Laparoscopic and open transhiatal oesophagectomy for corrosive stricture of the oesophagus: An experience.

BACKGROUND: Oesophagectomy for corrosive stricture of the oesophagus (CSE) is rarely performed due to high risk of iatrogenic complications. The aims of this study were to review our experience of transhiatal oesophagectomy (THE) in patients with CSE as well as to compare results of open and laparoscopic methods.
MATERIALS AND METHODS: This is a retrospective analysis of prospectively maintained data of patients with CSE who underwent open transhiatal oesophagectomy (OTE) or laparoscopic-assisted transhiatal oesophagectomy (LATE) by a single surgical team from 2012 to 2016. All study patients had either failed endoscopic dilatation or had a long stricture which was not amenable to endoscopic dilatation.
RESULTS: Totally, 35 patients were included in the study, of which 19 (54.3%) were female. OTE was performed in 20 (57%) patients, and LATE was performed in 15 (43%) patients. Gastric and colonic conduits were used in 23 (65.7%) and 10 (34.3%) patients, respectively. Demographic and clinical parameters were comparable between LATE and OTE groups (P > 0.05). Median intra-operative blood loss, post-operative requirement of analgesic and hospital stay were lower in LATE group (P ≤ 0.05). There was no hospital mortality (30 days), but three patients (8.6%) died during a median follow-up of 36 months.
CONCLUSION: THE is a safe procedure for patients with CSE, and LATE may be an alternative approach in selected patients.

INTRODUCTION

Corrosive stricture of the oesophagus (CSE) is an important cause of dysphagia in adults of the Indian subcontinent. CSE often requires substitution of oesophagus by stomach or colon conduit if stricture is not amenable to endoscopic dilatation. Retrosternal oesophageal bypass procedure is commonly performed surgical procedure for CSE because oesophagectomy is considered to be a high-risk procedure.[12] However, leaving behind the scarred oesophagus is associated with lifelong risk of development of mucocele and malignancy in a scarred oesophagus.[345] Gupta and Gupta have established feasibility and safety of open transhiatal oesophagectomy (OTE) in a series of 51 patients with CSE.[6] We hereby report our experience of transhiatal oesophagectomy (THE) in patients with CSE along with comparison of open and laparoscopic approaches.

MATERIALS AND METHODS

This is a retrospective analysis of 35 patients with CSE who underwent either OTE or laparoscopy-assisted transhiatal oesophagectomy (LATE) by a single surgeon (the corresponding author) at a tertiary care centre in North India, from January 2012 to January 2016. Patients with a history of exploratory laparotomy, feeding jejunostomy (FJ) and/or scarred stomach were considered for OTE; otherwise, they were considered for LATE. Written informed consent was obtained from all the patients. As per prevailing guidelines, approval from the Ethics Committee was not necessary for this study. Pre-operative assessment included oral contrast study, upper gastrointestinal endoscopic study and blood investigations [Figure 1]. All the study patients either had failure of endoscopic dilatation or had a long stricture which was not amenable to endoscopic treatment. Endoscopic failure was defined as inability to maintain a luminal diameter for 4 weeks after the attainment of target diameter (14 mm) or inability to dilate up to 14 mm over five sessions at 2 weeks interval.[7]

Figure 1

Barium swallow study depicting corrosive stricture of oesophagus

Open transhiatal oesophagectomy

Midline abdominal incision extending from epigastrium up to 3–5 cm below the umbilicus was used for the laparotomy. Gastric conduit was utilised if condition of stomach was satisfactory otherwise colonic conduit was used. Ultrasonic shear (Ethicon Endo-Surgery, LLC, Guaynabo, Puerto Rico, USA) was used for dissection and division of the vessels up to 3 mm in size; vessels >3 mm size were ligated before division. Gastric conduit was based on the right gastroepiploic artery and GIA stapler 75/100 mm (Ethicon, LLC, Guaynabo, Puerto Rico, USA) was utilised for the division of the stomach. Colonic conduit was based on ascending branch of the left colic artery and GIA stapler 75 mm (Ethicon, LLC, Guaynabo, Puerto Rico, USA) was used for colonic transection. The lower part of the oesophagus was mobilised under vision after widening of the oesophageal hiatus of the diaphragm. An oblique cervical incision was used to mobilise cervical and upper thoracic oesophagus. Mobilisation of the mid-thoracic oesophagus was completed with both transhiatal and transcervical route in perioesophageal plane to avoid iatrogenic injury. The cervical oesophagus was divided 5–6 distal to cricoesophageal sphincter and specimen was pulled out through abdominal wound. Gastric or colonic conduit was placed in orthotopic position. Cervical oesophagogastric anastomosis (CEGA) or cervical oesophagocolic anastomosis was fashioned with interrupted 3-0 silk (Ethicon, India) sutures. In patients with colonic conduit, both cologastric and colocolic anastomoses were performed with the use of interrupted 3-0 polyglactin (Vicryl) sutures (Ethicon, India). Either pre-existing FJ was retained or a new FJ was fashioned in all the patients. Both pleural cavities were drained; a drain was also kept near oesophageal hiatus and brought out through left flank. Cervical wound was closed in layer after placement of an 18F suction drain (Romsons, India).

