TRANSHIATAL ESOPHAGECTOMIES - A TWO DECADE EXPERIENCE FROM A HIGH VOLUME CENTRE

Despite improvements in preoperative evaluation, anesthetic and operative techniques, and postoperative care, esophageal resection and reconstruction remain formidable operations in patients whose nutritional and pulmonary status have been compromised by cancer of esophagus. The necessity for a combined thoracoabdominal procedure in a debilitated patient and the disastrous results of disruption of an intrathoracic esophageal anastomosis are major contributing factors to the morbidity and mortality rates of esophageal replacement.

We have employed the technique of blunt transhiatal esophagectomy through the diaphragmatic hiatus and this report reviews our experience with esophagectomy and esophageal substitution with gastric tube without thoracotomy.

The efficacy of THE as a palliative procedure, as well as its ability to achieve apparent cure in patients with esophageal carcinoma, is emphasized.

The study has been conducted in the department of Cardiovascular and Thoracic Surgery, Sher-i-Kashmir Institute of Medical Sciences, Srinagar, and Kashmir. It was both retrospective  including patients operated from  year 1998 to   2011 as well as  prospective study including  patients operated from 2012 to 2021 by the same surgeon. The eligible patients had histologically confirmed carcinoma of the middle and distal esophagus or carcinoma of the gastroesophageal junction involving the distal esophagus, had no evidence of distant metastases, and did not have unresectable local disease. Retrospective data was collected from the Medical

Records Department of our tertiary care hospital.

Exclusion criteria were previous or coexisting cancer, previous gastric or esophageal surgery, recurrent laryngeal nerve palsy, and extension of the tumor into trachea-bronchial tree and other conditions that made it impossible for the surgeon to construct a gastric tube.

The pre-operative diagnostic workup consisted of endoscopy with biopsy and histological examination, ultrasonography of the abdomen, chest radiography, indirect laryngoscopy and bronchoscopy if tumor invasion of the upper airway was suspected. Computed tomography abdomen and chest was performed in all the patients. Operative risk analysis included standard blood examination, electrocardiography, pulmonary function tests and cardiopulmonary exercise tests (in higher risk patients). Surgery was offered to medically fit patients.  After obtaining written informed consent, patients were planned for surgery. All patients were operated by one particular consultant surgeon of our department with experience in esophageal surgery.

Operative technique:

 After induction of anesthesia nasogastric tube is inserted and stomach is decompressed. An initial laparotomy is performed through a midline incision to confirm tumour resectability and the stomach is assessed immediately for its suitability as an esophageal substitute. The peritoneum overlying the esophageal hiatus is incised and hiatus is opened by dividing the crus of the diaphragm. Finger dissection of lower esophagus is done after opening the hiatus to assess the resectibility of the tumour. The greater omentum along the greater curvature of stomach is retracted to the left and a clear space between stomach wall and the omentum is developed. Left gastroepiploic and high short gastric vessels are identified and ligated well away from stomach wall to avoid ischemic injury. Dissection is then carried inferiorly along the greater curvature of the stomach towards pylorus after ligating major vessels except right gastroepiploic artery which remains the main supply of mobilized stomach. When the entire greater curvature of stomach is freed from its adjacent omentum, adhesions between the stomach and retroperitoneum are carefully divided to ensure complete gastric mobility. Dissection is then started along the lesser curvature of stomach, the left gastric vein is identified, mobilized and ligated with 3-0 silk ties. The adjacent left gastric artery is identified at its origin from celiac axis, is doubly ligated towards celiac artery, clamped and divided reflecting adjacent lymph nodes to the left with the stomach. Utmost care was taken to avoid any trauma to vascular pedicle. After abdominal exploration and gastric mobilization had been performed, proximal portion of stomach is resected and 3-4 cm wide gastric conduit is fashioned in two layers, inner layer closed with continuous silk 3-0 sutures, and outer layer with continuous Polypropylene 4-0 sutures. The line of resection can be modified as per the extent of growth. In case of mid-thoracic esophageal growth minimum of stomach is sacrificed, however when growth is situated in distal esophagus and gastroesophageal junction, upper partial gastrectomy is mandatory with at least 6-10 cms distance from the tumor. Now, the esophageal hiatus is widened further and retractors are positioned to facilitate exposure of the intrathoracic esophagus upto the level of the carina. This enables enbloc resection of the esophagus and para-esophageal tissue including the crura and pleura (if indicated) under direct vision. Standard lymph node dissection involves removal of lymph nodes in the lower mediastinum, around the esophagogastric junction and along the lesser curvature of the stomach. A radical lymph node dissection is performed at the origins of the left gastric and common hepatic arteries; lymph nodes at the celiac axis were included when enlarged and resectable. As the esophagus was progressively mobilized from lower mediastinum attention is paid to the pleura on both sides. Invariably because of pleura being very thin, it gets breached and hence bilateral chest drains are warranted. During its mobilization  the esophagus is  grasped and “rocked’’ from side to side to verify that it is not fixed to adjacent spine, prevertebral fascia, or aorta. In this fashion mobilization of distal esophagus is achieved up to level of carina. Both main bronchi can be safeguarded by feeling their texture with the finger and thereby avoiding any bronchial/tracheal tear while dissecting the esophagus in the mediastinum. What actually is important is the patience of the surgeon during the entire procedure, especially once growth  or lymph nodes are adherent to vital structures. Azygous vein  on right side  or aorta on the left side can be torn by clumsy or  inexperienced hands and can cause major blood loss.

