In this study, we compared the blood glucose concentration, gestational weeks, weight, height and head circumference at birth, and sex and parity distributions among VD newborns and CS newborns (Table1). We found no statistically significant difference in all aforementioned demographic variables among newborns delivered through VD and CS. However, the cord blood glucose level was significantly higher in VD newborns than CS newborns (Figure1). A similar finding has been reported by Marom and his colleagues in which neonatal blood glucose concentration was higher in VD newborns than CS newborns.
Our special interest in this study was to determine whether the neonatal blood glucose difference among these groups is necessitating partitioned RIs. As clearly depicted in Table2, there were no outlier glucose values detected in both groups. At the same time, the distributions were Gaussian/normal in both groups. The RIs were then determined to be 2.46-6.85 mmol/l and 2.46-5.04 mmol/l for newborns delivered through VD and CS respectively. Harris and Boyd analysis suggested that partitioning should be done as the two RIs are far apart from each other. However, the lower limit of both RIs is almost identical. As a result, partitioning the reference intervals may not be relevant clinically specially in the diagnosis and management of neonatal hypoglycemia. Even though we failed to determine the glucose concentrations in the subsequent hours after delivery, it has been described in literatures that the difference in glucose level might not be reflected in subsequent hours (≥ 2 hours) after delivery as a result of larger glucose decrease in VD newborns from the action of residual insulin secretion and slight glucose increase in CS newborns[6, 10]. Hence, it might be important to apply the combined RI (2.24-6.48 mmol/l) for the interpretation of glucose results of infants older than two hours irrespective of the mode of delivery.
Since we have used a robust method for determinations of the RIs, the lower limits were in close proximity to 2SDs lower than the respective mean values. Hence, the lower limits can be used as statistical tools to define neonatal hypoglycemia that are associated with clinical signs leading to neuro-developmental squeal. Therefore, it is possible to statistically estimate that newborns with cord blood glucose level of lower than 2.24 mmol/l, irrespective of mode of deliveries, could be considered as on greater risk of developing symptomatic hypoglycemia that might be accompanied with clinical presentations.