The ability of CRIB score for prediction of very low birthweight infants
(800-1,500 g) in Songklanagarind Hospital
การใช้ clinical risk index for babies (CRIB) score ทำนายอัตราตายของทารกแรกเกิด น้ำหนัก 800-1,500 กรัม ในโรงพยาบาลสงขลานครินทร์
Prapatsorn Isarawattana, D. of Pediatrics, F. of Medicine, PSU.
Waricha Janjindamai, Asst. Prof., D. of Pediatrics, F. of Medicine, PSU.
Published : The 16th Annual Academic Meeting of the Faculty of Medicine,
Prince of Songkla University, 16-18 August 2000
Key words : clinical risk index for babies, baby, infant, mortality rate
Background : Neonatal death is the major public health problem in Thailand. Neonatal mor-tality is one of the indices for evaluation of the effectiveness of maternal and neonatal care. The lower birthweight of the neonate, the mortality risk is higher. Especially for the very low birthweight
infants, there are many factors associate with the mortality. There were many studies tried to develop the tool for predict the mortality rate of these infants such as the CRIB score. The previous study
found that the CRIB score was high accuracy, easy to perform and not take time to evaluate. In
Thailand, there have never had no previous study concerning the prediction of mortality of very low birthweight infants by using the CRIB score.
Objective : To study the ability of CRIB score for prediction of the hospital mortality compared to gestational age and birthweight and to study the mortality rate of very low birthweight infants in Songklanagarind hospital.
Methods : Medical records were reviewed for the demographic data and the factors for evaluation of 1) gestational age, 2) birthweight, 3) congenital malformation, 4) maximum base excess in first 12 hour 5) minimum appropriate FiO2 in first 12 hour 6) maximum appropriate FiO2 in first
12 hour. The summation scores in each item (the CRIB score) were analyzed. ROC curve was used to compare between CRIB score, gestational age and birthweight.
Results : A total of 122 patients enrolled in the study. The mortality rate of the patients with birthweight 800-1,500 grams was 21.3%. Areas under ROC curve of CRIB score, gestational age and birthweight were 0.78 0.68 and 0.65 respectively. The area under ROC curve between CRIB score
and gestational age and between CRIB score and birthweight were not significantly difference (p= 0.164 and p=0.060 respectively).
Conclusion : CRIB score can be used to predict the mortality of very low birthweight better
than gestational age and birthweight but non statistically significant.
BACK