Association between postnatal weight gain and severe retinopathy of prematurity in preterm babies with very low birth weight

Abstract

Objective To analyze the association of postnatal weight gain proportion of very low birth weight (BW) preterm babies and the onset of severe retinopathy of prematurity, and investigate the optimal cut-off points and predictive ability of postnatal weight gain (WG) proportion for the onset of retinopathy of prematurity (ROP). Methods A retrospective cohort study. 257 preterm infants underwent screening whose weight was less than 1500 g were enrolled in this study. Risk factors include BW, gestational age (GA), history of oxygen inhalation, need for blood transfusions, Apgar score in 1 to 10 minutes, embryo number, delivery mode, in vitro fertilization infants, and WG proportion within 6 weeks after birth and other systemic diseases were recorded. Their correlation with severe ROP is analyzed. Clinical outcomes were divided into severe ROP group (patients who suffered from ROP and required treatment) and mild and no ROP group (patients who suffered from ROP but do not require treatment and patient without ROP). The severe ROP group included 18 patients and mild and no ROP group included 239 patients. Multiple logistic regression and receiver operating characteristic (ROC) curve were used to determine if the WG proportion was independently related to severe ROP development and if it was capable of predicting severe ROP. This study determines the predict value by comparing the area under the ROC curve (AUC) of independent risk factors. Results GA (t=-4.835, P<0.001), BW(t=-5.192, P<0.001), history of oxygen inhalation (χ2=6.001, P=0.009), proportion of infants who had oxygen inhalation for more than 10 days(χ2=10.019, P=0.002), postnatal WG proportion at 1 week(t=-3.663, P<0.001), postnatal WG proportion at 2 weeks(t=-3.425, P=0.001) had significant difference between two groups. Multiple logistic regression analysis revealed that GA(β=-0.858, P=0.008), BW(β=-0.005, P=0.010), postnatal WG proportion at 2 weeks (β=-8.745, P=0.035)were correlated to severe ROP significantly. And their area under the ROC were 0.836[95% confidence interval (CI): 0.752 - 0.920], 0.826 (95%CI: 0.947 - 0.903), 0.744 (95%CI: 0.598 - 0.891) respectively. The optimal cut-off points of GA, BW, and postnatal WG proportion at 2 weeks were 28.41 weeks, 1241.96 g, 12.80% respectively. Conclusion Low WG proportion at 2 weeks of very low BW preterm babies is an important and independent risk factor for severe ROP and has certain predictive value of the onset of severe ROP. Key words: Retinopathy of prematurity/etiology; Birth weight/growth and development; Weight gain/growth and development; Forecasting