366
Ausr. N.Z..ISurg. . 1991,61, 366-369
IS PYLOROPLASTY NECESSARY FOLLOWING INTRATHORACIC TRANSPOSITION OF STOMACH? RESULTS OF A PROSPECTlVE CLINICAL STUDY
v.
T.K . CHATTOPADHYAY,~s. GUPTA,* A. K . PADHY3* AND K . KAP0OR4' Departments of Surgery and *Nuclear Medicine, All India Institute of Medical Sciences, New Delhi, India Twenty four patients underwent oesophagectomy for oesophageal cancer. The oesophagogastric anastomosis was performed in the neck in all patients. Following oesophagectomy and gastric mobilization patients were randomly selected into pyloroplasty and no pyloroplasty groups. Pre and postoperative gastric emptying of these patients evaluated by radioisotope technique were then compared. The results suggest significantly delayed postoperative gastric emptying in both the groups though it was less pronounced in the pyloroplasty group. All patients were then carefully followed until death (period varying between 6 months and 4 years) for ill effects of delayed gastric emptying which were present in some patients of both the groups. It was thus concluded that emptying of thoracic stomach is delayed and pyloroplasty fails to improve it completely. Postoperativelypatients behave much the same way with or without pyloroplasty. Key words: gastric emptying, intrathoric stomach, oesophageal cancer, pyloroplasty.
Introduction
Methods
Oesophagectomy with oesophagogastrostomy either in the chest or in the neck is an accepted procedure for carcinoma of the oesophagus. Complete vagotomy is an inevitable consequence of such an operation and may cause gastric stasis. Most authors therefore advocate some form of drainage procedure to prevent gastric stasis following this procedure. I-' However, other surgeons feel that drainage is unnecessary because stomach laced intrathoracically empties well without it.*-' l f Some authorities also feel that a drainage procedure is hazardous because of the high incidence of bile gastritis in these patients."*I2 Unfortunately, not many controlled studies are available to assist surgeons to decide whether to drain a vagotomized intrathoracic stomach. The present study is conducted on randomly selected groups of patients undergoing total oesophagectomy with intrathoracic placement of stomach in the presence or absence of pyloroplasty.
Twenty-four patients with carcinoma of the oesophagus subjected to total oesophagectomy and cervical oesophagogastrostomy by a synchronous cervical-thoracic-abdominal approach were included in the study.I3 In this approach, initially the abdomen and neck are explored. If there are no enlarged nodes in the neck and there is no intraabdominal spread, the thoracic surgeon proceeds with anterolateral thoracotomy through the right fifth intercostal space. Thoracic oesophagus containing the tumour and the cervical oesophagus are mobilized in the usual way. The abdominal surgeon mobilizes the stomach, preserving the right gastric and the right gastro-epiploic arteries. The oesophagus, with the tumour, is resected next and the gastric stump sutured in two layers. After oesophagectomy and gastric mobilization these 24 cases were randomly divided into 2 groups. In group I patients, no pyloroplasty was done; in group 2 it was. A Heineke-Mikulicz type of pyloroplasty was performed using a single layer interrupted OOO atraumatic silk suture. The oesophageal hiatus was then dilated by finger stretching so as to accommodate four of the surgeon's fingers and the stomach was then pulled up into the neck through the posterior mediastinum. Both groups of patients had gastric emptying time evaluated pre-operatively. Gastric emptying was re-evaluated in each patient 4-5 months after the operation in both groups of patients.
