骨科多发伤患者术后转入ICU风险预测模型构建

邓婷, 刘峰, 卿琪, 邓婷, 李丽莎

湖南师范大学学报医学版 ›› 2026, Vol. 23 ›› Issue (3) : 18-26.

PDF(2769 KB)
PDF(2769 KB)
湖南师范大学学报医学版 ›› 2026, Vol. 23 ›› Issue (3) : 18-26.
临床医学

骨科多发伤患者术后转入ICU风险预测模型构建

  • 邓婷1, 刘峰, 卿琪, 邓婷2, 李丽莎
作者信息 +

Construction of a Risk Prediction Model for Postoperative ICU Transfer in Orthopedic Patients with Multiple Trauma

  • DENG Ting1, LIU Feng, QING Qi, DENG Ting2, LI Lisha
Author information +
文章历史 +

摘要

目的 探讨影响骨科多发伤患者术后转入重症监护病房(intensive care unit,ICU)的风险因素,并据此构建风险预测模型。方法 回顾性选取我院2020年1月—2025年1月收治的骨科多发伤患者300例为研究对象,依据7∶3的比例采用简单随机分组法分为建模集(n=210)和验证集(n=90)。依据骨科多发伤患者术后是否转入ICU分为转入组、未转入组。筛选患者术后转入ICU的危险因素并构建预测模型。结果 建模集210例患者中,转入ICU61例(29.05%),验证集90例患者中,转入ICU26例(28.89%)。多因素Logistic回归分析显示,年龄、ASA分级、ISS评分、急诊手术、术中出血量、术中低血压或低氧血症、术后GCS评分、术后6 h尿量、术后6 h内的Hb、D-二聚体、SpO2、Lac、PT、ALB、Cr是骨科多发伤患者术后转入ICU的独立影响因素(P<0.05),各因素间未见显著多重共线性现象;经LASSO回归最终筛选出ASA分级、ISS评分、急诊手术、术中出血量、术后6 h内SpO2、Lac6项核心预测因子构建列线图模型。模型验证结果显示,建模集AUC为0.783(95%CI:0.677~0.889),验证集AUC为0.771(95%CI:0.554~0.823);Hosmer-Lemeshow检验显示建模集P=0.258,验证集P=0.222,模型拟合度良好;DCA曲线显示模型在合理阈值范围内具有良好的临床净收益。结论 本研究基于ASA分级、ISS评分、急诊手术、术中出血量、术后6 h内SpO2及Lac水平构建的骨科多发伤患者术后转入ICU风险预测模型,经校准曲线及净收益率分析验证具有良好预测效能,可为临床决策提供支持。

Abstract

Objective To identify risk factors influencing ICU transfer after orthopedic surgery in patients with multiple trauma and to develop a corresponding predictive model. Methods A total of 300 patients with orthopedic multiple trauma admitted to our hospital from January 2020 to January 2025 were retrospectively selected as the research subjects. They were divided into a modeling set (n=210) and a validation set (n=90) using simple random grouping in a 7∶3 ratio. Patients were classified into the transfer group and the non-transfer group based on whether they were transferred to the ICU after surgery. Risk factors for postoperative ICU transfer were screened, and a prediction model was constructed. Results Among the 210 patients in the modeling set, 61(29.05%) were transferred to the ICU, while 26 (28.89%) of the 90 patients in the validation set were transferred to the ICU. Multivariate Logistic regression analysis showed that age, ASA grade, ISS score, emergency surgery, intraoperative blood loss, intraoperative hypotension or hypoxemia, postoperative GCS score, urine output at 6 h postoperatively, and Hb, D-dimer, SpO2, Lac, PT, ALB, and Cr levels within 6 h postoperatively were independent factors associated with postoperative ICU transfer (all P<0.05). No significant multicollinearity was observed among these factors. LASSO regression further identified six core predictors, including ASA grade, ISS score, emergency surgery, intraoperative blood loss, SpO2 within 6 h postoperatively, and Lac, which were used to construct the nomogram. The AUC was 0.783 (95%CI: 0.677-0.889) in the modeling set and 0.771 (95%CI: 0.554-0.823) in the validation set. The model showed good calibration and clinical net benefit. Conclusions The ICU-transfer risk nomogram, based on ASA class, ISS score, emergency surgery, intra-operative blood loss, postoperative 6 h SpO2, and lactate level, exhibits good discriminative performance and calibration. It can serve as a practical tool to support clinical decision-making in orthopedic patients with multiple trauma.

