|
|
||||||||
Ann Thorac Surg 1995;60:915-920
© 1995 The Society of Thoracic Surgeons
Division of Thoracic Surgery, Department of Surgery, Université de Montréal, Hôtel-Dieu de Montréal, Montréal, Québec, Canada
| Abstract |
|---|
|
|
|---|
Methods. From 1990 through 1993, 27 consecutive patients (16 men, 11 women) underwent an uncut Collis-Nissen gastroplasty. Mean age was 59 years (range, 30 to 75 years). Three patients had a previous failed antireflux procedure. Indications for operation were gastroesophageal reflux disease resistant to medical treatment in 18 patients and symptomatic hiatal hernia in 9 patients. Fourteen patients had Barrett's esophagus and 4 had a peptic stricture. Complete esophageal function testing including barium swallow, endoscopy, manometry, and 24-hour pH recording was performed in 26 of 27 patients preoperatively and postoperatively.
Results. Five patients (19%) had complications, which included atelectasis in 2, cardiac dysrhythmia in 2, and prolonged ileus in 1. There were no operative deaths. Follow-up was complete in all patients and ranged from 8 to 45 months (mean, 22 months). Subjectively, symptoms of reflux were resolved in all patients. Six patients complain of slow esophageal emptying and 3 have occasional episodes of dysphagia. None required postoperative dilation. Ulcers and erosions healed in all 26 patients who underwent endoscopy but recurred in 2 at 21 and 36 months postoperatively. Mean lower esophageal sphincter gradient increased from 8.3 mm Hg preoperatively to 14.6 mm Hg (p = 0.0001). Total percent of acid exposure decreased from 8.0% preoperatively to 1.7% (p = 0.003).
Conclusions. We conclude that the uncut Collis-Nissen procedure provides acceptable short-term control of gastroesophageal reflux disease.
| Introduction |
|---|
|
|
|---|
Different antireflux operations have been designed for patients with complications of gastroesophageal reflux disease [1]. Standard antireflux repairs usually succeed in controlling mucosal damage caused by this condition. However, when esophageal pathology has resulted in peptic stricture and shortening of the esophagus, adequate control of reflux is more difficult when using a standard operation [2]. The difficulty in obtaining a sufficient length of intraabdominal esophagus has been proposed as the explanation for the failed control of reflux in this situation. The lengthening operation proposed by Collis in 1957 [3, 4] was modified by Pearson and associates in 1971 [5], who proposed to add a partial fundoplication to the gastroplasty. Henderson [6, 7] and Orringer and Sloan [8] subsequently proposed the use of a total fundoplication to better control the reflux.
Bingham [9, 10] and Demos [11, 12] and later Evangelist [13], Paris [14], Van Kemmel [15], Piehler [16], Payne [17], and their associates proposed to retain the benefits of a total fundoplication around a gastroplasty without transecting the gastric wall. They created a mucosal apposition of the anterior and posterior fundic walls by stapling them together and wrapping this gastroplasty with the remaining fundus. The reported results with this operation have suggested a control of reflux in 94% or more of the operated population [14, 16]. These results, however, were mostly subjective and lack the objectivity of functional and endoscopic reassessment after treatment.
Our aim in this work is to review the short-term results of the uncut Collis-Nissen gastroplasty using the most objective methods at our disposal to document the control of reflux disease.
| Material and Methods |
|---|
|
|
|---|
The 16 men and 11 women ranged in age from 30 to 75 years (mean age, 58.8 years), and their average height and weight were 164 cm and 58.6 kg, respectively. Follow-up ranged from 8 to 45 months (mean, 20 months). Three patients had previous failed antireflux procedures 13, 8 and 4 years before (3 Nissen fundoplications through an abdominal approach).
Indications for operation were gastroesophageal reflux disease with failure of medical antireflux therapy in 18 patients (Barrett's esophagus in 11 including 3 patients with associated peptic stricture, linear esophagitis with and without stricture in 1 and 5 patients, respectively, and gastroesophageal reflux without esophagitis producing symptoms refractory to medical management in 1 patient) and hiatal hernias in 9 patients (8 type III and 1 pure paraesophageal hernia after a previous antireflux procedure). None of the patients had evidence of a primary esophageal motor disorder, collagen vascular disease, neuromuscular disease, or chronic alcoholism. Two patients had diabetes.
