Comparison of end-to-side hand-sewn and side-to-side stapled cervical esophagogastric anastomosis in patients with lower thoracic esophageal cancer undergoing transhiatal esophagectomy: an Iranian retrospective cohort study

Esophageal cancer is one of the most prevalent and multifaceted gastrointestinal malignancies and the sixth leading cause of mortality among cancers [1, 2]. Various methods have been introduced as a mainstay of treatment, including surgical procedures and non-surgical palliative approaches, but the current standard for the management of esophageal cancer is esophagectomy [3‐5]. Esophagogastric anastomosis, including the popular variants hand-sewn and stapled anastomoses, is the most critical procedure during esophagectomy [6, 7]. Regardless of surgical approach, nonetheless, preventing anastomotic complications is necessary to minimize early morbidity and improve intervention outcomes [6].

Postoperative complications subsequent to esophagogastric anastomosis may lead to life-threatening situations, including anastomotic leakage, anastomotic stricture, and other rare complications, such as fistulas and abscesses. Anastomotic leakage occurs in more than 10% of patients undergoing esophagogastric anastomosis, with the condition accompanied by some complications, such as mediastinitis, nourishing discomfort, and anastomotic stricture as well as less common complications, including cervical osteomyelitis [6, 8, 9]. An important requirement therefore is to choose a surgical procedure that can accurately and effectively prevent and reduce post-anastomotic complications. An automatic stapling for intestinal anastomosis was introduced by Berthold et al. in 1980. Stapled techniques have been presented in the past years as a means of minimizing the risk of anastomotic leakage and stricture [10‐12]. Collard et al. used linear stapling devices in 1998 to carry out esophagogastric anastomosis [13], and Orringer applied structural modifications to previously developed techniques to improve results [14], but whether the improved versions are superior remains a matter of debate.

This deficiency is exacerbated by the fact that although some studies, including meta-analyses and systematic reviews, have been conducted to compare the effectiveness of hand-sewn and stapled anastomosis techniques (through the use of linear or circular staplers), the superiority of one approach over the other remains a controversial issue [15, 16]. To address this matter, we compared the operation times and postoperative outcomes of transhiatal esophagectomy and cervical esophagogastric anastomosis performed via hand-sewn and stapled techniques over a year of follow-up. The specific issues compared were symptoms, anastomotic leakage, anastomotic stricture, and the need for anastomotic dilatation in patients.

An initial sample of 575 consecutive patients with resectable esophageal cancer undergoing esophagectomy and regional lymphadenectomy from March 2010 to March 2016 were screened for possible inclusion in the study. These patients were those admitted into the Department of Thoracic Surgery at Imam Reza Hospital in Tabriz, Iran (the referral hospital in the northwestern region of the country).

The exclusion criteria were as follows: (i) patients who underwent esophagogastric anastomosis via circular stapling (n = 68), (ii) patients who had reconstruction using colonic or jejunal interposition (n = 44), (iii) patients with incomplete medical records (n = 10), and (iv) patients subjected to Ivor Lewis, McKeown, or thoracoabdominal esophagectomy (n = 20). On these bases, 142 patients were excluded, and the remaining 433 were enrolled in this study. The study protocol was approved by the Regional Ethics Committee headed by the Vice-Chancellor of Research and Development at Tabriz University of Medical Sciences (Ethical approval number: 91/1–1/4). Written informed consent was obtained from each patient during questionnaire administration for the collection and analysis of applicable clinical data. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki, as reflected in a priori approval by the institution’s human research committee.

Demographic information, including age, sex, preoperative clinical signs and symptoms, and intraoperative surgery-related variables, such as operation time, anastomotic approach, pathological stage, and postoperative outcomes, were extracted from medical records. The patients were evaluated postoperatively for anastomotic leakage and stricture on the 12th month of follow-up.

The stomach is a good alternative for patients with esophageal cancer undergoing esophagectomy. Traditionally, the accepted standard treatment for operable esophageal carcinoma is resection of the esophagus and lymph nodes with gastric pull-up and construction with cervical or intrathoracic esophagogastric anastomosis [6, 10, 17]. The main complications occurring after esophagectomy and esophagogastric anastomosis are anastomotic leakage and stricture, which may affect patients’ quality of life [17]. Recent studies have shown that anastomotic leakage and the type of mechanical stapler used in surgical procedures are critical to the development of anastomotic stenosis [18, 19].

The present research was conducted on 409 patients undergoing esophagectomy via end-to-side hand-sewn and side-to-side stapled cervical esophagogastric anastomoses. The results showed a significantly high prevalence of cervical esophagogastric anastomotic leakage following transhiatal esophagectomy through manual anastomosis. In a retrospective study Cooke et al. indicated that 1133 patients undergoing esophagectomy followed by esophagogastric anastomosis showed a significant reduction in postoperative complications and the prevalence of problems in anastomotic construction using mechanical anastomosis [7]. The literature reflected that the incidence rate of cervical esophagogastric anastomotic leak falls between 9 and 14% but that rates of 15 to 25% are commonly reported with respect to hand-sewn anastomosis [5, 20]. In the present study, the prevalence of anastomotic leakage in patients undergoing stapled and hand-sewn anastomoses were 4.9 and 14.02% respectively. These findings may be more acceptable than the levels derived in other studies. It is understandable that the difference in leakage rates between anastomotic techniques is difficult to distinguish given that leakage incidence is less than 3% when anastomosis is performed by experienced thoracic surgeons [5]. Similar to our results, those of Mishra et al. showed that the rate of anastomotic leakage was significantly higher in patients undergoing hand-sewn anastomosis than in patients treated via linear stapled anastomosis [21].

