随机森林模型结合流式细胞术数据可鉴定胸外伤后的肺部感染。

A random forest model using flow cytometry data identifies pulmonary infection after thoracic injury.

机构信息

From the Department of Surgery (R.B.G., C.J.D., T.G.B.), Emory University, Atlanta, Georgia; Uniformed Services University of the Health Sciences (S.S., E.G., E.E.), Walter Reed National Military Medical Center (E.E.), Surgical Critical Care Initiative (R.B.G., H.H., S.S., L.S., E.G., C.J.D., T.G.B., A.D.K., E.E.), Bethesda, Maryland; DecisionQ (H.H.), Arlington, Virginia; Department of Surgery (L.S., D.M., A.D.K.), Duke University, Durham, North Carolina; Department of Surgery, Trauma, Burns, and Surgical Critical Care (R.B.G.), University of Alabama at Birmingham, Birmingham, Alabama; and Henry M Jackson Foundation for the Advancement of Military Medicine (S.S., E.G.), Bethesda, Maryland.

出版信息

J Trauma Acute Care Surg. 2023 Jul 1;95(1):39-46. doi: 10.1097/TA.0000000000003937. Epub 2023 Apr 11.

DOI:10.1097/TA.0000000000003937

PMID:37038251

Abstract

BACKGROUND

Thoracic injury can cause impairment of lung function leading to respiratory complications such as pneumonia (PNA). There is increasing evidence that central memory T cells of the adaptive immune system play a key role in pulmonary immunity. We sought to explore whether assessment of cell phenotypes using flow cytometry (FCM) could be used to identify pulmonary infection after thoracic trauma.

METHODS

We prospectively studied trauma patients with thoracic injuries who survived >48 hours at a Level 1 trauma center from 2014 to 2020. Clinical and FCM data from serum samples collected within 24 hours of admission were considered as potential variables. Random forest and logistic regression models were developed to estimate the risk of hospital-acquired and ventilator-associated PNA. Variables were selected using backwards elimination, and models were internally validated with leave-one-out.

RESULTS

Seventy patients with thoracic injuries were included (median age, 35 years [interquartile range (IQR), 25.25-51 years]; 62.9% [44 of 70] male, 61.4% [42 of 70] blunt trauma). The most common injuries included rib fractures (52 of 70 [74.3%]) and pulmonary contusions (26 of 70 [37%]). The incidence of PNA was 14 of 70 (20%). Median Injury Severity Score was similar for patients with and without PNA (30.5 [IQR, 22.6-39.3] vs. 26.5 [IQR, 21.6-33.3]). The final random forest model selected three variables (Acute Physiology and Chronic Health Evaluation score, highest pulse rate in first 24 hours, and frequency of CD4 + central memory cells) that identified PNA with an area under the curve of 0.93, sensitivity of 0.91, and specificity of 0.88. A logistic regression with the same features had an area under the curve of 0.86, sensitivity of 0.76, and specificity of 0.85.

CONCLUSION

Clinical and FCM data have diagnostic utility in the early identification of patients at risk of nosocomial PNA following thoracic injury. Signs of physiologic stress and lower frequency of central memory cells appear to be associated with higher rates of PNA after thoracic trauma.

LEVEL OF EVIDENCE

Diagnostic Test/Criteria; Level IV.

摘要

背景

胸部损伤可导致肺功能损害，从而导致肺炎（PNA）等呼吸系统并发症。越来越多的证据表明，适应性免疫系统的中央记忆 T 细胞在肺部免疫中发挥关键作用。我们试图探讨使用流式细胞术（FCM）评估细胞表型是否可用于鉴定胸外伤后的肺部感染。

方法

我们前瞻性研究了 2014 年至 2020 年在 1 级创伤中心存活超过 48 小时的胸部损伤创伤患者。将入院 24 小时内采集的血清样本的临床和 FCM 数据作为潜在变量进行考虑。使用后向消除法建立随机森林和逻辑回归模型，以评估医院获得性和呼吸机相关性 PNA 的风险。使用留一法对内部分别验证模型。

结果

共纳入 70 例胸部损伤患者（中位数年龄 35 岁[四分位数范围（IQR），25.25-51 岁]；62.9%[44/70]为男性，61.4%[42/70]为钝器伤）。最常见的损伤包括肋骨骨折（52/70[74.3%]）和肺挫伤（26/70[37%]）。PNA 的发生率为 14/70（20%）。有和无 PNA 的患者的损伤严重程度评分中位数相似（30.5[IQR，22.6-39.3]与 26.5[IQR，21.6-33.3]）。最终的随机森林模型选择了三个变量（急性生理学和慢性健康评估评分、前 24 小时内最高脉搏率和 CD4+中央记忆细胞频率），可识别 PNA，曲线下面积为 0.93，灵敏度为 0.91，特异性为 0.88。具有相同特征的逻辑回归模型的曲线下面积为 0.86，灵敏度为 0.76，特异性为 0.85。