Operative Technique
All patients underwent an uncut Collis-Nissen gastroplasty through a left thoracotomy above the eighth rib. The esophagus is dissected from the aortic arch to the diaphragmatic hiatus. The proximal stomach is dissected free through the hiatus when a large hernia exists or it is dissected through a peripheral diaphragmatic incision when no hernia or a small hiatus is present. The whole esophagogastric junction is delivered into the chest. Meticulous dissection of the junction as for a supraselective vagotomy is performed. The fat pad is removed. The anterior and posterior vagi are left intact but away from the gastric wall. A no. 50 Maloney bougie is passed into the stomach and held in place along the smaller curvature. A 3-cm stapler (Linear stapler 30 Proximate; Ethicon, Inc, Somerville, NJ) is then used, pushing the pin through the anterior and posterior gastric walls to ensure linear stapling. Both the anterior and posterior wall of the remaining fundus are brought around the created gastroplasty tube and fixed with four sutures in front of the staple line to cover it. The repair is reduced under the diaphragm and fixed in place by three sutures passing through esophageal wall, apex of the fundoplication, and diaphragm. The right and left crus of the diaphragm are reapproximated behind the esophagus (Fig 1
).
|
Endoscopy
A standard fiberoptic system (Pentax FG34JH) was used to assess the esophagus and gastroesophageal junction. Mucosal lesions were classified using the MUSE system (metaplasia, ulcer, stricture, erosion) as proposed by Armstrong and colleagues [18]. Mucosal damage was graded according to the increasing severity of metaplasia, ulcers, stricture, and erosion, with a score of from 0 to 3 assigned for each of these aspects (Fig 2
). A single patient refused to have the postoperative endoscopic evaluation. Barrett's esophagus was defined as the presence of 3 cm or more of columnar epithelium in the distal esophagus or any biopsy showing incomplete intestinal metaplasia (specialized epithelium) in the distal esophagus despite its length.
|
Ten swallows of a 2-mL water bolus were recorded in the proximal and distal half of the esophagus. The esophageal resting pressure, the peak of contraction, and the type of contractions (primary peristalsis or tertiary waves) were assessed in both the proximal and distal esophagus. The lower esophageal sphincter resting pressure, closing pressure, and percentage of relaxation were assessed in the same fashion using a station pull-through technique in the high-pressure zone. When the intragastric pressures are substracted from the intrasphincteric pressures, this yields lower esophageal sphincter gradient pressures between esophagus and the stomach.
24-Hour pH Recording
A Sandhill ambulatory pH recorder (Sandhill, Littleton, CO) was used to register reflux events during 24-hour periods. An antimony electrode was placed 5 cm above the gastroesophageal junction, which had been identified manometrically. The total number of reflux episodes, the number of reflux episodes lasting more than 5 minutes, the number of minutes of reflux, and the percentage of time of exposure to acid were computed. Recordings were obtained in 26 patients before and in all 27 patients after their operation.
Statistical Analysis
A two-tailed Student's t test for paired continuous values and McNemar analysis for discontinuous values were used when appropriate. A p value of less than 0.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
|
Radiology
The radiologic observations are described in Table 2
. Hiatal hernias were observed in 19 patients. Ten were type I hernias. There was one paraesophageal hernia through the hiatus occurring after a Nissen fundoplication. Eight patients showed a type III hernia. Barrett's esophagus was later confirmed in 6 of 10 type I hernias and in one type III hernia, and had already been documented in the only iatrogenic paraesophageal hiatal hernia.
|
Initially 1 patient with a type I hernia showed esophageal stasis. After the operation 8 patients showed evidence of delayed barium emptying on the esophagogram. In 15 patients spontaneous gastroesophageal reflux was observed during the preoperative examination. In 2 patients gastroesophageal reflux was seen radiologically after the operation. Neither of these 2 patients had objective evidence of reflux.
Mucosal damage such as erosions and ulcers were suggested in 8 and 5 patients, respectively. Endoscopy confirmed the erosions in 4 of 8 patients, whereas ulcers were confirmed in 2 of 5 patients. Barrett's esophagus was suspected in 3 patients (1 with radiologically short esophagus plus a fine ulcer and 2 patients with short esophagus and stricture). Barrett's esophagus was endoscopically confirmed in the 2 patients with a short esophagus and stricture and in 12 additional patients.
Endoscopy
The endoscopic findings are outlined in Table 3
. Twelve of the 27 patients exhibited preoperatively a columnar-lined mucosa (11 circumferential and 1 with finger-like projections). Postoperative endoscopic assessment revealed that 14 of the 27 patients had Barrett's metaplasia. These 2 additional patients were underassessed, being interpreted as having a type III hiatal hernia for 1 and circumferential erosions in the second. Biopsies in all 14 patients showed fundic and specialized type epithelium without any evidence of epithelial dysplasia. The preoperative columnar metaplasia zone showed a mean length of 6.16 cm. The first endoscopic assessment after the operation showed a mean length of 4.57 cm. All these patients are being followed up in a surveillance program.
|
There were four strictures. Three of the four strictures showed associated Barrett's metaplasia. One patient with a 5-mm-diameter stricture required repeated preoperative guidewire dilations over a period of 1 year. Postoperatively, three more dilation sessions were required. Currently, the stricture has improved to a diameter greater than 9.5 mm as measured by free passage of the endoscope. The three other strictures have regressed.