'
' '
MS: Associate Professor. MS: Senior Resident. DRM. MNAMS; Associate Professor. MS; Associate Professor. 'Present address: Department of Surgical Gastroenterology. Sanjay Gandhi Postgraduate Institute of Medical Sciences. Lucknow. Uttar Pradesh. India. Correspondence: Dr T. K. Chattopadhyay, Associate Professor, Department of Surgery. All India Institute of Medical Sciences. New Delhi I10029, India. Accepted for publication 12 December 1990
367
PYLOROPLASTY AND GASTRIC EMPTYING
Gastric emptying time was measured using the radioisotope technique described by Choudhuri. l4 Patients were instructed to stop all medications, tea, coffee and smoking 48 h prior to testing. The study was conducted after overnight fasting in supine position. A liquid meal containing 200mL of water labelled with 3.7 X lo7 Bq of 9n”rc diethylenetriamine pentaacetic acid was used. Radioactivity was recorded by a Gamma camera over the stomach, in the abdomen and in the chest in the pre- and postoperative periods. A timeactivity curve was generated using a computer and suitable software. Gastric emptying was expressed as To (estimated time taken for initial stomach radioactivity to reach zero). Parameter TOwas taken instead of To.5, as the latter could not be estimated in many cases because the To.5 fell outside the entire study period. Since the emptying of liquid meal in all cases was found to be mono-exponential, To was estimated from the time-activity curve using the available computer software. All these patients were then followed up for between 6 months and 4 years. During follow-up, patients. were observed for early fullness, postprandial discomfort, eructations, nausea, vomiting, excessive flatulence, heartburn, dumping, diarrhoea and any respiratory or cardiovascular symp toms. The findings were then correlated with the postoperative gastric emptying time evaluated earlier (4-5 months after operation). All data were statistically analysed during the Student’s r-test and paired 1-test.
Results GASTRIC EMPTYING
The emptying of liquid meal was mono-exponential in all cases, with visualization of entire stomach in early scans and the gradual appearance of activity in the intestines in later ones. In most subjects there was no substantial overlap of duodenal or even jejunal activity over the gastric area. The mean estimated pre-operative gastric emptying time ( * s.d.) (To) in group 1 (i.e. without pyloroplasty) was 37.70 k 3.33 min, whereas the postoperative To for the same group was 370.87 25.03 min. Thus, there was significant delay (P< 0.001) in gastric emptying after gastric pull up surgery without pyloroplasty (Table 1). The mean pre-operative value for group 2 patients was 38.76 k 2.89min. The corresponding postoperative value in this group (i.e. with pyloroplasty) was 161.20 k 23.56min. Although the extent of delay in gastric emptying was significantly less (P < 0.001) with the addition of pyloroplasty (Table I), nevertheless it was still prolonged compared with pre-operative values (P< 0.001).
*
Table 1. Mean pre-operative and postoperative To in groups 1 and 2
To(min) Pre-operative Postoperative
Group
1 (no pyloroplasty)* 37.70 +- 3.33
2 (pyloroplasty)*
38.76 k 2.89
370.87 +- 25.03** 161.20 k 23.56**
* n = 12.
** P < 0.001.
FOLLOW-UP RESULTS
All surviving patients were followed up for 6 months to 4 years (average follow-up: 14 months). During each postoperative visit the patients were evaluated by an independent observer for ill effects of delayed gastric emptying, as mentioned earlier. Details of follow-up studies are shown in Table 2. Although 1 patient in group 2 experienced early fullness after meals, it was not seen in any patient in group 1. This patient improved after 3 months. Post-prandial discomfort was observed in 1 patient in group 1 and in none in group 2. With dietary adjustment this patient also improved considerably (Viscik Grade 11). Three patients in group 1 and 4 patients in group 2 experienced regurgitation and responded only partially to conservative measures (Viscik Grade 111). Features suggestive of dumping syndrome were seen in 2 patients in group 2 and 1 patient in group 1. Diarrhoea was present in 2 patients of each group. Respiratory symptoms (breathing discomfort or tachypnoea) or cardiovascular symptoms after meals were not seen in any patient. Based on these results, a modified Viscik grading of the findin s was carried out (Table 3) as outlined by Angorn.
k
Table 2. Postoperative symptoms
Symptom
Early fullness Postprandial discomfort Regurgitation Vomiting Heartburn Dumping
Diarrhoea Respintory
Cardiovascular
No. patients Group 2 Group 1 (no pyloroplasty) (pyloroplasty) 0
1
I 3 0 3 I 2 0 0
0
4 0 2 2
2 0 0
CHA'ITOPADHYAY ET AL.
