| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2005;28:296-300
© 2005 Elsevier Science NL
Original articles |
a Department of Surgery, Saint George University Teaching Hospital, Seregélyesi ut 3, H-8000 Székesfehérvár, Hungary
b Department of Surgery, National Medical Center, Budapest, Hungary
c Department of Gastroenterology, Medical and Health Science Centre, University of Debrecen, Debrecen, Hungary
Received 22 February 2005; received in revised form 26 April 2005; accepted 28 April 2005.
* Corresponding author. Tel./fax: +36 22 504 100. (Email: altorjay{at}mail.fmkorhaz.hu).
Abstract
Objective: Although stomach is the best choice for reconstruction after esophagectomy from the viewpoint of safety and ease, an intrathoracic stomach, nevertheless, is a poor long-term substitute. This anatomical configuration abolishes normal antireflux mechanisms and places the acid-excreting stomach subject to biliary reflux, moreover, in an adjacent position to the esophagus within the negative-pressure environment of the thorax. Methods: Between 1995 and 2002, 27 patients with high-grade neoplasia—as early Barrett's carcinoma—or non-dilatable peptic stricture underwent limited surgical resection of the distal esophagus and esophagogastric junction. In 11 of these cases, the reconstruction was performed with gastro-jejuno-duodenal interposition. The long-term functional results of this specially adapted form of interposition reconstruction have been evaluated. The postoperative follow-up period ranged between 24 and 95 months (mean 68 months). Nine patients (9/11=81.8%) have agreed to undergo endoscopy, radiographic contrast-swallow examination, and 24-h ambulatory esophageal pH and bilirubin monitoring. Results: Three out of nine patients (3/9=33%) demonstrated abnormal levels of esophageal acid exposure during the 24-h study period, whilst none had any evidence of bilirubin exposure in the esophageal remnant. Endoscopy revealed that three patients had reflux esophagitis in the remnant esophagus: Los Angeles A=2, C=1. No stomal or jejunal ulceration at the gastro-jejunal anastomosis could be observed. Histopathologic assessment of the squamous epithelial biopsies demonstrated microscopic evidence of inflammation: minor in two cases, moderate in one and major in one case; however, none of them had evidence of columnar metaplasia in the esophagal remnant at a median of 68 months after surgery. The majority of the patients have been doing well since the operation: 8/9 (88%)=Visick I–II. Conclusions: Gastro-jejuno-duodenal interposition represents an adequate ‘second-best’ method of choice if technical difficulties emerge with jejunal or colon interposition following limited resection of the esophagus performed due to early Barett's carcinoma or non-dilatable peptic stricture.
Key Words: Early Barrett's carcinoma • Peptic stricture • Columnar metaplasia • Limited esophageal resection • Interposition reconstruction
1. Introduction
The potential for limited surgical resection in patients with early tumour of the distal esophagus or the esophagogastric junction has recently been brought into the limelight [1–4]. In addition to the obviously reduced postoperative morbidity and mortality rates, a limited resection, will, however, only be advantageous to the patient, when this type of intervention is combined with a reconstructive procedure providing optimal alimentary function and prevention of gastroesophageal reflux. The options for reconstruction following limited resection include esophagogastrostomy, colon interposition, and interposition of a pedicled jejunal segment.
Although the use of the stomach is the best method of reconstruction after esophagectomy from the viewpoint of safety and ease, an intrathoracic stomach, nevertheless, is a poor long-term substitute [5–7]. This anatomical configuration abolishes normal antireflux mechanisms and places the acid-excreting stomach subject to biliary reflux, moreover, in an adjacent position to the esophagus within the negative-pressure environment of the thorax. In such patients, duodeno-gastro-esophageal reflux may thus result in the development of Barrett's metaplasia in the remnant esophagus. If early Barrett's carcinoma is present, the full-thickness removal of the entire segment of the distal esophagus affected with intestinal metaplasia together with regional lymphadenectomy is required. In such cases reconstruction with an interposed jejunal loop is preferred [4]. Colon interposition is an existing alternative for such patients; however, this procedure is less safe and less easy to perform [8,9].
There are certain conditions—like long-segment Barrett's metaplasia, peptic stricture accompanied by brachyesophagus, previous jejunal or colon resections, etc.—when interposition is technically not feasible. In such cases the only remaining option is gastric replacement. However, the question arises whether gastro-jejuno-duodenal interposition as a specially designed and adopted type of surgery can favourably influence the poor functional results associated with severe reflux observed after simple gastric substitution. We have looked for an answer to this question by evaluating the long-term functional results of gastro-jejuno-duodenal interposition surgery performed in the last 8 years. We have paid special attention to the fact whether columnar metaplasia has or has not developed in the esophageal remnant following this type of reconstruction.
