World J Surg DOI 10.1007/s00268-014-2478-2
Long-term Results of the Paraesophagogastric Devascularization with or without Esophageal Transection: Which Is More Suitable for Variceal Bleeding? He-yun Zhang • Wen-bin Li • Hua Ye • Zhi-yu Xiao • Yao-rong Peng • Jie Wang
Ó Socie´te´ Internationale de Chirurgie 2014
Abstract Background It has been reported that the paraesophagogastric devascularization with esophageal transection procedure, also known as the modified Sugiura procedure, was effective in the treatment of variceal bleeding. However, it was not widely accepted by other surgeons because of the high rate of rebleeding, complications, and mortality. To discover the effects of the paraesophagogastric devascularization procedure and the modified Sugiura procedure, we retrospectively analyzed the outcomes of these two procedures. Materials and methods During January 1990 and December 2009, 278 patients with variceal bleeding underwent devascularization after failed pharmacotherapy and endotherapy. In these 278 patients, 180 underwent paraesophagogastric devascularization without esophageal transection (group I), and the other 98 patients were subjected to the modified Sugiura procedure (group II). Results Postoperative mortality was 7.2 % in group I, and 9.2 % in group II (P = 0.563). The postoperative rebleeding rate in the two groups was 2.2 and 3.1 %, respectively (P = 0.474). After a mean follow-up of 67.9 ± 37.3 months and 67.4 ± 44.6 months, respectively, esophageal transaction-related morbidity (leak, bleeding,
He-yun Zhang and Wen-bin Li have contributed equally to this study. H. Zhang Department of Hepatobiliary Surgery, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, 510120 Guangzhou, China H. Zhang W. Li H. Ye Z. Xiao Y. Peng J. Wang (&) Guangdong Provincial Key Laboratory of Malignant Tumor Epigenetics and Gene Regulation, Sun Yat-sen Memorial Hospital of Sun Yat-Sen University, 510120 Guangzhou, China e-mail: [email protected]
and stricture) was 8.2 % (8/98) in group II and 0 % (0/180) in group I (P \ 0.001). The overall rebleeding rate was 27 % (41/152) in group I, and 27.2 % (22/81) in group II (P = 0.976). The overall mortality was 28.3 % (43/152) in group I, and 28.4 % (23/81) in group II (P = 0.986). Conclusions In the management of variceal bleeding, paraesophagogastric devascularization without esophageal transection is as effective and safe as devascularization with esophageal transaction, but with less esophageal transection-related morbidity.
Introduction Pharmacotherapy combined with endotherapy remains the first line treatment for the management of variceal bleeding, but repeated endotherapy procedures were required because of the high rebleeding rate, which ranged from 20 to 50 % after the first endotherapy [1–6], Patients who experienced rebleeding after the endotherapy, require surgical intervention as the second line salvage treatment for the emergent and long-term control of rebleeding before liver failure occurs. In 1984, Sugiura and Futagawa [7] proposed what has come to be called the Sugiura procedure, which included esophageal transection through the chest and paraesophagogastric devascularization through the chest and abdomen. However, because of the complexity of the procedure and the incidence of severe postoperative complications, some medical institutions improved the surgical procedure and renamed it the modified Sugiura procedure, in which both the paraesophagogastric devascularization and esophageal transection are performed in the abdomen [8]. The modified Sugiura procedure had been shown to be
123
World J Surg
n=12
n=4
n=38
Devascularization without stapler transection Devascularization with stapler transection Shunt
n=58
n=180
Splenectomy TIPS Liver transplantation
n=98
Fig. 1 The form of surgical intervention for patients whom had failed in endotherapy in our hospital (1990–2009)
effective for the management of variceal bleeding, but with a high rate of complications and rebleeding, ranging from 30 to 40 % in cirrhotic patients [9, 10]. The aim of the present study was to evaluate the safety and efficacy of paraesophagogastric devascularization with or without esophageal transection in the management of variceal rebleeding in patients who had accepted adequate endotherapy and pharmacotherapy. The results of these two procedures, paraesophagogastric devascularization without esophageal transection in one group (group I) esophageal transaction in a second group (group II), were analyzed in terms of their safety, complications, incidence of variceal rebleeding, and morbidity and mortality. Fig. 2 The main step of paraesophagogastric devascularization with or without esophageal transection [14]