Laterza et al. compared manual and mechanical anastomoses and found that patients treated using the latter exhibited a high prevalence of anastomotic leakage and benign stricture [22]. Other randomized controlled trials revealed a higher prevalence of anastomotic leakage and anastomotic stricture in manually operated individuals, suggesting the superiority of mechanical anastomosis as a technique for esophagogastric anastomotic construction [23‐25]. Even in intrathoracic esophagogastric anastomosis where linear stapling is conducted, a significant decrease in anastomotic leakage and stricture was observed compared with the levels occurring under hand-sewn anastomosis [19].

Sugimura et al. used a modified Collard technique and a linear stapler to construct the posterior wall in anastomosis and closed the anterior wall using the linear stapler twice. The authors showed that anastomotic leakage was less frequent in the modified Collard group than in the hand-sewn group but that the difference was not significant. Anastomotic stenosis occurred to a significantly lower extent in the modified Collard group, and the period between esophagectomy and initial dilatation was significantly shorter in the hand-sewn anastomosis group [26]. Similarly, Ishibashi et al. performed triple-stapled quadrilateral anastomosis to create esophagogastric anastomosis and reported no significant anastomotic leakage and stricture [27].

Some reviews indicated no significant difference between hand-sewn and stapled anastomosis techniques in terms of the prevalence of anastomotic stricture. However, our results showed a decreasing pattern in the rate of anastomotic stricture during the follow-up period in the stapled anastomosis group compared with the rate observed in the manual anastomosis patients. Comparably, Cooke et al. discovered a significant reduction in the prevalence of postoperative complications and morbidity in patients for whom mechanical anastomosis was carried out [7]. Price et al. found that although an anastomotic site was irrelevant to the likelihood of postoperative complications, such as anastomotic leakage and stenosis, patients treated via manual anastomosis experienced a higher incidence of anastomotic leakage and stricture [25].

The current study uncovered that the prevalence of benign anastomotic stricture during the 12th month of follow-up was significantly higher in the hand-sewn anastomosis group than in their stapled counterparts. This led us to conclude that mechanical anastomosis plays an important role in reducing the incidence of postoperative complications by creating a wider anastomotic space than that achieved with a hand-sewn technique.

There are several studies showing an increase in reflux symptoms following esophagectomy and gastric pull up in patients with esophageal cancer [28‐30]. The current study uncovered that the prevalence of reflux symptoms during the 12th month of follow-up was significantly higher in the hand-sewn anastomosis group than in their stapled counterparts. However, Ercan et al. reported no significant difference between stapled and hand sewn anastomoses in reflux symptoms. Despite the increase in the diameter of the anastomosis, the probability of reflux did not increase in the patients of their study [31]. Sugimura and colleagues showed that frequency of reflux esophagitis tended to be lower in the mechanical group than in the hand-sewn group prior to propensity score matching [26]. In our study, the length of the remaining cervical esophagus was longer than the hand-sewn group in order to create a proper anastomosis in the stapled group. Therefore, the location of the anastomosis was at the entrance to the chest; while in the hand-sewn method, the anastomotic location was performed approximately 3 cm below the cricopharynx level. Consequently, we hypothesized that the length of esophagus remnant may be a major factor contributing to the reduced prevalence of reflux symptoms in patients with stapled anastomosis.

We also evaluated endoscopic bougie dilatation to relieve benign anastomotic stenosis and recurrent dysphagia within the 12-month follow-up. The statistical results showed a significant difference between the patients who required esophageal dilatation. In a similar vein, Sugimura et al. found a significantly lower frequency of anastomotic stricture in the stapled anastomosis group than in the hand-sewn anastomosis group [26].

Hsu et al. compared the operating times entailed in manual and mechanical anastomoses and discovered that duration was significantly shorter in mechanical anastomosis technique than manual method [9]. However, some studies, including meta-analyses, found no significant difference in surgical times between manual and mechanical anastomosis procedures [16, 32]. The results of the present study showed that the effective time of using a stapled technique for cervical esophageal anastomosis was significantly shorter than the manual procedure, This difference may be attributed to a number of reasons: use of numerous hand-sewn techniques described by surgeons (single-layer vs. multilayer anastomosis, interrupted vs. running suture techniques), intraoperative mishaps (e.g., poor alignment of sutures), and the skills of surgeons performing operations [9].

Our research has certain limitations. First, this study was a retrospective study and therefore required clinical trials to achieve acceptable results. Second, the follow-up period spanned 12 months, but a longer time frame may generate different results. Third, all the operations were performed by three surgeons. Although the surgical team had sufficient experience in esophageal surgery and all anastomoses were performed in the neck, the composition of the team may still have resulted in undesirable bias.

Compared with hand-sewn anastomosis, cervical esophagogastric anastomosis using a side-to-side stapled technique involved a shorter operating time, caused lower anastomotic leakage and stricture rates, and required less frequent anastomotic dilatation. To conclude, we found stapling to be a safer and more superior approach for esophagogastric in patients with esophageal cancer undergoing transhiatal esophagectomy.