2. Materials and methods
From 31 May, 1995 to 30 October, 2002, a total of 27 patients (median age 50, range 29–63 years) with high-grade neoplasia—early Barrett's carcinoma—or non-dilatable peptic stricture due to combined duodeno-gastro-esophageal reflux underwent limited surgical resection of the distal esophagus and esophagogastric junction together with lymphadenectomy of the lower posterior mediastinum and upper abdominal compartment. In those cases where the jejunal segment did not prove sufficient to bridge the gap and the colon was not suitable for substitution, the reconstruction was performed with gastro-jejuno-duodenal interposition in order to prevent postoperative reflux. The overall patient number undergoing this procedure was 11 (seven Barrett's carcinoma pT1a–b (4.7±1.1cm in length), three peptic stricture, one high-grade dysplasia (6.7cm in length)). The total number of patients submitted to esophageal resections by the authors in the same period was 479.
The most critical part of the surgical technique was the preservation of the vascular arcade of the right gastric and gastroepiploic artery and vein in the process of antrum resection (Fig. 1 A). The antrum was resected exactly in the line of the angular notch. An average of 40-cm long isoperistaltic pedicled jejunal segment was placed in between the neo-esophagus—prepared from the gastric corpus—and the duodenum. The vascular pedicle was lying on the jejunal loop that was brought up behind the colon and this settling ensured that no strangulation would occur (Fig. 1B). Stapled (CDH 25–33, Ethicon, Johnson & Johnson) esophagogastric, and single layer hand-sewn end-to-end gastro-jejunal and jejuno-duodenal anastomoses were fashioned in all cases (3/0 Vicryl, 3/0 PDS, Ethicon, Johnson & Johnson). Proper surgical approach was achieved by left thoraco-laparotomy or laparotomy with right posterolateral thoracotomy in eight and three cases consecutively. The level of esophagogastric anastomosis was above the pulmonary vein in all cases. The postoperative follow-up period ranged between 24 and 95 months (mean 68 months).
|
The severity of reflux esophagitis was determined according to the Los Angeles classifications [10]. Squamous epithelial biopsies were taken on endoscopy, from at least two different locations, at 2cms above the staple-line, whilst gastric and jejunal biopsies were taken from 2cms above and below the gastro-jejunal anastomosis. All biopsies were taken using a standard pair of spiked forceps. Biopsy specimens were fixed and stained routinely using haematoxylin and eosin. Slides were viewed under light microscopy by two independent observers, using a double-blind review protocol regarding clinical details. Inflammation in the squamous epithelium was classified according to the extent of basal cell hyperplasia and inflammatory cell infiltration. In case of the gastric epithelium, the degree of inflammation was assessed in relation to the extent of neutrophil infiltration and lamina propria mononuclear cell infiltration.
Combined 24-h ambulatory pH and bilirubin studies were performed using precalibrated antimony pH and Bilitec probes lightly taped together and connected to their respective portable data recorders (Medtronic Synectics, Sweden). The tips of all catheters were placed 2cms above the esophagogastric anastomosis. This was achieved under direct vision.
The symptoms of the participating patients were evaluated according to the following Visick-classification [11]:
3. Results
We have observed neither suture-insufficiency, nor other serious complications in the perioperative period. Oral feeding was started on the 9th postoperative day. The average length of hospital stay was 12 days. All margins of the resected specimens proved to be negative for Barrett's on histology. At a median of 68 months after surgery, three of the nine patients demonstrated abnormal levels of esophageal acid exposure during the 24-h study period, whilst none of them had any evidence of bilirubin exposure in the esophageal remnant (Table 1 ). Endoscopy revealed that three patients had reflux esophagitis in the remnant esophagus. Severe esophagitis—i.e. grade C—was found only in one patient, and grade A was found in two. At the same time, no stomal or jejunal ulceration at the gastro-jejunal anastomosis could be observed.
|
One patient without esophagitis had taken antacid agents for reflux symptoms, whilst another patient, who was proven to have severe grade C esophagitis stated that he was completely symptom-free, took no antacid agents; however, a history of massive alcohol intake was revealed. Radiographic contrast-swallow examination verified unobstructed passage and well-dilating anastomoses in all cases (Fig. 2 A and B). Two patients mentioned transient meteorism, one of them proved to have gallstones. Due to the occurrence of postoperative abdominal hernia, abdominal wall-reconstruction was performed in one of the patients 3 years after the initial operation. Nobody reported dysphagia. The majority of the patients have been doing well since the operation: 8/9(88%)=Visick I–II.
|
The number of long-term survivors after esophagectomy has increased during this decade due to the advances in surgical technique, perioperative management and detection rate of early tumour stages among all resected esophageal adenocarcinomas [12–14]. The gastric tube has been widely used as an esophageal substitute after esophagectomy; however, there are no structures in place that could prevent gastro-esophageal reflux in these patients. Many of them complain of reflux symptoms, and they often have esophagitis in the residual esophagus when examined by endoscopy.