In 1 patient partial dehiscence of the proximal gastroplasty tube was documented at the first endoscopic reassessment after 15 months. He remains asymptomatic and without objective evidence of reflux.
Manometry
Table 4
details the preoperative and postoperative manometric values. The esophageal resting pressures in the proximal and distal esophagus did not change after the operation. The peak contraction pressures were similar in the proximal and distal esophagus and did not change after the total fundoplication uncut gastroplasty. Both in the proximal and distal esophagus, there is an increase in tertiary activity after voluntary swallows. This was considered nonsignificant.
|
24-Hour pH Recordings
Details of the pH recordings are given in Table 5
. In the 27 patients who underwent 24-hour pH assessment after their total fundoplication gastroplasty, the number of reflux episodes, the number of episodes lasting more than 5 minutes, the minutes of acid exposure, and the percentage of exposure time to acid were all found to be significantly decreased.
|
| Comment |
|---|
|
|
|---|
The technique has the combined advantages of lengthening the distal esophagus while providing an ``anchor'' for the fundoplication to reduce the incidence of anatomic hernia recurrence or slipping of the esophagus out of the wrap [16]. Mucosal apposition using the uncut technique raises the potential of dehiscence of the gastroplasty. Because mucosal apposition results in a nonhealing process, the sutures of the stapled line, either by progressively working their way through the gastric wall or by becoming undone, would result in the loss of anterior and posterior wall apposition. The disappearance of the gastroplasty could lead theoretically to an increased incidence of recurrent hernia or reflux. This in practice has not been a significant problem. Of 80 patients operated on for an intrathoracic stomach using a transthoracic uncut Collis-Nissen [19], only 1 patient (1.3%) required reoperation for a recurrent hernia 86 months postoperatively. In another series of 101 reflux-induced esophageal stenoses [17], Payne reported excellent to good functional results in 85% of a group of 55 patients treated with a transthoracic uncut Collis-Nissen. Leak at the staple line is another potential area of concern. A confined extravasation of contrast medium occurred in 3.9% in the intrathoracic stomach group [19] and 5% in the reflux-induced esophageal stenoses group [17]. All did heal well with conservative measures and resumed oral diet by the time of hospital dismissal.
Bingham [10] in 1977 reported his personal experience with 138 patients. Clinical and radiographic follow-up ranged from 1 to 3 years for 110 patients. Three patients were interpreted as presenting symptoms of recurrent reflux. Disruption of the gastroplasty was observed in 8 patients (7%). In 18 additional patients the complete plication wrapped around the gastroplasty lost its radiologic appearance. One of our patients showed a partial dehiscence of the proximal gastroplasty at endoscopy 14 months after the operation. He remains asymptomatic with disappearance of esophagitis and objective evidence of reflux control. This was due to a technical problem resulting from misalignment of the linear stapler when the pin was not pushed through the stomach wall. This complication has not been reported by other authors using this technique.
Demos [12] reported on 82 patients, 80 of whom were followed up for a period of 2 to 12 years. There were no deaths. The patients remained asymptomatic and there was no anatomic recurrence.
Paris and associates [14] reported on 48 patients who had total fundoplication gastroplasties. Thirty-three had uncut Collis-Nissen gastroplasties and 15 had a cut Collis-Nissen operation as described by Orringer and Sloan [8]. Without comparing the results of these two techniques, they have reported excellent symptom control for the total group.
Piehler and colleagues [16] reported the Mayo Clinic experience in the evaluation of 136 patients over a 4.5-year period. During the follow-up 56.8% of the patients had esophagograms, 23.1% had endoscopy, and 25.4% had esophageal motility studies. Symptoms of gastroesophageal reflux disease dissapeared in 85.5%. The results were considered highly acceptable for 94% of the group. Evangelist and associates [13] reported on 48 patients with an undivided gastroplasty tube with added total fundoplication created by using a modified GIA stapler. Their average follow-up was 27 months. All patients are reported as having complete clinical control of reflux symptoms.
