| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2002;21:534-540
© 2002 Elsevier Science NL
Division of Thoracic Surgery, Dalhousie University, QEII Health Sciences Centre, Victoria Building 7S-013, 1275 Tower Road, Halifax, NS B3H 2Y9, Canada
Received 12 September 2001; received in revised form 6 December 2001; accepted 20 December 2001.
* Corresponding author. Tel.: +1-902-473-2281; fax: +1-902-473-4426
e-mail: thoracic{at}dal.ca
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
Key Words: Reoperative antireflux surgery • Collis gastroplasty
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
This study examined patterns of failure following primary antireflux surgery in a consecutive series of patients referred for reoperation after long-term medical therapy. Anticipating significant adhesions, complex hiatal and esophagogastric anatomy, the primary purpose of this study was to critically evaluate the results of reoperation using a left thoracoabdominal approach. As all patients in this series were found to have a short esophagus, the subjective results of Collis gastroplasty and selective fundoplication were assessed as a secondary objective.
|
2. Materials and methods |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
A single primary antireflux procedure had been performed at various centres (multiple surgeons) from 1 to 33 years (mean, 15 years) previously. Usually described as a ‘hiatus hernia repair’, review of the operative notes determined that these initial procedures were variations of a transabdominal (Nissen-type) fundoplication (21 patients), or a partial (Belsey-type) fundoplication performed by left thoracotomy (ten patients). No patient in this series had undergone laparoscopic antireflux surgery.
All patients had clinically disabling symptoms, comprising heartburn (28 patients, 90%), regurgitation (17 patients, 55%), dysphagia (16 patients, 52%), epigastric pain distinct from heartburn (nine patients, 29%), and pulmonary symptoms secondary to aspiration (six patients, 19%). Two patients (7%) presented with profound anemia, each having a hemoglobin of 4 g/l (normal range 140–180 g/l). The mean duration of symptoms prior to surgical referral was 60 months (range, 12–120 months), and despite maximal medical therapy, all patients indicated that their symptoms were worsening.
Objective studies included esophagogastroduodenoscopy (EGD) (all patients), barium swallow (30 patients, 97%), esophageal motility studies (26 patients, 85%) and ambulatory 24-h pH studies (17 patients, 55%). EGD (with selective biopsy) was performed preoperatively by the operating surgeon under local anesthesia using flexible fibreoptic instrumentation. The anatomic esophagogastric junction (where the tubular esophagus dilates into stomach) was measured in all patients and recorded as centimetres (cm) from the incisors. Barium contrast studies were used to define foregut anatomy. One patient who required urgent surgery for suspected ischaemia secondary to organoaxial volvulus did not have a preoperative barium swallow. Hiatus hernia was classified as: Type I (sliding, associated with an intrathoracic EGJ); II (paraesophageal, with an intraabdominal EGJ); III (mixed sliding and paraesophageal, associated with an intrathoracic EGJ); and IV (herniation of other intraabdominal contents). A short esophagus was therefore defined when the EGJ was found to be intrathoracic and irreducible, based on both endoscopic measurement and correlative barium swallow. Esophagitis was classified as: Grade I (mucosal erythema); II (erythema with superficial linear ulceration); III (deep, confluent or circumferential ulceration); and IV (stricture). A columnar epithelium-lined (Barrett's) esophagus was diagnosed endoscopically when the squamocolumnar junction extended greater than 3 cm proximal to the anatomic EGJ, or histologically by the presence of specialized intestinal epithelium (goblet cells) at any level.
Esophageal motility studies were used to evaluate peristalsis in the body of the esophagus (for planning fundoplication), to exclude associated primary esophageal motor disorders, and to define the lower esophageal sphincter (LES). These studies were not technically complete for two patients (the catheter could not be advanced beyond the intrathoracic stomach); two patients refused this investigation; and were not attempted for one patient who required urgent surgery.
Ambulatory 24-h pH studies were used to quantitate acid reflux, and were complete for only 17 (55%) patients. Of the 14 remaining patients, seven patients refused to have pH studies (inability to tolerate the naso-esophageal probe), the probe could not be positioned satisfactorily in six patients (inability to define the upper border of the LES using manometry), and one patient required urgent surgery.
