Scholarlink

<p>Malnutrition and metabolic abnormalities are common in patients with heart failure (HF) requiring intensive care unit (ICU) admission and are associated with poor outcomes. This study evaluated the association between triglyceride–cholesterol–body weight index (TCBI) and in-hospital mortality in critically ill patients with HF and developed a machine learning–based model to assess its predictive value. This multicenter retrospective study included 1,537 HF patients in a derivation cohort and 3,537 patients in an external validation cohort. Kaplan–Meier, Cox regression, and restricted cubic spline analyses were used to assess the association and nonlinearity. An XGBoost model integrating TCBI and clinical variables was developed for risk prediction. Among 1,537 patients in the derivation cohort, 295 (19.2%) died during hospitalization. Mortality decreased across increasing TCBI tertiles (22.6% vs. 18.4% vs. 16.6%, P = 0.043). Patients in the highest tertile had lower mortality risk than those in the lowest (HR 0.58, 95% CI 0.39–0.85). A nonlinear inverse association was observed (P for nonlinearity = 0.001) and confirmed in the external cohort (HR 0.75, 95% CI 0.59–0.94). The model achieved area under the curve values of 0.791, 0.744, and 0.709 in the training, internal validation, and external validation cohorts, outperforming SAPS II and SOFA. Decision curve analysis indicated superior net clinical benefit across a range of threshold probabilities. Lower TCBI was independently associated with higher in-hospital mortality in critically ill patients with HF. Incorporating TCBI improved risk prediction and may aid early risk stratification in this high-acuity population.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Association of triglyceride–cholesterol–body weight index with in-hospital mortality in critically ill patients with heart failure

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月27日

查看全文
  • Yachen Xie,
  • Ke Tong

摘要

Malnutrition and metabolic abnormalities are common in patients with heart failure (HF) requiring intensive care unit (ICU) admission and are associated with poor outcomes. This study evaluated the association between triglyceride–cholesterol–body weight index (TCBI) and in-hospital mortality in critically ill patients with HF and developed a machine learning–based model to assess its predictive value. This multicenter retrospective study included 1,537 HF patients in a derivation cohort and 3,537 patients in an external validation cohort. Kaplan–Meier, Cox regression, and restricted cubic spline analyses were used to assess the association and nonlinearity. An XGBoost model integrating TCBI and clinical variables was developed for risk prediction. Among 1,537 patients in the derivation cohort, 295 (19.2%) died during hospitalization. Mortality decreased across increasing TCBI tertiles (22.6% vs. 18.4% vs. 16.6%, P = 0.043). Patients in the highest tertile had lower mortality risk than those in the lowest (HR 0.58, 95% CI 0.39–0.85). A nonlinear inverse association was observed (P for nonlinearity = 0.001) and confirmed in the external cohort (HR 0.75, 95% CI 0.59–0.94). The model achieved area under the curve values of 0.791, 0.744, and 0.709 in the training, internal validation, and external validation cohorts, outperforming SAPS II and SOFA. Decision curve analysis indicated superior net clinical benefit across a range of threshold probabilities. Lower TCBI was independently associated with higher in-hospital mortality in critically ill patients with HF. Incorporating TCBI improved risk prediction and may aid early risk stratification in this high-acuity population.