368
Table 3. Modified Viscik grading of results No. patients Group 1
Group 2
111
4 5 3
IV
0
5 3 4 0
Grade
I I1
Discussion Total gastric vagotomy with resultant gastric stasis is an inevitable result of oesophagectomy. Most authors therefore routinely perform pyloroplasty following such a procedure. Unfortunately, these recommendations are based not on studies of intrathoracic stomach but on the results of truncal vagotomy for peptic ulcer disease. We undertook this prospective randomized study on the emptying pattern of intrathoracic stomach with and without pyloroplasty with the aim of ascertaining: (i) whether pyloroplasty improves gastric emptying and (ii) whether pyloroplasty has any bearing on the clinical behaviour of these patients. In the present study we observed that vagotomy following total oesophagectomy caused delayed emptying of liquids from intrathoracic stomach. Pyloroplasty was not effective in preventing this delay, although gastric emptying was shortened more in patients with pyloroplasty than in those without it. A similar delay has been observed by other authors However, Mannel et af. could not show delayed gastric emptying even though there was a high incidence of gastric food residue on endoscopy." On the contrary, Huang suggested rapid emptying of intrathoracic stomach." It is possible that an upright position played an important role in both the studies. In view of these conflicting reports it is difficult to explain delayed emptying in the present study. It is presumed that the vagotomized stomach dilates more in response to negative intrathoracic pressure and/ or undergoes flaccid receptive relaxation. The addition of pyloroplasty does not seem to affect these factors and, hence, is associated with delayed gastric emptying, although of a lesser magnitude. The study also aimed to assess whether this delayed gastric emptying had any deleterious effect on the clinical behaviour of patients. To ascertain this, all patients were carefully followed up for ill effects of delayed gastric emptying, for 6 months to 4 years postoperatively. Some symptoms were attributable to gastric stasis in patients of both groups. The difference was not statistically significant. Pyloroplasty does not seem to influence the clinical outcome favourably. In explanation, it is worthwhile meantioning that, postoperatively, all
patients were instructed to avoid supine postures as far as practicable, and were also advised to have small meals at a time, raise their heads at bedtime and avoid drinking water during meals. It is reasonable to conclude that all these factors equally, rather than a drainage procedure alone, are important in avoiding the problems of delayed gastric emptying. Thus, we conclude that the intrathoracic vagotomized stomach is associated with delayed gastric emptying. Postoperatively, patients in either group behave in much the same way. Patients may have some symptoms due to this delayed emptying but pyloroplasty fails to prevent them.
References I . BELSEYR. H. R. (1980)Palliative management of oesophageal carcinoma. Amer. J . Surg. 139,789-94. 2. WONC J. (1981) Management of carcinoma of oesophagus: art or science? J. R. Coll. Surg. Edinb. 26, 138-49. 3. O~~G.B.&K~0~~K.H.(l969)TheLewis-Tanner operation for cancer of the oesophagus. J. R. Coll. Surg. Edinb. 14,3-19. 4. GROVES L. K. (1965)Carcinoma of the oesophagus. Ann. Thorac. Surg. 1,416-31. 5 . MACILL T. G. & SIMMONS R. L. (1967)Resection of cardio-oesophageal carcinoma. Arch. Surg. 94,865-70. 6. ELLISR. H.JR (1984)Cancer of the oesophagus and cardia: role of surgery in palliation. Postgrad. Med. 75, 139-48. 7. MANNELL A,, MCKNICHT A. & ESSERJ. D. (1990) Role of pyloroplasty in the retrostemal stomach: results of a prospective, randomized controlled trial. Brit. J . Surg. 77. 57-9. 8. ELLISF. H. & GIBESS. P. (1979)Oesophagogastrectomy for carcinoma. Current hospital mortality and morbidity rates. Ann. Surg. 190. 699-705. 9. COLLIS J. L. (1972)The long-term clinical state after resection with gastro-oesophagostomy. In: Surgery of the Oesophagus. (Eds R. A. Smith & R. G. Smith), pp. 19-21. Butterworths, London. 10. ANGORN I. B. (1975)Oesophagogastrostomy without a drainage procedure in oesophageal carcinoma. Brit. J . Surg. 62,601-4. 1 1 . HINDER R. A. (1976)The effect of posture on the emptying of the intrathoracic vagotomised stomach. Brit. J . Surg. 63,581-4. 12. LAMK. H., LIMS. T. K., WONCJ.. LAMS. K. & ONCG. B. (1979)Gastric histology and function in patients with intrathoracic stomach replacement after oesophagectomy. Surgery 85, 283-90. 13. CHAITOPADHYAY T. K. (1986)Squamous cell carcinoma of the oesophagus: current perspective. Trop. Castroenterol. 7. 94- 115. 14. CHOUDHURY T. K. (1974)Use of ""Tc DTPA for measuring gastric emptying time. J. Nucl. Med. 15, 391-5. 15. GUPTAS., CHAITOPADHYAY T. K., GOPINATH P. G., KAPOOR V. K. & SHARMA L. K. (1989)Emptying of the intrathoracic stomach with and without pyloroplasty. Amer. J . Castro. 84, 921-3.
369
PYLOROPLASTYAND GASTRIC EMPTYING 16. CHURCHILL, E. D. & SWEET R. H.(1942) Transthoracic resection of tumors of stomach and oesophagus. Arch. Surg. 94,865-70. 17. CHENC H.C., Siu K . F. & WONCJ. (1987) Is pyloro-
plasty necessary in oesophageal replacement by stomach? A prospective randomized controlled trial. Surgery 102, 19-24. 18. MANNEL A., HINDER R. A. & SAN-GARDE B. A. (1984) The thoracic stomach: a study of gastric emptying,
bile reflux and mucosal change. Brir. J . Surg. 71, 438-41. 19. HUANG G. J., GANG D. C. & ZHANCD. W. (1985) A comparative study of resection of carcinoma of the oesophagus with and without pyloroplasty. In: Esophageal Disorders: Parhophysiology and Therapy (Eds T. R. De Meester & D. B. Skinner), pp. 383-8. Raven Press, New York.
Ausr. N.Z..ISurg. . 1991,61, 366-369
IS PYLOROPLASTY NECESSARY FOLLOWING INTRATHORACIC TRANSPOSITION OF STOMACH? RESULTS OF A PROSPECTlVE CLINICAL STUDY
v.
T.K . CHATTOPADHYAY,~s. GUPTA,* A. K . PADHY3* AND K . KAP0OR4' Departments of Surgery and *Nuclear Medicine, All India Institute of Medical Sciences, New Delhi, India Twenty four patients underwent oesophagectomy for oesophageal cancer. The oesophagogastric anastomosis was performed in the neck in all patients. Following oesophagectomy and gastric mobilization patients were randomly selected into pyloroplasty and no pyloroplasty groups. Pre and postoperative gastric emptying of these patients evaluated by radioisotope technique were then compared. The results suggest significantly delayed postoperative gastric emptying in both the groups though it was less pronounced in the pyloroplasty group. All patients were then carefully followed until death (period varying between 6 months and 4 years) for ill effects of delayed gastric emptying which were present in some patients of both the groups. It was thus concluded that emptying of thoracic stomach is delayed and pyloroplasty fails to improve it completely. Postoperativelypatients behave much the same way with or without pyloroplasty. Key words: gastric emptying, intrathoric stomach, oesophageal cancer, pyloroplasty.
Introduction
Methods
Oesophagectomy with oesophagogastrostomy either in the chest or in the neck is an accepted procedure for carcinoma of the oesophagus. Complete vagotomy is an inevitable consequence of such an operation and may cause gastric stasis. Most authors therefore advocate some form of drainage procedure to prevent gastric stasis following this procedure. I-' However, other surgeons feel that drainage is unnecessary because stomach laced intrathoracically empties well without it.*-' l f Some authorities also feel that a drainage procedure is hazardous because of the high incidence of bile gastritis in these patients."*I2 Unfortunately, not many controlled studies are available to assist surgeons to decide whether to drain a vagotomized intrathoracic stomach. The present study is conducted on randomly selected groups of patients undergoing total oesophagectomy with intrathoracic placement of stomach in the presence or absence of pyloroplasty.
