Perfusion speed of indocyanine green in the stomach before tubulization is an objective and useful parameter to evaluate gastric microcirculation during Ivor-Lewis esophagectomy

Background: Anastomotic leakage (AL) during Ivor-Lewis esophagectomy (ILE), owing to gastric conduit (GC) ischemia, is a serious complication. Measurement parameters during intraoperative ICG fluorescence angiography (ICG-FA) are unclear. We aimed to identify objective ICG-FA parameters associated with AL. Study design: Patients > 18 years with an indication for ILE were enrolled. ICG-FA was performed at the abdominal and thoracic stage, and data, such as time of fluorescence appearance, speed of ICG perfusion, quality of GC perfusion (good, poor, ischemic), blood pressure, baseline patient characteristics, GC dimensions, and other intraoperative parameters were collected. On postoperative day 4 to 6, Gastrografin swallow radiography was performed. AL development was classified based on the Clavien–Dindo and SISG severity classifications. Univariate analysis with a 95% confidence level (p < 0.05) was performed. Factors with p < 0.05 were included in the multivariate analysis. Results: 100 patients were enrolled. During ICG-FA, evaluation of subjective perfusion was a very specific test (94.1%) with good negative predictive value (NPV 71.9%, p 0.034), but not powerful enough to detect patients at risk of leak (sensibility 21.8%, PPV 63.6%). The GC perfusion speed (cm/s) after gastric vascular isolation and before tubulization showed a significant association with AL (p < 0.003). Median arterial blood pressure in the thoracic stage (p < 0.001) or use of inotropic (p < 0.033) was associated with AL development. Conclusion: GC perfusion speed at ICG-FA is an objective parameter that could predict AL risk. Other results emphasize the importance of the microcirculation in the development of AL.