Computed tomography measurements to predict need for robotic transversus abdominis release: a single institution analysis.

Abstract

The radiographic rectus width to hernia width ratio (RDR) has been shown to predict ability to close fascial defect without additional myofascial release in open Rives-Stoppa abdominal wall reconstruction (AWR), but it has not been studied in robotic AWR. We aimed to examine various CT measurements to determine their usability in predicting the need for transversus abdominis release (TAR) in robotic AWR. We performed a single-center retrospective review of 137 patients with midline ventral hernias over a 5-year period who underwent elective robotic retrorectus AWR. We excluded patients with M1 or M5 hernias, lateral/flank hernias, and hybrid repairs. The CT measurements included hernia width (HW), hernia width/abdominal width ratio (HW/AW), and RDR. Univariate, multivariate and area under the curve (AUC) analyses were performed. 58/137 patients required TAR (32 unilateral, 26 bilateral). Patients undergoing TAR had a significantly higher average HW and HW/AW and lower RDR. Multivariate analysis revealed that prior hernia repair was independently associated with need for TAR (p = 0.03). ROC analysis and AUC values showed acceptable diagnostic ability of HW, HW/AW and RDR in predicting need for TAR. Cutoffs of RDR ≤ 2, HW/AW > 0.3, and HW > 10cm yielded high specificity in determining need for any TAR (97.5% vs. 96.2% vs. 92.4%) or bilateral TAR (95.5% vs. 94.6% vs. 92.8%). History of prior hernia repair was a risk factor for robotic TAR. CT measurements have some predictive value in determining need for TAR in robotic AWR. Further prospective analysis is needed in this patient population.