Poor results after an antireflux operation are due either to inadequate reflux control or to side effects resulting from the technique. Dysphagia is the most frequently reported side effect of both the cut and uncut forms of Collis-Nissen gastroplasty. The grading of dysphagia remains difficult to establish from the existing literature. Demos [12] and Paris and associates [14] found ``transient mild dysphagia'' to be the most frequent immediate postoperative complication, although it usually subsided by the seventh or eighth week after the operation. Bingham [10] and Evangelist and associates [13] did not report the exact prevalence of dysphagia during the postoperative follow-up. Piehler and colleagues [16] reported dysphagia in 40% of their patients (sporadic in 29%, at least once a week in 6.9%, and daily in 3.8%). This observation led them to reduce the length of the gastroplasty from 5 to 3 cm.
Stirling and Orringer [20], reporting on 261 patients who had a cut Collis-Nissen gastroplasty, reported dysphagia in 17% of their patient population. These patients required regular (9%) or occasional (8%) dilations. Persistent dysphagia was more common in patients with a complicated hernia or reflux problem (type III hernias, stricture, previous operation, scleroderma). Stirling and Orringer also reported mild dysphagia not requiring dilation in an additional 25% of their patients. This led them to reduce their total fundoplication from 6 to 3 cm around their gastroplasty.
After creating a total fundoplication of 3 cm around an uncut gastroplasty we observed 11 patients retaining some form of dysphagia. For 8 patients this was a slow emptying sensation; this was evident in 6 patients after 12 months. One patient had a hard stricture, which improved slowly after postoperative dilations. Two more patients have intermittent episodes of dysphagia. These observations led us to conclude that an uncut Collis-Nissen gastroplasty creates significant resistance to emptying at the distal end of the esophagus at least during the initial year after the operation.
As for other antireflux repairs, our manometric studies show a significant increase in the high-pressure zone of the esophagogastric junction after the uncut Collis-Nissen operation. There was no change observed in the contraction amplitude of the esophageal body. Demos [12] reported a reduction in the incidence of tertiary esophageal contractions from 74% before the operation to 18% after operation. We observed a slight increase in tertiary waves, although it was not considered significant. This might be explained by increased resistance to esophageal emptying by the operation.
Endoscopy and 24-hour pH recordings showed definite improvements in mucosal damage, whereas a significant decrease in acid exposure was recorded. Still, 2 patients showed recurrent linear erosions on their esophageal mucosa, 1 of them with an abnormal 24-hour pH study (10%). Two patients revealed abnormal acid exposure (12% and 6%) without symptoms and without mucosal lesions. Barrett's esophagus remains underdiagnosed in the initial assessment of hiatal hernias and esophagitis patients [21, 22]. This metaplasia is possibly reduced in length by effects of the operation itself. Otherwise it is not modified by the disappearance of reflux.
In conclusion, we think it is well established that on short-term follow-up the uncut Collis-Nissen gastroplasty affords good protection against reflux disease and allows complete healing of ulcerative and erosive lesions in the esophagus. When present, a columnar-lined esophagus remains unchanged.
| Acknowledgments |
|---|
|
|
|---|
| Footnotes |
|---|
|
|
|---|
Address reprint requests to Dr Duranceau, Department of Surgery, Hôtel-Dieu de Montréal, 3840, St-Urbain, Montréal, Québec, Canada H2W 1T8.
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
L.-Q. Chen, D. Nastos, C.-Y. Hu, T. S. Chughtai, R. Taillefer, P. Ferraro, and A. C. Duranceau Results of the Collis-Nissen gastroplasty in patients with Barrett’s esophagus Ann. Thorac. Surg., September 1, 1999; 68(3): 1014 - 1020. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. F. Trastek, C. Deschamps, M. S. Allen, D. L. Miller, P. C. Pairolero, and A. M. Thompson Uncut Collis-Nissen fundoplication: learning curve and long-term results Ann. Thorac. Surg., November 1, 1998; 66(5): 1739 - 1744. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. P. Ritter, J. H. Peters, T. R. DeMeester, M. Gadenstatter, S. Oberg, M. Fein, J. A. Hagen, P. F. Crookes, and C. G. Bremner Treatment of Advanced Gastroesophageal Reflux Disease With Collis Gastroplasty and Belsey Partial Fundoplication Arch Surg, May 1, 1998; 133(5): 523 - 529. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Deschamps, V. F. Trastek, M. S. Allen, P. C. Pairolero, J. O. Johnson, and D. R. Larson LONG-TERM RESULTS AFTER REOPERATION FOR FAILED ANTIREFLUX PROCEDURES J. Thorac. Cardiovasc. Surg., March 1, 1997; 113(3): 545 - 551. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |