- Email: [email protected]
Endoscopic stapling of pharyngeal pouch: A 10-year review of single versus multiple staple rows
Otolaryngology–Head and Neck Surgery (2009) 140, 245-249
ORIGINAL RESEARCH–GENERAL OTOLARYNGOLOGY
Endoscopic stapling of pharyngeal pouch: A 10-year review of single versus multiple staple rows Jason A. Roth, BSc (Med), MBBS, Elizabeth Sigston, MBBS, and Neil Vallance, MBBS, Melbourne, Australia OBJECTIVE: To compare the outcomes obtained in patients undergoing endoscopic stapling of pharyngeal pouches with single versus multiple rows of staples. STUDY DESIGN: A retrospective, 10-year review. SUBJECTS AND METHODS: Review of medical records in 38 patients who underwent endoscopic pharyngeal pouch repair. RESULTS: Patients who underwent stapling with multiple rows had a higher postoperative leak rate than patients who were stapled with a single row (36% vs 0%, P ⬍ 0.05). Patients with multiple rows also had a more prolonged length of stay and a slower return to both clear fluids and solid diet (P ⬍ 0.05). There was no difference in recurrence rate or patient satisfaction between the two groups. CONCLUSION: The technique of endoscopic pharyngeal pouch stapling has the potential to achieve excellent results. The application of more than one row of staples may be necessary to divide the common wall. However, in our series this is associated with a significantly increased risk of esophageal or pouch perforation. Care should be taken during the placement of multiple rows of staples. No sponsorships or competing interests have been disclosed for this article. © 2009 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved.
Z
enker diverticulum is a protrusion of the pharyngeal mucosa through the Killian triangle, a weak area between the inferior constrictor and cricopharyngeus.1 Management of these diverticulae has undergone an evolution from open procedures to less invasive endoscopic procedures over the course of the 20th century. Endoscopic management was first described by Mosher2 in 1917. However, following initial success, he subsequently abandoned the approach owing to a significant incidence of complications. Dohlman and Mattson3 reintroduced and popularized the technique in 1960, describing electrocautery for division of the common wall. In 1993, Collard4 presented the technique of endoscopic stapling of the diverticulum. Collard’s approach solved early problems associated with lack of a seal along the divided mucosal edges, with subsequent salivary leak and mediastinitis. The most recent evidence suggests that endoscopic stapling is superior to open tech-
niques in the management of this condition.5,6 Compared with open techniques, the procedure is minimally invasive, has a reduced operative time, a lower rate of complications, and shorter hospital stay.5,6 It also has excellent rates of symptom resolution and a low rate of recurrence.5,6 A single row of endoscopic staples may not be sufficient to completely divide the common wall of the pharyngeal pouch. On observing this, a second or third row can be applied in series with the first row to obtain a more complete division. During subsequent staplings the surgeon aims to leave the more proximal stapled mucosal edges undisturbed and not advance the stapler too far into the pouch, causing trauma to its distal end. Although this procedure is performed infrequently in our unit, it was observed that, when multiple rows of staples were required to obtain complete division of the dividing wall, there was an increased incidence of postoperative complications and a prolonged hospital stay. To ascertain if this observation was statistically significant, we designed a study to compare patients who had undergone pharyngeal stapling with either a single row of staples or multiple rows of staples. Our hypothesis was that patients undergoing endoscopic pharyngeal pouch stapling with multiple rows of staples would have a higher complication rate and prolonged hospital stay compared with patients in whom a single row of staples was placed.
METHODS A retrospective study was conducted of all patients who underwent endoscopic pharyngeal pouch surgery between July 1, 1997 and December 1, 2007 in the Southern Health network, Melbourne, Australia. Approval for the review was granted by the Southern Health Human Research Ethics Committee. Funding was obtained from an otolaryngology departmental research fund. Patients were identified by operation code from a computerized database with the earliest point of review being
Received May 23, 2008; revised October 7, 2008; accepted October 22, 2008.
0194-5998/$36.00 © 2009 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved. doi:10.1016/j.otohns.2008.10.035
246
Otolaryngology–Head and Neck Surgery, Vol 140, No 2, February 2009
the date from which the database began. A spreadsheet was designed to record patient baseline demographics and outcomes of interest. Patient files were then individually reviewed to obtain the data. Primary outcomes of interest were number of rows of staples applied to the bridging wall, length of stay, and complications and their management. The operative technique was analyzed in comparison with Collard’s4 initial approach and any modifications noted. Secondary outcomes of interest were operating time, use of nasogastric tube, time to clear fluid diet, time to normal diet, and postoperative instructions. Postoperative patient satisfaction with the surgery was recorded as either full or partial resolution of symptoms. A patient was recorded as having full resolution of symptoms if they no longer complained of any dysphagia or regurgitation at their postoperative visits. A partial resolution was recorded if the patient was still symptomatic. Patients who had partial symptom resolution at the first visit who subsequently became asymptomatic were recorded as having full resolution of symptoms. Recurrence of symptoms at a later date after full symptom resolution and whether these patients required a repeat procedure were recorded. All analyses were performed by a statistician using SAS version 8.2. (SAS Institute Inc, Cary, NC). Tests of significance for categorical data were obtained with the Fisher exact test. Nonparametric data were analyzed with Wilcoxon rank sum tests.