                Beginning at a point 1.5-2 cm on the gastic side of pylorus and crossing across the pylorus  towards duodenum, pyloromyotomy  is routinely  performed. A feeding jejunostomy is  done 15-25cm away from ligament of Treitz in all patients.With gastric mobilization, dissection of esophagus from lower mediastinum, pyloromyotomy and feeding jejunostomy, abdominal part of procedure ends and attention was turned to cervical part of the surgery. During cervical phase of procedure, 5-7cm oblique incision was  made parallel to anterior border of sternocleidomastoid muscle on left side and extended 2-3cm superior to cricoid cartilage, dissection was carried deep into tracheoesophageal groove and circumferential mobilization of esophagus is completed and  portion of esophagus along with tumour is delivered into neck wound taking care of recurrent laryngeal nerve which was routinely examined. All possible efforts were taken to avoid any demage to it.  Fashioned stomach tube was brought through esophageal hiatus into posterior mediastinum after applying two stay sutures on two sides of gastric tube and corresponding sides of lower esophageal end to avoid any subsequent torsion. An end- to- side hand sewn double  layer esophagogastric anastomosis is made in the neck using 3-0 silk posteriorly,  3-0 vichry for inner layer by  continuous technique .  Pedicled omentum from greater curvature of stomach , designed as per the length of the tube, is routinely sutured around the anastomosis using continuous Polypropylene 4-0 sutures around the esophageogastric anastomosis for completion of the anterior layer (2^nd layer (outer ) . Two anchoring sutures are taken between prevertebral fascia and gastric tube to keep the anastomosis in place. Both pleural cavities are  routinely drained by intercostal tubes  to avoid any pneumothorax. Soft rubber drain  is kept just  lateral to the anastomosis  but never inside the mediastinum and incision was closed back in layers. A drain inserted deep inside the mediastinum leads to mediastinal / subcutaneous emphysema. Another rubber drain was inserted near the hiatus and brought through abdominal wall. Abdominal wound was closed in 3 layers at the end of the main procedure..After extubation, patient was transferred to a high dependency ward for monitoring during initial 24-48 hours and then shifted to general ward. Nutrition was maintained by jejenostomy feed. However, many patients required total parenteral nutrition. Dye study was performed on 10^th postoperative day to look for any leak and nasogastric tube was removed on 11^th postoperative day. Liquid diet was started if dye study was reported normal and patients were discharged on 13^th or 14^th post operative day depending upon progress on jejenostomy tube for feeding.

