|
|
||||||||
Ann Thorac Surg 2007;83:265-271
© 2007 The Society of Thoracic Surgeons
a Unit of Upper Gastro-Intestinal Surgery, Saint-Luc Academic Hospital, Brussels, Belgium
b Department of Radiology, Saint-Luc Academic Hospital, Brussels, Belgium
Accepted for publication July 26, 2006.
* Address correspondence to Dr Collard, Unit of Upper G-I Surgery, Saint Luc Academic Hospital, Hippocrate Ave, 10, B-1200-Brussels, Belgium (Email: collard{at}chir.ucl.ac.be).
| Abstract |
|---|
|
|
|---|
METHODS: An intrathoracic fundoplication was performed on 84 patients suffering from gastroesophageal reflux disease with the junction between upper gastric folds and the unwrinkled esophageal mucosa remaining above the diaphragm while the esophageal body was quite straight on barium swallow study. Particular attention was paid to the following steps: further enlargement of the hiatal sling to avoid any strangulation of the stomach, very careful manipulation of gastric tissues with the fingers rather than with forceps, and meticulous anchoring of the wrap to the hiatus with numerous sutures while mimicking diaphragmatic movements that arise on cough. Results were assessed by personal interview (n = 84; median follow-up, 51.5 months), barium swallow study (n = 84), 24-hour esophageal pH monitoring (n = 65), and esophageal stationary manometry (n = 56).
RESULTS: No patient had any symptoms of reflux; 5 (5.9%) had episodes of dysphagia, which were frequent in 2; and 31 (37%) had some degree of flatulence, which interfered with social life in 5. The mean percentage of total time that esophageal pH was below 4 at esophageal pH monitoring dropped significantly (p < 0.001) from 12.3% before fundoplication to 0.5% after. Lower esophageal sphincter resting pressure increased significantly (p < 0.0001) from 6.9 mm Hg to 20.6 mm Hg. Nine patients (10.7%) were reoperated on for spontaneous (n = 1) or anti-inflammatory druginduced (n = 1) gastric perforation, further herniation of the stomach (n = 3), herniation of the colon (n = 3), or both (n = 1), into the chest.
CONCLUSIONS: Intrathoracic periesophageal fundoplication for short esophagus is amazingly effective for treating reflux. Strict observance of some critical technical details makes spontaneous gastric perforation very unlikely. Any sudden increase in abdominal pressure at early follow-up is to be avoided, and anti-inflammatory drugs are strictly forbidden.
| Introduction |
|---|
|
|
|---|
Various surgical procedures have been used to create an effective barrier against the reflux of gastric contents into the lumen of a short esophagus. These include forward transposition of the hiatus [1], transabdominal perigastric fundoplication according to Maillet technique [2], intrathoracic Nissen fundoplication [3, 4], transthoracic Collis-Belsey fundoplication [5], transthoracic Collis-Nissen fundoplication [6], duodenal diversion procedures [7], and even distal esophagectomy [8, 9]. Even though they leave acid-secreting gastric tissue proximal to the fundoplication, the Collis-Nissen and Collis-Belsey operations, either transthoracic [6] or laparoscopic [10], have been claimed to be the "gold standard" of treatment [10]. Although very good long-term clinical outcomes have been achieved after these procedures [5], several authors reported clinical failure rates as high as 25% [6, 11, 12], with the residual presence of a pathological acidity proximal to the wrap in as many as 50% of patients [10].
The intrathoracic Nissen operation, the only procedure that creates an antireflux barrier above the GE junction, has been put into disrepute since the report of life-threatening mechanical complications in the 1970s and early 1980s [1318]. In 1991, however, we reported encouraging clinical and pH results, together with the absence of major technical complications [4], using an intrathoracic periesophageal fundoplication technique that included the following steps: further enlargement of the hiatal sling to avoid any strangulation of the stomach, very careful manipulation of gastric tissues with the fingers, meticulous anchoring of the wrap to the hiatus to prevent any tension on the gastric wall on diaphragmatic movements, and effective decompression of the gastric cavity until bowel movements resume postoperatively.