Materials and methods This study was a retrospective analysis of patients with variceal bleeding (n = 708), who received endotherapy and pharmacotherapy as the first line treatment in our hospital during January 1990 and December 2009. Among them, 390 patients (55.1 %) had rebleeding after endotherapy and pharmacotherapy. Of the 278 patients who underwent devascularization procedures, 180 were subjected to devascularization without esophageal transection (group I), and the other 98 patients underwent devascularization with esophageal transection, a procedure known as the modified Sugiura procedure (group II) (Fig. 1). Devascularization with or without esophageal transection is the choice of second line salvage treatment for patients who experiences rebleeding after endotherapy and pharmacotherapy. Devascularization with esophageal transection was applied if it was technically practical and the esophagus was supple enough. In other cases, when it was difficult to perform transection, either for technical
123
reasons or when staplers were not suitable or not available, or the esophagus was edematous and thickened, friable or fibrosed, then the devascularization procedure alone was performed. Section of both vagus nerves and pyloroplasty were performed during devascularization in all patients. Clinical data of patients were studied from the patient records. Esophageal varices were graded by Conn’s classification through endoscopy [11], and the grades of varices at each endoscopy were noted. The cirrhotic grades of patients were catalogued according to the Child-Pugh classification. The number of bleeding episodes, the interval between endotherapy and rebleeding, and the operative procedures performed were documented. Postoperative data were retrieved, and included complications, encephalopathy, rebleeding episodes, and deaths. Follow-up details were documented at the end of the study, and included rebleeding episodes, complications of the surgery, liver dysfunction–related diseases, and deaths.
World J Surg
Operative procedures
Table 1 The comparisons and statistical analyzes of preoperative clinical details between the two groups
The paraesophagogastric devascularization procedure was described by Yang and Qiu [12], and the paraesophagogastric devascularization with stapler esophageal transection, also termed the modified Sugiura procedure, was described by Sugiura and Futagawa [7], and by Mathur et al. [13] through an abdominal incision. The main surgical procedures are shown in Fig. 2 [14]. The paraesophagogastric devascularization procedure included complete devascularization of at least half of the lesser curvature of the stomach and more than two-thirds of the greater curvature proximally, and 6–8 cm of the lower esophagus. The left gastric and short gastric vessels, coronary vein, retrogastric veins, as well as the periesophageal collaterals were dissected during this step. An EEA circular stapler (SDH 25–28 mm; Ethicon), was then used for transection and end-to-end anastomosis of the lower esophagus about 3 cm above gastroesophageal junction. Splenectomy was done by the common method. Pyloroplasty was required in the patients who underwent esophageal transection.
Total number of patients (n = 278)
Follow-up The schedule of the follow-up visits was every 3 months during the first year, every 6 months for the next 5 years, and once every year thereafter. All patients were scheduled for upper GI endoscopy once a year after surgery, according to their clinical situations.
Statistical analysis
Group I (n = 180)
Group II (n = 98)
5–77
8–70
47.4 ± 13.1
46.3 ± 13.1
Male
147
80
Female
33
18
Age (years) Range Mean ± SD Gender
Criteria used in the study We defined the study criteria as follows for the present study: (1) operative mortality was considered as death during hospitalization or within the first 6 weeks postoperatively [13]; (2) uncontrolled variceal bleeding referred to either continuing variceal bleeding, even after adequate pharmacotherapy and endotherapy, or major rebleeding
0.495a 0.994b
0.451b
Prior endotherapy EST
33
15
EST and EVL
60
40
EVL
87
43 0.095b
Time frame 1990–1999
66
46
2000–2009
114
52
Hepatitis B related
134
69
0.469b
Hepatitis C related
4
6
0.174c
Alcohol related
29
17
0.79b
Etiology
0.728b
Type of portal hypertension Cirrhotic
167
92
Non-cirrhotic
13
6 0.574b
Child-Pugh classification Grade A
93
47
Grade B
72
45
Grade C
15
6
Timing of surgery Emergency Elective
The statistical software SPSS 15.0 was used for the statistical analysis. The quantitative data of patients were described as the mean ± standard error and range. The qualitative data were described as frequencies. Complications, mortality, rebleeding rate, ascites, and encephalopathy between the two groups were analyzed by the v2 test and Fisher’s exact test. Comparisons of quantitative data were analyzed by t test. Statistical significance was noted when the P value \ 0.05.
