The outbreak occurred at a public high school that accommodates boarding students. It is segregated into junior and senior high school sections, which are located at considerable distances from each other. The school had 3,710 total students, including 1,860 senior high students distributed across 36 classes (average 52 students / class). The teaching buildings at the senior high school were configured in a “B” shape (Figure 1). The school’s boarding program was exclusively for senior high students, with a total of 1,435 boarding students accommodated. Two student dormitory buildings were present, with the lower floors (first to fourth) designated for boys and the upper floors (fifth to seventh) for girls. Each dormitory housed eight students.

The outbreak was confined to the senior high school division. We retrospectively reviewed the outbreak survey data, investigation records, and medical records. This included detailed information on the students and school, such as gender, classroom and residential status, classroom and dormitory distributions, tuberculin skin test (TST) results, and clinical information of TB cases. Additionally, we interviewed physicians and nurses from the school health clinic. We also conducted in-person interviews of case-students or telephoned their parents to gather fundamental case history, time of disease onset, how medical services were sought, diagnosis, and treatment.

All TB cases, including laboratory-confirmed cases, clinically diagnosed cases, and suspected cases, were defined and classified following the Diagnostic Criteria for Pulmonary TB (WS 288–2017) and Classification of TB (WS 196-2017) in China (21, 22). Among these, suspected cases were defined as individuals presenting radiological features consistent with TB, representing pre-diagnostic cases requiring further evaluation.

The initial case of active TB identified within the school was defined as the index case. Subsequently, contacts were categorized into three groups—close contacts, general contacts, and occasional contacts—based on the mode, degree, and duration of their exposure to the index case. Among them, close contacts referred to individuals who had had direct contact with TB patients, including teachers and students sharing the same classroom and the same dormitory (23).

For TST, an induration with a mean diameter <5 mm was classified as negative; 5–<10 mm as generally Mantoux positive; 10–<15 mm as moderately Mantoux positive. A strong Mantoux positive (SMP) reaction was defined as an induration ≥15 mm and/or with blisters, necrosis, and lymphangitis. A positive TST result was described as an induration ≥5 mm.

Five contact screenings were performed according to the school TB prevention and control work regulations (2017 edition) (23), which included symptom screening, TST, and chest radiography. Individuals under the age of 15 underwent symptom screening and TST initially. Subsequently, chest radiography was conducted on those exhibiting suspected symptoms of TB or SMP. Individuals aged 15 and older had symptom screening, TST, and chest radiography conducted concurrently. TST were performed by staff in accordance with standard guidelines. A qualified nurse administered 0.1 mL (5 IUs) of Purified Protein Derivative (PPD) derived from BCG (Chengdu Institute of Biological Products, Chengdu, China) via intradermal injection into the medial aspect of the left forearm. The size of the induration was assessed 72 h post-injection.

For individuals with suspected symptoms of TB, SMP, or abnormal chest radiography, three sputum specimens were examined by microscopy and cultured. Strain identification and drug susceptibility testing were performed on the cultured positive strains. Suspected cases of TB with negative pathogen results can be further evaluated using computed tomography (CT) and fiberoptic bronchoscopy (FOB). During the outbreak investigation period, given the potential measurement errors in TST testing, Interferon-Gamma Release Assay (IGRA) assays were performed on contacts with moderately Mantoux positive results to more accurately identify those meeting preventive therapy indications. QuantiFERON-TB Gold (Qiagen, Germany) was employed for IGRA testing per manufacturer’s instructions (24). Briefly, whole blood samples were collected in heparinized tubes, incubated with TB-specific antigens (ESAT-6, CFP-10, TB7.7), and plasma interferon-γ concentrations were measured by ELISA.

According to laboratory test procedures for TB (25), sputum smears were analyzed using the Ziehl-Neelsen staining technique; Mycobacterium TB was cultivated through L-J solid culture and/or BACTEC-MGIT960 liquid culture; drug susceptibility testing was detected by the Roche proportion method, and strains were identified via the traditional biochemical method. Concurrently, GeneXpert was employed for the swift molecular detection of sputum specimens.