Patients were selected for reoperation based on severity of symptoms, and objective studies as summarized above. Full informed consent was obtained. From January 1991, representing the first appointment of the senior author, the reoperative strategy has remained consistent for patients shown to have a short esophagus: (1) use of a left thoracoabdominal incision to obtain accurate anatomic exposure; (2) esophageal lengthening (Collis gastroplasty) to restore an intraabdominal EGJ; (3) selective fundoplication based on preoperative esophageal function studies; (4) reapproximation of the crura; and (5) postoperative epidural analgesia.
2.2. Operative technique
A thoracic epidural catheter was placed by the anesthetist. After induction of general anesthesia, the left lung was excluded by placement of a double-lumen endotracheal tube. Patients were positioned semi-lateral for a left thoracoabdominal incision, which extended antero-medially from a point 2 cm below the tip of the scapula, across the costal margin to the midline, to a point corresponding to the upper one-third between the xiphoid and umbilicus. The diaphragm was incised circumferentially preserving the left phrenic nerve, leaving at least a 2 cm margin adjacent to the chest wall for reconstruction. After division of the inferior pulmonary ligament, the thoracic esophagus was encircled above the level of the inferior pulmonary vein, taking care to identify and preserve the vagi. By simultaneous dissection in the upper abdomen and left chest, all adhesions were divided, any hernia sac excised, the diaphragmatic crura and hiatus clearly defined, and the previous fundoplication taken down to accurately identify the anatomic EGJ. In this series, all patients were confirmed to have a short esophagus at surgery, and therefore a Collis gastroplasty was created. A Maloney bougie, tailored to the esophageal diameter (usually 48–58 French), was placed by the anesthetist, and guided by the surgeon across the EGJ. While held firmly against the lesser curvature of the stomach, a cut gastroplasty was created by application of a mechanical stapling instrument (linear cutter, GIA). The staple line was oversewn with a running 3-0 continuous non-absorbable silk suture. The overall length of gastroplasty was variable, but a minimum length of 5 cm was usually required to position the neoesophagus in a tension-free, intraabdominal location.
For seven patients with non-propagated or significantly reduced esophageal peristalsis, a partial (270° Belsey-type) fundoplication was created with two to three tiers of interrupted 3-0 silk plicating sutures. For all remaining patients (including the five patients who did not have preoperative motility studies, and who did not have dysphagia clinically), a complete (360° Nissen-type) fundoplication was created. The fundoplication was 2 cm in length, incorporated both vagi, and was considered tension-free or ‘floppy’. With the bougie still in place, the crura were reapproximated posteriorly with interrupted non-absorbable No. 2 silk sutures. Pleural drainage tubes were positioned above the reapproximated left hemidiaphragm, adjacent to, but not touching, the upper margin of the gastroplasty.
The only modification to this procedure was reconstruction of the costal margin. Up to 1995, primary approximation of the costal margin was attempted, but subsequently, a 2 cm portion of costal cartilage was resected, and only the underlying musculature was closed.
Postoperative management comprised continuous nasogastric drainage to decompress the stomach, EKG monitoring for a minimum of 24 h, early ambulation and chest physiotherapy. A water soluble contrast study was performed routinely on the fifth to seventh postoperative day, and if no leakage from the gastroplasty was seen, the study was repeated using dilute barium. After removal of the nasogastric tube, patients were progressed from water to semi-solids over 2–3 days. If patients reported satisfactory swallowing with a soft diet at the first (3 week) follow-up visit, a regular diet was instituted.
2.3. Follow-up
Demographic data, including clinical presentation, results of objective studies, operative findings, surgical procedures, and postoperative complications, were collected prospectively. Follow-up was obtained by personal patient interview, phone interview or by contact with the primary care physician. A structured questionnaire was used to evaluate dysphagia (none, solids, semisolids, liquids), severity of reflux (episodes of heartburn, regurgitation, and medication use), weight, and duration and intensity of post-thoracotomy pain (mild if requiring occasional non-narcotic analgesia, moderate if requiring regular analgesia, and severe if requiring narcotics or intercostal nerve blocks). Global patient satisfaction with the outcome of surgery was assessed by a simple four-point subjective scale: very satisfied, generally satisfied, disappointed, and ‘would not have had surgery’.
