Transmissibility of tuberculosis among school contacts: an outbreak investigation in a boarding middle school, China

Abstract

Tuberculosis (TB) outbreak occurred in a boarding middle school of China. We explored its probable sources and quantified the transmissibility and pathogenicity of TB. Clinical evaluation, tuberculin skin testing and chest radiography were conducted to identify TB cases. Mycobacterium tuberculosis isolates underwent genotyping analysis to identify the outbreak source. A chain-binomial transmission model was used to evaluate transmissibility and pathogenicity of TB. A total of 46 active cases were ascertained among 258 students and 15 teachers/staff, an attack rate of 16.8%. Genetic analyses revealed two groups of M. tuberculosis cocirculating during the outbreak and possible importation from local communities. Secondary attack rates among students were 4.1% (2.9%, 5.3%) within grade and 7.9% (4.9%, 11%) within class. An active TB case was estimated to infect 8.4 (7.2, 9.6) susceptible people on average. The smear-positive cases were 28 (8, 101) times as infective as smear-negative cases. Previous BCG vaccination could reduce the probability of developing symptoms after infection by 70% (1.4%, 91%). The integration of clinical evaluation, genetic sequencing, and statistical modeling greatly enhanced our understanding of TB transmission dynamics. Timely diagnosis of smear-positive cases, especially in the early phase of the outbreak, is the key to preventing further spread among close contacts.

Keywords: Chain-binomial transmission model; Genome sequencing; Outbreak; Transmission; Tuberculosis.

Fig. 1

Epidemic curve for the TB outbreak among students and staff attending a boarding middle school in China. The question marks represent indetermination of sputum smear status.

Fig. 2

Hierarchical clustering of patterns of Mycobacterial Interspersed Repetitive Unit-Variable Number Tandem Repeats (MIRU-VNTRs) and Spacer Oligonucleotide Typing (spoligotyping). Six isolates from Shijiazhuang, Hebei Province (HB) and 14 isolates from the Yushu County, Qinghai Province (QH) collected between 2009 and 2011 were used to root the MIRU-VNTR tree. MIRU loci are listed along the top of the panel, with the values in each column indicating the allele present.

Fig. 3

Phylogenetic analysis of Whole-Genome-Derived Genetic Distances of 13 Mycobacterium tuberculosis outbreak isolates (MT) and 4 historical isolates (HB). The whole genome analysis of single-nucleotide polymorphisms of the outbreak isolates reveals two lineages: group A composed mostly of the outbreak isolates and group B mainly from the endemic isolates in the community.

Fig. 4

Goodness of fit of model-predicted numbers (colored curves) to the observed numbers (histogram) of symptom onsets for a variety of settings of tuning parameters. (a) The four earliest cases are primary cases. (b) The two earliest cases are primary cases. The vertical dashed lines indicate the beginning date of the summer break, the ending date of the summer break, and the intervention starting date, in temporal order. The colors and line types are coded as follows, where φ₁ determines the shape of the incubation period, D_max is the maximum incubation period, and RR_prim is the relative infectiousness level of primary cases. Red solid line: φ₁ =10, D_max =14, RR_prim =5; Blue dotted line: φ₁=3, D_max =14, RR_prim =5; Blue dashed line: φ₁ =15, D_max =14, RR_prim =5; Orange dotted line: φ₁=10, D_max =8, RR_prim =5; Orange dashed line:φ ₁ =10, D_max =20, RR_prim =5; Green dotted line: φ₁ =10, D_max =14, RR_prim =1; Green dashed line: φ₁ =10, D_max =14, RR_prim=10.