Evaluation of 5-year results of laparoscopic transhiatal oesophagectomy as a single-centre experience.

INTRODUCTION: Minimal Invasive Surgery of oesophageal cancers is gaining popularity. We have published our Thoracoscopic Esophagectomy results. The present study focuses on our expertise of TransHiatal Esophagectomy.
MATERIALS AND METHODS: 287 patients underwent Esophagectomies for Cancer of Esophagus at Galaxy Care Laparoscopy Institute from January 2010 to December 2014 after thorough assesment. Out of these, 81 patients underwent laparoscopic trans hiatal esophagectomies. Their charts were reviewed retrospectively for intraoperative and postoperative results. The median follow up was 28 months.
RESULTS: Out of 81 patients,76 patients had R0 resection and 5 had R1. The average lymphnode yield was 20,average survival was 28months. 3 patients had local recurrence,18 had regional recurrence and 30 had distant recurrence. Average operating time was 140 min,mean blood loss was 80 ml. Average Post-operative ICU stay was 1 day and hospital stay 7 days.
CONCLUSION: Classic THE has limitations which can be overcome by the use of laparoscopic techniques. Laparoscopic approach for THE has better magnified vision facilitating better clearance under vision. Hence we recommend laparoscopic technique for THE to minimize morbidity and improve oncologic results.

INTRODUCTION

Oesophageal cancer is one of the common gastrointestinal malignancies.[1] Patients usually present in advanced stage and have low survival rates.[2] Although multimodality management is gaining popularity, surgery is the mainstay of therapy and offers the only chance for cure. Three surgical approaches to oesophagectomy have been described such as thoracoabdominal, transhiatal and transthoracic. However, all three are associated with high incidence of intraoperative and postoperative complications.[3] In a given patient, the approach is decided by tumour location, co-morbidities and surgeon's preference. Minimal access surgery has the potential to reduce the morbidity associated with the procedure and can improve the oncologic outcomes.

MATERIALS AND METHODS

A total of 287 patients who underwent laparoscopic/thoracoscopic transhiatal/transthoracic oesophagectomy at Galaxy Care Hospital, Pune, Maharashtra, India, between January 2010 and December 2014, were enrolled in the study. The charts of the patients who underwent oesophageal resections were reviewed. Out of the 287 patients, 81 patients underwent laparoscopic transhiatal oesophagectomy (THE). All patients were biopsy-proven cases of oesophageal cancer. All patients underwent thorough medical examination to assess fitness for surgery. All patients had oral water-soluble dye study; contrast-enhanced computed tomography (CT) scan of the chest and abdomen and upper gastrointestinal endoscopy. Bronchoscopy was done in patients with carcinoma of middle-third oesophagus. Positron emission tomography scan was done only if there was suspicious lesion on CT scan either in the chest or abdomen. An informed consent for surgery was obtained including the possibility of conversion to open or to do thoracotomy. The criteria for THE were

Tumour at the gastro-oesophageal junction

Middle 1/3rd tumours 3 cm below the carina

No nodes in the chest on CT scan

No distant metastasis

No extra-oesophageal spread

Good general condition.

Surgical technique

The patient was placed in the modified Lloyd–Davies position and the operating surgeon stood between the legs of the patient. Camera port (10 mm) was placed in midline at junction of middle and lower third from xiphisternum to umbilicus. Two working ports (5 mm) were placed at lateral border of rectus, just above the camera port. Fourth port was placed in the epigastrium (5 mm) for liver retraction.

Diagnostic laparoscopy was done first to rule out peritoneal/liver metastasis or significant gastric involvement. The procedure started with gastric mobilisation at greater curvature. Greater omentum was divided below the gastroepiploic arcade, lesser omentum was divided from the liver and the regional nodes were taken towards the specimen. The left gastric artery was clipped at the origin; the right gastric and right gastroepiploic arteries were preserved. Dissection was continued towards hiatus and then into the posterior mediastinum. Mediastinal dissection was done and all the perioesophageal lymph nodes were removed with the specimen. All the direct aortic branches were clipped.

