文献速递急腹症与外科感染围术期管理与 ERAS

ACS NSQIP手术风险计算器在意大利急诊腹部手术中的验证

Can the American College of Surgeons NSQIP Surgical Risk Calculator Accurately Predict Adverse Postoperative Outcomes in Emergency Abdominal Surgery? An Italian Multicenter Analysis.
PubMed 文献 · J Am Coll Surg · 2022-10-03

本研究纳入意大利六家机构的2749例急诊腹部手术,评估ACS NSQIP手术风险计算器对30天不良结局的预测价值。结果显示该工具对死亡等结局区分度良好,死亡AUC为0.932,且风险比>3.00时预测死亡阴性预测值达99%,推荐用于急诊腹部手术的多学科管理。

要点

  • ACS NSQIP计算器在意大利急诊腹部手术中对死亡预测区分度良好,AUC为0.932。
  • 风险比>3.00预测死亡敏感性86%、特异性77%、阴性预测值99%。
  • 多数非死亡结局存在风险低估,使用外科医生调整评分可改善校准。

核心结论

  • ACS NSQIP手术风险计算器在意大利急诊腹部手术中能可靠预测术后不良结局,尤其对死亡预测准确。

研究设计与样本

  • 多中心验证研究,纳入意大利六家机构的2749例急诊腹部手术,纳入诊断包括急性胆囊炎、阑尾炎、胃肠道穿孔或梗阻。

主要发现

  • 肾衰竭、心脏并发症、肺炎、静脉血栓栓塞、严重并发症及任何并发症的AUC均>0.8。
  • 死亡观察事件比与预期事件比为1.0,出院至机构为0.8;其他多数变量存在风险低估。
  • 风险比>3.00预测死亡敏感性86%、特异性77%、阴性预测值99%。

关键数值与推荐

  • 死亡观察事件比与预期事件比为1.0,出院至机构为0.8。
  • 风险比>3.00预测死亡敏感性86%、特异性77%、阴性预测值99%。

临床怎么用

  • 可用于急诊腹部手术患者术前风险评估,识别需临床优化的高危患者并支持知情同意。
  • 建议在多学科诊疗中常规使用该计算器,尤其关注死亡风险预测。
  • 对非死亡结局需注意风险低估可能,可结合外科医生调整评分改善校准。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: The American College of Surgeons NSQIP surgical risk calculator provides an estimation of 30-day postoperative adverse outcomes. It is useful in the identification of high-risk patients needing clinical optimization and supports the informed consent process. The purpose of this study is to validate its predictive value in the Italian emergency setting. STUDY DESIGN: Six Italian institutions were included. Inclusion diagnoses were acute cholecystitis, appendicitis, gastrointestinal perforation or obstruction. Areas under the receiving operating characteristic curves, Brier score, Hosmer-Lemeshow index, and observed-to-expected event ratio were measured to assess both discrimination and calibration. Effect of the Surgeon Adjustment Score on calibration was then tested. A patient's personal risk ratio was obtained, and a cutoff was chosen to predict mortality with a high negative predicted value. RESULTS: A total of 2,749 emergency procedures were considered for the analysis. The areas under the receiving operating characteristic curve were 0.932 for death (0.921 to 0.941, p < 0.0001; Brier 0.041) and 0.918 for discharge to nursing or rehabilitation facility (0.907 to 0.929, p < 0.0001; 0.070). Discrimination was also strong (area under the receiving operating characteristic curve >0.8) for renal failure, cardiac complication, pneumonia, venous thromboembolism, serious complication, and any complication. Brier score was informative (<0.25) for all the presented variables. The observed-to-expected event ratios were 1.0 for death and 0.8 for discharge to facility. For almost all other variables, there was a general risk underestimation, but the use of the Surgeon Adjustment Score permitted a better calibration of the model. A risk ratio >3.00 predicted the onset of death with sensitivity = 86%, specificity = 77%, and negative predicted value = 99%. CONCLUSIONS: The American College of Surgeons NSQIP surgical risk calculator has proved to be a reliable predictor of adverse postoperative outcomes also in Italian emergency settings, with particular regard to mortality. We therefore recommend the use of the surgical risk calculator in the multidisciplinary care of patients undergoing emergency abdominal surgery.

来源:J Am Coll Surg,查看原文。版权归原作者所有。

引用

  1. Scotton G, La Greca A, Lirusso C, et al. Can the American College of Surgeons NSQIP Surgical Risk Calculator Accurately Predict Adverse Postoperative Outcomes in Emergency Abdominal Surgery? An Italian Multicenter Analysis. J Am Coll Surg. 2022. PMID: 36648267. doi:10.1097/XCS.0000000000000445

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。