Twenty-four patients with carcinoma of the oesophagus subjected to total oesophagectomy and cervical oesophagogastrostomy by a synchronous cervical-thoracic-abdominal approach were included in the study.I3 In this approach, initially the abdomen and neck are explored. If there are no enlarged nodes in the neck and there is no intraabdominal spread, the thoracic surgeon proceeds with anterolateral thoracotomy through the right fifth intercostal space. Thoracic oesophagus containing the tumour and the cervical oesophagus are mobilized in the usual way. The abdominal surgeon mobilizes the stomach, preserving the right gastric and the right gastro-epiploic arteries. The oesophagus, with the tumour, is resected next and the gastric stump sutured in two layers. After oesophagectomy and gastric mobilization these 24 cases were randomly divided into 2 groups. In group I patients, no pyloroplasty was done; in group 2 it was. A Heineke-Mikulicz type of pyloroplasty was performed using a single layer interrupted OOO atraumatic silk suture. The oesophageal hiatus was then dilated by finger stretching so as to accommodate four of the surgeon's fingers and the stomach was then pulled up into the neck through the posterior mediastinum. Both groups of patients had gastric emptying time evaluated pre-operatively. Gastric emptying was re-evaluated in each patient 4-5 months after the operation in both groups of patients.
'
' '
MS: Associate Professor. MS: Senior Resident. DRM. MNAMS; Associate Professor. MS; Associate Professor. 'Present address: Department of Surgical Gastroenterology. Sanjay Gandhi Postgraduate Institute of Medical Sciences. Lucknow. Uttar Pradesh. India. Correspondence: Dr T. K. Chattopadhyay, Associate Professor, Department of Surgery. All India Institute of Medical Sciences. New Delhi I10029, India. Accepted for publication 12 December 1990
367
PYLOROPLASTY AND GASTRIC EMPTYING
Gastric emptying time was measured using the radioisotope technique described by Choudhuri. l4 Patients were instructed to stop all medications, tea, coffee and smoking 48 h prior to testing. The study was conducted after overnight fasting in supine position. A liquid meal containing 200mL of water labelled with 3.7 X lo7 Bq of 9n”rc diethylenetriamine pentaacetic acid was used. Radioactivity was recorded by a Gamma camera over the stomach, in the abdomen and in the chest in the pre- and postoperative periods. A timeactivity curve was generated using a computer and suitable software. Gastric emptying was expressed as To (estimated time taken for initial stomach radioactivity to reach zero). Parameter TOwas taken instead of To.5, as the latter could not be estimated in many cases because the To.5 fell outside the entire study period. Since the emptying of liquid meal in all cases was found to be mono-exponential, To was estimated from the time-activity curve using the available computer software. All these patients were then followed up for between 6 months and 4 years. During follow-up, patients. were observed for early fullness, postprandial discomfort, eructations, nausea, vomiting, excessive flatulence, heartburn, dumping, diarrhoea and any respiratory or cardiovascular symp toms. The findings were then correlated with the postoperative gastric emptying time evaluated earlier (4-5 months after operation). All data were statistically analysed during the Student’s r-test and paired 1-test.
Results GASTRIC EMPTYING
The emptying of liquid meal was mono-exponential in all cases, with visualization of entire stomach in early scans and the gradual appearance of activity in the intestines in later ones. In most subjects there was no substantial overlap of duodenal or even jejunal activity over the gastric area. The mean estimated pre-operative gastric emptying time ( * s.d.) (To) in group 1 (i.e. without pyloroplasty) was 37.70 k 3.33 min, whereas the postoperative To for the same group was 370.87 25.03 min. Thus, there was significant delay (P< 0.001) in gastric emptying after gastric pull up surgery without pyloroplasty (Table 1). The mean pre-operative value for group 2 patients was 38.76 k 2.89min. The corresponding postoperative value in this group (i.e. with pyloroplasty) was 161.20 k 23.56min. Although the extent of delay in gastric emptying was significantly less (P < 0.001) with the addition of pyloroplasty (Table I), nevertheless it was still prolonged compared with pre-operative values (P< 0.001).
