|ORIGINAL RESEARCH REPORT
|Year : 2019 | Volume
| Issue : 3 | Page : 81-86
Management of long segment corrosive esophageal stricture in children and adults: A 5 years' review
Ismail Mohammed Inuwa1, Jameel Ahmad Ismail1, Nurein Tunde Oyebanji1, Lofty-John Chukwuemeka Anyanwu2, Sani Ali Aji3, Mohammad Aminu Mohammad2, Abubakar Umar4, Nuhu Garba5, Ibrahim Aliyu6
1 Department of Surgery, Cardiothoracic Unit, Aminu Kano Teaching Hospital, Kano, Nigeria
2 Department of Surgery, Paediatric Surgery Unit, Aminu Kano Teaching Hospital, Kano, Nigeria
3 Department of Surgery, Urology Unit, Aminu Kano Teaching Hospital, Kano, Nigeria
4 Department of Surgery, Cardiothoracic Unit, Usman Dan Fodio University Teaching Hospital, Sokoto, Nigeria
5 Department of Paediatrics, Federal Medical Centre, Nguru, Yobe State, Nigeria
6 Department of Paediatrics, Aminu Kano Teaching Hospital, Bayero University, Kano, Nigeria
|Date of Web Publication||3-Jul-2019|
Dr. Ismail Mohammed Inuwa
Department of Surgery, Cardiothoracic Unit, Aminu Kano Teaching Hospital, Kano
Source of Support: None, Conflict of Interest: None
Introduction: The aim of this study is to highlight our experience in the management of long segment corrosive esophageal stricture using the left colon bypassing the strictured segment of the esophagus thereby maintaining continuity and restoring normal swallowing. Subjects and Methods: This was a retrospective study of patients with long-segment corrosive esophageal stricture who were managed by left colon bypass surgery in our center over a 5-year period between January 2012 and December 2016. Hospital records were reviewed and relevant information such as age, sex, clinical presentations, and operative procedure were obtained. The left colon grafts were taken under the pedicle of left colic artery in all cases, and three anastomoses were made at colocolic, cologastric, and colo-esophageal points. Feeding gastrostomy was done for all patients for early feeding. Results: There were 12 males (60%) and 8 females (40%) and age ranging from 2 years to 42 years; among them, there were 13 children (65%) and 7 adults (35%); caustic soda was the most common corrosive substance in 14 cases (70%) and acid in 2 cases (10%); the exact nature of the substances in 4 (20%) was not clear. Accidental ingestion was found in 90% of the cases. Psychiatric illness was associated in 2 adults (10%). Conclusion: The treatment of long segment corrosive esophageal stricture with left colon bypass graft surgery under the pedicle of the left colic artery is very effective with minimal postoperative complications; we, therefore, recommend this technique for relief of dysphasia from this form of injury.
Keywords: Corrosive esophageal stricture, left colic artery, left colon bypass
|How to cite this article:|
Inuwa IM, Ismail JA, Oyebanji NT, Anyanwu LJC, Aji SA, Mohammad MA, Umar A, Garba N, Aliyu I. Management of long segment corrosive esophageal stricture in children and adults: A 5 years' review. J Clin Sci 2019;16:81-6
|How to cite this URL:|
Inuwa IM, Ismail JA, Oyebanji NT, Anyanwu LJC, Aji SA, Mohammad MA, Umar A, Garba N, Aliyu I. Management of long segment corrosive esophageal stricture in children and adults: A 5 years' review. J Clin Sci [serial online] 2019 [cited 2021 Feb 25];16:81-6. Available from: https://www.jcsjournal.org/text.asp?2019/16/3/81/262068
| Introduction|| |
Long segment corrosive esophageal structure is difficult to treat without surgical intervention. The surgical technique using the colon was initiated over 100 years ago by Kelling and Vuillet. Surgery using the left colon bypass is not commonly used in Nigeria. Corrosive esophagitis affects all age groups. It is mostly accidental in both children and adult. The line of management for corrosive ingestion is usually resuscitation, and then possibly feeding gastrostomy in some patients before definitive surgery. Different centers have adopted varied surgical approach to the treatment of long segment corrosive oesophagitis.,,, Some surgeons use the right colon for esophageal replacement under the pedicle of middle colic artery , while others use the stomach. However, we prefer using the left colon under the pedicle of left colic artery which is a branch of inferior mesenteric artery for obvious reasons; the left colon is more like the native esophagus in terms of its diameter and nature of the content it propels, (solid) it is also isoperistaltic and less bulky. Furthermore, the blood supply to left colon is more constant then the right colon. Therefore, our study hopes to determine the outcome of left colon bypass approach practiced in our center.
