result of progression of CF respiratory disease two years after completion of therapy. A further participant died of gastrointestinal bleeding and renal insufficiency eight years after lung transplant which followed clearance of NTM infection (negative NTM cultures were maintained until death). Eight participants gained weight (range 3.30 kg to 14.00 kg), while three participants lost weight (range -0.90 kg to -6.00 kg). Investigators additionally reported body mass index values in three participants, which decreased minimally in two participants and increased slightly in the third participant.

Authors' conclusions: The very low-certainty evidence identified in this review suggests that antimicrobial treatment may lead to sputum clearance of NTM in people with CF, but may result in variable clinical response in terms of lung function. Very low-certainty evidence also suggests that adverse events may be common, necessitating close monitoring. This review highlights the need for larger, more standardised studies in order to make meaningful comparisons between treatment regimens. Although microbiological clearance seems feasible, studies should be powered to detect relevant clinical outcomes as well.

Funding: Cochrane CF received funding from the Cystic Fibrosis Foundation for a series of reviews on NTM, of which the update of this review is one.

Registration: The protocol for this updated version of the review was registered at PROSPERO in November 2023.