The present paper reports a 20-year experience of surgical management of short esophagus using this intrathoracic periesophageal fundoplication technique as regards clinical and pH control of reflux, restoration of a high-pressure zone at the GE junction, and mechanical complications.
| Patients and Methods |
|---|
|
|
|---|
|
Disease Assessment
Basically, the diagnosis of short esophagus was settled by barium swallow study, including radiographs taken in both the upright and supine positions [19]. The esophagus was considered short whenever the junction between the upper gastric folds and the unwrinkled esophageal mucosa was above the diaphragm while the esophageal body was quite straight (Fig 2). All the radiological examinations were done by the same radiologist (L.G.).
|
Twenty-four-hour esophageal pH monitoring was performed after discontinuation of any antacid medication for a 10-day period in 46 patients [20]. The percentage of time (mean ± SEM) that esophageal pH was less than 4 was 11.9 ± 1.3, 12.0 ± 1.6, and 9.9 ± 1.3 for the total, upright, and supine periods of recording, respectively.
Lower esophageal sphincter (LES) resting pressure at esophageal stationary manometry using a three-channel assembly in 79 patients ranged from 0 mm Hg to 28 mm Hg (mean, 7.4 mm Hg). Esophageal body motility was normal in 72 patients (91.1%), whereas slight dysmotility was observed in 7 others (8.9%). No patient had esophageal dyscontractility, defined as pressure waves less than 20 mm Hg in amplitude.
Surgical Technique
Seventy-four patients (88.1%) were operated on by left thoracotomy at once, and 10 patients (11.9%) in whom irreducibility of the GE junction was unclear on preoperative upper GI series had an abdominal approach (laparotomic, n = 4; laparoscopic, n = 6) first with subsequent conversion to a left thoracotomy after surgical assessment of the exact location of the GE junction.
The operation is performed through a left thoracotomy in the seventh interspace. Esophageal dissection is carried out from the hiatus to the aortic arch. In patients with a history of unsuccessful antireflux operation, the residual repair, if any, is taken down. Care is taken to avoid injury to the vagi, especially in the presence of severe periesophagitis, a condition that makes dissection extremely difficult. The GE junction is clearly identified (Fig 3A), and its irreducibility below the diaphragm confirmed. The hiatus, which is already enlarged because of the presence of a hiatal hernia, is widened further by either division of the left crus or a radial diaphragmatic incision from the anterior margin of the hiatal sling (Fig 3B). The fundus is dissected off the spleen through the hiatus by division of the short gastric vessels to bring a large amount of gastric tissue into the chest and to create a very floppy wrap. The stomach is handled carefully with the fingers, avoiding the use of forceps, especially the traumatizing Babcock clamps.
|
The wrap is anchored circumferentially to the crura by placement of numerous nonabsorbable sutures, taking the extramucosal layers of the gastric wrap and the pleuromuscular layers of the crura (Fig 3D). To do this, the surgeon pushes the left part of the diaphragm down with the left hand, mimicking diaphragmatic movements like those that arise on cough, so that the anchoring sites can be precisely marked out on the fundoplication and no tension develops in the postoperative course. Because in some patients belonging to the first half of the present series, herniation of the splenic flexure of the colon developed or there was further herniation of the stomach into the chest, the number of anchoring sutures was progressively increased over time so that now about 15 sutures are placed on the hiatal sling with great care.
Postoperatively, the nasogastric catheter is removed only when bowel movements resume, to prevent acute distension of the freshly constructed fundoplication. Carbonated beverages and anti-inflammatory drugs are strictly forbidden at follow-up.
Postoperative Work-Up
All patients underwent barium swallow study 1 month postoperatively. One month later, 24-hour esophageal pH monitoring and esophageal stationary manometry were performed in 65 and 56 patients, respectively. In all, this made a total of 34 patients who underwent both preoperative and postoperative esophageal pH studies and a total of 53 patients having motility of their esophagus assessed before and after operation.