| Subjects and Methods|| |
This was a retrospective study of all patients who had long segment corrosive esophageal structure and were managed by colon bypass surgery using the left colon under the pedicle of the left colic artery over 5 years from January 2012 to December 2016. Hospital records were reviewed and relevant information such as age, sex, duration of symptoms, radiological findings, initial resuscitative measures, with or without feeding tube, operative procedure, complications, and follow-up over at least 1 year were extracted. Ethical approval was obtained from the Research and Ethics Committee of Aminu Kano Teaching Hospital, Kano.
Our surgical protocol
Feeding gastrostomy was done before the surgery for those patient who presented late, over 3 months from the unset of corrosive ingestion; the aim was to improve the weight of the patients to reach near normal for age at the time of surgery. However, early presenters had feeding gastrostomy inserted during surgery. Routine protocol for surgical preparation was observed, and bowel preparation was done using 10% mannitol infusion which was given orally or through the feeding gastrostomy tube at the rate of 5 ml/kg/h until patient purged clear fluid; at the same time, dextrose/saline or pediatric saline was given parenterally, with maintenance potassium at the dose of 1 mmol/1 kg added to the infusion. No additional mechanical bowel preparation was required in all cases. Antibiotics were given at induction of anesthesia (ceftriaxone and metronidazole). Standard monitoring protocol using pulse oximeter, electrocardiography, capnography, and noninvasive blood pressure monitoring were used. General anesthesia with endotracheal tube in situ connected to ventilator using close system was adopted. The neck, chest, and abdomen were cleaned and draped in continuity using standard protocol. Patients were operated in supine position with extended neck and the head rotated to the right side; this exposed the left side of the neck where the proximal anastomosis was sited using hockey stick skin incision along the medial border of the left sternocleidomastoid muscle; and the curved part was situated at the suprasternal notch. After dissecting the neck muscles, the thyroid and trachea were retracted medially, and the carotid sheath was retracted laterally to expose the esophagus which was identify by passing a nasogastric tube. The esophagus was opened longitudinally on its lateral side. The abdomen was opened through upper midline incision from xiphoid to the midway between umbilicus and pubic symphysis; this created easy access to explore the abdomen, transverse colon, splenic flexure, and the left colon can easily be mobilized. Marginal arteries were assessed, and middle colic artery was divided below the Y junction [Figure 1] to maintained blood supply from the left colic artery [Figure 2]. The length of the graft was measured with a suture from the tip of the esophageal incision to the root of the inferior mesenteric artery. This was used to measure the length of colon to be used from 5 cm below the attachment of the left colic artery to the descending colon. Using intestinal clamps distally, the colon was divided and gas was expelled before remeasurement to ascertain the length before dividing the proximal colon. Colocolic end-to-end anastomosis was done in two layers for adults and single layer for children. The protective feeding gastrostomy was constructed if it was not in situ before the surgery. Then, cologastric anastomosis was constructed between the distal colonic graft end and the anterior gastric wall at the antral portion distal to the site of feeding tube. A plane was created under the xiphoid through the diaphragm into the anterior mediastinum to connect to the suprasternal notch, through which the colon was routed to the neck [Figure 3]; finally, a colo-esophageal end to side anastomosis was constructed and appendectomy was done if the ascending colon was mobilized to get adequate length; as was seen in two cases in this study. This was to avoid future diagnostic confusion because of its displaced position. Vicryl 2/o for adult or 3/o for children was used for anastomosis and single layer in children was used. The pylorus was checked for stenosis. The abdomen and neck were closed in layers. All patients were on Nil per oral (NPO) for 10 days and if no leak was noticed afterward, oral feeding was commenced; usually from the 3rd to 4th day feeding through feeding gastrostomy was initiated. The feeding tube was removed during the first clinic visit after discharge if swallowing was normal. Patients who developed pneumothorax during mediastinal tunneling had chest tube inserted after confirmation by chest X-ray. Those patients who developed respiratory distress postoperatively were admitted to intensive care unit (ICU) for close monitoring.
