Comparison of short-term outcomes between transthoracic and robot-assisted transmediastinal radical surgery for esophageal cancer: a prospective study

Successful treatment options for esophageal cancer include chemotherapy and radiotherapy, but surgical treatment also plays an important role [1]. However, a radical esophagectomy entails postoperative complications and can result in postoperative mortality. According to the National Clinical Database (a large-scale database in Japan), the rate of postoperative complications following an esophagectomy is approximately 43%, and the postoperative mortality rate is approximately 3% [2]. Video-assisted thoracoscopic esophagectomy is regarded as a minimally invasive surgical technique for esophageal cancer [3‐5] and has the advantage of less postoperative pain and faster lung capacity recovery than seen in transthoracic esophagectomy (TTE) [6, 7]. However, a study comparing TTE with video-assisted thoracoscopic esophagectomy based on the aforementioned National Clinical Database showed no significant decrease in the length of hospital stay or any significant reduction in postoperative mortality. Furthermore, no significant reduction in the postoperative pneumonia rate was observed (15.5% for TTE and 15.0% for video-assisted thoracoscopic esophagectomy). Additionally, TTE reportedly increased postoperative complications requiring reoperation [2]. Despite being considered a minimally invasive procedure, video-assisted thoracoscopic esophagectomy had no significant positive impact on the pulmonary complication rate. Transhiatal esophagectomy is also a favored choice and is associated with lower perioperative morbidity, but the oncological outcome of this approach is generally considered to be inferior because fewer lymph nodes can be harvested than in the transthoracic approach [8].

We developed transmediastinal esophagectomy (TME), a nontransthoracic esophagectomy with radical mediastinal lymphadenectomy combining a robotic transhiatal approach with the video-assisted cervical approach reported previously by the authors [9]. In our previous retrospective study, we confirmed the practicability and safety of this procedure, demonstrated the equivalence of its oncological outcome with that of TTE in terms of the number of harvested lymph nodes and surgical margin pathology, and assessed the reduction in respiratory complications [10]. In the present, retrospective study, postoperative pneumonia was not observed in the TME group but was observed in 15% patients in the TTE group, and the length of postoperative hospital stay also significantly decreased.

The present study aimed to assess the lower invasiveness and the utility of the robot-assisted transmediastinal radical esophagectomy by prospectively evaluating the postoperative outcomes, changes in serum cytokine levels, and respiratory function over time.

The study patients were recruited at the Department of Gastrointestinal Surgery of the University of Tokyo Hospital between April 2015 and March 2017. The inclusion criteria were: (1) histologically proven esophageal cancer; (2) a T0–3 N0–2 M0 stage tumor according to the TNM Classification of Malignant Tumors, 7th edition; (3) age 20 years or older to 85 years or younger; (4) European Clinical Oncology Group Performance Status (ECOG-PS) ≦1; (5) good enough general health to tolerate a conventional open esophagectomy; (6) no concomitant malignancies; and (7) no preoperative radiotherapy. Patients with a history of surgery for other malignancies were excluded.

All the patients were offered the option of surgery (TTE or robot-assisted TME), and robot-assisted TME was performed in patients who elected to undergo this procedure despite lack of coverage under National Health Insurance. TTE was performed for all the remaining patients. The perioperative outcomes, serum cytokine levels, and respiratory function between the TME and TTE groups were compared.

To measure the serum cytokine levels, blood samples were collected preoperatively, immediately after surgery, and on postoperative days 1, 3, 5, and 7. The samples were immediately stored at 4 °C and centrifuged for 10 min at 3000 rpm, and the supernatant was cryopreserved at − 80 °C. After thawing, serum IL-6, IL-8, and IL-10 levels were measured using a BD CBA Human Inflammatory Cytokines Kit (BD Biosciences Inc., NJ, USA). All the samples were analyzed using BD Accuri C6 (BD Biosciences Inc., NJ, USA).

For respiratory function assessment, vital capacity (VC), forced vital capacity (FVC), and forced expiratory volume at 1.0 s (FEV 1.0) were measured preoperatively and at postoperative six months. The two groups were then compared in terms of pre- and postoperative changes.

The present, prospective study was approved by The University of Tokyo’s institutional review board. All 60 study participants provided written informed consent. This trial was registered in the UMIN Clinical Trial Registry (UMIN000017565 14/05/2015).