RESULTS Thirty-eight patients who had undergone endoscopic pharyngeal pouch surgery during the period of review were identified. Twenty-seven patients had undergone procedures in which a single row of staples was applied to the common wall. Eleven patients had more than one row of staples applied to the common wall. Of these patients, nine patients had two rows of staples applied and three patients had three rows of staples applied. The mean patient age in the single-row group was 74 years compared with a mean age of 68 years in the multiplerow group. There were twice as many males compared with females in the single-row group compared with the multiple-row group. There was no difference in the timing of operation dates in the two groups. None of these results showed any statistically significant difference between the two groups. Each procedure was undertaken or supervised by one of the nine consultant surgeons who worked in the unit. A registrar or fellow was listed as the primary operator in six of the cases with a single row of staples and in two of the cases with multiple rows. There were no complications in the group of patients for whom the registrar was listed as the primary operator. All patients underwent surgery within a single tertiary institution. Patients underwent general anesthesia, a dental
guard was used, and then a Weerda (Karl Stortz, Tuttlingen, Germany) bivalved laryngoscope was introduced and used to expose the common wall between the esophagus and diverticulum. The longer lip was placed in the esophagus and the short one in the diverticulum. Photodocumentation was achieved with the use of 0- or 30-degree telescopes connected to a camera and recording device. The diverticulum was then examined for any suspicious lesions before proceeding. The Endo-GIA 30 (TYCO Health Care) stapler blade with 2.5-mm staples was then introduced over the common wall and its position checked with a telescope. After applying the stapler and removing it, the length of the residual common wall was assessed and a decision made as to whether a second or third application of staples would be necessary to completely divide the common wall. No particular algorithm appears to have been followed by the operating surgeons as to what length of residual common wall required a further row of staples. The telescope was then used to examine the esophageal and diverticular wall for any evidence of perforation or loose staples. The mean operating time overall was 31 minutes. There were no reported intraoperative complications in either group. A nasogastric tube was inserted in 30 percent of patients in the single row group and 46 percent of patients in the multiple rows group. These differences were not statistically significant. There was some variability in the postoperative instructions between groups; however, none of these results were significant. All patients were sent to recovery with instructions to assess for fever, chest or back pain, hemoptysis, tachycardia, or respiratory distress. Patients in the singlerow group were given a clear fluid diet before a solid diet in 67 percent compared with 82 percent in the multiple row group. Twenty-one percent of patients were ordered a routine postoperative chest X-ray. Five percent of patients were booked a routine gastrograffin swallow. The mean length of stay was 2.1 days (range 1-10 days) in the single-row group and 7.1 days (range 1-29 days) in the multiple-row group, a difference of 5 days between groups (P ⫽ 0.017). In the single-row group, two patients had a length of stay greater than 3 days. One patient stayed 10 days to undergo treatment for other unrelated medical problems. The other patient stayed 7 days for investigation of a suspected perioperative myocardial infarction. The mean time until a clear fluid diet was achieved was 1.1 days (range 0-5 days) in the single-row group and 5.5 days (range 1-22 days) in the multiple-row group (P ⫽ 0.003). The mean time until normal diet was achieved was 1.5 days (range 1-5 days) in the single-row group and 5.8 days (range 1-25 days) in the multiple row group (P ⫽ 0.024). There were four cases of postoperative pouch or esophageal perforation. These all occurred in patients who had multiple rows of staples applied to divide the common wall. No cases occurred in patients who had a single row of
Roth et al
Endoscopic stapling of pharyngeal pouch: A 10-year . . .
staples applied. This equates to a statistically significant leak rate of 36 percent in the multiple-row group compared with zero percent in the single-row group (P ⫽ 0.004). No perforation was noted intraoperatively in any of the cases that had a postoperative leak. There was no significant difference between the leak group and the overall patient group in terms of age, sex, surgeon, use of nasogastric tube, or operation time. However, patients who had multiple staple rows had larger pouch sizes (based on surgeon intraoperative description) than those who had a single row applied (P ⫽ 0.041). The management of the four patients who had leaks postoperatively varied. Two of the patients were managed conservatively with nothing by mouth and a nasogastric tube. They were discharged home after 6 and 14 days, respectively. One patient was taken back to the operating theater for open closure of the perforation and was discharged on day 9. One patient developed an empyema and required a thoracotomy. These two patients were subsequently discharged after 22 days. Length of follow-up was similar in both groups. The recurrence rate was 29.6 percent in the single-row group and 18.2 percent in the multiple-row group. This difference was not significant. There was no significant difference in patient satisfaction between groups.
