Prevalence of pneumococcal carriage and risk factors for pneumonia and carriage among under-5 children in Malaysia: findings from the MY-Pneumo study

Paediatric pneumonia remains a significant public health concern globally, and the risk factors for pneumonia and carriage vary across different populations. The accuracy of pneumococcal urinary antigen test (PUAT) remains unclear. We aimed to determine the prevalence of pneumococcal carriage, and the risk factors associated with pneumonia and carriage among children aged under 5 years in Peninsular Malaysia, and to evaluate the performance of the PUAT at detecting pneumococcal carriage. A total of 500 cases of pneumonia and 500 age-matched controls at three tertiary hospitals across three states in Peninsular Malaysia were recruited from October 2021 to August 2023. Nasopharyngeal swabs (NPS) were taken to test for Streptococcus pneumoniae (Spn) using culture and multiplex-PCR while urine samples were collected for PUAT with BinaxNOW®. Logistic regression analyses were fitted for paediatric pneumonia and pneumococcal carriage risk factors. The overall prevalence of nasopharyngeal pneumococcal carriage was 15.4% (n = 154) based on culture and optochin susceptibility, with identical rates observed among children with pneumonia and healthy controls. PUAT positivity was higher among pneumonia cases (11.8%, 59/500) than the controls (8.8%, 44/500) but was not statistically significant (p = 0.119). The carriage prevalence was significantly higher in the east coast (19.5%) compared to the central west coast (9.5%) (p < 0.001). Significant predictors of pneumonia were having siblings aged under 5 years (p = 0.002), pre-existing illness (p < 0.001), exposure to indoor smoking (p = 0.01), overcrowding (p = 0.012), underweight (p = 0.001) and obesity (p = 0.023). The predictors of pneumococcal carriage were recruiting sites and pneumococcal vaccination status. Compared to culture and PCR, the sensitivity and specificity of BinaxNOW® to detect Spn in urine were similar at about 31% and 93%, respectively. Pneumococcal carriage prevalence was similar in pneumonia cases and healthy controls, and the true burden of pneumococcal pneumonia in Malaysia remains unknown. Our findings identified risk factors for childhood pneumonia and pneumococcal carriage that are consistent with those reported elsewhere. The BinaxNOW® showed low sensitivity and moderate specificity indicating its limited utility as a standalone diagnostic tool for Spn detection in clinical settings. A focus should be made to recognise risk profiles to support early identification, timely intervention, and prevention strategies for respiratory infections in vulnerable paediatric populations. There is a need for continued Spn carriage surveillance and ongoing research to improve diagnostics to guide public health strategies for the prevention of pneumococcal infections.