| 何正超,陈嘉玲,徐春蕾,王 震,陆文婷,李玲红,储玉琴,余艳伟,李 京.PICU患儿压力性损伤的临床特征、影响因素及预测模型的构建与评价[J].现代生物医学进展英文版,2023,(13):2444-2449. |
| PICU患儿压力性损伤的临床特征、影响因素及预测模型的构建与评价 |
| Construction and Evaluation of Clinical Characteristics, Influencing Factors and Predictive Models of Pressure Injury in Children with PICU |
| Received:February 06, 2023 Revised:February 21, 2023 |
| DOI：10.13241/j.cnki.pmb.2023.13.009 |
| 中文关键词: 儿科重症监护室 压力性损伤 临床特征 影响因素 预测模型 |
| 英文关键词: Pediatric intensive care unit Pressure injury Clinical characteristics Influencing factors Predictive model |
| 基金项目:上海交通大学医学院科研基金项目（Jyhz2014）；上海市交通大学医学院2021年学科人才队伍建设项目（jdhlrc202001）；上海市科学技术委员会科研计划项目（1761951713） |
|
| Hits: 606 |
| Download times: 367 |
| 中文摘要: |
| 摘要 目的：分析儿科重症监护室（PICU）患儿压力性损伤（PI）的临床特征和影响因素，并构建和评价PICU患儿PI预测模型。方法：选取2020年1月～2022年5月我院PICU收治的387例患儿，收集所有PICU患儿临床资料并根据其是否发生PI分为PI组（63例）和非PI组（324），分析PI患儿临床特征。采用单因素、多因素Logistic回归分析PICU患儿发生PI的影响因素并构建多指标联合应用预测模型，应用受试者工作特征（ROC）曲线分析多指标联合应用预测模型对PICU患儿发生PI的预测价值，采用H-L检验多指标联合应用预测模型的拟合优度。结果：387例PICU患儿PI发生率为16.28％（63/387），共76处，美国国家压疮咨询委员会（NPUAP）分期以1期为主，其中58.73％（37/63）为黏膜PI，41.27％（26/63）为器械相关PI，64.47％（49/76）发生在头面部，18.42％（14/76）发生在下肢，10.53％（8/76）发生在骶尾部，3.95％（3/76）发生在上肢，2.63％（2/76）发生在躯干。单因素分析显示，PI组心血管疾病、神经系统疾病、昏迷/嗜睡比例和医疗设备数量高于非PI组，入住PICU时间长于非PI组，且Braden-Q量表评分低于非PI组（P＜0.05）。多因素Logistic回归分析显示，心血管疾病、神经系统疾病、昏迷/嗜睡、医疗设备数量增加为PICU患儿发生PI的独立危险因素，Braden-Q量表评分升高为其独立保护因素（P＜0.05）。Braden-Q量表评分与多指标联合应用预测模型的ROC-AUC分别为0.702、0.910，多指标联合应用预测模型的预测效能更高。经H-L检验，多指标联合应用预测模型拟合效果良好。结论：PICU患儿是PI高危人群，NPUAP分期以1期且PI损伤部位多集中于头面部。心血管疾病、神经系统疾病、昏迷/嗜睡、医疗设备数量、Braden-Q量表评分是PICU患儿PI的影响因素，根据以上影响因素构建的PICU患儿PI多指标联合应用预测模型，拟合效果良好、预测价值较高。 |
| 英文摘要: |
| ABSTRACT Objective: To analyze the clinical characteristics and influencing factors of pressure injury (PI) in children with pediatric intensive care unit (PICU), and to construct and evaluate a PI prediction model for children with PICU. Methods: 387 children who were admitted to our PICU from January 2020 to May 2022 were selected, and clinical data of all children with PICU were collected, and they were divided into PI group (63 cases) and non-PI group (324 cases) according to whether they had PI. Clinical characteristics of children with PI were analyzed. Univariate and multivariate Logistic regression were used to analyze the influencing factors of PI in children with PICU, and multi-indicator combined application prediction model was constructed. Receiver operating characteristic (ROC) curve was used to analyze the predictive value of multi-indicator combined application prediction model for PI in children with PICU. The H-L was used to test the goodness fitting of multi-indicator combined application prediction model. Results: The incidence rate of PI in 387 children with PICU was 16.28% (63/387), with a total of 76 sites. The national pressure ulcer advisory panel (NPUAP) had mainly stage 1, of which 58.73% (37/63) was mucous membrane PI, and 41.27% (26/63) was instrument-related PI, and 64.47% (49/76) occurred in the head and face, 18.42% (14/76) in the lower extremities, 10.53% (8/76) in the sacral caudal, 3.95% (3/76) in the upper extremities, and 2.63%(2/76) in the trunk. Univariate analysis showed that cardiovascular disease, nervous system disease, comatose/lethargy ratio and the number of medical equipment in the PI group were higher than those in the non-PI group, and the time to stay in PICU was longer than that in the non-PI group, and the Braden-Q scale score was lower than that in the non-PI group(P<0.05). Multivariate Logistic regression analysis showed that cardiovascular disease, nervous system disease, comatose/lethargy and the increased in the number of medical equipment were independent risk factors for PI in children with PICU, and the increased of Braden-Q scale score was independent protective factor(P<0.05). The ROC-AUC of Braden-Q scale score and multi-indicator combined application prediction model were 0.702 and 0.910, respectively, and the prediction efficiency of multi-indicator combined application prediction model was higher. H-L test showed that the combined application of multiple indexes had a good fitting effect. Conclusion: The children with PICU are at high risk of PI. NPUAP is divided into stage 1 and the injured parts of PI are mainly in the head and face. Cardiovascular disease, nervous system disease, comatose/lethargy, number of medical equipment, and Braden-Q scale score are the influencing factors of children with PICU of PI. The multi-indicator combined application prediction model of children with PICU of PI constructed according to the above influencing factors is applied jointly, with good fitting effect and high prediction value. |
|
View Full Text
View/Add Comment Download reader |
| Close |
|
|
|