DISCUSSION The results obtained in this study suggest that in our unit, endoscopic pharyngeal pouch surgery in which more than one row of staples is applied to the common wall is associated significantly with an increased risk of esophageal or pouch perforation. Patients who had more than one row of staples applied took significantly longer to tolerate both clear fluid and normal diets, and had a longer length of stay. Patient satisfaction and recurrence rates were similar in both groups. These data highlight the increased risk associated with the placement of subsequent rows of staples and questions the benefit of placing more than one row of staples in this procedure. A number of mechanisms may explain why multiple applications of the Endo-GIA stapler are associated with an increased risk of perforation. After the first row of staples is introduced, the tension sutures lose their retraction ability and it may be difficult to apply the second and subsequent staple rows in precisely the correct plane. There is a risk of mucosal laceration each time the endostapler is introduced.7 When the anvil is inserted into the diverticulum it may cause a perforation by pushing too far into the distal end of the pouch.8 Approximation of the common wall mucosal edges may also fail because of excessive width of the common wall or technical failure of the staple gun, or because of too much tissue being divided with insufficient stapling.9,10 It is unclear to what extent the common wall has to be divided to achieve complete symptom resolution and min-
247
imize the risk of recurrence. Kaoy and Baates11 have suggested that deciding the length of common wall to divide is a matter of judgment and is probably related to experience. They suggest that it is safer to under-divide rather than risk a perforation. A case report by Arunachalam and Cameron12 cautions against over-stapling after a patient developed a persistent foreign body sensation and pharyngeal pain from a clump of staples. Hilton and Brightwell9 described a case of a defect in the posterior wall of the esophagus that occurred after a single row of endoscopic stapling owing to trauma from the end of the stapling gun. Other papers describing endoscopic stapling of pharyngeal pouches have suggested that if two rows of staples could not be applied, then there was a significant chance of incomplete symptom resolution. Collard4 suggested that stapling should be abandoned if the tissue bridge was not large enough to allow two rows of cartridges. Koay and Bates11 described residual symptoms in two of five patients who had a single row of staples applied. There is evidence that stapling of smaller pouches has a higher complication rate. de Casso et al5 described a leak rate of 14.6 percent with 57 percent of leaks occurring in patients who had small pouches. It may be that in small pouches either trauma from the stapler or the knife cutting too close to the end of the tissue bridge may be causing a higher rate of perforation. Patients who had multiple rows of staples applied to their pouch had larger pouch sizes according to surgeon intraoperative description. It could therefore be argued that the higher complication rate is a consequence of having a large pouch rather than of the stapling procedure itself. There is no previous reported evidence that larger pouches when stapled have higher complications. Applying the stapler to a larger pouch is often easier in the first instance owing to the increased room to maneuver the stapler in the pouch lumen. It seems most likely that the stapling itself rather than simply the increased size of the pouch is the risk factor for postoperative leaks. It is not clear whether the insertion of a nasogastric tube following this procedure is beneficial. A number of series describe its use as “optional” and have inserted them in only early cases.11,13,14 In our series there was no statistically significant benefit associated with insertion. However, it may be beneficial to place a nasogastric tube in patients who are at higher risk of perforation such as when multiple rows of staples are placed. Richtsmeier and Monzon15 examined the use of alternative endoscopic staple devices in this procedure after two patients leaked after an Endopath ETS was use to insert 3.5-mm staples. The authors suggested that the ETS45 stapler with a 2.5-mm cartridge may offer an improved seal at the distal end of the incision compared with the same device with a 3.5-mm cartridge. They discuss that this device is also 15 mm longer than an EndoGIA-30, allowing a longer cut and fewer stapler firings. However, this series consisted of just 14 patients, two of whom leaked, and the
248
Otolaryngology–Head and Neck Surgery, Vol 140, No 2, February 2009
EndoGIA-30 remains the most widely accepted device currently in use. Pharyngeal pouch stapling is a procedure potentially associated with a high morbidity and mortality rate. A number of publications have supported the belief that there is a significant learning curve associated with the procedure.5,16-18 The report of the National Confidential Enquiry into Peri-operative Deaths commissioned by the Royal College of Surgeons of England16 recommends that it should be undertaken by one otolaryngologist in each center who has developed a special interest in it. Southern Health is one of Australia’s largest area health networks and services a large population. However, this procedure was performed only 38 times over approximately 10 years. Dividing these cases between nine consultant surgeons equates to approximately one case every 2 to 3 years. The leak rate of 36 percent is higher than leak rates published in other larger series (range 0%-10%)11 and suggests that the infrequency with which our unit is performing this procedure may be contributing to a higher rate of complications. There is good evidence to suggest that residual or recurrent symptomatic pouches can be treated endoscopically without any increase in perioperative morbidity or mortality.19 This finding should be kept in mind when attempting to achieve complete division of the common wall. If the patient has only partial symptom resolution or recurrence from the degree of stapling performed, a further procedure can be safely performed.