With the exception of surveillance for Barrett's metaplasia (five patients), postoperative objective studies were not performed routinely. Selective investigations (EGD, 14 patients; barium swallow, five patients; motility studies, one patient) were performed to investigate symptoms if clinically indicated.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
|
There was no perioperative mortality. The median length of hospitalization was 8 days (range, 6–17 days), and was uncomplicated for 18 (58%) patients. Postoperative morbidity was minimal, and comprised left lower lobe infiltrates (six patients, 16%) diagnosed by chest radiography and treated by physiotherapy and antibiotics; atrial fibrillation (three patients, 10%) detected by routine EKG monitoring and treated by antiarrhythmic medication; urinary tract infection (one patient, 3%) treated by removal of the urinary catheter and antibiotics; a superficial wound infection (one patient, 3%) treated by opening a portion of the incision, saline dressings and oral antibiotics; aspiration (one patient, 3%) treated by chest physiotherapy; nausea and vomiting (one patient, 3%) which was thought to be medication-induced.
Two patients were lost to follow-up, which was otherwise complete for 29 patients to June 2001 (median 42 months; range, 6–105 months postoperatively). Overall, the majority of evaluable patients (27/29; 93%) were satisfied with the results of the surgery, Fig. 2 . One patient was disappointed with the surgical outcome in terms of reflux symptoms, and one patient ‘would not have had surgery’ because of postoperative pain.
|
Antacids (including proton pump blockers) were taken intermittently by all patients for the first 6 months postoperatively. At last follow-up, 20 (69%) patients had no reflux symptoms and were taking no antacids. However, nine (31%) patients still experienced occasional heartburn (reported to be significantly less than before surgery) which was controlled by simple antacids or by low-dose H2 blockers. Of the 19 patients who underwent EGD postoperatively, no patient was found to have esophagitis or a disrupted fundoplication.
Post-thoracotomy pain was prevalent for 3–6 months postoperative, but was judged to be moderate to severe in only six patients. Although post-thoracotomy pain had resolved for all patient up to 18 months, one patient who required repeated intercostal nerve block (local anaesthetic), indicated he ‘would not have had surgery’, despite satisfactory swallowing and improvement of reflux symptoms. Two (7%) patients reported the symptom of movement at the costal margin. A malunion was found in one patient whose costal margin was reapproximated initially by primary suture, and regrowth of cartilage was found in the other patient whose costal margin was initially excised. Both were treated by wide local re-excision of costal cartilage.
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
Associated anatomic findings in all patients were hiatus hernia, and a disrupted fundoplication. The latter is thought to account for recurrent reflux symptoms, and wrap disruption is reported to generally occur within the first postoperative year [7]. Although all our patients reported reflux-related symptoms, these did not generally occur within the 1st year postoperatively, suggesting a more complex etiology. The consistent finding of extensive adhesions, and in ten patients a mixed (Type III) hiatus hernia, would likely provide an anatomic basis for symptoms of epigastric pain and dysphagia.
Objective studies were essential to evaluate all patients preoperatively. We found EGD to be the most useful investigation to diagnose a short esophagus [12], based on preoperative measurements and with a correlative barium swallow. In contrast to recent reports, we did not find esophageal manometry useful in the diagnosis of a short esophagus [9,10]. Barium contrast studies were considered complimentary to EGD, especially to establish the diagnosis of a hiatus hernia, a disrupted fundoplication [7], and in planning the operative approach. In keeping with other studies, we found a high percentage of patients to have functional disorders of the foregut. Although technical factors and patient refusal limited routine use of ambulatory 24 h pH studies, 88% of patients evaluated had objective evidence of pathologic acid reflux. Motility studies similarly demonstrated motor disorders in the esophageal body of 36% of patients evaluated, and were used in planning the fundoplication.