Cervical dissection was done by making a transverse incision on the left side, just above the clavicle. Subplatysmal flaps were raised, two heads of sternocleidomastoid were separated and omohyoid was divided. Carotid sheath was retracted laterally. We avoided use of any metal retractor in the tracheo-oesophageal groove to avoid inadvertent injury to the left recurrent laryngeal nerve (RLN). oesophagus was mobilised, divided and distal end was tied to an umblical tape.

A minilaparotomy incision (4 cm incision) was made, and the stomach along with the oesophagus was delivered into the wound. A linear cutter was used to make stomach tube and also to detach the specimen. Gastric tube was then pulled up in the neck, and anastomosis was performed in four layers (inner with continuous 3-0 Vicryl and outer with 3-0 Silk). Feeding jejunostomy was done in all cases. Intercostal drains were inserted only if post-operative X-ray showed pneumothorax.

Perioperative care

Aggressive chest physiotherapy and bronchodilators were given to all patients to minimise chest infection. All patients were started on jejunostomy tube feeding from the 2nd post-operative day. Ryle's tube was removed on the 7th post-operative day, and oral feeding was started on the 8th post-operative day.

Follow-up period

All patients were called 7 days after discharge from the hospital, and the histopathology report was discussed in the tumour board meeting. In case no adjuvant therapy was decided, patients’ have advised three monthly follow-ups for 1 year and then six monthly for the subsequent years.

RESULTS

A total of 81 patients underwent laparoscopic THE from January 2010 to December 2014. Male: female ratio was 1.89 and median age was 58 years. The most common presenting complaint was dysphagia and tobacco addiction was present in 53 out of 81 patients (65%). About 56 patients (69%) had squamous cell carcinoma and 25 patients (31%) had adenocarcinoma. Clinicopathologic details are presented in Table 1. Most commonly found tumour stage was T3 in 38 patients (47%) and incidence of nodal metastasis increased with higher tumour stage. In T1 Stage, none had nodal metastasis. Out of 24 patients in T2 Stage, 11 patients (45.8%) had nodal metastasis. Out of 38 patients of Stage T3, 29 patients (76%) had nodal metastasis. All 11 patients of T4 Stage had nodal metastasis. About 38 patients had lower-third and 43 patients had middle-third tumour [Table 2].

Table 1

Analysis of data

Parameters	Results
Gender
Male	53
Female	28
Median Age	58 years (50-81yrs)
Chief Complaint
Dysphagia	53
Anorexia	21
Vomiting	7
Comorbidities
Hypertension	37/81
Diabetes	33/81
COPD/B. Asthma	21/81

Table 2

Pathology

	Number of patients (%)
Pathology
Squamous Cell Ca	56/81 (69%)
Adenoca	25/81 (31%)
Level of Growth
Mid	43/81 (53%)
Lower	38/81 (47%)
Tumour stage	Number of patients (%)	Nodal Metastasis
1	8 (9.8%)	0/8 (0%)
2	24 (29.6%)	11/24 (45.8%)
3	38 (47%)	29/38 (76.3%)
4	11 (13.5%)	11/11 (100%)

Surgical and clinical outcomes are shown in Table 3. Mean operative time was 140 min and average blood loss was 80 ml. All patients were kept in the Intensive Care Unit overnight after surgery. Five (6.1%) patients had post-operative chest infections. This prolonged the hospital stay. Three patients had anastomotic leak. This was clinical leak with discharge from the drain site in the neck. They were managed conservatively. Ten patients had post-operative stricture which was seen 3 months after surgery. These were managed by endoscopic dilatation. Two patients (2.46%) had RLN palsy, which recovered in 6 months. We had no intra-operative or 30-day mortality.