CONCLUSION Patients undergoing endoscopic stapling of pharyngeal pouches with more than one row of staples may be at increased risk of postoperative complications including esophageal perforation and leak. These patients have a more prolonged postoperative hospital admission and take longer to resume either a clear fluid or normal diet. Care should be taken during the insertion of staples, and surgeons should be aware of the risks associated with the placement of additional rows.
ACKNOWLEDGEMENTS Michael Bailey, PhD, MSc, BSc(hons), Department of Epidemiology and Preventive Medicine, Faculty of Medicine, Nursing and Health Sciences, Monash University: assistance in performing the statistical analysis of our data.
AUTHOR INFORMATION From the Department of Otolaryngology, Head and Neck Surgery, Southern Health, Melbourne (Drs Roth, Sigston, and Vallance); and the Department of Surgery, (Monash Medical Centre), Faculty of Medicine, Monash University (Drs Sigston and Vallance), East Bentleigh.
Corresponding author: Jason Roth, Department of Otolaryngology, Head and Neck Surgery, Monash Medical Centre Moorabbin Campus, Centre Road, East Bentleigh VIC 3165, Australia. E-mail address: [email protected]. Data from this paper were presented at the Australian Society of Otolaryngology Head and Neck Surgery Scientific Meeting, Perth, Western Australia, April 8-11, 2008.
AUTHOR CONTRIBUTIONS Jason A Roth: data collection, manuscript preparation; Elizabeth Sigston: review of manuscript; Neil Vallance: study design.
FINANCIAL DISCLOSURE None.
REFERENCES 1. Killian G. La boudre de l’oesophage. Annales Mal Orelle Larynx 1907;34:1. 2. Mosher HP. Webs and pouches of the esophagus: their diagnosis and treatment. Surg Gynecol Obstet 1917;25:175– 87. 3. Dohlman G, Mattson O. The endoscopic operation for hypopharyngeal diverticula. Arch Otolaryngol 1960;71:744 –52. 4. Collard JM, Otte JB, Kestens PJ. Endoscopic stapling technique of esophagodiverticulostomy for Zenker’s diverticulum. Ann Thoracic Surg 1993;56:573– 6. 5. Casso C, Lalam M, Ghosh S, et al. Endoscopic stapling diverticulotomy: an audit of difficulties, outcome and patient satisfaction. Otolaryngol Head Neck Sur 2006;134:288 –93. 6. Sen P, Bhattacharyya AK. Endoscopic stapling of pharyngeal pouch. J Laryngol Otol 2004;118:601– 6. 7. Scher RL, Richtsmeier WJ. Long-term experience with endoscopic staple-assisted esophagodiverticulostomy for Zenker’s diverticulum. Laryngoscope 1998;108:200 –5. 8. Koay CB, Commins D, Bates GJ. The role of endoscopic stapling diverticulotomy in recurrent pharyngeal pouch. J Laryngol Otol 1998; 112:954 –5. 9. Hilton M. Esophageal perforation during stapling of a pharyngeal pouch: adverse clinical incident report. J Laryngol Otol 2000;114: 549 –50. 10. Kubba H, Hussain SS. Stapler failure in pharyngeal diverticulectomy: a suggested modification in surgical technique. J R Coll Surg Edinb 1998;43:31–2. 11. Koay CB, Bates GJ. Endoscopic stapling diverticulotomy for pharyngeal pouch. Clin Otolaryngol 1996;21:371– 6. 12. Arunachalam PS, Cameron DS. Persistent foreign body sensation and pharyngeal pain due to retention of staples: an interesting sequelae of endoscopic stapling procedure. J Laryngol Otol 2001;115:425–7. 13. Narne S, Cutrone C, Bonavina L, et al. Endoscopic diverticulotomy for the treatment of Zenker’s diverticulum: results in 102 patients with staple-assisted endoscopy. Ann Otol Rhinol Laryngol 1999;108: 810 –5. 14. Scher Rl, Richtmeier WJ. Long-term experience with endoscopic staple-assisted esophagodiverticulostomy for Zenker’s diverticulum. Laryngoscope 1998;108:200 –5.
Roth et al
Endoscopic stapling of pharyngeal pouch: A 10-year . . .
15. Richtsmeier WJ, Monzon JR. Postendoscopic Zenker esophagodiverticulostomy leaks associated with a specific stapler cartridge. Arch Otolaryngol Head Neck Surg 2002;128:137– 40. 16. Resouly A. Pharyngeal pouch surgery. In: Royal College of Surgeons for England, editors. The report of the national confidential enquiry into perioperative deaths 1996/1997. London: Royal College of Surgeons of England; 1998. p 32.
249
17. Thaler ER, Weber RS, Goldberg AN, et al. Feasibility and outcome of endoscopic staple-assisted esophagodiverticulostomy for Zenker’s diverticulum. Laryngoscope 2001;111:1506 – 8. 18. Mirza S, Dutt SN, Minhas SS, et al. A retrospective review of pharyngeal pouch surgery in 56 patients. Ann R Coll Surg Engl 2002;84:247–51. 19. Scher RL. Endoscopic staple diverticulostomy for recurrent Zenker’s diverticulum. Laryngoscope 2003;113:63–7.