|Figure 1: The transverse colon showing the middle colic artery and marginal arteries|
Click here to view
|Figure 2: Harvested colonic graft with the pedicle of the left colic artery|
Click here to view
The data were analyzed using Statistical Package for Social Sciences version 16 (SPSS Inc., Chicago, Illinois, USA). Quantitative variables were summarized using mean and standard deviation while qualitative variables were summarized using frequencies and percentages. The relationship between categorical variables was tested using Fisher's exact tests and P < 0.05 was considered as statistically significant.
| Results|| |
There were 12 males (60%) and 8 females (40%). The age ranged from 1.5 years to 42 years with mean of 12.2 ± 12.8; among them, there were 13 children (65%) and 7 adults (35%). Caustic soda corrosive esophagitis was witnessed in 13 cases (65%) and acid esophagitis was observed in 3 cases (15%), whereas in 4 (20%) cases, the exact nature of the substances involved was not known. Accidental ingestion was found in 18 (90%) cases which involved all the children and five adults. Psychiatric illness was associated in 2 of the adult cases (10%); while a patient had systemic lupus erythematosus (5%). The average duration of surgery was 4 h ± 30 min longer in adults. All patients had successful pedicle graft, except 1 patient (5%) who had graft necrosis. Five patients (25%) developed reflux which was noted during follow-up; while 15 (95%) did not. Four patients (20%) had leakage at proximal anastomosis (colo-esophageal), which was managed conservatively, and this prolonged the duration of NPO to 3 weeks instead of 10 days; however, 16 (80%) did not develop any leakage. Four patients (20%) had right-sided pneumothorax and had chest tube inserted; while 16 (80%) were free of pneumothorax, 18 patients (90%) had relief of dysphasia after surgery. One patient (5%) died following graft necrosis; esophagostomy and feeding gastrostomy was done after removal of the graft. The patient died 6 days' postoperatively from sepsis. One child (5%) had dysphagia from stenosis at the proximal anastomosis which necessitated the use of feeding tube; and the patient was lost to follow-up. None of the patients has any indication for ICU admissions. Only 2 (10%) patients had appendectomy. Time of commencement of oral feeding ranged from 8 to 30 days with a mean of 12.4 ± 7.1. Most of the cases were within 8–14 days (15, 78.9%) subgroup, while the >14 days' subgroup was 4 (21.1%). Only 7 (35%) of the cases had feeding tube inserted before surgery; while 13 (65%) had no feeding tube inserted before surgery. The duration of symptoms before presentation ranged from 3 weeks to 108 weeks with mean of 12.1 ± 22.9 weeks. Most cases presented >28 days (11, 55%) after onset of symptoms; followed by those presenting within 14–28 days (6, 30%) then those within 8–14 days (3, 15%).
In [Table 1], majority of children and adults commenced tube feeding postsurgery in the 2nd week after surgery; furthermore, almost half of the children (46%) had feeding tube passed before surgery; while only a single adult had feeding tube before surgery. However, these observations were not statistically significant (Fisher's exact test = 0.377, P = 0.603; Fisher's exact test = 2.031, P = 0.177, respectively).
|Table 1: Comparing the age classification with the time taken for the commencement of tube feeding; and passage of feeding tube before surgery|
Click here to view
In [Table 2], almost equal proportion of children and adults had complications; however, pneumothorax was most common in children, and the only documented postoperation infection was recorded in a child; though this observation was not statistically significant (Fisher's exact test = 1.316, P = 1.000).
|Table 2: Comparing the age classification and the presence of complications|
Click here to view
In [Table 3], most adults presented late to the hospital; however, caustic corrosive esophageal injury was the most common in both groups; though these observations were not statistically significant (Fisher's exact test = 5.041, P = 0.084; Fisher's exact test = 3.231, P = 0.203 respectively).
|Table 3: Comparing the age classification with duration of illness before presentation to the hospital; and the type of corrosive injury|
Click here to view
In [Table 4], the only death recorded in this review was in the child group; however, postoperative reflux complication was noticed mostly among the adult group, and this observation was statistically significant (Fisher's exact test = 5.934, P = 0.031).