Table 3

Operative analysis

Parameter	Result
Duration of Surgery	140 min (110-245 min)
Mean Blood Loss	80 ml (50-250ml)
Post op ICU stay	1 Day (0.5-4 Days)
Post op Hospital Stay	7 Days (5-11 Days)
30-Day Mortality	0

A total of 76 patients had R0 resection, while five patients had R1 resection. Those patients with R1 resection had peritoneal involvement at the level of hiatus. The average lymph nodal yield was 20 (range: 13–31). The survival in months ranged from 9 to 43 months with average of 28 [Figure 1]. Three patients out of 81 patients had local recurrence (3.7%). About 18 patients (22%) had regional and 30 patients (37%) had distant recurrence (37%). The recurrences were pulmonary and liver [Table 4].

Figure 1

Survival curve of the present patient series

Table 4

Surgical outcome

Resection Status	Results
R0	74/81 (91.4%)
R1	5/81 (6.1%)
R2	0/81 (0%)
Lymph Node Yield	20 (13-31)
Survival	28 months (9-43)
Patterns of Failure	Results
Local	3/81 (3.7%)
Regional	18/81 (22%)
Distant	30/81 (37%)
Unknown	30/81 (37%)

DISCUSSION

According to the projections of cancer cases in India, during 2010–2030, oesophageal cancer is the second most common gastrointestinal malignancy. There will be approximately 42,000 cases by the year 2020.[1] The prognosis remains poor, with 5 year survival rate being only 19%.[4] It is more common in males and the incidence rises with age. Tobacco is an important aetiologic factor. Squamous cell carcinoma is more common than adenocarcinoma. The incidence of adenocarcinoma was 31%, while that of squamous cell carcinoma was 69%. The incidence of adenocarcinoma was higher than in other series from India.[5] This may be due to more oesophageal–gastric junction tumours in our series.

Multimodality management for oesophageal cancer is being used more and more frequently, but surgery still remains the mainstay of therapy. Out of three surgical approaches to oesophagectomy (THE, transthoracic oesophagectomy and Ivor–Lewis Surgery), THE is the only approach which avoids a thoracotomy. That is why many surgeons prefer this approach in patients of lower oesophageal cancer with poor general condition (because of associated comorbidities). As thoracotomy is not required, pulmonary complications can be avoided and recovery is easier.

Open THE has higher incidence of morbidity with more chances of intra-operative complications. This may be due to poor visualisation of structures in the thorax. Orringer et al. have reported 12 cases of massive intra-operative bleeding (including four cases resulting in death) that occurred during transhiatal mobilisation of oesophagus in a series of 2000 patients. They have also reported eight cases (<1% incidence) of tracheobronchial tears, 91 cases (4.5% incidence) of RLN injury (as manifested in post-operative period with hoarseness), chylothorax in 25 (1%), clinically significant pneumonia or atelectasis in 2%, anastomotic leaks of 12% and an overall mortality of 3%.[6] We did not have single tracheobronchial injury or intraoperative haemorrhage. The decrease rate of complications may be due to magnification and better visualisation during the procedure.

Minimally invasive surgery is gaining popularity as incisions are smaller and recovery is smoother. Magnification provided by the laparoscope gives a better view of anatomy and decrease intra-operative complications [Tables 5 and 6].

Table 5

Complications

Complications	Number of patients (%)	Grade of Complications (Clavien-Dindo Classification)
Bleeding	0	-
Intra-operative Hypotension	0	-
Chest Infection	5 (6.1%)	Grade II
Anastomotic Leak	3 (3.7%)	Grade II
Stricture	10 (13%)	Grade IIIb
RLN palsy	2 (2.46%)	Grade II
Chyle Leak	0	-
Tracheo-Bronchial Injury	0	-

