Comparing Treatment Outcomes between Confirmed and Non-Confirmed Childhood TB Cases: Study from a Major Tertiary Center in Thailand

Abstract Background Tuberculosis (TB) in children is challenging to diagnose due to its non-specific symptoms and difficulties in obtaining confirmatory laboratory results. This study aimed to compare the treatment outcomes of confirmed and non-confirmed TB in children. Methods The medical records of children aged ≤ 15 years diagnosed with TB at Songklanagarind Hospital from January 2007 to December 2018 were examined. The TB cases were classified into three groups, confirmed cases (positive culture and/or polymerase chain reaction), probable cases (positive histopathological and/or acid-fast bacilli results) and presumptive cases (clinical manifestations and imaging findings compatible with TB and good response to anti-TB drugs but without microbiological confirmation). Results Of 177 children, 66 (37.3%) had a confirmed diagnosis, 57 (32.2%) a probable diagnosis, and 54 (30.5%) a presumptive diagnosis. The successful treatment rates of the confirmed, probable and presumptive TB groups were 80.3%, 87.7% and 92.6%, respectively (p = 0.21). Of the 54 presumptive cases, a history of household contact with TB led to a diagnosis of asymptomatic pulmonary TB in 10 (18.5%) cases. Patients with a presumptive diagnosis were less likely to have disseminated and/or miliary TB (1.9% vs. 7.0% of probable vs. 22.7% of confirmed, p < 0.01). Conclusions Only one-third of the study patients had a confirmed TB diagnosis. The successful treatment and mortality rates did not significantly differ among the three groups. To increase the detection rate of early diagnosis of childhood TB, physicians should perform active TB contact investigations in household members of the index case.