Preoperative hiatal hernia in oesophageal adenocarcinoma: An impact on patient outcomes?

Abstract

Abstract Objective Uncontrolled gastroesophageal reflux and the often-associated hiatal hernia (HH) are frequently encountered in oesophageal adenocarcinoma patients. Previous data suggest unfavourable long-term oncologic outcomes in the presence of a HH, but the evidence remains scarce. The aim of this study was to assess the potential impact of preoperative HH on histologic response after neoadjuvant treatment (NAT), as well as on overall and disease-free survival. Methods All patients operated for an adenocarcinoma of the oesophagus or gastro-oesophageal junction (GOJ) between 2012-2018 were assessed. Baseline endoscopy and CT-scan images were retrospectively reviewed to identify the presence of a clinically significant HH (≥3cm). Response to neoadjuvant treatment (Mandard TRG grade), postoperative outcomes and survival were compared between HH and non-HH patients. Categorical variables were compared with the x2 or Fisher’s test, whereas continuous ones with the Mann-Whitney-U test. Survival analyses were performed with the Kaplan-Meier method and log-rank test. Results Overall, 101 patients were included (84.1% male, median age 63 years); among them, 33 (32.7%) had a HH ≥ 3cm at diagnosis of oesophageal cancer. There were no significant baseline differences in demographics and tumour stages between the two groups. NAT was used in 80.9% of non-HH versus 81.8% HH patients (P = 0.910), most often chemoradiation (57.3% in non-HH versus 63.6% in HH patients, P = 0.423). Surgical approach and postoperative complication rates were similar in all patients. Good response to NAT (TRG 1-2) was observed in 32.3% of non-HH, versus 33.3% of HH patients (P = 0.297), whereas R0 resection was achieved in 94.1% vs 90.9% of patients respectively (P = 0.551). Overall survival was comparable between HH (median 28 mo, 95%CI 22-NA) and non-HH patients (median 41mo, 95% CI 29-NA) (P = 0.605). Disease-free survival was also similar (median 18 mo, 95%CI 12-NA for HH, vs 34mo, 95%CI 14-NA for non-HH patients, P = 0.283), although HH patients experienced higher rates of distant (51.6% vs 29.2% for non-HH, P = 0.033), but not locoregional recurrence. Conclusion A clinically significant HH is encountered in almost a third of patients with oesophageal adenocarcinoma. However, in our study, it was not associated with a worse response to NAT, nor did it lead to a worse overall and disease-free survival.