The conventional approach favored by thoracic surgeons for reoperative antireflux surgery has generally been the left thoracotomy [3,4,6,8]. Extension of the incision into the abdomen is considered by many to be unnecessary, but if required, division of the diaphragm peripherally is recommended [6]. The left thoracoabdominal approach, or thoracolaparotomy, has traditionally been considered to be a painful incision, despite excellent exposure of the upper abdomen and chest [14,15]. This approach has been reported infrequently for reoperative antireflux surgery. In 1981, Henderson reported his personal experience with thoracoabdominal total fundoplication gastroplasty in 121 patients with recurrent hiatus hernia and gastroesophageal reflux, achieving excellent results in over 90% of patients [5]. Despite these impressive results in terms of swallowing and control of reflux, postthoracotomy pain, in the era before widespread use of epidural analgesia, was not discussed. In a 35 year review (1960–1995) of 185 patients who underwent reoperative antireflux surgery at the Mayo Clinic (multiple surgeons), only 25 patients were reoperated using a left thoracoabdominal incision [8]. The primary indication for this approach was to improve exposure for more extensive antireflux procedures, such as antrectomy or esophageal resection.
We report a consecutive series of 31 patients who underwent reoperative antireflux surgery using a left thoracoabdominal approach. This incision was primarily utilized to provide excellent anatomic exposure of the esophagus, stomach, vagi, and associated structures in what we considered to be technically challenging reoperative surgery. With epidural analgesia, this incision was well tolerated and facilitated early mobilization. Median length of hospital stay (8 days), and postoperative morbidity was minimal. Although postthoracotomy pain was prevalent after discharge, symptoms were comparable to patients undergoing thoracotomy alone, and in all patients, resolution by 18 months was usual. However, one patient who required intercostal nerve blocks indicated initially that he ‘would not have had surgery’, despite improved reflux symptoms and satisfactory swallowing.
The antireflux procedure consistently used comprised a cut Collis gastroplasty to lengthen the esophagus, and a selective fundoplication based on preoperative manometry. Although not the primary purpose of this study, we evaluated intermediate-term results of reoperative antireflux surgery (median 42 months follow-up) for 29 patients as a secondary objective. Subjectively, the majority of patients were satisfied with the results of reoperation. It is well recognized that subjective results do not necessarily correlate with objective findings, and clearly one limitation of this study was that only 19 of 29 evaluable patients had postoperative objective investigations, which were not performed routinely. In this series, nine (31%) patients required long-term antacid therapy for occasional heartburn, but this was easily controlled with simple antacids and low-dose H2 blockers. No patient had evidence of ongoing esophagitis, and the fundoplication was judged to be intact. Although the results of re-operation are generally reported to be less satisfactory, very few surgical studies have reported antacid use as an end point [16].
The most disturbing postoperative symptom was dysphagia, which was reported by five patients within the first postoperative month. One patient, who presented with an impacted food bolus above the fundoplication, improved following one esophageal dilation, suggesting that early postoperative edema was the etiology of the esophageal obstruction. It is likely that esophageal motor dysfunction was the primary etiology of dysphagia in the remaining four patients, as no anatomic obstruction was demonstrated objectively. As Nissen-type fundoplications were performed on two patients with minimal non-specific motility alterations (judged secondary to reflux), and on two additional patients who did not have preoperative esophageal manometry, it is likely that the complete fundoplication resulted in a functional obstruction at the neoesophagus. Dysphagia improved with further dilation (two to six times), and had resolved completely by 6 months postoperatively.