*
Table 1. Mean pre-operative and postoperative To in groups 1 and 2
To(min) Pre-operative Postoperative
Group
1 (no pyloroplasty)* 37.70 +- 3.33
2 (pyloroplasty)*
38.76 k 2.89
370.87 +- 25.03** 161.20 k 23.56**
* n = 12.
** P < 0.001.
FOLLOW-UP RESULTS
All surviving patients were followed up for 6 months to 4 years (average follow-up: 14 months). During each postoperative visit the patients were evaluated by an independent observer for ill effects of delayed gastric emptying, as mentioned earlier. Details of follow-up studies are shown in Table 2. Although 1 patient in group 2 experienced early fullness after meals, it was not seen in any patient in group 1. This patient improved after 3 months. Post-prandial discomfort was observed in 1 patient in group 1 and in none in group 2. With dietary adjustment this patient also improved considerably (Viscik Grade 11). Three patients in group 1 and 4 patients in group 2 experienced regurgitation and responded only partially to conservative measures (Viscik Grade 111). Features suggestive of dumping syndrome were seen in 2 patients in group 2 and 1 patient in group 1. Diarrhoea was present in 2 patients of each group. Respiratory symptoms (breathing discomfort or tachypnoea) or cardiovascular symptoms after meals were not seen in any patient. Based on these results, a modified Viscik grading of the findin s was carried out (Table 3) as outlined by Angorn.
k
Table 2. Postoperative symptoms
Symptom
Early fullness Postprandial discomfort Regurgitation Vomiting Heartburn Dumping
Diarrhoea Respintory
Cardiovascular
No. patients Group 2 Group 1 (no pyloroplasty) (pyloroplasty) 0
1
I 3 0 3 I 2 0 0
0
4 0 2 2
2 0 0
CHA'ITOPADHYAY ET AL.
368
Table 3. Modified Viscik grading of results No. patients Group 1
Group 2
111
4 5 3
IV
0
5 3 4 0
Grade
I I1
Discussion Total gastric vagotomy with resultant gastric stasis is an inevitable result of oesophagectomy. Most authors therefore routinely perform pyloroplasty following such a procedure. Unfortunately, these recommendations are based not on studies of intrathoracic stomach but on the results of truncal vagotomy for peptic ulcer disease. We undertook this prospective randomized study on the emptying pattern of intrathoracic stomach with and without pyloroplasty with the aim of ascertaining: (i) whether pyloroplasty improves gastric emptying and (ii) whether pyloroplasty has any bearing on the clinical behaviour of these patients. In the present study we observed that vagotomy following total oesophagectomy caused delayed emptying of liquids from intrathoracic stomach. Pyloroplasty was not effective in preventing this delay, although gastric emptying was shortened more in patients with pyloroplasty than in those without it. A similar delay has been observed by other authors However, Mannel et af. could not show delayed gastric emptying even though there was a high incidence of gastric food residue on endoscopy." On the contrary, Huang suggested rapid emptying of intrathoracic stomach." It is possible that an upright position played an important role in both the studies. In view of these conflicting reports it is difficult to explain delayed emptying in the present study. It is presumed that the vagotomized stomach dilates more in response to negative intrathoracic pressure and/ or undergoes flaccid receptive relaxation. The addition of pyloroplasty does not seem to affect these factors and, hence, is associated with delayed gastric emptying, although of a lesser magnitude. The study also aimed to assess whether this delayed gastric emptying had any deleterious effect on the clinical behaviour of patients. To ascertain this, all patients were carefully followed up for ill effects of delayed gastric emptying, for 6 months to 4 years postoperatively. Some symptoms were attributable to gastric stasis in patients of both groups. The difference was not statistically significant. Pyloroplasty does not seem to influence the clinical outcome favourably. In explanation, it is worthwhile meantioning that, postoperatively, all
patients were instructed to avoid supine postures as far as practicable, and were also advised to have small meals at a time, raise their heads at bedtime and avoid drinking water during meals. It is reasonable to conclude that all these factors equally, rather than a drainage procedure alone, are important in avoiding the problems of delayed gastric emptying. Thus, we conclude that the intrathoracic vagotomized stomach is associated with delayed gastric emptying. Postoperatively, patients in either group behave in much the same way. Patients may have some symptoms due to this delayed emptying but pyloroplasty fails to prevent them.