ORIGINAL RESEARCH–GENERAL OTOLARYNGOLOGY
Endoscopic stapling of pharyngeal pouch: A 10-year review of single versus multiple staple rows Jason A. Roth, BSc (Med), MBBS, Elizabeth Sigston, MBBS, and Neil Vallance, MBBS, Melbourne, Australia OBJECTIVE: To compare the outcomes obtained in patients undergoing endoscopic stapling of pharyngeal pouches with single versus multiple rows of staples. STUDY DESIGN: A retrospective, 10-year review. SUBJECTS AND METHODS: Review of medical records in 38 patients who underwent endoscopic pharyngeal pouch repair. RESULTS: Patients who underwent stapling with multiple rows had a higher postoperative leak rate than patients who were stapled with a single row (36% vs 0%, P ⬍ 0.05). Patients with multiple rows also had a more prolonged length of stay and a slower return to both clear fluids and solid diet (P ⬍ 0.05). There was no difference in recurrence rate or patient satisfaction between the two groups. CONCLUSION: The technique of endoscopic pharyngeal pouch stapling has the potential to achieve excellent results. The application of more than one row of staples may be necessary to divide the common wall. However, in our series this is associated with a significantly increased risk of esophageal or pouch perforation. Care should be taken during the placement of multiple rows of staples. No sponsorships or competing interests have been disclosed for this article. © 2009 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved.
Z
enker diverticulum is a protrusion of the pharyngeal mucosa through the Killian triangle, a weak area between the inferior constrictor and cricopharyngeus.1 Management of these diverticulae has undergone an evolution from open procedures to less invasive endoscopic procedures over the course of the 20th century. Endoscopic management was first described by Mosher2 in 1917. However, following initial success, he subsequently abandoned the approach owing to a significant incidence of complications. Dohlman and Mattson3 reintroduced and popularized the technique in 1960, describing electrocautery for division of the common wall. In 1993, Collard4 presented the technique of endoscopic stapling of the diverticulum. Collard’s approach solved early problems associated with lack of a seal along the divided mucosal edges, with subsequent salivary leak and mediastinitis. The most recent evidence suggests that endoscopic stapling is superior to open tech-
niques in the management of this condition.5,6 Compared with open techniques, the procedure is minimally invasive, has a reduced operative time, a lower rate of complications, and shorter hospital stay.5,6 It also has excellent rates of symptom resolution and a low rate of recurrence.5,6 A single row of endoscopic staples may not be sufficient to completely divide the common wall of the pharyngeal pouch. On observing this, a second or third row can be applied in series with the first row to obtain a more complete division. During subsequent staplings the surgeon aims to leave the more proximal stapled mucosal edges undisturbed and not advance the stapler too far into the pouch, causing trauma to its distal end. Although this procedure is performed infrequently in our unit, it was observed that, when multiple rows of staples were required to obtain complete division of the dividing wall, there was an increased incidence of postoperative complications and a prolonged hospital stay. To ascertain if this observation was statistically significant, we designed a study to compare patients who had undergone pharyngeal stapling with either a single row of staples or multiple rows of staples. Our hypothesis was that patients undergoing endoscopic pharyngeal pouch stapling with multiple rows of staples would have a higher complication rate and prolonged hospital stay compared with patients in whom a single row of staples was placed.
METHODS A retrospective study was conducted of all patients who underwent endoscopic pharyngeal pouch surgery between July 1, 1997 and December 1, 2007 in the Southern Health network, Melbourne, Australia. Approval for the review was granted by the Southern Health Human Research Ethics Committee. Funding was obtained from an otolaryngology departmental research fund. Patients were identified by operation code from a computerized database with the earliest point of review being
Received May 23, 2008; revised October 7, 2008; accepted October 22, 2008.
0194-5998/$36.00 © 2009 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved. doi:10.1016/j.otohns.2008.10.035
246
Otolaryngology–Head and Neck Surgery, Vol 140, No 2, February 2009
the date from which the database began. A spreadsheet was designed to record patient baseline demographics and outcomes of interest. Patient files were then individually reviewed to obtain the data. Primary outcomes of interest were number of rows of staples applied to the bridging wall, length of stay, and complications and their management. The operative technique was analyzed in comparison with Collard’s4 initial approach and any modifications noted. Secondary outcomes of interest were operating time, use of nasogastric tube, time to clear fluid diet, time to normal diet, and postoperative instructions. Postoperative patient satisfaction with the surgery was recorded as either full or partial resolution of symptoms. A patient was recorded as having full resolution of symptoms if they no longer complained of any dysphagia or regurgitation at their postoperative visits. A partial resolution was recorded if the patient was still symptomatic. Patients who had partial symptom resolution at the first visit who subsequently became asymptomatic were recorded as having full resolution of symptoms. Recurrence of symptoms at a later date after full symptom resolution and whether these patients required a repeat procedure were recorded. All analyses were performed by a statistician using SAS version 8.2. (SAS Institute Inc, Cary, NC). Tests of significance for categorical data were obtained with the Fisher exact test. Nonparametric data were analyzed with Wilcoxon rank sum tests.
