Skip to main navigation Skip to search Skip to main content

Detecting the metabolic transition to personalize nutritional timing: model development and preliminary validation in a large ICU cohort

  • Yonatan Gargi
  • , Neriya Levran
  • , Jacob Vine
  • , Amir Cohen
  • , Dana Weiner
  • , Dorit Stein
  • , Ori Levi
  • , Dor Cohen
  • , Julia Klein
  • , Hamutal S. Taube
  • , Maxim Glebov
  • , Teddy Lazebnik
  • , Mor Saban
  • , Shaked Efrat
  • , Elad Drori
  • , Yael Haviv
  • , Eran Segal

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

Background: The metabolic transition from catabolism to anabolism is a key determinant of recovery in critical illness and should guide nutritional therapy. However, no validated clinical marker currently exists to identify this transition, and current clinical practice relies on calendar-based recommendations. We aim to develop and preliminarily validate a physiology-based, trajectory-driven model to detect the metabolic transition window in critically ill patients. Methods: We conducted a retrospective cohort study in a tertiary-care general ICU. A daily insulin resistance index (IRI) was computed from glucose and insulin data and corrected for steroid exposure. Transition was defined as a ≥ 30% sustained drop in IRI after its peak, together with ≥ 2 of 8 physiologic recovery criteria (lactate, noradrenaline, vasopressin, adrenaline, inflammatory markers-WBC, %neutrophils, CRP, and albumin). Associations with 90-day mortality and caloric exposure were evaluated using Kaplan–Meier analysis and multivariable landmark Cox models. Results: The cohort included 2,350 patients (age 59.8 ± 16.5 years; SOFA 11.7 ± 3.9; 82% mechanically ventilated). A metabolic transition was identified in 94% of patients, with ~ 60% transitioning by ICU day 3. Transition by day 3 was associated with lower 90-day mortality (HR 0.72, 95% CI 0.65–0.81). Patients who never transitioned had substantially higher mortality. High caloric delivery (≥ 1.0 kcal/kg/h for ≥ 24 h) before transition was independently associated with increased 90-day mortality (OR 1.25, 95% CI 1.01–1.55), with a dose–response pattern. In contrast, high caloric delivery based on calendar timing failed to demonstrate a similar pattern. Conclusions: We developed and validated a physiological model for detecting metabolic transition in critical illness and showed that transition occurs early and strongly predicts survival. Higher caloric delivery before transition is associated with increased mortality, supporting nutrition strategies aligned with physiological recovery rather than fixed calendar days.

Original languageEnglish
Article number132
JournalCritical Care
Volume30
Issue number1
DOIs
StatePublished - Dec 2026
Externally publishedYes

Keywords

  • Catabolism
  • Critical illness
  • Insulin resistance
  • Intensive care
  • Metabolism
  • Nutrition

Fingerprint

Dive into the research topics of 'Detecting the metabolic transition to personalize nutritional timing: model development and preliminary validation in a large ICU cohort'. Together they form a unique fingerprint.

Cite this