Technical complications that occurred either in the immediate postoperative period or at follow-up were analyzed in reference to those in our prevously published [4] historical series of 16 intrathoracic fundoplications performed between 1976 and 1981. At that time, less care was taken regarding further enlargement of the hiatus, careful handling of the gastric wall with the fingers, creation of a floppy wrap with use of a rather large amount of gastric tissue, and meticulous placement of numerous anchoring sutures to the diaphragm while mimicking diaphragmatic movements that arise on cough.
Long-Term Interview
Seventy-seven patients (91.7%) were carefully interviewed for the presence of heartburn, regurgitation, dysphagia, and flatulence by the first author (F.V.), a resident from Geneva, Switzerland who acted as an independent observer. For the remaining 7 patients (8.3%) who had died in the mean time, information from interim analyses [21] was taken into account. Overall, follow-up time ranged from 1 to 228 months (mean, 65.6; median, 51.5), for a total of 5,515 months.
Statistical Analysis
Categorical variables were analyzed using Fishers exact or
2 tests, and continuous variables using Wilcoxon nonparametric rank sum test. A p value less than 0.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
Reflux Control
Postoperative barium swallow study showed an intrathoracic wrap encircling the lower segment of the esophagus with no GE reflux of the contrast medium in all the 84 patients (Fig 4).
|
None of the 65 patients who underwent postoperative esophageal pH monitoring over a 24-hour period had abnormal pH results. As shown in Figure 5, the percentage of time that esophageal pH was below 4 (mean ± SEM) in the 34 patients investigated before and after fundoplication dropped significantly (p < 0.001) from 12.3% ± 1.6% preoperatively to 0.5% ± 0.1% after the operation for the total period, from 12.0% ± 1.9% to 0.8% ± 0.2% for the upright period, and from 12.3% ± 1.9% to 0.1% ± 0.04% for the supine period. Likewise, the DeMeesters score (mean ± SEM) dropped significantly (p < 0.001) from 42.8 ± 5.2 before fundoplication to 2.0 ± 0.5 2 months postoperatively.
|
|
|
Seven patients (8.3%) were reoperated on by either thoracotomy (n = 2) or laparotomy (n = 5) because of further herniation of the stomach into the chest (n = 3), intrathoracic migration of the splenic flexure of the colon alongside the fundoplication (n = 3), or both (n = 1). Conditions leading to gastric or colic herniation into the chest were strain while awaking from anesthesia and strain under heavy loads in 2 and 5 patients, respectively. There was only one herniation in the second half of the current series, namely, after we increased the number of anchoring sutures to the hiatal sling. Altogether, this made a total of 9 patients (10.7%) who required reoperation for a technical complication.
| Comment |
|---|
|
|
|---|
Short esophagus (brachyesophagus) is to be distinguished from Barretts esophagus (endobrachyesophagus); the latter consists of a columnar metaplasia of the distal esophageal mucosa with subsequent elevation of the squamous-glandular mucosal junction (z-line) but without any shortening of the esophageal muscle coat. However, both anatomical abnormalities may coexist in the same patient, as was the case in the 17 Barretts patients of the present series and in 20 of a personal series of 55 patients operated on for Barretts adenocarcinoma [22]. Misestimation of the actual location of the GE junction when performing a transabdominal fundoplication in such patients explains, in part at least, why the results of antireflux surgery are less satisfactory in Barretts patients than in patients having erosive esophagitis only [24].
Anatomical shortness of the esophagus is best assessed by barium swallow study [19] with particular focus on both the level where gastric folds meet the unwrinkled esophageal mucosa and the straightness of the esophageal body on radiographs taken in the erect position, with a clear superiority of barium swallow study over upper GI endoscopy and esophageal manometry [12]. Marginalization of radiology in favor of upper GI endoscopy together with the development of laparoscopic antireflux surgery by abdominal surgeons not used to thoracic esophageal procedures [25] explains why the concept of short esophagus has been scotomized in the numerous meetings devoted to antireflux surgery during the last 15 years, so far as it was even taken for granted that all GERD patients could be operated on from the abdomen.