RESULTS Thirty-eight patients who had undergone endoscopic pharyngeal pouch surgery during the period of review were identified. Twenty-seven patients had undergone procedures in which a single row of staples was applied to the common wall. Eleven patients had more than one row of staples applied to the common wall. Of these patients, nine patients had two rows of staples applied and three patients had three rows of staples applied. The mean patient age in the single-row group was 74 years compared with a mean age of 68 years in the multiplerow group. There were twice as many males compared with females in the single-row group compared with the multiple-row group. There was no difference in the timing of operation dates in the two groups. None of these results showed any statistically significant difference between the two groups. Each procedure was undertaken or supervised by one of the nine consultant surgeons who worked in the unit. A registrar or fellow was listed as the primary operator in six of the cases with a single row of staples and in two of the cases with multiple rows. There were no complications in the group of patients for whom the registrar was listed as the primary operator. All patients underwent surgery within a single tertiary institution. Patients underwent general anesthesia, a dental
guard was used, and then a Weerda (Karl Stortz, Tuttlingen, Germany) bivalved laryngoscope was introduced and used to expose the common wall between the esophagus and diverticulum. The longer lip was placed in the esophagus and the short one in the diverticulum. Photodocumentation was achieved with the use of 0- or 30-degree telescopes connected to a camera and recording device. The diverticulum was then examined for any suspicious lesions before proceeding. The Endo-GIA 30 (TYCO Health Care) stapler blade with 2.5-mm staples was then introduced over the common wall and its position checked with a telescope. After applying the stapler and removing it, the length of the residual common wall was assessed and a decision made as to whether a second or third application of staples would be necessary to completely divide the common wall. No particular algorithm appears to have been followed by the operating surgeons as to what length of residual common wall required a further row of staples. The telescope was then used to examine the esophageal and diverticular wall for any evidence of perforation or loose staples. The mean operating time overall was 31 minutes. There were no reported intraoperative complications in either group. A nasogastric tube was inserted in 30 percent of patients in the single row group and 46 percent of patients in the multiple rows group. These differences were not statistically significant. There was some variability in the postoperative instructions between groups; however, none of these results were significant. All patients were sent to recovery with instructions to assess for fever, chest or back pain, hemoptysis, tachycardia, or respiratory distress. Patients in the singlerow group were given a clear fluid diet before a solid diet in 67 percent compared with 82 percent in the multiple row group. Twenty-one percent of patients were ordered a routine postoperative chest X-ray. Five percent of patients were booked a routine gastrograffin swallow. The mean length of stay was 2.1 days (range 1-10 days) in the single-row group and 7.1 days (range 1-29 days) in the multiple-row group, a difference of 5 days between groups (P ⫽ 0.017). In the single-row group, two patients had a length of stay greater than 3 days. One patient stayed 10 days to undergo treatment for other unrelated medical problems. The other patient stayed 7 days for investigation of a suspected perioperative myocardial infarction. The mean time until a clear fluid diet was achieved was 1.1 days (range 0-5 days) in the single-row group and 5.5 days (range 1-22 days) in the multiple-row group (P ⫽ 0.003). The mean time until normal diet was achieved was 1.5 days (range 1-5 days) in the single-row group and 5.8 days (range 1-25 days) in the multiple row group (P ⫽ 0.024). There were four cases of postoperative pouch or esophageal perforation. These all occurred in patients who had multiple rows of staples applied to divide the common wall. No cases occurred in patients who had a single row of
Roth et al
Endoscopic stapling of pharyngeal pouch: A 10-year . . .
staples applied. This equates to a statistically significant leak rate of 36 percent in the multiple-row group compared with zero percent in the single-row group (P ⫽ 0.004). No perforation was noted intraoperatively in any of the cases that had a postoperative leak. There was no significant difference between the leak group and the overall patient group in terms of age, sex, surgeon, use of nasogastric tube, or operation time. However, patients who had multiple staple rows had larger pouch sizes (based on surgeon intraoperative description) than those who had a single row applied (P ⫽ 0.041). The management of the four patients who had leaks postoperatively varied. Two of the patients were managed conservatively with nothing by mouth and a nasogastric tube. They were discharged home after 6 and 14 days, respectively. One patient was taken back to the operating theater for open closure of the perforation and was discharged on day 9. One patient developed an empyema and required a thoracotomy. These two patients were subsequently discharged after 22 days. Length of follow-up was similar in both groups. The recurrence rate was 29.6 percent in the single-row group and 18.2 percent in the multiple-row group. This difference was not significant. There was no significant difference in patient satisfaction between groups.