References I . BELSEYR. H. R. (1980)Palliative management of oesophageal carcinoma. Amer. J . Surg. 139,789-94. 2. WONC J. (1981) Management of carcinoma of oesophagus: art or science? J. R. Coll. Surg. Edinb. 26, 138-49. 3. O~~G.B.&K~0~~K.H.(l969)TheLewis-Tanner operation for cancer of the oesophagus. J. R. Coll. Surg. Edinb. 14,3-19. 4. GROVES L. K. (1965)Carcinoma of the oesophagus. Ann. Thorac. Surg. 1,416-31. 5 . MACILL T. G. & SIMMONS R. L. (1967)Resection of cardio-oesophageal carcinoma. Arch. Surg. 94,865-70. 6. ELLISR. H.JR (1984)Cancer of the oesophagus and cardia: role of surgery in palliation. Postgrad. Med. 75, 139-48. 7. MANNELL A,, MCKNICHT A. & ESSERJ. D. (1990) Role of pyloroplasty in the retrostemal stomach: results of a prospective, randomized controlled trial. Brit. J . Surg. 77. 57-9. 8. ELLISF. H. & GIBESS. P. (1979)Oesophagogastrectomy for carcinoma. Current hospital mortality and morbidity rates. Ann. Surg. 190. 699-705. 9. COLLIS J. L. (1972)The long-term clinical state after resection with gastro-oesophagostomy. In: Surgery of the Oesophagus. (Eds R. A. Smith & R. G. Smith), pp. 19-21. Butterworths, London. 10. ANGORN I. B. (1975)Oesophagogastrostomy without a drainage procedure in oesophageal carcinoma. Brit. J . Surg. 62,601-4. 1 1 . HINDER R. A. (1976)The effect of posture on the emptying of the intrathoracic vagotomised stomach. Brit. J . Surg. 63,581-4. 12. LAMK. H., LIMS. T. K., WONCJ.. LAMS. K. & ONCG. B. (1979)Gastric histology and function in patients with intrathoracic stomach replacement after oesophagectomy. Surgery 85, 283-90. 13. CHAITOPADHYAY T. K. (1986)Squamous cell carcinoma of the oesophagus: current perspective. Trop. Castroenterol. 7. 94- 115. 14. CHOUDHURY T. K. (1974)Use of ""Tc DTPA for measuring gastric emptying time. J. Nucl. Med. 15, 391-5. 15. GUPTAS., CHAITOPADHYAY T. K., GOPINATH P. G., KAPOOR V. K. & SHARMA L. K. (1989)Emptying of the intrathoracic stomach with and without pyloroplasty. Amer. J . Castro. 84, 921-3.
369
PYLOROPLASTYAND GASTRIC EMPTYING 16. CHURCHILL, E. D. & SWEET R. H.(1942) Transthoracic resection of tumors of stomach and oesophagus. Arch. Surg. 94,865-70. 17. CHENC H.C., Siu K . F. & WONCJ. (1987) Is pyloro-
plasty necessary in oesophageal replacement by stomach? A prospective randomized controlled trial. Surgery 102, 19-24. 18. MANNEL A., HINDER R. A. & SAN-GARDE B. A. (1984) The thoracic stomach: a study of gastric emptying,
bile reflux and mucosal change. Brir. J . Surg. 71, 438-41. 19. HUANG G. J., GANG D. C. & ZHANCD. W. (1985) A comparative study of resection of carcinoma of the oesophagus with and without pyloroplasty. In: Esophageal Disorders: Parhophysiology and Therapy (Eds T. R. De Meester & D. B. Skinner), pp. 383-8. Raven Press, New York.