Table 6

Comparative studies

Parameter	Current Series	Cash[10]	Bann[11]	Avital[12]
Number of Cases	81	33	20	22
Duration of Surgery	140 min (110-245 min)	274 min	415	380 min
Mean Blood Loss	80 ml (50-250ml)	No Transfusions	300	220 ml
Post op ICU stay	1 Day (0.5-4 Days)	-	-	-
Post op Hospital Stay	7 Days (5-11 Days)	10 Days	17 Days	8 (5-46)
Complications
Conversion	0	2 (6.1%)	2 (10%)	1
Chest Infection	5 (6.1%)	2 (6%)	4 (20%)	2
Anastomotic Leak	3 (3.7%)	3 (9%)	2 (10%)	1
Stricture	10 (13%)	-	-	1
RLN palsy	2 (2.46%)	-	4 (20%)	1
Chyle Leak	0	-	-	-
Tracheo-Bronchial Injury	0	-	-	-
30-Day Mortality	0	0	0	1
R + resection	5/81 (6.1%)	-	-	-
Lymph Node Yield	20 (13-31)	24	5	14.3
Median Survival	28 months	26 months	-	61% at 30 months

Laparoscopic THE was first described by DePaula et al. in 1995, and since then, many others have done similar publications. DePaula et al. have reported minimal morbidity in terms of blood loss, hoarseness and chest complications and no mortality in a small series of 12 patients.[7]

In our series of 81 patients of laparoscopic, THE average duration of surgery was 140 min (110–245 min), average blood loss was 80 ml (50–250 ml) and no patient was converted to open surgery. All surgeries were done by the single surgeon; this has not only minimised the operative time and blood loss but has also decreased the conversion rates to zero.

We do oesophagogastric anastomosis in four layers (outer layer with 3-0 silk and inner layer with 3-0 vicryl, both as applied in a continuous manner), this has resulted in zero leak rates, but ten cases developed anastomotic stricture. All were managed by endoscopic dilatation and did well on that.

No pyloromyotomy/plasty was done in our patients. We compensate by doing good tubular gastric conduit. Bemelman et al. reported their experience with making of tubulised stomach without pyloroplasty and have reported the incidence of delayed gastric emptying to be around 3%. They claim that the most important factor in decreasing the incidence of delayed gastric emptying is making of a good tubulised stomach only and not pyloroplasty. Pyloroplasty not only increases the operative time but also poses the risk of leak.[8] We also hold a similar opinion.[9]

We had two cases (2.46%) of RLN palsy which manifested as hoarseness in the post-operative period. The incidence of RLN injury during oesophagectomy can be minimised by doing dissection of supra-azygos segment carefully and avoiding any metallic retractor in trachea–oesophageal groove. We did not have any case of chyle leak in this series of 81 patients. We did not have any tracheobronchial injury. The present STUDY did not have ant 30 day mortality.

The average hospital stay was 7 days. In five patients, due to chest infection, the hospitalisation was 9–11 days. Aggressive chest physiotherapy and spirometry before and after surgery has led to low incidence of post-operative chest infection.

We had no conversion to open surgery. This may be due to correct patient selection. This is also supported by the fact that 76 patients (93.8%) had R0 resection, and only five patients (6.2%) had R1 resection. We did not have any R2 resection. Average lymph nodes retrieved in this series were 20 (13–31).

This is a high-volume centre performing all types of minimal invasive oncosurgeries. The team is trained and the steps are standardised. This had led to decrease in operative time and mortality.

CONCLUSION

Classic THE has received some criticism because of blind nature of surgery. These can be removed by use of laparoscopic techniques, and the incidence of tracheobronchial injury, intra-operative haemorrhage and intra-operative hypotension can be brought down to almost zero and 3.7%, respectively. From oncologic point of view, laparoscopic THE is better than open procedure as peri-oesophageal lymph nodes are taken towards the specimen under vision. Also with help of magnification, resection margins are also more likely to be negative. Hence, we recommend the use of laparoscopic technique in THE to minimise morbidity and to improve oncologic results.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.[12]