Anastomotic leak after esophagectomy

Background: Esophageal anastomotic leakage (AL) remains a frequent and feared postoperative complication, associated with high mortality and impaired quality of life. The aim of this study was to assess AL rates after esophagectomy with anastomosis at neck for esophageal and gastroesophageal junction cancer (GEJ), and compare the impact of AL on oncological outcome. Methods: Patients with squamous cell carcinoma and adenocarcinoma of esophagus/ gastroesophageal junction who underwent surgery between 2001-2018 were analyzed for cervical anastomotic leak. Results: 419 patients underwent esophagectomy with anastomosis placed at neck during 2001-2018. AL rate was 16%. AL was not found to be associated with anastomotic technique, surgical approach and technique, organ of conduit and route of conduit. A subgroup of patients (n=93) who had undergone neoadjuvant chemoradiation followed by surgery had AL of 30% vs 12% in rest of the treatment modality group (p<0.001). Median survival was 26 months and 34 months in patients with AL and without AL, respectively (p=0.03). AL was managed successfully in all patients. Conclusion: Cervical AL after esophagectomy for cancer of esophagus and GEJ can be treated successfully without major complications.


Introduction
Esophageal cancer is one of the most aggressive of gastrointestinal malignancies.Esophagectomy has always remained the mainstay of treatment, usually in combination with chemoradiation.Even when esophageal cancer is resectable, esophagectomy carries a high risk of death (3.6-4.5%)compared with most surgically treated cancers. 1 Many efforts have been made to improve the esophagectomy technique and to reduce postoperative complications, but esophageal anastomotic leakage (AL) remains a frequent and feared postoperative complication, associated with high mortality and impaired quality of life.However, improvement of surgical techniques and management of complications has led to a steady decrease in postoperative mortality over the years. 2Some factors have been associated with AL development, such as patients' nutritional status and comorbidities, cancer stage, surgical procedure, and neoadjuvant therapy, but there are some controversies in the literature about the significant risk factors for this adverse event. 3Correspondence Dr Binay Thakur , Dept. of Surgical Oncology (Thoracic Unit), BP Koirala Memorial Cancer Hospital, Bharatpur, Nepal.Email: binaythakur@hotmail.com.Phone: +977-9855055931.
The aim of this study was to assess AL rates after esophagectomy with anastomosis at neck for esophageal and gastroesophageal junction cancer (GEJ), and compare the impact of AL on oncological outcome.

Patients:
Patients with cancer of the thoracic esophagus or gastroesophageal junction (GEJ) seeking surgical treatment (Thoracic Unit) at BP Koirala Memorial Cancer Hospital (BPKMCH) between 2001 and 2018 were evaluated.This was a retrospective analysis of prospectively maintained database.
The study was approved by the Institutional Review Committee, BPKMCH.Because individual patients could not be identified, the need for patients' consent was waived.

Staging:
The preoperative workup included physical examination, standard laboratory tests, pulmonary function test, ECG, Echocardiography and anesthesiological assessment.Esophagogastroduodenoscopy (OGD) was performed to properly locate the tumor and to obtain biopsy.CT scan of chest and abdomen was performed for staging of the disease.Clinical and final pathological staging was done as per AJCC/ UICC 8th edition. 4Only clinical stages I-IVa patients with ECOG 0-1 were considered for surgery.

Treatment:
Treatment protocol varied from surgery alone to multimodality treatment.In initial years, patients (both SCC and adenocarcinoma) with resectable disease were subjected to upfront surgery.In adenocarcinoma of GEJ with >T2 or N+, adjuvant chemoradiation was used as per MacDonald's regimen.For locally advanced SCC, neoadjuvant chemoradiation (Cisplatin + 5-FU and RT  41.4-50.4Gy) or 2 cycles of chemotherapy alone (cisplatin + 5-FU) was used before surgery. 5For adenocarcinoma, MAGIC protocol was used. 6In recent years, for locally advanced SCC and adenocarcinoma, treatment was initiated as per CROSS protocol and FLOT protocol, respectively. 7,8rgical management was transthoracic (McKeown's/ left thoracotomy + neck/ left throcaolaparotomy + neck), and transhiatal.In three incision minimally invasive surgery (MIS), 3-4 ports were used in thorax and esophagus and nodes were excised en-block.During laparoscopy, five ports were used.In both MIS and open surgery, stomach was preferably used for reconstruction.Colon was used if stomach was not available.

Anastomotic technique:
We have used various techniques for gastroesophageal anastomosis in the neck.
1. Single layer: anastomosis was fashioned using single layered Gambee stiches with PDS 4-0 2. Classical 2 -layer: with PDS 4-0, the full thickness inner layer was stitched.The outer layer was Lambert stiches.3. Totally stapled: side to side stapled anastomosis with closure of anterior layers with stapler as well.4. Orringer's technique: side to side stapled anastomosis with anterior manual stiches was used.