DISCUSSION The results obtained in this study suggest that in our unit, endoscopic pharyngeal pouch surgery in which more than one row of staples is applied to the common wall is associated significantly with an increased risk of esophageal or pouch perforation. Patients who had more than one row of staples applied took significantly longer to tolerate both clear fluid and normal diets, and had a longer length of stay. Patient satisfaction and recurrence rates were similar in both groups. These data highlight the increased risk associated with the placement of subsequent rows of staples and questions the benefit of placing more than one row of staples in this procedure. A number of mechanisms may explain why multiple applications of the Endo-GIA stapler are associated with an increased risk of perforation. After the first row of staples is introduced, the tension sutures lose their retraction ability and it may be difficult to apply the second and subsequent staple rows in precisely the correct plane. There is a risk of mucosal laceration each time the endostapler is introduced.7 When the anvil is inserted into the diverticulum it may cause a perforation by pushing too far into the distal end of the pouch.8 Approximation of the common wall mucosal edges may also fail because of excessive width of the common wall or technical failure of the staple gun, or because of too much tissue being divided with insufficient stapling.9,10 It is unclear to what extent the common wall has to be divided to achieve complete symptom resolution and min-
247
imize the risk of recurrence. Kaoy and Baates11 have suggested that deciding the length of common wall to divide is a matter of judgment and is probably related to experience. They suggest that it is safer to under-divide rather than risk a perforation. A case report by Arunachalam and Cameron12 cautions against over-stapling after a patient developed a persistent foreign body sensation and pharyngeal pain from a clump of staples. Hilton and Brightwell9 described a case of a defect in the posterior wall of the esophagus that occurred after a single row of endoscopic stapling owing to trauma from the end of the stapling gun. Other papers describing endoscopic stapling of pharyngeal pouches have suggested that if two rows of staples could not be applied, then there was a significant chance of incomplete symptom resolution. Collard4 suggested that stapling should be abandoned if the tissue bridge was not large enough to allow two rows of cartridges. Koay and Bates11 described residual symptoms in two of five patients who had a single row of staples applied. There is evidence that stapling of smaller pouches has a higher complication rate. de Casso et al5 described a leak rate of 14.6 percent with 57 percent of leaks occurring in patients who had small pouches. It may be that in small pouches either trauma from the stapler or the knife cutting too close to the end of the tissue bridge may be causing a higher rate of perforation. Patients who had multiple rows of staples applied to their pouch had larger pouch sizes according to surgeon intraoperative description. It could therefore be argued that the higher complication rate is a consequence of having a large pouch rather than of the stapling procedure itself. There is no previous reported evidence that larger pouches when stapled have higher complications. Applying the stapler to a larger pouch is often easier in the first instance owing to the increased room to maneuver the stapler in the pouch lumen. It seems most likely that the stapling itself rather than simply the increased size of the pouch is the risk factor for postoperative leaks. It is not clear whether the insertion of a nasogastric tube following this procedure is beneficial. A number of series describe its use as “optional” and have inserted them in only early cases.11,13,14 In our series there was no statistically significant benefit associated with insertion. However, it may be beneficial to place a nasogastric tube in patients who are at higher risk of perforation such as when multiple rows of staples are placed. Richtsmeier and Monzon15 examined the use of alternative endoscopic staple devices in this procedure after two patients leaked after an Endopath ETS was use to insert 3.5-mm staples. The authors suggested that the ETS45 stapler with a 2.5-mm cartridge may offer an improved seal at the distal end of the incision compared with the same device with a 3.5-mm cartridge. They discuss that this device is also 15 mm longer than an EndoGIA-30, allowing a longer cut and fewer stapler firings. However, this series consisted of just 14 patients, two of whom leaked, and the
248
Otolaryngology–Head and Neck Surgery, Vol 140, No 2, February 2009
EndoGIA-30 remains the most widely accepted device currently in use. Pharyngeal pouch stapling is a procedure potentially associated with a high morbidity and mortality rate. A number of publications have supported the belief that there is a significant learning curve associated with the procedure.5,16-18 The report of the National Confidential Enquiry into Peri-operative Deaths commissioned by the Royal College of Surgeons of England16 recommends that it should be undertaken by one otolaryngologist in each center who has developed a special interest in it. Southern Health is one of Australia’s largest area health networks and services a large population. However, this procedure was performed only 38 times over approximately 10 years. Dividing these cases between nine consultant surgeons equates to approximately one case every 2 to 3 years. The leak rate of 36 percent is higher than leak rates published in other larger series (range 0%-10%)11 and suggests that the infrequency with which our unit is performing this procedure may be contributing to a higher rate of complications. There is good evidence to suggest that residual or recurrent symptomatic pouches can be treated endoscopically without any increase in perioperative morbidity or mortality.19 This finding should be kept in mind when attempting to achieve complete division of the common wall. If the patient has only partial symptom resolution or recurrence from the degree of stapling performed, a further procedure can be safely performed.
CONCLUSION Patients undergoing endoscopic stapling of pharyngeal pouches with more than one row of staples may be at increased risk of postoperative complications including esophageal perforation and leak. These patients have a more prolonged postoperative hospital admission and take longer to resume either a clear fluid or normal diet. Care should be taken during the insertion of staples, and surgeons should be aware of the risks associated with the placement of additional rows.