Intraoperative and early postoperative complications occurring in the patients undergoing procedure were noted. A complete proforma containing post-operative events occurring in each patient was filled. After discharging the patients, they were asked to follow the outpatient department of our institute. Jejenostomy tube was removed after 3 weeks if there was no indication of delayed leak from cervical anastomosis. The histopathological findings of resected specimen were analysed. Patients were planned for adjuvant therapy if indicated. All patients were followed at intervals of two weeks during first three months, three to four months during the first two years and every six months for three more years. The follow-up data was also obtained by telephone from the patient or his/her family. The patients were followed for any procedure related complication or recurrence during his/her post operative visit. Recurrence of disease was suspected on clinical grounds. However, whenever a relapse was suspected, radiologic, endoscopic and histologic confirmation was sought.

The present study was carried out in the department of Cardiovascular and Thoracic Surgery (CVTS), Sher-i-Kashmir Institute of Medical Sciences, Srinagar, J&K. The study was designed as prospective and retrospective study and included 315 patients operated for carcinoma esophagus, studied retrospectively from November  1998 to December 2011 and 490 prospectively from January 2012 to October 2021. The study revealed following results after analysis.

The mean age of the study population was 55 years with standard deviation of 5 years. 51.94% of the patients were falling in the age group of 51-60 years as shown in Table – 1, Figure – 1.

              Out of 805 patients, 539 were males and 266 were females as given in  Table –2.

Table - 2: Sex distribution of patients

Most of the operated patients were from rural areas (80%) as shown in Table – 3, Figure – 3.

Table – 3: Dwelling status of patients

Dysphagia was the most common presenting complaint, followed by weight loss. Some patients presented with multiple symptoms as given in Table – 4.

Table – 4: Clinical presentation of patients.

Mid esophagus was most common site of tumor location followed by distal esophagus. There were multiple lesions in some patients as shown in Table – 5, Figure – 5.

Table – 5: Distribution of patients in terms of location of tumor.

Most of operated patients had TNM stage 3 & TNM stage 2 diseases at the time of surgery due to late presentation which is very common in this part of world. (Table 6, Fig 6)

Table 6 - TNM Staging of Operated Patients.

Mean operating time from start of skin incision to closure of wound was 160+/-10 minutes and average blood loss was 500+/-20 ml.

Table - 7:Mean Operating Time and Mean Blood Loss

Post operative histological study of resected specimen revealed that squamous cell carcinoma was most common tumour followed by adenocarcinoma. (Table - 8)

Table - 8: Tumor Histology

Neo-adjuvant chemotherapy was used in 120 patients out of 480 patients diagnosed having squamous cell carcinoma. Three cycles of Cisplatin plus 5-Florouracil based chemotherapy was used in these patients. Out of 277 patients of adenocarcinoma 93 patients received neo-adjuvant chemotherapy. Three cycles of Epirubicin plus Cisplatin plus 5-Florouracil based chemotherapy was used in these patients. These comparisons are shown in Table – 9.

Table - 9: Neo-adjuvant Chemotherapy

All patients were sent for post-operative chemotherapy and/or radiotherapy as dictated by the individualized parameters and oncologists opinion.

Major complications which occurred in our study included respiratory tract complications, anastomotic leaks, recurrent laryngeal nerve palsy, significant hemorrhage, per-operative &postoperative arrhythmias, postoperative strictures, cardiovascular complications & wound infections. (Table - 10, Figure - 8) 

Table- 10: Complications.

Table - 11: Estimated 3 year Survival

Table - 12:Estimated 5 year Survival after Transhiatal Esophagectomy

Overall survival was defined as the time interval from the date of operation until the date of death or most recent follow up. Disease free survival was defined as the time interval from the date of operation until the date of disease recurrence or most recent follow up. Survival curves were calculated according to the Kaplan-Meier method. 

Despite oncological advances, surgical resection is the only treatment that has repeatedly shown to prolong survival in esophageal cancers. Esophagogastric resection can be performed in many ways and all of them pose considerable challenges to both the surgeon and patient. The technical approaches vary in one or more of following: type of incision, extent of resection, conduit for reconstruction and type of anastomosis. The concept of resection of the esophagus without thoracotomy had been proposed initially in 1913 by the German anatomist, Denk. In1933, the British surgeon, Turner, carried out the first successful Transhiatal blunt esophagectomy for carcinoma. THE is often advocated as the preferred surgical approach in patients with benign disease or early tumors or those patients with more advanced disease who would not tolerate a thoracotomy. The advantages of THE are:

Avoidance of thoracotomy and therefore less postoperative respiratory distress, and

Cervical esophagogastric anastomosis as the consequences and complications of this anastomotic leak are much less and easy to manage.