Definition and management of AL:
An anastomotic leak was defined as a "full thickness gastrointestinal defect involving the esophagus, anastomosis, staple line, or conduit irrespective of presentation or method of identification" according to the Esophagectomy Complications Consensus Group definition. 9AL has been graded as grade I, II and III.Diagnosis was made principally on clinical grounds: tenderness at neck incision site with fullness and crepitus which revealed pus, saliva or air on bed-side exploration of the wound.CT/ oral contrast study was not used.Esophagogastroscopy was used only selectively if the leak did not show improvement after 2 weeks.AL was managed with parenteral antibiotics, daily dressing, nil per mouth and enteral feeding through jejunostomy tube.Oral diet was gradually started once there was complete healing of neck wound.In case of development of stricture, early esophageal dilatation was done using serial Savary -Gilliard bougies at 2-3 weeks interval till a satisfactory dilatation of anastomotic lumen > 13 mm was achieved. 10Follow-up: Patients were followed up every 4 months for first 2 years then every six months for next three years.

Statistical analysis:
A detailed analysis of anastomotic leak was done.Association of leak with anastomotic technique, preoperative chemoradiation and various surgical approaches was explored.Categorical variables were compared using the Chi square test, and continuous data were ana-lyzed using the Mann-Whitney U test.Survival was estimated using Kaplan-Meier survival curves and compared using the log-rank test.P<.05 was considered significant.SPSS version 26.0 was used for analysis.Median survival was 26 months and 34 months in patients with AL and without AL, respectively (p=0.03) (Fig. 1).

Discussion
In a metanalysis of 174 studies reporting 74226 patients showed a pooled overall leak rate of 11% (range: 0-49%). 11This study identified AL were associated with increased pulmonary and cardiac complications, in-hospital mortality, and significantly longer length of hospital stay.This review identified that underlying cardiovascular disease such as hypertension, ischaemic heart disease, cardiac arrhythmia, vascular diseases, and underlying aortic or coeliac axis artery calcification were associated with AL.There was lesser incidence of AL in intrathoracic anastomosis in comparison to cervical anastomosis (OR = 0.48(0.36-0.64),p<0.001).
In a recently published retrospective analysis of 119 patients, AL was observed in 21.8% (cervical AL = 22%, intrathoracic AL = 25%; p=0.98).Specific emphasis has been given to endoscopic treatment (stenting/ clipping) and surgical revision, but no differentiation between the management of intrathoracic and cervical leak has been made. 12st of the studies represent combined analysis of both cervical and intrathoracic AL with specific emphasis on intrathoracic AL.Literature is scarce for the management of cervical AL.We analyzed specifically cervical AL in a relatively high cohort (n=419).Incidence of AL was 16%, which is in acceptable range.Though most of the studies do not show increased rate of AL after neoadjuvant treatment, it is important to highlight that few studies have specifically looked at leak rates when comparing anastomoses performed within and outside the radiation field.Juloori et al. examined the location of anastomosis in relation to the radiation field. 13This study found that the rates of AL were significantly higher when the anastomosis was performed in the irradiated area (39% vs 2.6% for in-field vs out-of-field anastomosis, p<0.001).
A Japanese retrospective analysis of 686 esophageal cancer patients also showed that neoadjuvant chemoradiation was an independent predictor for postoperative complications with an anastomotic leak rate of 28% in patients who received neoadjuvant radiation compared to 16.5% in patients who had surgery alone (p<0.05). 14In our study as well, there was significantly higher AL after neoadjuvant chemoradiation (30% vs 12%, p <0.001).With this knowledge, when possible, surgeons should avoid constructing the anastomosis within the radiotherapy field.In our study, the cause of in-hospital mortality was multifactorial and in none of them, isolated AL was detected.Excluding the mortality, all the patients recovered from AL on conservative management, drainage, dressing and jejunostomy feeding.There was higher rate of anastomotic stricture requiring dilatation after AL (21% in AL group vs 0.9% in non-AL, p <0.001).AL rate was not associated with organ of conduit (stomach vs colon), anastomotic technique, surgical approach (TT vs THE), surgical technique (open vs MIS) and route of reconstruction (transmediastinl vs retrosternal).Most importantly, median OS was higher in non-AL than in AL group: 26 months vs 34 months, p=0.03).
The limitation of our study is its retrospective nature.The causes of AL were not addressed in our study, thus any specific recommendations could not be made to minimize AL.To best of our knowledge, this kind of study is first to be reported from Nepal in large number of patients.

Conclusion
Cervical AL after esophagectomy for cancer can be successfully managed conservatively with dressing, drainage and nil per mouth and jejunostomy feeding.Median OS is lesser after AL, hence further studies are needed to identify the modifiable risk factors in order to minimize AL.