P value
a
0.738b 11
7
169
91
t-test
b
Pearson’s Chi squared test
c
Fisher’s exact test
EVL endoscopic variceal ligation, EST endoscopic variceal ligation plus sclerotherapy
within 5 days after the first episode of variceal bleeding [15]; (3) failure of emergency endotherapy was defined as failure to control variceal bleeding, or as variceal rebleeding after two separate endotherapy sessions during the hospitalization or less than two weeks apart [15]; (4) failure of chronic endotherapy was defined as at least two episodes of life-threatening rebleeding occurring during the regular follow-up after chronic endotherapy; (5) variceal rebleeding referred to one new episode of upper gastrointestinal hemorrhage causing a hemoglobin drop of more than 2 g/L within 24 h or occurring within 2 weeks after the initial bleeding had been stabilized [16–18]; (6) early rebleeding
123
World J Surg Table 2 The early outcomes (B 6 weeks) of the surgery Result
Group I (n = 180)
Group II (n = 98)
P value
Bleeding from esophageal stoma
–
1/98 (1.0 %)
0.353b
Esophageal leak
–
3/98 (3.1 %)
0.043b*
Gastric retention
2/180 (1.1 %)
5/98 (5.1 %)
0.025b*
Respiratory infection
5/180 (2.8 %)
2/98 (2.0 %)
0.526b
Wound infection Ischemic necrosis of the stomach wall
4/180 (2.2 %) 1/180 (0.6 %)
2/98 (2.0 %) 0/98 (0 %)
0.643b 0.647b
Variceal rebleeding
4/180 (2.2 %)
3/98 (3.1 %)
0.474b
Intra-abdominal bleeding
6/180 (3.3 %)
2/98 (2.0 %)
0.420b
Portal vein thrombosis
4/180(2.2 %)
2/98(2.0 %)
0.643b
Acute pulmonary embolism
1/180(0.6 %)
1/98(1.0 %)
0.582b
Mortality for emergency surgery
3/11(27.3 %)
5/7(71.4 %)
0.145b
Overall mortality
13/180(7.2 %)
9/98(9.2 %)
0.563a
Liver failure
5/180(2.8 %)
3/98(3.1 %)
0.580b
Postoperative infection
1/180(0.6 %)
1/98(1.0 %)
0.582b
Intra-abdominal bleeding
4/180(2.2 %)
2/98(2.0 %)
0.643b
Variceal rebleeding
3/180(1.7 %)
2/98(2.0 %)
0.583b
Acute pulmonary embolism and DIC
1/180(0.6 %)
1/98(1.0 %)
0.582b
Esophageal transection-related
General complications
Cause of death
a
Pearson’s Chi squared test
b
Fisher’s exact test
*Statistically significant
was defined as rebleeding that occurred within 6 weeks of operation, and late rebleeding was that which occurred beyond 6 weeks after operation; (7) symptomatic hypersplenism referred to splenomegaly with malaise, anemia, thrombocytopenia (PLT \100 9 109/L), WBC \4 9 109/ L, as well as coagulation disorder [19, 20]; (8) emergency surgery referred to surgical procedures performed only if acute variceal bleeding could not be controlled by endotherapy combined with pharmacotherapy [8].