Construction of a fundoplication in the chest, a surgical maneuver used for more than 80 years to cover intrathoracic esophageal sutures [32], is very effective in the control of GERD as attested to by none of our 84 consecutive patients having any reflux symptoms at follow-up; all the 65 patients who underwent postoperative 24-hour esophageal pH monitoring had a normal esophageal acid exposure, and postoperative LES resting pressure in 53 patients was threefold higher than that measured preoperatively.
Reduction of the GE junction into the abdomen without undue tension, performance of a fundoplication around the intra-abdominal segment of the esophagus, and approximation of the diaphragmatic crura to obviate postoperative herniation of the wrap into the chest have been recommanded to make an antireflux operation successful [33]. However, the present experience of intrathoracic fundoplication for short esophagus shows that the most important step is the placement of the fundoplication proximal to the GE junction, around the lower esophagus. The study demonstrates indeed that a fundoplication can function in the chest, namely, in a negative-pressure environment. It also confirms that approximation of the crura plays a marginal role in the control of reflux compared with the fundoplication itself [33].
In this regard, intraoperative findings in our 19 patients who experienced failure of a transabdominal antireflux operation demonstrate that any attempt to construct an antireflux mechanism around an organ that has lost or never had room in the abdomen leads to failure (Fig 8) [2628]. Of course, exertion of undue downward traction on a short esophagus allows the construction of the fundoplication proximal to the cardia from the abdomen in some patients. But the spontaneous tendency of the short esophagus to go back to its natural location in the lower mediastinum explains why, most of the time, such a maneuver results in partial breakdown of the crura closure and early herniation of the fundoplication into the chest with the subsequent onset of dysphagia [29].
|
Obviously, surgical management of short esophagus, one of the most severe forms of benign GERD, can not be done in a simple way. Any surgical technique carries a risk of either technical complication or failure to control GE reflux that may require redo surgery [1317, 31]. That was the case for 10% of patients in one of the largest series of Collis-Nissen operations ever reported [6] and in 10.7% of those in the present series of intrathoracic fundoplications. In this respect, the present paper shows that some critical technical rules must be respected when performing an intrathoracic fundoplication. Indeed, contrary to what was reported 25 years ago [1317] and what we observed in our own historical series [4], namely, when surgical technique was less refined, only 1 patient of the present series experienced postoperative spontaneous gastric perforation [34]. Spontaneous gastric perforation in this patient was a consequence of the misplacement of a hemostatic 8-stitch on the gastric wall leading to local ischemia. In fact, further enlargement of the hiatal sling prevents postoperative strangulation of the intrathoracic wrap.
Careful handling of gastric tissue with the fingers rather than with forceps does not tear out the gastric wall. The use of a rather large amount of gastric tissue allows the construction of a floppy wrap without undue tension. Likewise, placement of numerous anchoring sutures on the hiatal sling while the surgeon is mimicking diaphragmatic movements like those that arise on cough obviates any tension on the antireflux repair and gastric tear at follow-up. Regarding the latter point, the need for a reoperation for further gastric herniation or herniation of the colon alongside the fundoplication in 7 patients emphasizes the critical importance of the placement of a very large number of anchoring sutures on the hiatal sling (only one reoperation in the second half of the series). In addition, prevention of any sudden increase in abdominal pressure by smooth recovery from anesthesia and no strain under heavy loads at early follow-up may contribute to reducing the risk of transhiatal herniation of an abdominal organ [25].
The remaining concern with the intrathoracic periesophageal fundoplication is the potential of anti-inflammatory drugs to exert a harmful action on the fundus owing to the risk of entrapment of the pill into the mucosal folds. The use of anti-inflammatory medicines by 3 of our patients, although they had been warned against this practice, led to the development of a gastric ulcer in 2 and to fundic perforation in 1. As a consequence, intrathoracic fundoplication is contraindicated in GERD patients also suffering from chronic inflammatory disease.
Reflux-related short esophagus is difficult to manage surgically. For want of being the "gold standard" of treatment, the intrathoracic periesophageal fundoplication, because it is amazingly effective on reflux, constitutes a "gold option." However, strict observance of critical technical details, prevention of any sudden increase in abdominal pressure at early follow-up, and no intake of anti-inflammatory drugs are of utmost importance to make the procedure safe and effective.
| 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 |