The study conducted by Mehta, et al. [1] is indeed, one of its own kind and has a good internal validity, although following issues need attention

It is not mentioned on what basis a dose for zinc (1 mg/ kg/day) was decided; and how the exact dose of 1 mg/kg of zinc was administered in each infant by using a 10 mg dispersable tablet (almost all weighing less than 3 kg). The safety of administering zinc to an infant who needed to be nil per oral has also not been described. The breast feeding status of the infants in the two groups have not been mentioned. There is no mention of the average age at which the infants were enrolled in the study. The inclusion of other markers of infection like procalcitonin and blood culture (BACTEC) would have given more specificity in identifying sepsis patients. Some kind of sickness assessment score (PRISM / CRIB) to determine the severity of illness could help better understand the status of the given cohort. Duration of antibiotic treatment is not clear. The diagnosis of sepsis is described to be one of the three criteria. However, in the absence of first two criteria the presence of only third criteria is unlikely to warrant the need of antibiotics as X-ray findings could be non-specific for neonatal pneumonia.

Studies have shown that zinc supplementation is benificial in reducing the mortality of small for gestation age (SGA) infants [2]. It will be prudent to sub-group the cohort as appropriate for gestational age (AGA) and SGA and then analyse the results. Demographic, clinical, microbiological data of the infants who died need to be compared with the rest of the group.

1. Mehta K, Bhatta NK, Majhi S, Shrivastava MK, Singh RR. Oral zinc supplementation for reducing mortality in probable neonatal sepsis: A double blind randomized placebo controlled trial. Indian Pediatr. 2013;50:390-3.

2. Sazawal S, Black RE, Menon VP, Dinghra P, Caulfield LE, Dhingra U, et al. Zinc supplementation in infants born small for gestational age reduces mortality: a prospective, randomized, controlled trial. Pediatrics. 2001;108:1280-6.