|Table 4: Comparing the age classification with the outcome at one postoperation follow-up visit; and the presence of reflux|
Click here to view
| Discussion|| |
The management of long-segment corrosive esophageal stricture using colon bypass surgery is one of the options for definitive surgery. There are other treatment options, but the left colon for esophageal replacement is preferable because it maintains better continuity and passage of food to the stomach. Other centers have documented using the right colon while others stretch the stomach serving as the conduit. Advantages of using the right colon under the pedicle of middle colic artery are: the lumen is wider compared to the left colon and therefore it may be more effective for African food bolus, furthermore the ileocecal valve acts as a gate preventing regurgitation. While using the stomach has the advantage of having only to perform a single gastroesophageal anastomosis; more so, the stomach has a better blood supply, and therefore, healing is faster at the anastomotic site. However, in stomach leaks, gastric and pancreatic secretions may result in chemical mediastinitis resulting in increased morbidity and mortality. In our cases, we used the left colon because it is more structurally similar to the native esophagus in terms of diameter and type of contents it propels (solid), when compared to the right colon which is semi-solid and also it has more constant blood supply-left colic artery is more constant than the middle or right colic artery. According to Wilkins, the outcome of left colon is better than right colon when used as graft  furthermore mucous fistula from coloesophageal anastomosis is easier to treat then gastric fistula when the stomach is used. Fürst et al. also reported good outcome when they used modification of colonic graft using right colon but under the pedicle of left colic artery avoiding left flexure release which will reduce the risk of graft necrosis. Again stenosis is more common in proximal anastomosis when the right colon is used compared to the left colon. Only a single case of stenosis was documented in our study.
Patient with esophageal stricture usually present with difficulty in swallowing progressing from solid to liquid and finally with drooling of saliva if they have absolute dysphagia due to total occlusion of the lumen, there may be cachexia, dehydration, and recurrent chest infection from aspiration.
The retrosternal route, through anterior mediastinal space was our surgical approach. Other route through the esophageal bed removes the native esophagus; this is preferable in cases associated with malignancies, but our cases were benign lesions. The retrosternal approach involved less dissection, and this will reduce morbidity and mortality, but the preserved native esophagus has a negligible risk of malignant transformation.,,, The subcutaneous route has been abandoned because of poor surgical outcome; patients will need to manually facilitate downward movement of the food using their hands. In our approach, three anastomoses were created: colocolic end-to-end anastomosis to maintain the continuity of large bowel, cologastric end to side anastomosis and coloesophageal end to side anastomosis.
The colo-esophageal anastomosis was done after tunneling the transplanted colon through the anterior mediastinum using long artery forceps to the neck. If during this process, the pleura is breached, a chest tube drain shall be inserted on the affected side often the right side is affected because the pleura crosses the midline on the right side. Three of our patients had right-sided pneumothorax, and all had chest tube inserted for 3–4 days.
Appendectomy was done if the right colon was mobilized for the purpose of getting additional length from the transverse colon; otherwise, we preserved the appendix routinely, only 2 patients (10%) had appendectomy in this study.
Children were mostly represented in this study for obvious reasons; corrosive oesophagitis is mostly accidently which has a higher tendency to occur in children, especially during the toddlerhood, when they often explore their environment; mouthing of object is prevalent at this age  and if poorly supervised the chances of aspiration or ingestion are heightened. Majority of our cases were accidental ingestion, this was similar to those of Eze et al. but this differed from those of Attipou et al., who reported suicidal intent as the main reason in their study. The age disparity in both studies may explain this variation. Furthermore, Attipou et al. reported that the use of caustic agents was mostly associated with suicidal intent while the use of acid was accidental, but this was not the case in our study. Similar to the studies of Eze et al. and Attipou et al., most of the cases in this study were males and caustic agents were mostly implicated as the cause of corrosive esophagitis, though Eze et al. used the right colon for the transposition, their outcome was similar to our study, such as less cases of complications such as pneumothorax and dysphagia at 1-year postoperation. Similarly, other centers have recorded favorable outcomes irrespective of their chosen surgical technique; Adegboye et al., while using the posterior mediastinal, transhiatal oesophagectomy approach reported good outcome; their record was substantiated by Bassiouny  and Pompeo.
