Severe airway obstruction due to pulmonary tuberculosis in a premature infant needing decompression of mediastinal lymph nodes

Abstract

Tuberculosis (TB) in infants below 3 months of age is referred to as perinatal TB and includes both congenital and postnatal TB. Airway obstruction due to pulmonary TB has been well described in infants and in children, especially those younger than 2 years. The true incidence of TB airway obstruction remains unknown, but the reported incidence varies from 8% to 38% in children < 15 years of age.1 Airway obstruction due to pulmonary TB has not been reported in premature babies, especially those requiring surgical intervention. A 27-week premature female infant was delivered via caesarean section due to foetal distress, with a birth weight of 960 g. The mother was 35 years old, G3 P3, with a high BMI and known to be HIV-negative. The pregnancy was extrauterine, with the placenta implanted on the colon (Figure 1). The baby's Apgar scores were 3, 4, and 7 at 1, 5, and 10 min. She required intubation at birth due to poor respiratory effort, was then extubated and placed on continuous positive airway pressure; no surfactant replacement was needed. During hospitalization, she developed medical necrotizing enterocolitis, requiring invasive ventilation, but no surgical intervention. On Day 68 of life (weight 1.6 kg) the baby developed apnoea and again required intubation, oral caffeine, and third-line antibiotics meropenem. The chest X-ray demonstrated the development of a progressive confluent air-space process in the right lung on a background of coarse lung markings. There was also narrowing of the bronchus intermedius (BI) and splaying of the carina in keeping with lymphadenopathy in the right hilar and subcarinal regions (Figure 2). Tracheal aspirate viral, bacterial, and fungal studies were negative. A bronchoscopy was performed with a 2.2 Olympus ultrafine flexible bronchoscope on Day 71 of life. Severe narrowing of the right BI was confirmed with 90% narrowing of the BI, 75% narrowing of the left main bronchus and a very broad carina. The chest CT scan confirmed bronchoscopy findings with large subcarinal and right paratracheal lymph nodes, which caused compression of the BI, 2 days after bronchoscopy. A four-drug TB treatment was started after bronchoscopy and sending tracheal aspirates for Xpert mycobacterium tuberculosis (MTB)/rifampicin (RIF) and TB culture. Due to the severe airway obstruction, intravenous Methylprednisolone 2 mg/kg 8-hourly for the first 48 h was administered and then changed to oral prednisolone. The Xpert MTB/RIF done on the tracheal aspirate was confirmed positive. The mother was tested for TB, but the results was negative. She had abdominal sepsis postcaesarean section, but the aspirates and biopsy remained negative. The mother also had a chest and abdominal CT scan with no evidence of TB. No other nosocomial contact could be identified. The infant remained ventilator dependent with persistent clinical and radiological signs of airway compression, necessitating surgical decompression on Day 75 of life. A right thoracotomy through the 4th intercostal space was performed. The lung was retracted anteriorly to expose the mediastinum. The right paratracheal and subcarinal lymph nodes were identified and capsules opened. Pus and glandular tissue were removed and sent for histology and TB culture. There was significant improvement of airway calibre as seen on intraoperative bronchoscopy. A follow-up chest X-ray after surgery demonstrated return of the BI to near normal calibre and improvement in the airspace process with reappearance of the right cardiac margin. The baby was extubated on Day 8 postoperatively. Lymph node histology demonstrated lymphoid tissue with large, confluent necrotizing granulomatous changes and intact granulomas comprising epithelioid histiocytes, neutrophils, and lymphocytes along the periphery. Lymph node cultures confirmed TB and indicated drug sensitivity. There was no contact identified as a possible source case, and the infant was nursed in an incubator, which further would have reduced exposure, but probably increased the likelihood that the infant acquired TB nosocomially from a health care worker. We describe a case of severe airway obstruction due to TB in a very small premature baby needing surgical decompression, showing that even in extremely small babies, surgical decompression is a safe and lifesaving intervention. In a high TB incidence country such as South Africa, the incidence ratio for smear positive TB in primary health care workers showed an incidence rate more than double that of the general population.2 Nosocomial exposure and spread have been reported in both the NICU and kangaroo care units. Ahn et al. reported that 4 of 108 infants developed TB infection after exposure to a nurse with active TB.3 Nosocomial TB exposure can be high in endemic areas and vigilance and contact investigation should be rapid, include health care workers and family members, and not be delayed if found.4, 5 Although rare, nosocomial transmission between infants due to contaminated equipment has also been described.6 It is difficult to distinguish between congenital TB and postnatally acquired TB. Congenital TB occurs due to haematogenous spread in utero through the umbilical vein or ingestion/aspiration of Mycobacterium TB (Mtb)-infected amniotic fluid during birth. In contrast, postnatal TB, which occurs by the inhalation of Mtb bacilli spread by the airborne route from a mother or possible health care worker with infectious pulmonary TB early after birth.7 The chest X-rays in these two groups may appear similar in postnatal acquired TB and can look similar to congenital TB, but mediastinal lymphadenopathy, large airway compression, lobar or unilateral hyperinflation, or collapse due to partial or complete large airway obstruction by enlarged mediastinal lymph nodes, respectively, and Ghon foci are more common in postnatal acquisition.8 In this case, the baby had normal chest X-rays for more than a month after birth and only presented at Day 68 with respiratory disease, including expansile pneumonia and mediastinal lymph nodes, which makes acquired TB more likely. The outcome of infants < 6 months needing ventilation for respiratory failure due to TB has been reported to be good. In all these cases, the chest X-rays were highly suggestive of TB.9 Goussard et al. described a large cohort of children with severe airway obstruction due to TB in whom nearly 34% needed surgical decompression of which 29% required an urgent procedure. Transthoracic decompression can be safely performed with low complication, failure, and fatality rates even in young infants.10 Although difficult to prove, it is possible that nosocomial acquired TB, as a result of contact with a health care worker, was the source of infection in this case. This highlights the importance of TB screening in neonates with respiratory deterioration in the NICU. This case also shows the feasibility of surgical lymph node decompression in small infants. Pierre Goussard: Conceptualization; methodology; investigation; writing—original draft; writing—review and editing. Lizelle van Wyk: Investigation; writing—original draft; writing—review and editing; formal analysis. Andre Gie: Investigation; writing—original draft; writing—review and editing; formal analysis. Carmen Jacobs: Investigation; formal analysis; writing—original draft; writing—review and editing. Shama Aashish Patel: Investigation; writing—original draft; writing—review and editing. Shyam Venkatakrishna: Writing—original draft; writing—review and editing; formal analysis; investigation. Savvas Andronikou: Writing—original draft; writing—review and editing; formal analysis. Lars Ebert: Formal analysis; writing—original draft; writing—review and editing; investigation. Janette Verster: Writing—original draft; writing—review and editing. Gerhard Walzl: Investigation; writing—original draft; writing—review and editing. Odunola Sefiyat Adebiyi: Investigation; writing—original draft; writing—review and editing. Pawel Schubert: Investigation; writing—original draft; writing—review and editing. Jacques Janson: Investigation; writing—original draft; writing—review and editing; methodology. The authors have nothing to report. The authors declare no conflict of interest. The parents gave consent for the publication of this case report, and permission granted by the Health Research Ethics Committee (HREC C24/05/018) of the Faculty of Medicine and Health Science of Stellenbosch University. Due to privacy restrictions, patient data is available on request.