Red cell distribution width and its association with mortality in neonatal sepsis

Abstract

Objective: Neonatal sepsis is a major cause of mortality in the developing countries. However, with current severity scores and laboratory parameters, predicting outcomes of neonatal sepsis is a serious challenge. Red cell distribution width (RDW) is a readily available pragmatic means to predict outcomes of various comorbidities in adults and children, without causing any additional blood loss. However, its utility in neonates remains unexplored. Hence, the objective of the present study was to evaluate the association of RDW with neonatal sepsis and its role as a predictive marker for mortality.

Methods: This Prospective observational study was carried out in a Level IIIB NICU for a period of 3 years. It involved comparison of RDW values of septic neonates with those of controls (matched for gestational age and birth weight) with an equal allocation ratio. A total of 251 septic neonates along with 251 controls >28 weeks of gestational age were enrolled. The RDW was derived from complete blood count done within first 6 hours of life. After arranging the RDW (median; interquartile range (IQR)), the values were categorized as those above the 50th percentile i.e. ≥20% and those below the 50th percentile i.e. <20%. The cumulative survival rates of the above two groups were assessed using the Kaplan–Meier curve and the log rank test.

Results: RDW levels were significantly higher among the neonatal sepsis cases (19.90%) as compared to the controls (18.90%) with a p value of < .001. RDW was significantly higher amongst the nonsurvivors than survivors (p < .003). Kaplan–Meier curve showed that septic neonates having RDW values ≥20% had significantly increased mortality (p < .02) with a hazard ratio of 0.5.

Conclusions: High RDW is associated with neonatal sepsis and is an independent outcome predictor for mortality associated with neonatal sepsis.

Keywords:

• Mortality
• neonatal sepsis
• RDW

Acknowledgements

We thank Indian Council of Medical Research (ICMR), New Delhi, for providing Senior Research Fellowship to SLM. The authors are also thankful to Dr. Nandkishor Kabra for his guidance in statistical analysis of the data.

Disclosure statement

The authors declare no conflict of interest.