Association of Healthcare and Aesthetic Procedures with Infections Caused by Nontuberculous Mycobacteria, France, 2012‒2020

Abstract

We describe nontuberculous mycobacteria (NTM) infections during 2012-2020 associated with health care and aesthetic procedures in France. We obtained epidemiologic data from the national early warning response system for healtcare-associated infections and data on NTM isolates from the National Reference Center for Mycobacteria. We compared clinical and environmental isolates by using whole-genome sequencing. The 85 original cases were reported after surgery (48, 56%), other invasive procedures (28, 33%) and other procedures (9, 11%). NTM isolates belonged to rapidly growing (73, 86%) and slowly growing (10, 12%) species; in 2 cases, the species was not identified. We performed environmental investigations for 38 (45%) cases; results for 12 (32%) were positive for the same NTM species as for the infection. In 10 cases that had environmental and clinical samples whose genomes were similar, the infection source was probably the water used in the procedures. NTM infections could be preventable by using sterile water in all invasive procedures.

Keywords: France; aesthetic procedure‒associated infections; bacteria; case reports; comparative genomics; epidemiology; healthcare-associated infections; mycobacteria; nontuberculous mycobacteria infections; respiratory infections; retrospective study; tuberculosis and other mycobacteria.

Figure 1

Onset of clinical signs in 85 reported cases of infection with nontuberculous mycobacteria associated with healthcare and aesthetic procedures, by year, France, January 2012‒June 2020. Blue line indicates linear fit for 2012‒2019 after excluding incomplete year 2020.

Figure 2

Genomic comparison of nontuberculous mycobacteria isolates by using whole-genome sequencing phylogenetic analysis and maximum parsimony trees. A) Mycobacterium chimaera, B) M. fortuitum, C) M. chelonae, D) M. marinum, E) M. mucogenicum, F) M. neoaurum. Environmental isolates are indicated in blue, and clinical isolates are indicated in red. Additional information for the 6 Mycobacterium species tested is provided in the Appendix. Ref, referent. Panel A: Based on 19,621 single-nucleotide polymorphisms (SNPs) generated from comparison of 16 genomes using square root scaling. Isolates were 1) clinical isolates from 2 patients who had M. chimaera disseminated disease after open-heart surgery (A1 and A2 from the first patient), A3 from the second patient; 2) a reference genome from a previously described epidemic patient (8); 3) 10 heater-cooler unit water samples from the hospital where the second patient underwent surgery, collected 4 years after the report (A4‒A13); 4) clinical sample from a patient who had a breast prosthesis infection (B1); and 5) the environmental isolate collected in the hospital’s water network 1 year after the report of the breast prosthesis infection (B2). Panel B: Based on 27,796 SNPs generated from comparison of 4 genomes using square root scaling. Isolates were 1) clinical isolate from a patient who had M. fortuitum breast prosthesis infection (C1) and 2) environmental isolate collected 1 month after the report in the water supply network in the shower of the patient’s hospital room (C2, C3). Panel C: Based on 67,759 SNPs generated from comparison of 24 genomes using square root scaling. Isolates were 11 clinical isolates: 1 from an infection after face lift surgery (D1), 3 from 2 skin and soft tissue infections after a tattoo (E1‒E2 from the same patient and E3 from another patient who was tattooed in the same tattoo parlor); 2 from 2 mesotherapy infections from 2 nonrelated reports (F1 and G1); and 5 using supply networks were obtained for investigations including: 1 isolate from a surgical sink (D2), 4 isolates from a doctor’s office sink (F2‒F3, G5‒G6), 2 isolates from a tattoo parlor’s sink used to dilute the ink (E4‒E5) and 5 isolates from a patient’s home (F4‒F5, G2‒G4). Panel D: Based on 24,757 SNPs generated from comparison of 4 genomes using square root scaling. Isolates were 1) a clinical isolate from a skin and soft tissue infection caused by contamination after a bath in a balneotherapy swimming pool (I1 and I2) 2 environmental isolates from the swimming pool (I2 and I3). Panel E: Based on 53,551 SNPs generated by comparison of 8 genomes using square root scaling. Isolates were 1) 3 clinical isolates from catheter-associated infections (3 patients J1‒J3) and 2) 4 environmental isolates from hospital water supply networks (J4‒J7). Panel F: Based on 58,473 SNPs generated by comparison of 4 genomes using square root scaling. Isolates were 1) 2 clinical isolates from the blood culture (K1 and K2) of 1 patient and 2) 1 isolate from microbiological control after an autologous stem cell transplant (K3).