Massive and Lengthy Clonal Nosocomial Expansion of Mycobacterium abscessus subsp. massiliense among Patients Who Are Ventilator Dependent without Cystic Fibrosis

Abstract

Nontuberculous mycobacterial infections are generally believed to be independently acquired from the environment. Although person-to-person transmission of nontuberculous mycobacteria, especially Mycobacterium abscessus subsp. massiliense, is a serious concern among individuals with cystic fibrosis (CF), evidence of its spread among patients without CF has never been established. We unexpectedly found a number of M. abscessus subsp. massiliense cases among patients without CF in a hospital. This study aimed to define the mechanism of M. abscessus subsp. massiliense infection among patients who were ventilator dependent and without CF who had progressive neurodegenerative diseases in our long-term care wards from 2014 to 2018 during suspected nosocomial outbreaks. We conducted whole-genome sequencing of M. abscessus subsp. massiliense isolates from 52 patients and environmental samples. Potential opportunities for in-hospital transmission were analyzed using epidemiological data. M. abscessus subsp. massiliense was isolated from one air sample obtained near a patient without CF who was colonized with M. abscessus subsp. massiliense but not from other potential sources. Phylogenetic analysis of the strains from these patients and the environmental isolate revealed clonal expansion of near-identical M. abscessus subsp. massiliense isolates, with the isolates generally differing by fewer than 22 single nucleotide polymorphisms (SNPs). Approximately half of the isolates differed by fewer than nine SNPs, indicating interpatient transmission. Whole-genome sequencing revealed a potential nosocomial outbreak among patients who were ventilator dependent and without CF. IMPORTANCE The isolation of M. abscessus subsp. massiliense from the air, but not from environmental fluid samples, may suggest airborne transmission. This was the first report to demonstrate person-to-person transmission of M. abscessus subsp. massiliense, even among patients without CF. M. abscessus subsp. massiliense may spread among patients who are ventilator dependent without CF through direct or indirect in-hospital transmission. The current infection control measures should address potential transmission among patients without CF, particularly in facilities that treat patients who are ventilator dependent and patients with preexisting chronic pulmonary diseases, such as CF.

Keywords: Mycobacterium abscessus subsp. massiliense; nontuberculous mycobacteria; progressive neurodegenerative disease; transmission; whole-genome sequencing.

FIG 1

Phylogenetic association of isolates in the current study with other isolates in the global Mycobacterium abscessus subsp. massiliense population. A core genome alignment of 376 isolates from five regions (Australia, East Asia, Europe, North America, and South America) was used to determine the phylogeny of the global M. abscessus subsp. massiliense population, including NBP isolates. The complete genome sequence of JCM15300 was used as a reference. The recombination-free alignment in the core genome was used with RAxML to construct a maximum likelihood tree with 1,000 bootstrap replicates. The 12 monophyletic clusters (MAS-GL1 to MAS-GL12) identified using TreeGubbins are presented. Each color box corresponds to the region from which the clinical isolate was isolated. The corresponding patients’ disease statuses (cystic fibrosis [CF] in yellow and non-CF in white) are presented. The presence (black) and absence (white) of macrolide or aminoglycoside resistance-associated mutations are indicated. The scale bar indicates the mean number of nucleotide substitutions (SNPs) per site on the respective branch.

FIG 2

Distribution of pairwise genetic distances among Mycobacterium abscessus subsp. massiliense isolates. (A) Histogram of pairwise single nucleotide polymorphism (SNP) distances among M. abscessus subsp. massiliense isolates consecutively obtained from the same patients. As described in Fig. 1, the whole-genome alignment was used to calculate SNP distances among isolates using snp-dists (https://github.com/tseemann/snp-dists). The 99th, 98th, and 50th percentiles of the SNP distances among isolates are presented. The purple line indicates the cumulative density of these SNP distances. (B) Comparison of pairwise SNP distances within NBP strains and those within clinical isolates constitutively isolated from the same patients.

FIG 3

Possible clonal expansions among Mycobacterium abscessus subsp. massiliense isolates. rPinecone software was used to detect possible clonal expansions (CEs) in the M. abscessus subsp. massiliense population. The phylogenetic tree described in Fig. 1 was converted to a single nucleotide polymorphism (SNP)-scaled tree using pyjar. This resulting tree and the SNP threshold (20 SNPs, accounting for 98.5% of SNPs among M. abscessus subsp. massiliense isolates consecutively obtained from the same patients [Fig. 2]) were used as inputs for rPinecone. Blue boxes indicate detected CEs. The indicated phylogenetic tree was estimated as described in Fig. 1. The 12 MAS-GL clusters, the disease status (cystic fibrosis [CF] or non-CF) of the corresponding patients, and the region from which the clinical isolate was obtained are shown in Fig. 1.

FIG 4

Genomic signatures shared among Mycobacterium abscessus subsp. massiliense NBP isolates. (A) The presence (black) and absence (white) of genes that distinguish NBP strains from other strains are indicated. Phylogenetic trees were determined as described in Fig. 1. The 12 MAS-GL clusters, the disease status (cystic fibrosis [CF] or non-CF) of the corresponding patients, and the region from which the clinical isolate was obtained are presented in Fig. 1. (B) Uniting pairs containing substitutions between NBP isolates (n = 51) and other M. abscessus subsp. massiliense strains; (C) genomic island exclusively shared among NBP clinical and environmental isolates. Arrows indicate genes annotated with DFAST-core (27). Orthologous genes between clinical isolates are presented in red. Thirteen genes exclusively associated with NBP clinical and environmental isolates are presented in yellow.

FIG 5

Opportunities for patient-to-patient transmission among NBP patients. (A) History of patient room transfers. A single-headed arrow denotes patients moving between rooms, and a double-headed arrow denotes patients in the same room at the same time. The number beside the double-headed arrow indicates the single nucleotide polymorphism distance. The timing of Mycobacterium abscessus subsp. massiliense isolation is displayed by different colors as follows: blue (patients 1 and 2) in 2014, green (from patients 3 to 6) in 2015, red (from patients 7 to 20) in 2016, purple (patients 21 and 22) in 2017, and orange (patients 23 to 52) in 2018. The dotted line for room 506 indicates contacts at different times. (B) Minimum-spanning network of M. abscessus subsp. massiliense NBP clinical and environmental isolates. Using the same alignment as in Fig. 1, a minimum-spanning network within CE18 and CE19 was constructed using the igraph R package. Each circle’s outer or inner color represents the year and ward in which the corresponding clinical or environmental isolate was isolated. The number in each circle represents the patient number (listed in Table S1) from which each clinical isolate was isolated. The red lines indicate that there are records of two corresponding patients admitted to the same room, the black lines indicate that there are records of two corresponding patients admitted to the same or adjacent floor, and the gray lines indicate that the corresponding patients were admitted to distant floors.