CT-derived subcutaneous adipose tissue thickness and circulating adipokines in relation to neurological outcome after in-hospital cardiac arrest: two parallel cohort analyses.
Zhou T, Ma W, et al. • Frontiers in neurology • 2026
CT-derived posterior subcutaneous adipose tissue thickness and circulating adipokines showed exploratory associations with neurological outcome in two independent IHCA cohorts, but these findings are hypothesis-generating and require external validation.
Key Findings
Results
In the CT Imaging Cohort, unadjusted differences in post-arrest posterior subcutaneous adipose tissue (pSAT) thickness were seen between neurological outcome groups, but none remained significant after FDR correction in the overall analysis.
Anterior and posterior SAT thickness were measured on pre-arrest and post-arrest thoracic CT.
Outcomes were survival and neurological outcome at hospital discharge.
Multiple comparisons were controlled with the Benjamini-Hochberg false discovery rate procedure.
Results
In the ventilation-matched sensitivity analysis, greater pSAT thickness at selected thoracic levels remained associated with favorable neurological outcome after multiple-comparison correction.
The association was seen only at selected thoracic levels.
The finding emerged in a sensitivity analysis, not the overall analysis.
Results
In the Biochemical Cohort, CTRP6, CTRP9, Omentin-1, and Vaspin concentrations were higher in patients with favorable neurological outcomes, and Chemerin was lower, after FDR correction.
Plasma concentrations of 12 adipokines measured during cardiac arrest were compared between neurological outcome groups.
Five of the 12 adipokines remained significant after Benjamini-Hochberg FDR correction.
Results
The clinical model (CPR duration and lactate) had an apparent AUC of 0.876, and adding CTRP6 raised it to 0.938.
Clinical model apparent AUC 0.876 (95% CI, 0.795-0.957).
Exploratory model with CTRP6 apparent AUC 0.938 (95% CI, 0.882-0.993).
Models were assessed with discrimination, calibration, bootstrap internal validation, and Firth regression sensitivity analysis.
Results
The improvement in discrimination from adding CTRP6 was not statistically significant.
ΔAUC was 0.062 (95% CI, -0.003 to 0.126; P = 0.062).
The optimism-corrected AUC of the extended model was 0.930.
Conclusions
The study design cannot establish a direct anatomical-biochemical relationship between adipose thickness and adipokines.
CT measurements and adipokines came from different patients in two parallel cohorts.
The authors describe the findings as hypothesis-generating and requiring external validation.
What This Means
This study asked whether body fat, measured on chest CT scans, and blood levels of fat-derived hormones called adipokines are linked to brain recovery after cardiac arrest in the hospital. It used two separate groups of patients: one with CT scans before and after the arrest, and one with blood samples taken during the arrest.
In the CT group, differences in the thickness of fat under the skin of the back were seen between patients with better and worse brain outcomes, but these did not hold up after statistical correction for multiple testing. They held up only at some chest levels in an analysis matched for ventilation. In the blood group, five adipokines (CTRP6, CTRP9, Omentin-1, Vaspin and Chemerin) differed between outcome groups. Adding CTRP6 to a model using CPR duration and lactate improved prediction numerically, but the improvement was not statistically significant.
This research suggests that fat tissue and its hormones may be related to brain outcomes after cardiac arrest, but the evidence is preliminary. The two groups included different patients, so the study cannot show that fat thickness and adipokine levels are directly connected. The findings need confirmation in other studies before any practical use.
Zhou T, Ma W, He Y, Tang S, Yao P, Gan L, et al.. (2026). CT-derived subcutaneous adipose tissue thickness and circulating adipokines in relation to neurological outcome after in-hospital cardiac arrest: two parallel cohort analyses.. Frontiers in neurology. https://doi.org/10.3389/fneur.2026.1898760