We performed mycobacterial interspersed repetitive variable numbers of tandem repeats analyses (MIRU-VNTR) on Mycobacterium TB isolates to determine genetic relationships among isolates, with 9 loci (QUB-11b, QUB-18, QUB-26, MIRU-26, MIRU-31, MIRU-40, Mtub21, Mtub4, and VNTR2372) and 3 hypervariable loci (VNTR3820, VNTR3232, and VNTR4120). We did a cluster analysis of the repeat numbers of 9 VNTR loci of different strains. If 2 or more patients were infected with identical 9 locus genotypes of Mycobacterium TB, they were initially identified as clustered strains. Subsequently 3 hypervariable sites were added to further confirm the recent transmission. If the isolated strains possess specific genotypes at 12 loci, they are classified as a single strain (26). H37Rv standard strain was utilized as a quality control strain.

Statistical analyses were performed using the R (version 4.5.0) statistical software. The Chi-square test or Fisher’s exact test was employed, as appropriate, to analyze changes in TST positivity rates between the two screenings and differences in TB incidence rates between Class 1 and Class 2. Differences were declared significant for p-value < 0.05.

Between December 2018 and December 2019, a total of six students were diagnosed with TB in two classes of the high school, comprising four boys and two girls. Out of these, five cases were confirmed cases, while one was a clinically diagnosed case (Table 1). The cases included three residential students and three non-residential students, with the dormitories of the residential students being adjacent to each other. Cases B, C, D and E were in the same class (Class 2) as the index case (case A), and the other one (case F) was in the neighboring class (Class 1). The seats of 5 cases in the Class 2 were concentrated on one side of the classroom, and cases C and E were the class cadres (Figure 2). Except for case A, who received home isolation treatment after the diagnosis of TB, cases B-F were isolated at home and received further treatment after being diagnosed as suspected cases.

In September 2019, these students moved from the second year of senior high school to the third. The class number, classmates, dormitory, and roommates remain unchanged; only the classroom location has gone up two floors. The two classes where TB cases were diagnosed have been adjacent all the time, and they were two classrooms located in the horizontal line of the letter “B” (Figure 1). Given that these classrooms were blocked by the classrooms and the offices, we presumed they might have been poorly ventilated.

On September 17, 2018, the index case (case A) exhibited clinical symptoms (coughing and expectoration). From October 11th to 27th, due to the worsening symptoms, she went to hospital A for hospitalization and was diagnosed with a suspected case of TB when she was discharged. The hospital reported it on November 11th. Between November 3rd and 9th, Case A was hospitalized at the designated TB hospital B for treatment of a persistent cough and chest pain. Upon discharge, her diagnosis certificate indicated a “lung infection with no active TB present.” On December 11th, the student was diagnosed with TB by hospital B, as the sputum culture tested positive. From the onset of symptoms to the diagnosis of TB, which took nearly three months, Case A continued to attend school regularly, except for the time spent in the hospital. After the diagnosis, the case A was suspended from school and did not return to school until she was cured.

From December 2018, when the first case was diagnosed, to December 2019, a total of five screenings for TB among contacts were conducted (Table 2; Figure 3).

In the first round of screening, which commenced on December 17, 2018, 67 close contacts (including 56 classmates and 11 teachers) of Case A were screened. None of them presented with symptoms of TB. TST was performed on 66 close contacts, among whom 3 had SMP. Chest radiography was conducted on 67 close contacts, and two were suspected of having TB. The participation rate of TST screening was 98.51%, and that of chest radiography screening was 100%. Ultimately, this screening identified one clinically diagnosed case (Case B) and one confirmed case (Case C) of TB.

The second round of screening was carried out on January 4, 2019, and the scope of screening was further expanded. Given that case C was a member of the student union, 389 contacts on the same floor and fellow student union members underwent screening. The participation rate of TST screening was 98.20%, and that of chest radiography screening was 100%. No case of TB was found.

