论文部分内容阅读
目的在反应范围模型指导下探讨住院早产儿喂养不耐受(FI)发生的风险因素,构建风险预测模型,为有效防治早产儿FI提供实证依据。方法采用横断面研究,选择2014年8月至2015年1月入住新生儿科的早产儿。在反应范围模型指导下设计临床观察表,从预测性稳态、反应性稳态、稳态超负荷及稳态失败等方面对发生FI及未发生FI者进行分析,通过logistic回归分析建立早产儿FI发生风险预测模型,并检验模型的有效性。结果 1207例早产儿中,男性125例,女性82例,胎龄27+2~37周,平均(33.48±1.66)周;出生体质量830~3 120g,平均(2 019.55±334.38)g。早产儿FI发生率为33.8%;FI主要发生在开奶72h内,早期早产儿FI的临床表现以胃潴留为主,晚期早产儿FI则以呕吐为主要临床表现。2单因素分析显示,胎龄、出生体质量、宫内窘迫、宫内感染、氨茶碱应用、母乳喂养及两次大便间隔时间与早产儿FI有关;logistic回归分析显示,胎龄与出生体质量是FI的保护因素,胎龄越大,出生体质量越重,FI发生的风险越低。宫内窘迫、氨茶碱应用、两次大便间隔时间超过3d是FI发生的高危因素。3经回代预测,回归方程的回代率为92.73%,预测FI的敏感度为97.14%,特异度为88.32%,准确度为91.30%。结论胎龄低、出生体质量低、宫内窘迫、使用氨茶碱及两次大便间隔超过3d是FI发生的重要风险因素,适用于logistic回归构建预测FI发生风险模型;该模型能够为临床确定FI的高危人群和及时进行干预提供客观可靠的依据。
Objective To explore the risk factors of feeding intolerance (FI) in hospitalized preterm infants under the guidance of response model and to construct a risk prediction model to provide evidence for the effective prevention and treatment of premature infants. Methods A cross-sectional study was conducted to select preterm infants admitted to the neonatology department from August 2014 to January 2015. Under the guidance of the reaction range model, a clinical observation table was designed to analyze the occurrence of FI and non-occurrence of FI from predictive steady-state, steady-state response, steady-state overload and steady state failure, and to establish premature infants by logistic regression analysis FI risk prediction model, and test the validity of the model. Results Among 1207 preterm infants, 125 were male and 82 were female. The gestational age ranged from 27 + 2 to 37 weeks (mean, 33.48 ± 1.66 weeks). The birth weight ranged from 830 to 320g (mean, 2019.55 ± 334.38) g. The incidence of FI in preterm infants was 33.8%. The FI mainly occurred within 72 hours after opening milk. The clinical manifestations of FI in early premature infants were gastric retention, and the late preterm infants were vomiting. Univariate analysis showed that gestational age, birth weight, intrauterine distress, intrauterine infection, aminophylline application, breastfeeding and stool interval were correlated with FI in preterm infants. Logistic regression analysis showed that gestational age and birth Quality is a protective factor for FI, the larger the gestational age, the heavier the birth weight, the lower the risk of FI occurring. Intrauterine distress, aminophylline application, twice stool interval more than 3d is a risk factor for the occurrence of FI. 3 After the regression prediction, the regression rate of the regression equation was 92.73%, the sensitivity of prediction FI was 97.14%, the specificity was 88.32% and the accuracy was 91.30%. Conclusions Low gestational age, low birth weight, intrauterine distress, the use of aminophylline and more than three stools stool interval is an important risk factor for FI, which is suitable for the construction of logistic regression model to predict the risk of FI. This model can be clinically determined FI high-risk groups and prompt intervention to provide objective and reliable basis.