Mathur N B, Arora Daisy
Department of Pediatrics, Maulana Azad Medical College, Lok Nayak Hospital, New Delhi 110002, India.
Acta Paediatr. 2007 Feb;96(2):172-5. doi: 10.1111/j.1651-2227.2007.00006.x.
To evaluate role of TOPS (a simplified assessment of neonatal acute physiology) in predicting mortality in transported neonates and to compare it with SNAP II.
Prospective study.
Referral Neonatal Unit of a teaching hospital.
175 neonates > 1000 g admitted to referral nursery.
Temperature, oxygenation, capillary refill time (proxy for perfusion) and blood sugar (given the acronym of TOPS) were recorded at admission. SNAP II was also recorded for first 12 hours following admission.
All the TOPS variables had significant correlation with fatality on univariate analysis. Fatality was 100% when all the four TOPS variables were deranged. The sensitivity, specificity, positive and negative predictive values of derangements of two or more TOPS parameters in predicting mortality were 81.6%, 77.39%, 65.3% and 89%, respectively All the variables of SNAP II significantly correlated with mortality. A cut-off score of 24 was derived from the ROC curve. The sensitivity, specificity, positive and negative predictive values of SNAP II score > 24 in predicting mortality were 78.3%, 86.1%, 74.6% and 88.4%, respectively. The area under ROC curve for derangements of 2 or more TOPS parameters was 0.89 while that for SNAP II was 0.88. Both models were well calibrated (p value for Hosmer Lemeshow goodness of fit for TOPS was 0.75 while that for SNAP II was 0.80). The total correct classification rate for TOPS was 81.7%, while that for SNAP II was 83.4%.
TOPS has an equally good prediction for mortality as SNAP II and can be used as a simple and useful method of assessment of risk of fatality that can be assessed immediately, at admission.
评估新生儿急性生理学简化评估法（TOPS）在预测转运新生儿死亡率中的作用，并将其与新生儿急性生理学评分系统Ⅱ（SNAP II）进行比较。
前瞻性研究。
一家教学医院的转诊新生儿病房。
175名体重超过1000克且入住转诊新生儿病房的新生儿。
入院时记录体温、氧合、毛细血管再充盈时间（灌注指标）和血糖（TOPS的首字母缩写）。入院后头12小时内也记录SNAP II。
单因素分析时，所有TOPS变量与死亡率均显著相关。当所有四个TOPS变量均紊乱时，死亡率为100%。两个或更多TOPS参数紊乱在预测死亡率时的敏感度、特异度、阳性预测值和阴性预测值分别为81.6%、77.39%、65.3%和89%。SNAP II的所有变量均与死亡率显著相关。通过ROC曲线得出截断值为24。SNAP II评分>24在预测死亡率时的敏感度、特异度、阳性预测值和阴性预测值分别为78.3%、86.1%、74.6%和88.4%。两个或更多TOPS参数紊乱的ROC曲线下面积为0.89，而SNAP II的为0.88。两种模型校准良好（TOPS的Hosmer Lemeshow拟合优度检验p值为0.75，而SNAP II的为0.80）。TOPS的总正确分类率为81.7%，而SNAP II的为83.4%。
TOPS在预测死亡率方面与SNAP II同样出色，可作为一种简单且有用的评估死亡风险的方法，在入院时即可立即进行评估。
