Improving surgical risk prediction through integrating automated body composition analysis: a retrospective study on colectomy surgery

Background: Body composition measurements may predict surgical outcomes, but current approaches are not optimized across body regions and outcome types. The added value of these measurements beyond established clinical factors remains unclear, highlighting a critical gap. Methods: This retrospective cohort study conducted across three hospitals within a tertiary health system evaluated whether computed tomography (CT)-derived body composition metrics predicted postoperative outcomes after colectomy, alone or in combination with clinical variables or the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) Surgical Risk Calculator. Of 3560 colectomy patients (2010–2023), 1621 met the inclusion criteria (a preoperative CT within 90 days with L3 visible). More than 300 features were extracted from preoperative CTs across multiple vertebral levels, including skeletal muscle area, density, fat area, and inter-tissue metrics. The primary outcome was 1-year all-cause mortality, assessed using Cox proportional hazards modeling, the concordance index (C-index), and the integrated Brier score. Secondary outcomes included complications, unplanned readmission, blood transfusion, and severe infection, and were evaluated using the area under the receiver operating characteristic curve (AUC) and Brier score from logistic regression. Results: The development subset included 474 patients (2010–2015), and the validation subset included 1147 patients (2016–2023). Optimal anatomic locations for body composition measurements varied by outcome. Sex-normalized L3 skeletal muscle density was the strongest independent predictor of 1-year mortality [odds ratio (OR): 0.42, 95% confidence interval (CI): 0.30–0.58; AUC: 0.71], whereas T12 muscle density better predicted complications (OR: 0.62, 95% CI: 0.49–0.78; AUC: 0.63). Integrating image-based frailty scores with clinical variables improved prediction over clinical variables alone for mortality (C-index 0.80 vs. 0.75) and complications (AUC 0.70 vs. 0.64). Adding sex-normalized muscle density to the NSQIP Calculator modestly but significantly improved mortality prediction (C-index 0.87 vs. 0.86, P < 0.01). The largest benefit was in NSQIP-classified low-risk patients (predicted mortality <5%), where the C-index increased by 4% ( P < 0.001), suggesting that image-based frailty scores can identify at-risk individuals who would otherwise be overlooked. Conclusions: CT-derived body composition metrics strongly predict colectomy outcomes, particularly in patients considered low risk preoperatively. Integrating these metrics with existing risk calculators can improve surgical risk stratification and guide preoperative decision-making.

Authors

Institutions

Publication Details

Journal
International Journal of Surgery
Published
2026-11-17
DOI
https://doi.org/10.1097/js9.0000000000005737
Citations
3
Primary Topic
Body Composition Measurement Techniques
Type
article
Field-Weighted Citation Impact
7.51
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Improving surgical risk prediction through integrating automated body composition analysis: a retrospective study on colectomy surgery

Christopher R. Mantyh
3 citations
International Journal of Surgery
Body Composition Measurement Techniques
7.51
article

Improving surgical risk prediction through integrating automated body composition analysis: a retrospective study on colectomy surgery

Christopher R. Mantyh
article en
3 citations

Abstract

Background: Body composition measurements may predict surgical outcomes, but current approaches are not optimized across body regions and outcome types. The added value of these measurements beyond established clinical factors remains unclear, highlighting a critical gap. Methods: This retrospective cohort study conducted across three hospitals within a tertiary health system evaluated whether computed tomography (CT)-derived body composition metrics predicted postoperative outcomes after colectomy, alone or in combination with clinical variables or the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) Surgical Risk Calculator. Of 3560 colectomy patients (2010–2023), 1621 met the inclusion criteria (a preoperative CT within 90 days with L3 visible). More than 300 features were extracted from preoperative CTs across multiple vertebral levels, including skeletal muscle area, density, fat area, and inter-tissue metrics. The primary outcome was 1-year all-cause mortality, assessed using Cox proportional hazards modeling, the concordance index (C-index), and the integrated Brier score. Secondary outcomes included complications, unplanned readmission, blood transfusion, and severe infection, and were evaluated using the area under the receiver operating characteristic curve (AUC) and Brier score from logistic regression. Results: The development subset included 474 patients (2010–2015), and the validation subset included 1147 patients (2016–2023). Optimal anatomic locations for body composition measurements varied by outcome. Sex-normalized L3 skeletal muscle density was the strongest independent predictor of 1-year mortality [odds ratio (OR): 0.42, 95% confidence interval (CI): 0.30–0.58; AUC: 0.71], whereas T12 muscle density better predicted complications (OR: 0.62, 95% CI: 0.49–0.78; AUC: 0.63). Integrating image-based frailty scores with clinical variables improved prediction over clinical variables alone for mortality (C-index 0.80 vs. 0.75) and complications (AUC 0.70 vs. 0.64). Adding sex-normalized muscle density to the NSQIP Calculator modestly but significantly improved mortality prediction (C-index 0.87 vs. 0.86, P < 0.01). The largest benefit was in NSQIP-classified low-risk patients (predicted mortality <5%), where the C-index increased by 4% ( P < 0.001), suggesting that image-based frailty scores can identify at-risk individuals who would otherwise be overlooked. Conclusions: CT-derived body composition metrics strongly predict colectomy outcomes, particularly in patients considered low risk preoperatively. Integrating these metrics with existing risk calculators can improve surgical risk stratification and guide preoperative decision-making.

International Journal of Surgery
Duke University (US), Duke Medical Center (US)
Good health and well-being
Openalex Percentile: Top 4%
Body Composition Measurement Techniques
7.51
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon