Laparoscopic ischaemic conditioning of the stomach may reduce gastric-conduit morbidity following total minimally invasive oesophagectomy.

OBJECTIVE Oesophagectomy, whether open or minimal access, is associated with a significant incidence of gastric-conduit-related complications. Previous animal and human studies suggest that ischaemic conditioning of the stomach prior to oesophagectomy improves perfusion of the gastric conduit. We have adopted laparoscopic ligation of the left gastric artery 2 weeks prior to minimally invasive oesophagectomy, having identified a relative high incidence of gastric-tube complications through a cumulative summation (CUSUM) analysis. METHODS This study included 77 consecutive patients who underwent a Total MIO (thoracoscopic oesophageal mobilisation, laparoscopic gastric tube formation, cervical anastomosis). The ligation group comprised 22 consecutive patients, excluding those with middle-third squamous tumours or early-stage adenocarcinoma, who underwent ligation 2 weeks prior to MIO at staging laparoscopy. The control group comprised 55 patients who did not undergo ischaemic conditioning in this way. We have defined conduit-related complications as: leak managed conservatively (L); tip necrosis requiring resection and re-anastomosis (TN) and conduit necrosis needing resection and oesophagostomy (CN). The values are reported as medians. The effect of ligation of the left gastric artery was followed with a CUSUM analysis. RESULTS Ligation was performed 15.5 days pre-operatively (median). There were no complications and the length of hospital stay was 1 day. Although gastric mobilisation at MIO was technically more difficult after ligation, there was no significant difference in operating time (ligation, 407 min; control, 425 min) or blood loss (ligation and control, 500 ml). There was less gastric-conduit morbidity in the ligation group (two of 22, 10%; one L, one CN) compared with the control group (11 of 55, 20%; four L, five TN, two CN), but these differences did not reach statistical significance (p=0.211 and p=0.176 Fisher's exact test). The CUSUM analysis showed that during ligation of the left gastric artery, conservatively treated gastric-conduit-related morbidity (leak, resection and re-anastomosis or conduit necrosis) remained within safe limits (10%). Conduit-related-morbidity increased after stopping ligation. CONCLUSION In this non-randomised clinical setting, our results suggest that ischaemic conditioning of the stomach prior to MIO is safe. There is a trend to reduced morbidity related to gastric-conduit ischaemia, which was demonstrated by a CUSUM analysis. A randomised trial is needed before ligation of the left gastric artery can be routinely recommended.