On October 1, 2019, a classmate (case D) of case A sought medical care due to symptoms of fever and cough. He was diagnosed as a suspected case of TB by hospital C on October 4 and then transferred to a designated TB hospital (hospital D). He was diagnosed with a confirmed case of TB on October 18.

In the third round of screening, which began on October 19, 2019, 64 close contacts of Case D were screened. Six close contacts exhibited suspected symptoms of TB. TST was performed on 61 close contacts, and none had SMP. Chest radiography was conducted on 61 close contacts, and two were suspected of having TB. The participation rates for both TST screening and chest radiography screening reached 100%. Ultimately, one confirmed case (Case E) of TB was identified in this screening.

On November 1, 2019, the screening scope was broadened to include 178 contacts from the same teaching building and dormitory floor. TST and chest radiography were conducted on 176 contacts. Four contacts had SMP, two had abnormal findings on chest radiography. The participation rates of both TST screening and chest radiography screening were 98.88%. Ultimately, one case (Case F) of TB was diagnosed.

On December 9, the scope of screening was expanded again. 118 contacts in the class near the floor were screened. Both TST screening and chest radiography screening achieved full participation rates of 100%, and no TB was found.

The two classes that found TB cases underwent two rounds of screening in this outbreak. The first screening time of Class 2 was in December 2018 and the second was October 2019. The first screening time of Class 1 was January 2019 and the second was November 2019. The positive rate of TST in the second time of class 2 (35.85, 95% CI: 23.49–19.25%) was higher than that in the first time (8.93, 95% CI: 3.33–20.37%), and the difference was statistically significant (χ² = 11.493, p < 0.001). The positive rate of TST in the second time of Class 1 (27.78, 95% CI: 16.86–41.86%) was also higher than that of the first time (14.81, 95% CI: 7.06–27.67%), but the difference was not statistically significant (χ² = 2.707, p = 0.100) (Table 3). Five students in Class 2 were diagnosed with TB, yielding an attack rate of 8.77% (95% CI: 2.91–19.30%). Class 1 had 1 confirmed TB case, with an attack rate of 1.85% (95% CI: 0.05–9.89%). The risk of TB in Class 2 was 5.10 times (95% CI: 0.58–45.12) that of Class 1, though this difference did not reach statistical significance (p = 0.207). Among the confirmed cases, only 3 students resided in school dormitories, and they were distributed across different dormitories (Dormitory B205, B206, and B207, each housing 8 students), resulting in an attack rate of 12.5% (95% CI: 0.31–52.65%) among dormitory students. No cases were found in other dormitories.

A total of five confirmed cases were reported during this outbreak, among which the sputum-positive culture specimen of one case (Case F) was not obtained. Only 4 cases (case A, C, D, and E) were genotyped by MIRU-VNTR typing. The results showed that all four strains belonged to non-Beijing genotypes. Cases A and C exhibited an identical genotype, and so did Cases D and E. Except for VNTR3232, the copy numbers of 11 loci were identical among Cases A, C, D, and E, indicating their belonging to the same transmission chain (Table 4).

Following two rounds of screenings conducted between December 2018 and January 2019, five contacts with SMP were found. Regrettably, all declined to undergo preventative medication. From October to December 2019, 29 contacts were moderately positive and 4 were SMP by three screenings. In order to further clarify the infection status, 29 moderately positive contacts were given IGRA, and 10 contacts were positive. Fourteen contacts who were SMP or IGRA positive were strongly recommended and mobilized to take prophylactic medication. Only one student accepted and completed prophylactic medication. For this outbreak, the overall tuberculosis preventive treatment (TPT) acceptance rate was 5.26% (1/19), including an 8.33% (1/12) acceptance rate among LTBI conversion participants. Teachers and students who declined prophylactic medication were mandated to undergo chest radiography at the conclusion of the 3rd, 6th, and 12th months post-screening. Continue to strengthen the follow-up observation of the classes where the outbreak occurred, no new related cases were found until November 2024.