Development and external validation of a 90-day mortality prediction model for comatose sepsis patients: impact of cerebrovascular disease and dementia - Takeaways - MDSpire

Development and external validation of a 90-day mortality prediction model for comatose sepsis patients: impact of cerebrovascular disease and dementia

  • By

  • Yaoxi Tan

  • Yadong Shen

  • Yanting Zhang

  • July 2, 2026

  • 0 min

Share

  • 1

    The study analyzed 841 comatose sepsis patients to develop a 90-day mortality prediction model using LASSO and logistic regression.

  • 2

    Ninety-day mortality among the patients was found to be 22.83%, with independent risk factors including age and neurological comorbidities.

  • 3

    Cerebrovascular disease and dementia significantly modulated the prognostic value of the Glasgow Coma Scale in predicting mortality.

  • 4

    The final prediction model achieved a C-index of 0.81, outperforming existing models like SOFA and APACHE II.

  • 5

    External validation confirmed the model's transportability with a C-index of 0.75, indicating its applicability in contemporary clinical settings.

Original Source(s)

Development and external validation of a 90-day mortality prediction model for comatose sepsis patients: impact of cerebrovascular disease and dementia

  • by Yaoxi Tan, Yadong Shen, Yanting Zhang

  • July 2, 2026

Related Content

Comparative performance of mitral E/e’ versus inferior vena cava variability for predicting fluid responsiveness in mechanically ventilated adults with shock

Integrating chemokine signatures and multi-omic biomarkers to predict immunotherapy response in non-small cell lung cancer: a comprehensive narrative review

Interpreting elevated FRAP in critical illness: beyond antioxidant capacity