104. RISK FACTORS FOR ANASTOMOTIC LEAKAGE AFTER ESOPHAGEAL SUBTOTAL RESECTION AND GASTRIC TUBE RECONSTRUCTION U USING ICG FLUORESCENCE

Abstract Background One of the significant causes of gastrointestinal anastomotic leakage is decreased blood flow to the organs, and the ICG fluorescence method is considered useful for assessing blood flow during gastrointestinal surgery. However, there are few reports highlighting important considerations when employing the ICG fluorescence method. This study aims to investigate risk factors for anastomotic leakage, including data obtained from ICG fluorescence. Methods We included 268 patients who underwent subtotal esophageal gastric tube reconstruction from January 2018 to August 2024. We excluded 24 cases due to evaluation difficulties from retrograde staining and an additional three challenging cases, resulting in 241 cases for assessment. Cases with Clavien-Dindo Grade 2 or higher anastomotic leakage were classified as having leakage. For ICG fluorescence of the gastric tube serosa, the rapidly fluorescing upper part was defined as the 1st line, the faintly fluorescing part toward the tip within seconds as the 2nd line, and the portion fully stained after about 20 seconds as the terminal part. Results 22/241 cases (8.4%) had anastomotic leakage(AL). ICG examination findings showed distances from the gastric tube tip: 1st line was 12.3 ± 2.4 (AL group)/11.4 ± 2.1 (nAL group) (P = 0.039), 2nd line was 9.2 ± 2.1/8.0 ± 2.2 (P = 0.016), and terminal was 4.9 ± 2.8/3.6 ± 2.3 (P = 0.017). The speed from the 1st to the 2nd line (cm/second) was 0.9 ± 0.3/1.1 ± 0.4) (P = 0.041). Multivariate analysis indicated that the speed from the 1st to the 2nd line was an independent risk factor for anastomotic leakage (OR: 0.69, 95% CI: 0.006–0.775, P = 0.030). Conclusion The results of the multivariate analysis indicated that the speed from the 1st line to the 2nd line was the only independent risk factor for anastomotic leakage. No significant difference was found when comparing the data obtained from ICG to the actual site of anastomosis. Additionally, no significant differences were observed in other patient background factors, surgical factors, or perioperative factors.