However, other variables studied, such as the presence of complications, duration of illness before presentation to the hospital, and the type of corrosive injury, and the outcome at 1 year postoperative follow-up visit, did not show any age predilection.
| Conclusion|| |
Surgical treatment of long segment corrosive esophageal stricture using the left colon, under the pedicle of left colic artery is very effective in relieving dysphagia. Majority of the cases had no significant complication, and only a single mortality was recorded.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Fürst H, Hartl WH, Löhe F, Schildberg FW. Colon interposition for esophageal replacement: An alternative technique based on the use of the right colon. Ann Surg 2000;231:173-8.
Aghaji MA, Chukwu CO. Oesophageal replacement in adult Nigerians with corrosive oesophageal strictures. Int Surg 1993;78:189-92.
Eze JC, Ezemba N. Corrosive injury of upper gastrointestinal tract. In: Nwankwo OE, editor. Essentials of Trauma Management and Related Principles of Surgery. Uwani, Enugu, Nigeria: Ezu Books Ltd.; 2011. p. 319-30.
Anyanwu CH, Okonkwo PO. Oesophageal strictures induced by herbal preparations. Trans R Soc Trop Med Hyg 1981;75:864-8.
Agagua NE. Colon bypass for corrosive esophageal strictures in Nigerian children. Int J Peadiatr Surg Sci 1988;2:7-10.
Adegboye VO, Adebo OA, Brimo IA. Esophagectomy without thoracotomy for corrosive esophageal stricture. Niger J Surg 1995;2:62-6.
Hiebert CA, Brendenberg CE. Selection and placement of conduits. In: Pearson FG, Cooper JD, Deslauriers J, Ginsberg RJ, Hiebert CA, Patterson GA, et al.,
editors. Esophageal Surgery. New York: Churchill Livingstone; 1995. p. 649-56.
Orringer MB, Orringer JS. Esophagectomy without thoracotomy: A dangerous operation? J Thorac Cardiovasc Surg 1983;85:72-80.
Wormuth JK, Heitmiller RF. Esophageal conduit necrosis. Thorac Surg Clin 2006;16:11-22.
Sa YJ, Kim YD, Kim CK, Park JK, Moon SW. Recurrent cervical esophageal stenosis after colon conduit failure: Use of myocutaneous flap. World J Gastroenterol 2013;19:307-10.
Sikorszki L, Kalmar K, Pavlovics G, Papp A, Sajjadi SG. Resection or bypass in the treatment of corrosive esophageal stricture? Malignant transformation as a late complication in both methods. Eur Surg 2012;44:299-303.
Mansour KA, Hansen HA 2nd
, Hersh T, Miller JI Jr., Hatcher CR Jr. Colon interposition for advanced nonmalignant esophageal stricture: Experience with 40 patients. Ann Thorac Surg 1981;32:584-91.
Kim YT, Sung SW, Kim JH. Is it necessary to resect the diseased esophagus in performing reconstruction for corrosive esophageal stricture? Eur J Cardiothorac Surg 2001;20:1-6.
Davids PH, Bartelsman JF, Tilanus HW, van Lanschot JJ. Consequences of caustic damage of the esophagus. Ned Tijdschr Geneeskd 2001;145:2105-8.
Aliyu I. Esophageal foreign body in a child mimicking a primary respiratory disease. J Med Trop 2013;15:159-61. [Full text]
Eze JC, Onyekwulu FA, Nwafor IA, Etukokwn K, Orakwe O. Right colon interposition in corrosive esophageal long segment stricture: Our local experience. Niger J Clin Pract 2014;17:314-9.
] [Full text]
Attipou K, Dossed D, Abousalem A, Sodji C, Komlavi J. Caustic stenosis of the oesophagus at Centre Hôpital D'Université (CHU) of Lome: Epidemiological and therapeutic aspects. Niger J Surg Res 2006;8:38-43.
Adegboye VO, Brimmo A, Adebo OA. Transhiatal esophagectomy in children with corrosive esophageal stricture. Afr J Med Med Sci 2000;29:223-6.
Bassiouny IE, Al-Ramadan SA, Al-Nady A. Long-term functional results of transhiatal oesophagectomy and colonic interposition for caustic oesophageal stricture. Eur J Pediatr Surg 2002;12:243-7.
Pompeo E, Coosemans W, De Leyn P, Denette G, Van Raemdonck D, Lerut T. Esophageal replacement with colon in children using either the intrathoracic or retrosternal route: An analysis of both surgical and long-term results. Surg Today 1997;27:729-34.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3], [Table 4]