Computed Tomography-Based Limb Volume Measurements for Isolated Limb Infusion in Melanoma

Background: Despite advances in cross-sectional imaging, chemotherapeutic dosing for isolated limb infusion (ILI) in melanoma is currently calculated through cumbersome and potentially imprecise manual measurements. The primary objective of this study was to examine the feasibility of using computed tomography (CT) to calculate limb volume, its concordance with manual measurement, and its ability to predict clinical response and toxicity in patients undergoing ILI. Methods: A retrospective analysis of all patients undergoing lower extremity ILI at Duke University Medical Center between 2003 and 2014 was performed. Data pertaining to manually measured limb volume, chemotherapeutic dosing, and patient outcome was obtained. CT-based measurements of limb volume were performed in all patients for whom imaging was available and subsequently compared with manually measured values. Results: CT data were sufficient for measurement in 73 patients. The mean measurement time was 4.61 ± 2.13 min. Although average CT-based measurements were 1.20 L higher in the case of lower limbs, they correlated well with those obtained manually (r² = 0.90). Unlike manual measurement, patients with complete responses to chemotherapy had smaller limb volumes than those with disease progression as measured by CT (9.3 vs. 10.7 L; p = .038). Patients suffering grade 3 and 4 toxicities also had statistically lower limb volumes as measured by CT than those who did not (p < .05). Conclusions: CT-based limb volume measurement is feasible for chemotherapy dosing in patients undergoing ILI for melanoma and has predictive value with respect to clinical response and toxicity.