Colorectal cancer surgery carries a significant risk of early complications, and identifying which patients are most vulnerable before the operation is a central challenge in surgical planning. While factors like age and emergency presentation are already known to matter, they often do not fully capture a patient's true physiological reserve.
Two processes that colorectal cancer frequently triggers are secondary sarcopenia (progressive loss of muscle mass and quality) and systemic inflammation. Cancer accelerates muscle breakdown through altered metabolism and reduced physical activity, while inflammation can worsen muscle wasting and disrupt the body's ability to recover from surgical stress.
Because skeletal muscle is the body's largest protein reservoir, patients with depleted muscle stores may have reduced capacity to mount an immune response and repair tissue after major surgery. This biological overlap between poor muscle health and heightened inflammation provides a strong rationale for measuring both together when assessing preoperative risk.
Skeletal Muscle Gauge (SMG) is a composite CT-derived metric calculated by multiplying the Skeletal Muscle Index (SMI) - the cross-sectional muscle area at the third lumbar vertebra adjusted for height - by mean muscle attenuation measured in Hounsfield units (HU). Lower HU values reflect fatty infiltration of the muscle, a condition called myosteatosis, which signals reduced muscle quality beyond what mass alone can show.
Pan-Immune-Inflammation Value (PIV) is calculated from a routine complete blood count using the formula: (neutrophil count × platelet count × monocyte count) divided by lymphocyte count. By combining four cell types into a single number, PIV captures the balance between pro-inflammatory immune cells and lymphocytes more completely than simpler ratios like the neutrophil-to-lymphocyte ratio.
Both markers are derived from tests that are already routinely performed during cancer staging and preoperative workup. This means they add potential risk-stratification value without requiring additional procedures or costs, which makes them practical candidates for clinical integration.
This was a retrospective, single-center observational study conducted at the Second Clinic of Surgery at the Emergency County Clinical Hospital in Targu Mures, Romania. Consecutive adult patients who underwent major colorectal cancer surgery between January 2022 and November 2025 were screened for eligibility. The final analytic cohort comprised 190 patients, including both elective and emergency presentations.
CT-based muscle assessment was performed at the L3 vertebral level using 3D Slicer open-source software, with predefined attenuation thresholds of -30 to +150 HU to isolate skeletal muscle tissue. Measured muscles included the rectus abdominis, psoas major, erector spinae group, and abdominal obliques. PIV was calculated from blood tests drawn within 1 to 2 days before surgery.
The primary outcome was 30-day major complications, defined as Clavien-Dindo grade IIIb or higher - meaning complications requiring surgical, radiological, or endoscopic reintervention under general anesthesia, or death. Optimal cutoff values for SMG and PIV were identified using ROC curve analysis and the Youden index, which balances sensitivity and specificity.
Patients were then classified into four combined host phenotypes based on whether their SMG was low or normal and whether their PIV was high or low. Multivariable logistic regression controlled for age, surgical urgency, and procedure type to test whether SMG and PIV provided independent predictive value beyond established clinical risk factors.
In univariable analysis, low SMG was the strongest single predictor of 30-day major complications, with an odds ratio of 6.50 (95% CI 3.24 to 13.05, p less than 0.001). This means patients below the SMG cutoff of 867.9 were more than six times as likely to experience a serious postoperative complication. High PIV was also associated with major complications (OR 3.51, 95% CI 1.77 to 6.99).
In multivariable analysis adjusting for age, surgical urgency, and procedure type, low SMG remained an independent predictor (adjusted OR 4.17, 95% CI 1.91 to 9.09, p less than 0.001). Emergency surgery was the only other independent predictor (adjusted OR 3.48). High PIV showed a positive association after adjustment (adjusted OR 2.05) but did not reach statistical significance, suggesting it may partly overlap with the clinical severity already captured by emergency status.
Regarding model discrimination, a baseline clinical model using age, surgical urgency, and procedure type achieved an AUC of 0.739. Adding low SMG improved the AUC to 0.784, while further adding high PIV raised it only slightly to 0.791. This pattern confirms that SMG is the primary contributor to improved risk prediction, with PIV providing a modest additional signal.
When patients were grouped into four host phenotypes based on their combined SMG and PIV status, striking differences in outcomes emerged. The most favorable group - patients with low PIV and normal SMG - had a 30-day major complication rate of only 14.4% (13 out of 90 patients).
The least favorable phenotype, defined by high PIV combined with low SMG, comprised 23 patients and had a major complication rate of 78.3% (18 out of 23). The median Comprehensive Complication Index in this group was 100, compared to 0 in the most favorable group, reflecting an extreme gradient in overall postoperative burden.
ICU admission rates followed the same gradient: 18.9% in the low PIV/normal SMG group, rising progressively to 65.2% in the high PIV/low SMG group. Importantly, length of hospital stay did not differ significantly across phenotypes, suggesting the complications were severe but not necessarily prolonged in all cases.
The clinical value of identifying low SMG depends on how much time is available before surgery. In elective settings, low SMG may represent a modifiable target: patients identified before surgery could be referred for structured prehabilitation, nutritional optimization, or dietitian consultation to try to improve their muscle reserve and reduce operative risk.
In emergency or urgent presentations, where there is no time for meaningful preoperative intervention, the SMG and PIV phenotype serves a different purpose. Identifying patients with the combined high PIV/low SMG profile can guide triage decisions, prompting earlier escalation planning, lower thresholds for postoperative workup (such as assessment for anastomotic leak or sepsis), and proactive intensive care unit readiness.
Because PIV is not itself directly treatable, its primary role is as a marker of heightened immune-inflammatory burden rather than a therapeutic target. Combined with SMG in a phenotype framework, it helps clinicians identify the subset of patients who warrant the most intensive perioperative resources and postoperative surveillance, supporting risk-adapted Enhanced Recovery After Surgery (ERAS) planning.
Several limitations temper the conclusions. The retrospective, single-center design limits generalizability, and the cohort included a high proportion of emergency and complicated presentations - including bowel obstruction and perforation - which likely inflated both inflammatory burden and complication rates. Results may not directly apply to predominantly elective surgical populations.
The SMG cutoff of 867.9 was derived from the study cohort itself using ROC analysis and was not sex-specific or externally validated. Because muscle mass and quality differ between men and women, a universal threshold may not be optimal. Future studies should derive and validate sex-specific cutoffs in independent cohorts before applying this tool more broadly.
Key laboratory markers including CRP, albumin, and total protein had substantial missing data, limiting their inclusion in adjusted analyses. The model was also not validated in an external dataset. Prospective multicenter studies with standardized prehabilitation pathways are needed to determine whether phenotype-guided optimization can translate the risk stratification insights from this study into measurable improvements in patient outcomes.