 This approach has been criticized because of the lack of a formal two field lymphadenectomy , failure to completely resect tumor under direct vision and because of concern regarding inadequate hemostasis in mobilizing the esophagus  from the mediastinum “blindly’’. Over the past 30 years, the technique of THE has been progressively redefined and is regarded as approach of choice in patients requiring esophageal resection and reconstruction for both benign and malignant conditions. The technical refinements in mobilizing and handling stomach, dissecting esophagus transhiatally, preparing gastric conduit, positioning the stomach into posterior mediastinum and constructing esophagogastric anastomosis have definitely improved the surgical outcome of this procedure over a period of time. Most often patients requiring esophageal resection and reconstruction suffer from obstruction which not only has nutritional consequences from impaired caloric intake but is also associated with pulmonary sepsis due to impaired swallowing. Transhiatal esophagectomy has been the favoured operative approach in our institute since 1998 for managing both carcinoma of the esophagus below the level of carina and tumors of gastroesophageal junction. It has also been utilized for benign lower esophageal diseases, high grade dysplasia and oesophageal perforations which can’t be repaired primarily.   

Oesophagial cancer has been one of the leading cancers prevalent in our area[2].This study evaluates our experience and outcomes with transhiatal esophagectomy. Between 1998 and 2021, a total of 805 patients, underwent transhiatal esophagectomy for carcinoma esophagus. Dysphagia, as a symptom was present in almost all the patients (96.10%). Significant weight loss was present in less than half number of patients (42.85%). Dysphagia and weight loss were the presenting features in many studies. Both were described as usual symptoms in one study by Orringer [4]. Most of the operated patients were from rural areas (79.87%). More than half patients were between 51 and 60 years of age (51.94%), the next common group being 61 to 70 years of age (31.81%).

Mid oesophagus was most common site involved by tumor (46.75%), followed by distal oesophagus 71 (35.71%). Most of the operated patients had TNM stage III and II disease. The patients having stage I disease were far less than that described in literature because of less health care services available and lack of routine screening for oesophageal carcinoma in our part of the world. Histopathological examination of specimens revealed squamous cell carcinoma in most cases (59.74%) followed by adenocarcinoma (34.41%). The tumor behavior and long-term prognosis of adenocarcinoma and squamous cell carcinoma are considered to be comparable. Neoadjuvant chemotherapy in selected patients resulted in  higher resectability rate and R0-resection[20].

Esophageal reconstruction was done by gastric tube and esophagogastric anastomosis was done in all patients in our study although colon and jejunum have been used as esophageal substitutes in many studies and functional results of various substitutes have been compared for procedure related complications such as dysphagia, regurgitation and diarrhea. In our study patients undergoing colonic or jujenal interposition were excluded.

In our study mean operating time from skin incision to closure of wound was 160+/_20 minutes which is almost same for other studies such as study conducted by Gupta [19] and Goldminc et al [8] in which the average operating time was 3 to 4 hours . In our study, average blood loss was 500±20 ml, which has been decreasing over the years as our experience with THE has increased, much of esophageal mobilization is performed under direct vision through the retracted diaphragmatic hiatus using long right angle clamps applied to para-esophageal tissues which allows more mobilization of the esophagus and hemostasis under direct vision. This amount is much less than the average blood loss of 1350ml reported by Mark B Orringer et al [15] but significantly more than average blood loss of 370 ml as reported by same author in a study conducted in year 2007[1]. This again signifies the fact that as the experience with THE increases, the amount of blood loss significantly decreases and the safety of the procedure increases[6].

The anastomotic leakage in our study was 7.14% which is significantly lower as compared to a study conducted by Jauch et al [9] which reported anastomotic leak of 39%.  Another study by Pommier et al [10] showed anastomotic leak of 13% for THE resection. Several technical aspects during the procedure help to minimize the incidence of anastomotic leak. Avoiding vascular trauma is the key to minimizing cervical esophagogastric anastomotic leak. The dictum “pink in abdomen (after complete gastric mobilization) and pink in the neck (after transposing the stomach into the cervical wound) is a current guiding principle. No traction drains, sutures or bags are affixed to the mobilized stomach, which is manually manipulated upward through the hiatus and mediastinum. Use of  pedicled omental flap to reinforce the esophagogastric anastomosis has certainly played a vital role in minimizing the leaks. Use of 4-0 prolene sutures for construction of gastric tube and esophagogastric anastomosis has also decreased the leaks in the anastomosis . In our study only one patient required exploration of the cervical wound and repair of esophago-cutaneous fistula. All other patients had minor leaks and were managed conservatively. A side-to-side stapled cervical esophagogastric anastomosis reduces the incidence of anastomotic leak and subsequent stricture formation. About 9% of patients undergoing THE had significant hemorrhage and needed post op transfusions. This can be explained by the fact that blood loss in THE is mostly hidden. Vocal cord paralysis was seen in 3.9% patients who underwent THE. In literature, high incidence of vocal cord paralysis was mentioned after transhiatal esophagectomy, indicating that the recurrent nerve is mainly at risk during the cervical dissection and the construction of the esophagogastric anastomosis. Tilanus et al [11] also showed more recurrent laryngeal nerve injuries after THE. Incidence of vocal cord palsy in our study was significantly lower than this study and comparable to other studies . One study by  Orringer [1] reported incidence of only 2% and another study by Yannopoulos [12] showed incidence of 1.33%. In our study the perioperative cardiac complications like arrythmias were seen in 6% patients and wound infection occurred in 4.5% of patients. Chylothorax was seen in 2.59 % of patients undergoing THE. The in-hospital mortality was 3.85% which is comparable to other major studies [1]. Out of 31 deaths, 9 patients died of respiratory insufficiency. Other causes included myocardial infarction, pulmonary embolism, haemorrhage, sepsis.1 patient died of fecal peritonitis as a result of feeding jujenostomy leak. Mortality rates following oesophagectomy for carcinoma oesophagus in literature vary widely (1% to 28%) and decrease with increasing experience [17]. Rindani and coworkers[13], who reviewed the literature from 1986 to 1996 found in hospital mortality rates of 6.3% for transhiatal resections. The in-hospital mortality according to Horstmann O et al [14] was 15% in the transhiatal procedure. The mean postoperative hospital stay was 12± 3days) similar to study by Andrew et al [3] and Kothari et al [7]. Ninh et al [18] in their study showed mean post operative hospital stay of 22.3 ± 16.1days for THE. Around 50 (6.49%) patients developed post operative anastomotic strictures requiring serial endoscopic dilatations. It was much lesser than 36.4% reported by Bruns et al [5].

Disease free and overall survival was calculated by Kaplan Meier curves. The Estimated three-year disease free survival rate was 30.27% whereas three-year overall survival rate was 40.24%. Also estimated five-year disease free survival rate was 20.60% whereas five-year overall survival rate was 28.42%. Rindani[13] and co-workers found approximately the same 5-year survival rates in their review (24%) for THE. A study conducted by Hulscher [16] showed an overall 5-year survival rate of 29%. Some western world studies have shown overall 5-year survival between 40-50% as most of the patients have early stage disease at the time of surgery whereas most of our study patients had advanced disease at the time of surgery.

     The authors of this study have established beyond doubt the efficacy and practicality of the transhiatal oesophagectomy. The centre being one of the biggest centres in the northern part of the country has had enough volume of the cases, and has concluded from its vast experience with the procedure that expertise and experience make THE a relatively safe and effective procedure to tackle the menacing cancer of oesoophagus .

1.     Two thousand transhiatal esophagectomies: changing trends, lessons learned. Orringer MB, Marshall B, Chang AC, Lee J, Pickens A, Lau CL. Ann Surg. 2007;246:363–72.discussion 72‐4

2.     High incidence of oesophageal and gastric cancer in Kashmir in a population with special personal and dietary habits.M S Khuroo 1, S A Zargar, R Mahajan, M A Banday Gut, 1992 Jan;33(1):11-5.