Gutshow and colleagues [15] noted that in the early stages following vagotomy, intraluminal acidity decreased in two-thirds of the patients, but the stomach did recover its normal pH profile with time. The problem of biliary reflux—as a consequence of pyloroplasty, for which the justification remains still open to debate—is even more complex in cases of substitution with the stomach. Romagnoli et al. [16] stated that the existence of a drainage procedure, such as a short pyloroplasty or pyloromyotomy, does not increase the exposure of the gastric transplant to bile. Rather, bile exposure, in the supine position at least, tends to be more prolonged in patients without any drainage procedure. Absence of a drainage procedure maintains a barrier to bile evacuation from the gastric cavity when the patient is lying down. In the erect position, by contrast, any resistance offered by the denervated pylorus is overcome by gravity.
Dresner [7] showed that 19 of 40 patients (47.5%) had evidence of columnar epithelium in the esophageal remnant at a median of 38 months after surgery. All of them had evidence of both abnormal acid and bilirubin exposure, and the majority of bilirubin reflux episodes occurred in an acid setting. The initial detection of esophageal columnar epithelium in these 19 patients was made at a median of 14 months after surgery. The remaining 21 patients who did not develop esophageal columnar epithelium demonstrated various degrees of squamous mucosal injury in the esophageal remnant on endoscopy, ranging from a normal appearance in seven patients (17.5%) to grades I–III esophagitis in 14 (35%). Shibuya et al. [6] reported that reflux esophagitis was present in 56.4% of patients with neck anastomosis and in 88.6% of patients with intrathoracic anastomosis. For that reason, gastric advancement should be avoided for reconstruction after esophagectomy in patients who have favourable long-term survival prospects or who have benign disease.
As mixed reflux induced injury in the residual squamous epithelium, it may progress to the development of columnar metaplasia, following limited esophageal resection due to early Barrett's carcinoma or peptic stricture, reconstruction with an interposed pedicled jejunal segment—as originally described by Merendino and Dillard [17]—is unquestionably the preferred method. The use of mechanical staplers for the esophago-jejunal anastomosis has made this procedure simple and safe. Low morbidity and mortality, combined with excellent long-term functional results for this procedure, have recently been reported by several researchers in the treatment of refractory or recurrent distal esophageal strictures and other benign lesions that required resection of the distal esophagus and cardia. Histologic studies of endoscopic biopsies of the interposed jejunal loops have confirmed the retention of a normal villous architecture with Paneth cell hyperplasia, but no evidence of metaplasia or dysplasia has been apparent at long-term follow-up. Only 12–16% of these patients have had evidence of esophagitis on endoscopy [4,18,19].
Sometimes during a limited resection of the esophagus, a neo-esophagus of required length cannot be formed from the jejunum due to its short mesenterium or due to its poor vascular arcade system. If, at the same time, the colon cannot be utilised either—for example due to previous resection, diverticulosis, etc.—the continuity of the alimentary tract can only be restored with the stomach. In such cases, we applied gastro-jejuno-duodenal interposition to prevent late complications induced by the almost inevitable reflux problem. After antrectomy the neo-esophagus formed from the corpus of the stomach can be pulled up even to the level of the chest-dome without tension; only its vascular pedicle requires special attention. Standard Roux-Y procedures have been demonstrated to be effective in eliminating alkaline reflux and in providing symptom relief. Having said that, as normal foregut physiology is severely altered and unacceptable side-effects such as the dumping syndrome, Roux stasis syndrome with bacterial overgrowth, diarrhoea, jejunal ulceration, and impaired pepsin-mediated protein digestion may result, we inserted an almost 40-cm long isoperistaltic jejunum-loop in between the corpus of the stomach and the duodenum [20–22]. Based on the results at follow-up, this—besides restoring normal foregut physiology—has effectively protected the residual esophagus from biliary reflux.
The number of the presented cases is undoubtedly not high, but the area of indication is also narrow, since in the majority of cases jejunal interposition is successful. The favourable value of gastro-jejuno-duodenal interposition as a special form of interposition surgery shows itself in the fact that even after many years following the operation no columnar metaplasia appears in the remnant esophagus since—as opposed to simple gastric advancement—no mixed reflux will develop after this type of reconstruction. On the basis of all this above, we think that gastro-jejuno-duodenal interposition represents an adequate ‘second-best’ surgical solution in the event of technical difficulties encountered with jejunal or colon interpositions following limited esophageal resections performed due to early Barett's carcinoma or non-dilatable peptic stricture. This way a safe neo-esophagus can be formed providing good long-term functional results, even if the resection must be unexpectedly extended on surgery.
Acknowledgments
The authors thank Mrs Nicola Pen Jackson MD (London) and Mrs Orsolya Salfay PhD (Budapest) and Gabor Kecskés MD for critical reading of the manuscript before its submission.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
