Lower thoracic ESCC has its own spread, anatomy, space relation and prognosis characteristics which make it a special type of esophageal cancer. Until now, there has been no consensus on the extent of esophageal resection. Although HA and LA have been widely applied in clinics, comparison between these anastomosis locations on long-term survival is still unclear. Based on our results, LA is associated with poor prognosis in lower thoracic ESCC with pT3-T4 status and should be avoided for this cohort.
In our study, we defined cervical anastomosis and high intrathoracic anastomosis as high anastomosis (HA) and sub-aortic anastomosis and anastomosis below the azygos arch as low anastomosis (LA). The basis of this classification was based on from findings of our pilot study which suggested that the prognosis of cervical anastomosis and high intrathoracic anastomosis was similar both in the entire cohort or subgroup analysis stratified by pT status (data not shown). Manabu Okuyama et al, in their randomized controlled trial reported that patients with cervical anastomosis and high intrathoracic anastomosis had similar surgical outcomes (11).
In this study, we firstly evaluated the impact of anastomosis choice in this entire cohort. Although LA was found to be correlated with shorter survival, it lacked statistical significance after controlling for other confounders. Then, we evaluated the impact of anastomosis choice in subgroup analysis stratified by pT status. We found that for patients with T1-2 status, the prognosis achieved by LA was similar to HA (median OS, 140.9 versus 124.8; P=0.345 by log-rank test), whereas for patients with T3-4 status, HA was associated with better outcome than LA, even after controlling for other confounders adjusted HR=0.726, 95%CI, 0.597 0.972, P=0.026). Furthermore, the survival advantage of HA was not significantly modified by nodal status (N-: adjusted HR=0.582, 95%CI, 0.369 0.920, P=0.021; N+: adjusted HR=0.567, 95%CI, 0.323 0.995, P=0.048). Therefore, the detrimental impact of LA might be concentrated on patients with advanced primary tumor stage.
For this issue, although the underlying mechanism is still unclear, it might be associated to two reasons. First, the existence of second multiple primary esophageal squamous cell carcinomas (SMPESCC). According to a previous study, SMPESCC is a common phenomenon in ESCC with the general incidence of 0.1–10% (12–14). A previous study reported that lower esophagus was the most common segment in which SMPESCC occurs (54.7%) (15). In addition, Predrag Pesko et al, in their study found out locally advanced patients were more likely suffered from multiple ESCC (16). This incidence would be even higher when the tumor is low thoracic location and locally advanced. Therefore, lower thoracic ESCC patients with T3-4 status might require a more extended esophagectomy to clear potential SMPESCC. Second, as is well known, lymph vessels exist in all layers of esophageal expect the epithelium and outer membrane (17). When the tumor invades deeper layers, the incidence of lymphatic metastasis or micro-metastasis would increase. Thus, for locally advanced ESCC, it would be beneficial to perform an extended esophagectomy.
Debate between right and left transthoracic approach for resectable ESCC has long been a hot topic for a long time. Recently, a series of studies have shown that the right thoracic approach was superior to left thoracic approach on extensive lymphadenectomy, especially in the upper mediastinum, which was critical for more accurate staging and long-term survival (10, 18). In this study, the proportion of patients undergoing HA was significantly higher in the right-thoracotomy group (97.8% versus 70.1%). The fact that extensive esophagectomy is more common in right thoracic approach might be another explanation for survival advantage of right thoracic approach.
Based on our results, LA for lower thoracic ESCC staged as pT3-4 may not be appropriate. For pT1-2 patients, low anastomosis should still be cautiously adopted due to its potentially detrimental impact on eliminating micrometastasis. Routine EUS examination before operation is greatly help for treatment decision (19).
There are some limitations should be considered in this study. First, this is a single-institution retrospective study. Second, we did not perform an external validation to validate the findings. Third, patients with neoadjuvant treatment were not included. Therefore, further validation from multi-center database is needed and meanwhile, the findings from this study should be cautiously interpreted.