关键词

骨科 / 多发伤 / 术后 / 转入重症监护病房 / 风险预测模型构建

Key words

orthopedics / multiple trauma / postoperative / ICU transfer / risk prediction model

引用本文

导出引用
邓婷, 刘峰, 卿琪, 邓婷, 李丽莎. 骨科多发伤患者术后转入ICU风险预测模型构建[J]. 湖南师范大学学报医学版. 2026, 23(3): 18-26
DENG Ting, LIU Feng, QING Qi, DENG Ting, LI Lisha. Construction of a Risk Prediction Model for Postoperative ICU Transfer in Orthopedic Patients with Multiple Trauma[J]. Journal of Hunan Normal University(Medical Science). 2026, 23(3): 18-26
中图分类号: R683    R459.7   

参考文献

[1] 胡云刚, 植林, 王艺雯, 等. 多发骨创伤伴脱套伤的多中心流行病学分析[J]. 骨科临床与研究杂志, 2025, 10(6): 372-376.
[2] 马颖宏, 孙志坚, 李庭, 等. 多发骨创伤的多中心流行病学分析[J]. 骨科临床与研究杂志, 2025, 10(2): 96-103.
[3] GRÖBLI L, KALBAS Y, KESSLER F, et al. Are the same parameters measured at admission and in the ICU comparable in their predictive values for complication and mortality in severely injured patients?[J]. Eur J Med Res, 2025, 30(1): 228.
[4] RUBANO A J, VOSSWINKEL A J, MCCORMACK E J, et al.Unplanned intensive care unit admission following trauma[J]. J Crit Care, 2016, 33(4): 174-179.
[5] 邱海波. 重症医学科建设管理规范[M]. 南京: 东南大学出版社, 2011.
[6] 顾立强. Pilon骨折的分类与功能评价[J]. 中华创伤骨科杂志, 2004, 6(8): 894-898.
[7] JU J, ZHANG P, WANG Y, et al.A clinical nomogram predicting unplanned intensive care unit admission after hip fracture surgery[J]. Surgery, 2021, 170(1): 291-297.
[8] ANIL U, ROBITSEK RJ, KINGERY MT, et al.Orthopedic pelvic and extremity injuries increase overall hospital length of stay but not in-hospital complications or mortality in trauma ICU patients: Orthopedic Injuries in Trauma ICU Patients[J]. Injury, 2024, 55(11): 843-855.
[9] HAN T, XIONG F, SUN B, et al.Development and validation of an artificial intelligence mobile application for predicting 30-day mortality in critically ill patients with orthopaedic trauma[J]. Int J Med Inform, 2024, 184(5): 383-392.
[10] JIANG W, LIU T, SUN B, et al.An artificial intelligence application to predict prolonged dependence on mechanical ventilation among patients with critical orthopaedic trauma: an establishment and validation study[J]. BMC Musculoskelet Disord, 2024, 25(4): 1089-1095.
[11] CELIK S, TASKIN YILMAZ F, ALTAS G.Medical device-related pressure injuries in adult intensive care units[J]. J Clin Nurs, 2023, 32(13-14): 3863-3873.
[12] GRAF A, WENDLER D, COURT T, et al.Acute clavicle fixation after blunt chest trauma: effect on pulmonary outcomes and patient disposition[J]. Eur J Orthop Surg Traumatol, 2023, 33(5): 1921-1927.
[13] BARANOWSKI P, PŁUSA T, BARANOWSKA A, et al. Analysis of mortality between 2019-2020 at the neuroorthopedic and traumatic orthopedic departments[J]. Pol Merkur Lekarski, 2022, 50(298): 227-231.
[14] MUNESHIGE K, MIYAGI M, INOUE G, et al.The relationship between falling distance and trauma severity among fall injury survivors who were transported to a trauma center[J]. Cureus, 2022, 14(5): e25099.
[15] ZHANG N, XUE Z, XU B, et al.Incidence and risk factors for unplanned intensive care unit transfer from the postanesthesia care unit: a case-control study[J]. BMC Anesthesiol, 2025, 25(1): 553.
[16] HAFNER T, KOLLMEIER A, LAUBACH M, et al.Care of geriatric patients with lumbar spine, pelvic, and acetabular fractures before and after certification as a geriatric trauma center DGU: a retrospective cohort study[J]. Medicina (Kaunas), 2021, 57(8): 794.
[17] BUTTS CC, MILLER P, NUNN A, et al.Rib fracture triage pathway decreases ICU utilization, pulmonary complications and hospital length of stay[J]. Injury, 2021, 52(2): 231-234.
[18] SAHNON ASA, TAHA M, OSMAN Y, et al.Timing of surgery and medical optimization in multisystem trauma: a systematic review of outcomes based on internal medicine co-management with orthopedic, neurosurgical, vascular, thoracic, and general surgical emergencies[J]. Cureus, 2025, 17(7): e87262.
[19] DORAN J, SALIH M, BELL A, et al.Major trauma patients and their outcomes - A retrospective observational study of critical care trauma admissions to a trauma unit with special services[J]. Injury, 2024, 55(8): 111622.
[20] WISE BT, CONNELLY D, ROCCA M, et al.Are deep infections that present before and after 90 days from orthopaedic trauma different? An analysis of the validity of the recent change in CDC criteria for infections[J]. Injury, 2022, 53(3): 912-918.

基金

湖南省卫生健康委员会科研计划项目 “骨科多发伤患者术后转入ICU风险预测模型构建”(D202304076456)

PDF(2769 KB)

Accesses

Citation

Detail

段落导航
相关文章

/

〈 〉