3.     Transhiatal esophagectomy in a high volume institution. Andrew R Davies, Matthew J Forshaw, Aadil A Khan, Alia S Noorani, Vanash M Patel, Dirk C Strauss and Robert C Mason. World Journal of Surgical Oncology 2008, 6:88

4.     Transhiatal esophagectomy without thoracotomy for carcinoma of the thoracic esophagus.Orringer MB.Ann Surg. 1984 Sep;200(3):282-8.

5.     Cervical anastomotic stenosis after gastric tube reconstruction in esophageal carcinoma. Evaluation of a patient sample 1989-1995. Bruns CJ, Gawenda M, Wolfgarten B, Walter M.Langenbecks Arch Chir. 1997;382(3):145-8.

6.     Transhiatal esophagectomy.Calotă F, Mănescu P, Georgescu I, Surlin V, Mărgăritescu D, Rotaru A.Chirurgia (Bucur). 2003 Sep-Oct;98(5):431-6

7.     Comparison of esophagectomy with and without thoracotomy in a low-resource tertiary care center in a developing country K C Kothari 1, C Krishnan Nair, P S George, M H Patel, R C Gatti, G C Gurjar  Dis Esophagus 2011 Nov;24(8):583-9.

8.     Oesophagectomy by a transhiatal approach or thoracotomy: a prospective randomized trial.M Goldminc 1, G Maddern, E Le Prise, B Meunier, J P Campion, B Launois.Br J of Surg.1993 Mar;80(3):367-70.

9.     Esophageal carcinoma: prognostic features and comparison between blunt transhiatal dissection and transthoracic resection. Jauch KW, Bacha EA, Denecke H et al (1992) Eur J Surg Oncol 18:553–562

10.   Pommier RF, Vetto JT, Ferris BL et al (1998) Relationships between operative approaches and outcomes in esophageal cancer. Am J Surg 175:422–4250.

11.   Tilanus HW, Hop WC, Langenhorst BL et al (1993) Esophagectomy with or without thoracotomy. is there any difference? J Thorac Cardiovasc Surg 105:898–903

12.   Esophagectomy without thoracotomy: 25 years of experience over 750 patients. Panagiotis Yannopoulos 1, Panagiotis Theodoridis, Konstantinos Manes2009 Jul;394(4):611-6. Langenbecks Arch Surg.

13.   Transhiatal versus Ivor-Lewis oesophagectomy: is there a difference? Rindani R, Martin CJ, Cox MR ,Aust N Z J Surg. 1999;69:187–194.

14.   Horstmann O, Verreet PR, Becker H et al (1995) Transhiatal esophagectomy compared with transthoracic resection and systematic lymphadenectomy for the treatment of esophageal cancer. Eur J Surg 161:557–567

15.   Esophagectomy without thoracotomy .Orringer MB, Sloan H.J Thorac Cardiovasc Surg. 1978 Nov;76(5):643-54

16.   Transthoracic versus transhiatal resection for carcinoma of the esophagus: a meta-analysis.J B Hulscher 1, J G Tijssen, H Obertop, J J van Lanschot.Ann Thorac Surg,2001 Jul;72(1):306-13.

17.   Effect of surgical experience on results of esophagectomy for esophageal carcinoma J D Miller 1, M K Jain, C J de Gara, D Morgan, J D Urschel .J Surg Oncol. 1997 May;65(1):20-1.

18.   Ninh T, David M, Bruce M et al (2000) Comparison of minimally invasive esophagectomy with transthoracic and transhiatal esophagectomy. Arch Surg 135:920–925

19.   Transhiatal versus Transthoracic Esophagectomy for Distal Esophageal Cancer Narendar Mohan Gupta,   Asian Cardiovascular and Thoracic Annals  Dec 1, 2000 Volume: 8 issue: 4, page(s): 347-352.

20.   Role of Preoperative Chemotherapy in Squamous Cell Carcinoma of Esophagus in Kashmir, a Cancer Belt - A Pilot Study .Ghulam Nabi Lone , Ajaz Aziz Sheikh, Zahoor Ahmad Sheikh