Challenging Weight-Tiered Antibiotic Prophylaxis in Obese Patients Undergoing Colorectal Surgery Using CT-Derived Body Composition
Manjunath P Pai, Aleksas Matvekas, June Sullivan, Shuhan Liu, Lu Wang, Brian Derstine, Brian Ross, Hari Tummala, Scott Regenbogen, John Byrn, Grace L Su, Stewart WangAbstract
Objective
To determine whether CT-derived body composition and kidney function better predict cefazolin concentrations at the surgical site than body weight alone in patients with obesity undergoing colorectal surgery.
Background
Current cefazolin prophylaxis guidelines recommend dose escalation from 2 g to 3 g in patients weighing ≥120 kg. However, body weight inconsistently reflects antibiotic distribution at the surgical site.
Methods
We conducted a prospective study of adults with obesity undergoing elective colorectal surgery who received cefazolin prophylaxis and had preoperative CT imaging. Cefazolin concentrations were measured in plasma, subcutaneous adipose tissue, and colorectal tissue during surgery. CT-derived body composition metrics were quantified using analytic morphomics. Modeling and simulations were used to identify predictors of target attainment in subcutaneous tissue. Agreement between revised physiologic dosing criteria and current weight-based guidance was assessed in both a derivation cohort and an independent external cohort.
Results
Among 153 patients, 58 were included in model development and 95 in prospective evaluation. Weight and BMI did not improve model performance. Subcutaneous fat area (SFA) predicted tissue exposure, while creatinine clearance (CLcr) predicted elimination. Current weight-based dosing achieved 78%–84% target attainment. A revised framework of SFA ≥500 cm2 or ≥300 cm2 with CLcr ≥90 mL/min identified an additional 10%–14% of patients for dose adjustment, improving identification of underdosing-related SSI risk without unnecessary antibiotic use.
Conclusions
Cefazolin exposure is better explained by local adiposity and kidney function than by the current ≥120 kg weight-based dosing threshold.