Laparoscopic-assisted transhiatal oesophagectomy

Patients were operated in French position and total four access ports (Ethicon Endo-Surgery, USA) were utilised to complete laparoscopic part of the procedure; an infraumbilical port (11 mm, for camera), an epigastric port (11 mm) and two pararectal ports (12 mm [right], 5 mm [left]) on either side of the umbilicus [Figure 2a]. Gastrocolic and gastrohepatic ligaments were divided along with preservation of whole right gastroepiploic artery with its arcade along the greater curvature of the stomach. Left gastric vessels were divided between LIGACLIPS (LT200, Ethicon Endo-Surgery, USA). Oesophageal hiatus of the diaphragm was widened and mobilisation of infracarinal oesophagus was completed under vision [Figure 2b] with the help of ultrasonic shear. Cervical part of the procedure was similar to as mentioned in earlier section. A small midline incision (~6 cm) incision was made in the epigastrium and transhiatal mobilisation of the mid-thoracic oesophagus was completed manually as mentioned earlier. Preparation of gastric conduit, route of conduit, anastomotic material, technique of anastomosis, placement of drains and closure of wounds were similar in both OTE and LATE groups [Figure 2c and d].

Figure 2

Intra-operative images: (a) Port position for laparoscopic-assisted transhiatal oesophagectomy, (b) transhiatal dissection, (c) gastric conduit formed, (d) post-operative mini-laparotomy scar

Statistical analysis

Data were analysed using SPSS version 20 (IBM, Chicago, IL, USA). Numerical variables were represented as median and range. Categorical variables were represented as percentages. The Chi-square test was used to compare categorical variables and Mann–Whitney U-test for non-parametric numerical variables. P < 0.05 was considered statistically significant.

RESULTS

The median age of patients was 23 (11–65) years with 19 females. Among 35 patients, 28 patients had accidental ingestion and seven patients had ingestion with suicidal intent. FJ was required before definitive surgery in 25 patients. Median level of stricture as measured on endoscopy was comparable between two groups [Table 1]. OTE was done in 20 (57.2%) and LATE was done in 15 (42.8%) patients. Stomach was used as a conduit in 23 (65.7%) patients and colon was used in 12 (34.3%) patients. Overall median (range) operation time, intra-operative blood loss and post-operative hospital stay were 330 (210–420) min, 200 (100–400) ml and 10 (7–30) days, respectively. On comparative analysis, reduced intra-operative blood loss (P = 0.025), decreased post-operative need of analgesia (P = 0.003) and shorter hospital stay (P = 0.020) were recorded in LATE group; however, operative duration was comparable in LATE and OTE groups (P = 0.852) [Table 2].

Table 1

Demographic characteristics of patients with corrosive stricture of oesophagus underwent oesophagectomy

Table 2

Comparison of outcomes following laparoscopy-assisted transhiatal oesophagectomy and open transhiatal oesophagectomy for corrosive stricture of oesophagus

One patient in OTE group had conduit necrosis (5%) and required re-laparotomy to replace necrotic conduit with colonic conduit. CEGA leak occurred in three patients (8.6%) (OTE:LATE = 2:1) and all were managed conservatively [Table 2]. Two patients had transient recurrent laryngeal nerve paresis which recovered in 6 weeks. Four patients (including three patients with leak of cervical anastomosis) developed anastomotic stricture (11.4%) (OTE:LATE = 3:1) and all were managed with endoscopic dilatation. There was no post-operative mortality. Total median (range) follow-up was 36 (6–48) months and total three patients died during this period. Out of three deaths, one had sepsis due to intestinal perforation, another had respiratory failure with sepsis and the last one had myocardial infarction.

DISCUSSION

Oesophagectomy in the presence of corrosive injury is considered difficult and is associated with complications.[128] Hence, a bypass procedure is preferred using a retrosternal route in the management of CSE. However, oesophagectomy not only permits the native and physiological route for conduit placement but also avoids complications related to retained diseased oesophagus. Oesophagectomy whether by open approach or by minimally invasive technique has been an accepted and established technique for malignant lesions of oesophagus.[9] Open transhiatal oesophageal resection and reconstruction has been safely done for extirpation of scarred oesophagus.[610] Compared to transthoracic approach, transhiatal approach avoids thoracotomy, reduces chest complications and provides for early post-operative recovery. Transhiatal laparoscopic- or hand-assisted oesophagectomy has been described with positive results in the past.[1112] After our initial experience with open THE in CSE, we resorted to LATE in selected group of patients and obtained significantly, and our initial experience (although short) has been encouraging.

The role of oesophagectomy in CSE has been linked to various peri- and post-operative complications such as increased risk of tracheobronchial injury and laryngeal nerve injury.[13] However, performing bypass procedure is associated with mucocele of remnant oesophagus[3] and a long-term risk of development malignancy in the native oesophagus.[45] Corrosive injuries are common in young patients whose expected life span is otherwise normal and regular surveillance is difficult. Furthermore, it has been reported that if malignancy occurs, it is often difficult to diagnose and usually diagnosed at an advanced stage.[14]

In the present study, none of the patients had tracheobronchial injury. Recurrent laryngeal nerve palsy was seen in two patients which was transient and managed conservatively and all patients recovered in average duration of 6 months. Aspiration was the major problem in these patients, managed with initial soft-solid foods, motivation and regular chest physiotherapy. Mediastinal dissection by laparoscopic method facilitates dissection of abdominal and lower thoracic oesophagus under direct vision and it helps in reduction of iatrogenic complication. Gupta et al.[6] have reported a low morbidity and mortality of transhiatal resection of corrosive oesophagus along with reconstruction with gastric or colonic conduit through transhiatal route; our results were also comparable. In our experience, LATE is advantageous in some aspects such as better cosmesis, low analgesic requirement and low blood loss. Shalaby et al.[15] have reported LATE being feasible in CSE with low morbidity and mortality which is in tune with our results. Small surgical incision, circumventing the need of thoracotomy and subsequent ventilatory support make the procedure noteworthy. However, retrospective nature of the study, small sample size and selection bias are limitations of our study.

CONCLUSION

In the current study, we have shown that both OTE and LATE approaches are safe and feasible for the removal of oesophagus after the development of stricture following corrosive ingestion. LATE may be an alternative to conventional open approach.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.