ACKNOWLEDGEMENTS Michael Bailey, PhD, MSc, BSc(hons), Department of Epidemiology and Preventive Medicine, Faculty of Medicine, Nursing and Health Sciences, Monash University: assistance in performing the statistical analysis of our data.
AUTHOR INFORMATION From the Department of Otolaryngology, Head and Neck Surgery, Southern Health, Melbourne (Drs Roth, Sigston, and Vallance); and the Department of Surgery, (Monash Medical Centre), Faculty of Medicine, Monash University (Drs Sigston and Vallance), East Bentleigh.
Corresponding author: Jason Roth, Department of Otolaryngology, Head and Neck Surgery, Monash Medical Centre Moorabbin Campus, Centre Road, East Bentleigh VIC 3165, Australia. E-mail address: [email protected]. Data from this paper were presented at the Australian Society of Otolaryngology Head and Neck Surgery Scientific Meeting, Perth, Western Australia, April 8-11, 2008.
AUTHOR CONTRIBUTIONS Jason A Roth: data collection, manuscript preparation; Elizabeth Sigston: review of manuscript; Neil Vallance: study design.
FINANCIAL DISCLOSURE None.
REFERENCES 1. Killian G. La boudre de l’oesophage. Annales Mal Orelle Larynx 1907;34:1. 2. Mosher HP. Webs and pouches of the esophagus: their diagnosis and treatment. Surg Gynecol Obstet 1917;25:175– 87. 3. Dohlman G, Mattson O. The endoscopic operation for hypopharyngeal diverticula. Arch Otolaryngol 1960;71:744 –52. 4. Collard JM, Otte JB, Kestens PJ. Endoscopic stapling technique of esophagodiverticulostomy for Zenker’s diverticulum. Ann Thoracic Surg 1993;56:573– 6. 5. Casso C, Lalam M, Ghosh S, et al. Endoscopic stapling diverticulotomy: an audit of difficulties, outcome and patient satisfaction. Otolaryngol Head Neck Sur 2006;134:288 –93. 6. Sen P, Bhattacharyya AK. Endoscopic stapling of pharyngeal pouch. J Laryngol Otol 2004;118:601– 6. 7. Scher RL, Richtsmeier WJ. Long-term experience with endoscopic staple-assisted esophagodiverticulostomy for Zenker’s diverticulum. Laryngoscope 1998;108:200 –5. 8. Koay CB, Commins D, Bates GJ. The role of endoscopic stapling diverticulotomy in recurrent pharyngeal pouch. J Laryngol Otol 1998; 112:954 –5. 9. Hilton M. Esophageal perforation during stapling of a pharyngeal pouch: adverse clinical incident report. J Laryngol Otol 2000;114: 549 –50. 10. Kubba H, Hussain SS. Stapler failure in pharyngeal diverticulectomy: a suggested modification in surgical technique. J R Coll Surg Edinb 1998;43:31–2. 11. Koay CB, Bates GJ. Endoscopic stapling diverticulotomy for pharyngeal pouch. Clin Otolaryngol 1996;21:371– 6. 12. Arunachalam PS, Cameron DS. Persistent foreign body sensation and pharyngeal pain due to retention of staples: an interesting sequelae of endoscopic stapling procedure. J Laryngol Otol 2001;115:425–7. 13. Narne S, Cutrone C, Bonavina L, et al. Endoscopic diverticulotomy for the treatment of Zenker’s diverticulum: results in 102 patients with staple-assisted endoscopy. Ann Otol Rhinol Laryngol 1999;108: 810 –5. 14. Scher Rl, Richtmeier WJ. Long-term experience with endoscopic staple-assisted esophagodiverticulostomy for Zenker’s diverticulum. Laryngoscope 1998;108:200 –5.
Roth et al
Endoscopic stapling of pharyngeal pouch: A 10-year . . .
15. Richtsmeier WJ, Monzon JR. Postendoscopic Zenker esophagodiverticulostomy leaks associated with a specific stapler cartridge. Arch Otolaryngol Head Neck Surg 2002;128:137– 40. 16. Resouly A. Pharyngeal pouch surgery. In: Royal College of Surgeons for England, editors. The report of the national confidential enquiry into perioperative deaths 1996/1997. London: Royal College of Surgeons of England; 1998. p 32.
249
17. Thaler ER, Weber RS, Goldberg AN, et al. Feasibility and outcome of endoscopic staple-assisted esophagodiverticulostomy for Zenker’s diverticulum. Laryngoscope 2001;111:1506 – 8. 18. Mirza S, Dutt SN, Minhas SS, et al. A retrospective review of pharyngeal pouch surgery in 56 patients. Ann R Coll Surg Engl 2002;84:247–51. 19. Scher RL. Endoscopic staple diverticulostomy for recurrent Zenker’s diverticulum. Laryngoscope 2003;113:63–7.