<i>Klebsiella Pneumoniae</i> Lower Respiratory Tract Infection in a South African Birth Cohort: A Longitudinal Study

Abstract

Background: Klebsiella pneumoniae (KP) causes neonatal sepsis and mortality, but its contribution to lower respiratory tract infection (LRTI) has not been well studied. We longitudinally investigated KP colonization and LRTI from birth through infancy in a South African birth cohort. Methods: We did a case-control study of infants who developed LRTI and age-matched controls, followed two weekly through infancy. Nasopharyngeal swabs (NP) were taken fortnightly from birth and at a LRTI for 33-multipex qPCR for organisms, including KP. Swabs were tested at LRTI and two weekly from 90 days prior to LRTI, and from controls over the equivalent period. Multivariate models were used to investigate factors associated with LRTI vs no LRTI, KP-LRTI vs non-KP-LRTI, or KP-LRTI vs no LRTI. Findings: Among 527 infants (26.9% HIV-exposed; 16.9% premature of which 61∙8% were > 34 weeks gestation), there were 439 LRTI events with a valid NP. KP found in 68(15.5%) of LRTI, was associated with cases (OR1∙93; 95%CI 1∙25-3∙03). The prevalence of KP at each two-weekly timepoint preceding LRTI was higher in those that developed LRTI vs. controls. Children with KP-LRTI were younger than those with non-KP-LRTI [median (IQR) 3∙7(2∙1-5∙9) vs 4∙7(2∙8-7∙9) months, p=0∙009]. Clinical features in KP and non-KP LRTI were similar except cough or rhinorrhea were less common in KP-LRTI. Amongst LRTI cases, 114 (26∙0%) were hospitalised, similar in KP and non-KP cases. Multivariate modelling found prematurity (aOR 11∙86; 95%CI 5∙22-26∙93), HIV-exposure (aOR 3∙32; 95%CI 1∙69-6∙53) or lower birthweight-for-age z-score 0∙68 (0∙51– 0∙91) associated with KP-LRTI vs no LRTI, while longer duration of breastfeeding was protective (aOR 0∙79; 95%CI 0∙65-0∙96). These factors and younger age were associated with KP-LRTI vs. non-KP LRTI. Amongst three deaths (mortality rate 0∙57%), two were due to LRTI of which one was KP-LRTI. Interpretation: KP-LRTI occurred in a substantial proportion of infants, associated with prematurity, HIV-exposure or short duration of breast feeding. Strategies for childhood pneumonia should consider empiric treatment of KP in premature or HIV-exposed infants and strengthened preventive interventions. Funding: Bill and Melinda Gates Foundation (OPP1017641); SA-MRC; NRF South Africa; NIH, H3Africa (1U01AI110466-01A1). Declaration of Interest: The authors have no conflicts of interest to declare. Ethical Approval: The study was approved by the Faculty of Health Sciences Research Ethics Committee, University of Cape Town and Western Cape Provincial Research committee.