In addition to fundoplication, several alternative approaches (i.e. vagotomy and antrectomy, duodenal diversion procedures) have been reported to be useful for re-operative antireflux surgery in selected patients [8]. However, recent studies have reported the feasibility of laparoscopic approaches for reoperative antireflux surgery [17–20]. Minimally invasive approaches currently appear well suited to repair a disrupted fundoplication in patients with an adequate length of intraabdominal esophagus, to reduce a paraesophageal herniation and reapproximate the crura. Although it is anticipated that further technical advances will expand the role of laparoscopy or thoracoscopy in reoperative antireflux surgery, it is also likely that selected patients will still require ‘open’ surgery, particularly in the setting of failed anti-reflux procedures secondary to short esophagus. We report the thoracoabdominal approach, with epidural analgesia, to be safe, well tolerated, and to provide excellent exposure of the esophagogastric junction for complex reoperative antireflux surgery.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
Appendix A. Conference discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
Dr Legare: It is difficult a little bit to tell from this group because the majority of the patients were 10–12 years after the original operation, and most of the data we had from the original operation, because they came from other centers, were essentially the operative report and some comments from the surgeon. It was therefore difficult to trace whether truly they had a short esophagus. But you are correct this is an important point.
Dr A. Lerut (Leuven, Belgium): Thank you very much for taking up such a difficult subject. Certainly today with the vast majority of the patients having operations performed laparoscopically, it is getting increasingly difficult in redo surgery to convince patients to have a thoracic approach, and you rightly pointed out the problem of post-thoracotomy pain. You showed your two most representative examples of a hiatal hernia, but it is not always that easy to diagnose of a hiatal hernia and/or esophageal shortening. So what is the definition of hiatal hernia and in how far can you preoperatively judge that you have to go from the thoracic side or perhaps can go from an abdominal side? That is the first question.
Secondly, I think in your follow-up you showed a number of patients that were on PPI, and I haven't seen any 24-h pH study in the follow-up.
Finally, there were a number of patients that had Barrett's metaplasia, and doing a lengthening plasty in that subset of patients I think is putting the patients at risk for further development of dysplasia and recurrent reflux. So have you seen any dysplasias and how did you perform your follow-up afterwards?
Dr Legare: To answer your first question, basically the definition for a short esophagus we used two diagnostic tests. Firstly esophagoscopy measurements were taken from the incisors down to the level of the diaphragm and the esophagogastric junction, and this, obviously, in an awake patient. Whether it was reducible or not, it was always unreducible, and assessed in the patient if you asked them to talk and you could see where the diaphragm marking were. Secondly this was always documented at barium swallow; as you saw, 30 out of 31 patients had barium swallow. And thirdly, at intraoperative findings, all patients were found after dissection of the anatomy to have an unreducible hiatal hernia. But, you are right, it is sometimes difficult to judge preoperatively. But those were the two studies that we used preoperatively.
In terms of your second question, very few patients have follow-up studies in terms of pH studies, and you are right, we tend to follow clinically if patients have problems. In patients who have Barrett's findings, there were no metaplastic changes, however, they were followed yearly and they continue to be followed yearly from that point of view.
Regarding the Collis gastroplasty, as you know, it is, again, a debatable issue, but they continue to be followed in that setting, and we haven't had any need for resection or increased metaplasia from that point of view.
In terms of a pH study or follow-up studies, 19 out of the total patients had a follow-up esophagoscopy and all have shown an intra-abdominal segment of stomach. All patients also prior to starting eating, had a Gastrografin followed by barium swallow postoperatively, which showed again a lengthening procedure and intra-abdominal stomach segments.
|
References |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
This article has been cited by other articles:
|
|
 |
|
  J.-P. Avaro, X.-B. D'Journo, D. Trousse, M. A. Ouattara, C. Doddoli, R. Giudicelli, P. A. Fuentes, and P. A. Thomas Long-term results of redo gastro-esophageal reflux disease surgery Eur. J. Cardiothorac. Surg., June 1, 2008; 33(6): 1091 - 1095. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. B. Furnee, W. A. Draaisma, I. A. M. J. Broeders, A. J. P. M. Smout, and H. G. Gooszen Surgical Reintervention After Antireflux Surgery for Gastroesophageal Reflux Disease: A Prospective Cohort Study in 130 Patients Arch Surg, March 1, 2008; 143(3): 267 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. A. Khan, G. Kanellopoulos, M. L. Field, K. R. Knowles, F. D. Beggs, W. E. Morgan, and J. P. Duffy Redo antireflux surgery--the importance of a tailored approach Eur. J. Cardiothorac. Surg., November 1, 2004; 26(5): 875 - 880. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
