Dear Editor,
The Bacillus Calmette–Guerin (BCG) vaccine, derived from wild-type Mycobacterium bovis, is administered in an attenuated form to prevent Mycobacterium tuberculous (MTB) infections in children residing in endemic regions. Since the introduction of the Singapore Tuberculosis Elimination Programme in 1997—specifying mandatory BCG-immunisation at birth—the incidence fell drastically to 32.6 per 100,000 population in 2021,1 with the paediatric population contributing 2.1% of infections.2
The vaccine is generally safe, but post-immunisation M. bovis infections (MBI) can occur, most commonly at the inoculation site forming localised MBI (L-MBI) such as cutaneous nodules, abscesses and lymphadenitis in immunocompetent children.3 MBI resolves spontaneously without the need for treatment with oral anti-TB drugs or surgery.4 Disseminated MBI (D-MBI) is rare and usually affects immunocompromised children in multiple sites including bones, joints, central nervous system, liver, spleen and lymph nodes. D-MBI also causes systemic complications of fever, weight loss, lymphadenopathy, hepatosplenomegaly and death.5
A third presentation variation described is the single site MBI (S-MBI), which affects a specific location remote from inoculation, without the presence of L-MBI or D-MBI, and has no correlation to the inoculation site. S-MBI involvement of the bone is estimated to occur in one per million vaccinations and affects the lower limbs, axial skeleton, upper limbs and multiple bones in decreasing frequency.6,7 It is thought to occur through haematogenous dissemination and is seen as a lytic lesion in the metaphysis of the long bone, with low or absence of inflammatory markers. Treatment regimens are highly varied, ranging from anti-TB drugs, chemotherapy and surgery, to undergoing no treatment at all. Musculoskeletal S-MBI in immunocompetent children is rare, and limited evidence-based literature is available describing its presentation and treatment. In this study, we aim to report the clinical characteristics of extremity bone and soft tissue S-MBI in immunocompetent children.
A retrospective review of children diagnosed with S-MBIs in a tertiary level paediatric hospital between 2017 and 2022 was performed with approval of the hospital ethics board. The inclusion criteria were children aged below 18, BCG vaccine administration following national immunisation guidelines, confirmed extremity musculoskeletal S-MBI, and absence of immunodeficiency conditions. L-MBIs were excluded. Data was collected on demographics; BCG vaccination details; S-MBI location and symptoms; and haematological, immunological and radiological investigations. Surgical details, anti-TB treatment regime, time to resolution, complications, readmissions and recurrence were recorded. Immunological tests for immunodeficiency included CD3, CD4, CD8 and CD20 levels; antibody testing for IgG, IgA, IgM and IgE levels; nitroblue tetrazolium levels; Mendelian susceptibility to mycobacterial diseases and immunological workup including interferon and gene sequence testing. MBI was confirmed through in-house laboratory mycobacterium culture or polymerase chain reaction (PCR) showing positive for M. bovis, and presence of acid-fast bacilli and necrotising granulomas on histology. PCR samples were sent for all patients. Mycobacterium cultures took on average 6 weeks for final results.
A total of 7 patients were included. Five were diagnosed with osteomyelitis, and 2 had deep soft tissue abscesses. The median age at presentation was 17.3 months (5.72–26.7). The median duration between symptom onset to seeking medical attention was 5 days (2–120). All reported pain, and the majority had localised swelling. Only 2 showed typical infective signs (redness and warmth), 1 had fever. All had normal immune workups. Two patients were subsequently diagnosed with thalassaemia and 1 with iron deficiency anaemia (Table 1).
The median drug treatment duration was 10 months (9–20) with Rifampicin (R), Isoniazid (H), Ethambutol (E), Levofloxacin (L) and Pyrazinamide (P) in combinations of RHEL, RHEZ or RHEZL for the first 2 to 3 months, and RH for the next 7 to 10 months (Table 1). The number of surgeries ranged from once (4 patients) to 5 times (1 patient). The median number of surgeries were 4.5 and 1 for soft tissue and osteomyelitis, respectively; and all in the former had recurrence while 1 in the latter group recurred. MBI recurrences occurred at sites different from the original. Advanced imaging, such as MRI and ultrasound, was employed for soft tissue infections, showing extensive involvement crossing multiple anatomical planes. The median time to recurrence after the last operation was 30 days (27–236). All the children were followed up for at least 24 months after the completion of anti-mycobacterial treatment, and no recurrences occurred.
The pathophysiology of MBI in immunocompetent children is unclear and presents a diagnostic challenge, making it difficult to select relevant investigations for accurate early diagnosis. In our patients, the subclinical nature and long latency (median 17.3 months) meant that a high degree of suspicion was required for diagnosis and institution of treatment. Yet, early recognition is important due to the potentially serious outcome, extended treatment period and risk of recurrences and chronicity that is complex to treat.
More severe disease and increased mortality were reported with high levels of C-reactive protein and erythrocyte sedimentation rate; there were also blood count abnormalities, including anaemia, leukopaenia, thrombocytopaenia and neutropaenia.8 Although we had 3 patients with anaemia, no correlation with infection severity was found. Our patients had no family history of autoimmune disease and had normal immune workups and hence deemed immunocompetent. Better characterisation of BCG-induced immune response can potentially help in understanding the variabilities in MBI occurrences. Controversy exists regarding the various hypotheses, and evidence points towards an unclassified inherited childhood immunodeficiency.9,10
Several notable differences between soft tissue abscesses and osteomyelitis were evident. The former demonstrated a shorter time to recurrence, higher number of surgeries and higher rate of recurrence. The sole osteomyelitis recurrence (patient 6) was postulated to be related to non-compliance to prescribed anti-mycobacterium medication regime. We also found that while soft tissue S-MBI was localised to a specific region, it was often poorly compartmentalised and spread to involve surrounding tissues. This was exemplified in patient 7 who had a recurrent extensive wrist abscess that crossed multiple tissue planes involving muscles, tendons and wrist joint. In such situations, treatment is largely dependent on multiple aggressive surgical debridement for clearance and source control, paired with extended medication.
Considering potential severity, we recommend mycobacterial cultures to be performed in atypical infection presentations. Early drug treatment could also be initiated in an attempt to obtain source control and limit spread. As the disease process remains poorly understood, more research on risk factors that might predict susceptibility to severe infection in immunocompetent children and treatment standardisation is needed to reduce MBI complications and improve quality of life for both children and their caregivers.
Declaration
The authors declare there are no affiliations with or involvement in any organisation or entity with any financial interest in the subject matter or materials discussed in this manuscript.
Ethics statement
The manuscript was approved by the SingHealth Centralised Institutional Review Board (CIRB 2023/2559). Waiver of consent was granted, as this was a retrospective deidentified review.
Correspondence: Dr Dawn Chia, Department of Hand & Reconstructive Microsurgery, Sengkang General Hospital, 110 Sengkang East Way, Singapore 544886. Email: [email protected]
REFERENCES
- Ministry of Health, Singapore. Update on Tuberculosis Situation in Singapore, 24 March 2023. https://www.moh.gov.sg/news-highlights/details/update-on-tuberculosis-situation-in-singapore-2023. Accessed 11 February 2024.
- Loh SW, Thoon KC, Tan NWH, et al. Paediatric tuberculosis in Singapore: a retrospective review. BMJ Paediatr Open 2018;2:e000308.
- Tural-Kara T, Ozdemir H, Erat T, et al. Local Cutaneous Complications After Bacille Calmette-Guerin Vaccine: Experience of a Single Center. Int J Clin Pediatr 2017;6:37-41.
- Villanueva P, Pittet LF, Curtis N. Management of Bacille Calmette-Guérin Lymphadenitis and Abscess in Immunocompetent Children: A Systematic Review. Pediatr Infect Dis J 2021;40:1037-45.
- Amanati A, Pouladfar G, Kadivar MR, et al. A 25-year surveillance of disseminated Bacillus Calmette–Guérin disease treatment in children in Southern Iran. Medicine (Baltimore) 2017;96:e9035.
- Lin WL, Chiu NC, Lee PH, et al. Management of Bacillus Calmette-Guérin osteomyelitis/osteitis in immunocompetent children—A systematic review. Vaccine 2015;33:4391-7.
- Mortensson W, Eklöf O, Jorulf H. Radiologic aspects of BCG-osteomyelitis in infants and children. Acta Radiol Diagn (Stockh) 1976;17:845-55.
- Sharifi Asadi P, Aghamohammadi A, Mahmoudi S, et al. Clinical, laboratory and imaging findings of the patients with disseminated bacilli Calmette–Guerin disease. Allergol Immunopathol 2015;43:254-8.
- Denis M, Forget A, Pelletier M, et al. Control of the Bcg gene of early resistance in mice to infections with BCG substrains and atypical mycobacteria. Clin Exp Immunol 1986;63:517-25.
- Al Busaidi N, Kp P, Al-Jardani A, et al. The Spectrum of Bacille Calmette–Guérin Diseases in Children—A Decade of Data from Neonatal Vaccination Settings. Vaccines (Basel) 2021;9:150.
The author(s) declare there are no affiliations with or involvement in any organisation or entity with any financial interest in the subject matter or materials discussed in this manuscript.