Seventy-eight patients with esophageal cancer between April 2015 and March 2017 who met the inclusion criteria were found. Among these patients, 18 had a history of surgery for another malignancy. In total, 60 patients were enrolled, and all were included in our analyses. Table 1 shows the background of 25 patients who underwent TME and 35 patients who underwent TTE. Cervical anastomosis was significantly more common in the TME group than in the TTE group (p < 0.0001). R0 resection (no residual tumor) was performed in 25 patients (100%) in the TME group and 34 patients (97.1%) in the TTE group. One patient (2.9%) in the TTE group underwent R1 resection, i.e., resection for a microscopic residual tumor. There was no case of R2 resection, i.e., resection for a macroscopic residual tumor, in either the TME or TTE group. Although there was no significant difference in the pathological stages between the groups, more patients in the TTE group had an advanced clinical and pathological tumor stage (cT and pT) (p = 0.0043, 0.01).

The present, prospective study compared perioperative outcomes, postoperative serum cytokine levels, and postoperative respiratory function in patients in a TME and TTE group who underwent radical surgery for esophageal cancer. Although the operation time was longer, the incidence of postoperative pneumonia was significantly lower, and the length of postoperative hospital stay was shorter, in the TME group than in the TTE group, suggesting that robot-assisted transmediastinal radical esophagectomy can be effective as a minimally invasive surgery for esophageal cancer. Differences in surgical approach and postoperative respiratory function might have contributed to the reduction in postoperative pneumonia. TME preserves the respiratory muscles and avoids adhesion associated with thoracotomy; this likely explains the postoperatively preservation of the respiratory function in TME. Furthermore, the results of multivariate analysis showed that older age and TTE were risk factors of postoperative pneumonia.

The present study revealed that postoperative serum cytokine levels decreased to a significantly greater degree in the TME group than in the TTE group. In studies comparing postoperative serum cytokine levels in TTE and video-assisted thoracoscopic esophagectomy, patients undergoing the latter procedure had significantly lower postoperative IL-6, IL-8, and IL-10 level [13‐16] and postoperative pneumonia incidence [15, 16]. The elevated serum IL-6 levels in TTE can reportedly be caused by exposure to oxygen in the lungs or by mechanical stimulation of the lungs [17‐19], both of which are reduced in video-assisted thoracoscopic esophagectomy. These two factors may be altogether absent in TME.

In the present study, the lower invasiveness of TME was assessed using serum IL-6, IL-8, and IL-10 levels as markers to indicate the degree of surgical invasiveness. Despite the longer surgical time required for TME, this study found that elevation in the level of all the cytokines immediately after surgery was inhibited in patients who underwent this procedure.

Respiratory function was better preserved after TME than after TTE. Previous studies on respiratory function after esophageal cancer surgery demonstrated that the reduction in the rate of post-thoracotomy VC and FEV 1.0 was 26 and 16%, respectively, whereas the corresponding values following video-assisted thoracoscopic esophagectomy were 15 and 8%, respectively [7]. In the present study, VC and FEV 1.0 after TTE decreased by 20.2 and 13.7%, respectively, in line with the findings of previous studies [7]. However, after TME, the VC and FEV 1.0 reduction rate was 6.3 and 1.0%, respectively, suggesting that TME is more effective in preserving respiratory function than video-assisted thoracoscopic esophagectomy. Respiratory muscle dysfunction following a laparotomy and thoracotomy is caused by reduced contraction efficiency of the dissected muscles, postoperative pain, and inhibition of phrenic nerve activity due to stimulation of the internal organs [17]. Respiratory function is most reduced immediately after surgery and recovers gradually [20]. In addition, physiological stress to the ventilated and non-ventilated lungs by differential lung ventilation is thought to reduce respiratory function during a thoracotomy. However, such factors are often minimized or avoided in TME. The preservation of postoperative respiratory function observed in the present study may have contributed to the reduced postoperative pneumonia rate.

Complications following TME included a higher incidence of anastomotic leakage than seen with TTE. Anastomotic leakage following surgery for esophageal cancer is associated with cervical anastomosis [21, 22], which is likely to be caused by tension applied to the site of anastomosis, ischemia, venous return insufficiency, and/or pressure of the gastric tube at the thoracic inlet [22]. In TME, all patients undergo cervical anastomosis, and the aforementioned factors were considered to have caused the high anastomotic leakage incidence in the present study. However, the length of postoperative hospital stay was significantly shorter in the TME group, suggesting that anastomotic leakage in this group might not have adversely affected the length of hospital stay in the present study.

The present study included a large number of patients in the TTE group with an advanced tumor stage. Although there was no significant difference in the lymph node stage or pathological stage, it is possible that the comparison was not oncologically equal. Moreover, in the present study, the assignment of patients to the surgical procedure was not randomized, and differences in the patient background may have affected the effectiveness of TME.

In conclusion, the present, prospective study indicated that TME might be a minimally invasive surgical procedure providing more, short-term benefits than TTE. However, additional studies should be conducted to evaluate the benefits of TME for patients with advanced esophageal cancer. Moreover, the present study did not compare TME with video-assisted thoracoscopic esophagectomy and included more confounding factors than a randomized trial.