High overall mortality of Mycobacterium genavense infections and impact of antimycobacterial therapy: systematic review and individual patient data meta-analysis.

Abstract Introduction Mycobacterium genavense is a fastidious slow growing mycobacterium (SGM) that causes disseminated infections in immunocompromised hosts. It has been described in HIV-positive individuals and increasingly in patients without HIV. The infections are difficult to treat and the optimal antimycobacterial regimen is still unknown. Methods An individual patient data meta-analysis was conducted aiming at including all hitherto published cases of infection with M. genavense. Clinical manifestations, microbiological data, dispositions and immunosuppression were recorded. Antimycobacterial therapies and mortality were analysed by logistic regression and time-to-event analysis. Results We included 223 patients with infection due to M. genavense published from 1992-2021. While the majority was HIV positive (n=223, 76.7 %), 52 patients were non-HIV-patients (23.3 %), 36 of whom received immunosuppressive therapy (69 %). We could confirm the bacterium's tropism for the gastrointestinal tract with abdominal pain, hepato-/splenomegaly and abdominal lymphadenopathy being major clinical manifestations. More than 90 % of patients received antimycobacterial therapy. The regimens consisted mainly of macrolides, rifamycins and ethambutol. Overall mortality was high, but in logistic regression and time-to-event analysis a macrolide containing regimen was associated with better outcomes. Conclusion In this first individual patient data meta-analysis of infections with M. genavense we confirm its tropism for the gastrointestinal tract. The high overall mortality underlines the clinical relevance of infection with this bacterium for the individual patient. In addition, our data give a hint that a macrolide containing regimen is associated with better survival.