Results The comparisons and statistical analyses of preoperative clinical details between the two groups are shown in Table 1. Early outcomes of surgery (within 6 weeks of surgery) Four patients in group II had complications associated with esophageal transection. One of them bled from the esophagus anastomotic stoma and recovered after surgical hemostasis. The other three patients had esophageal leaks. Two of them recovered after conservative treatment and
123
B-ultrasound-guided percutaneous drainage, but the other died of intra-abdominal infection and septic shock. Four patients in group I and three patients in group II experienced upper gastrointestinal rebleeding after surgery. Three patents in group I and two patients in group II died because of rebleeding. The others recovered after endotherapy and pharmacotherapy. Six cases of intra-abdominal bleeding occurred in group I. One patient had bleeding from the vascular stump and recovered after reoperation. Five other patients experienced bleeding from splenic bed errhysis. Four of them died of hemorrhagic shock, and the fifth patient recovered after conservative treatment. In group II, two patients had errhysis from the splenic bed and both died of hemorrhagic shock. Acute pulmonary embolism occurred in one patient of each group, and both of them died. Four patients in group I and two patients in group II developed portal vein thrombosis, and all of them recovered after administration of anticoagulants. Five patients in group I and three patients in group II died of progressive liver failure. Ischemic necrosis of the stomach at the greater curvature occurred in 1 patient in group I who died of septic shock after surgical repair. Other complications, such as respiratory infection,
World J Surg
developed hepatocellular carcinoma and 6 of them died, whereas 5 patients in group II developed HCC and 3 of them died. The overall mortality in the two groups was 27 % in group I (41/152 patients) and 27.2 % in group II (22/81 patients). The Kaplan–Meier survival test analysis (P [ 0.05) of the two groups is shown in Fig. 3. Four patients in group II developed esophageal stricture with symptoms; 2 patients in group I and 1 patient in group II had incisional hernia repairs. In group I, 18 patients required pharmacotherapy for ascites, as did 7 patients in group II. None of the patients had post-splenectomy sepsis at the end of the study. The long-term outcomes of the two groups are shown in Table 3.
Discussion Fig. 3 The Kaplan–Meier survival test of the two groups
wound infection, and gastric retention occurred in both groups. The early outcomes of the two groups are shown in Table 2. Long-term outcomes of the surgery Of the 167 survivors in group I, 91 % (152/167 survivors) were followed from 8 to 216 months (mean: 67.9 ± 37.3 months), and 91 % of the survivors in group II (81/89 survivors) were followed from 7 to 204 months (mean: 67.4 ± 44.6 months). In group I, the rebleeding rate was 5.9 % (9/152) in 1 year, 17.1 % (26/152) in the first 3 years, and 23.7 % in the first 5 years (36/152). The overall rebleeding rate was 27 % (41/152), and 35 cases of rebleeding were caused by recurrent varices; 25 of those patients were cured by endotherapy and pharmacotherapy, 2 were cured by portosystemic shunt. The other 8 died of uncontrolled bleeding. The other 6 cases of rebleeding were the result of portal gastropathy or hemorrhagic gastritis, and all were successfully managed by pharmacotherapy. In group II, the rebleeding rate was 6.2 % (5/81 patients) in the first year, 13.6 % (11/81 patients) in the first 3 years, and 21 % in the first 5 years (17/81 patients). The overall rebleeding rate was 27.2 % (22/81 patients). In 17 of the 22 patients, the cause of rebleeding was recurrent varices; 11 were cured by endotherapy and pharmacotherapy but 6 died of uncontrolled bleeding. In the other 5 patients, the bleeding was due to portal gastropathy or hemorrhagic gastritis, and all of them were successfully managed by pharmacotherapy. Progressive liver failure occurred in 20 patients in group I and 11 patients in group II. Eight patients in group I
Therapy for variceal bleeding in patients with portal hypertension has evolved over time, and it now it encompasses a spectrum of treatment modalities among which sequential therapies are often necessary [21, 22]. In many centers, the first-line treatment is pharmacotherapy and endotherapy. Once variceal bleeding occurs, the risk of recurrence can be as high as 70 % [23]. The challenge of management for variceal bleeding in patients with portal hypertension is how to prevent rebleeding while maintaining satisfactory liver function. Liver transplantation is beneficial for patients with endstage liver failure, but it is not an ideal treatment for variceal bleeding. Also, because of the limited supply of donor organs, and for economic reasons, liver transplantation cannot be available for all patients. For these reasons, only if the liver is in its decompensated stage should liver transplantation be considered for patients with variceal bleeding secondary to portal hypertension. There is accumulating evidences that variceal bleeding with compensated liver function (Child-Pugh class A and B?) should initially be treated with non-transplantation therapy [23]. Transjugular intrahepatic portosystemic shunt (TIPS) is increasingly used in patients with variceal hemorrhage caused by portal hypertension. Because TIPS is a nonoperative approach, it appears to be a desirable treatment when only temporary portal decompression is required, such as in liver transplantation candidates who fail endotherapy and pharmacotherapy. However, the high rate (40–50 %) of shunt stenosis or shunt thrombosis within the first year will lead to variceal rebleeding, and this risk limits the use of TIPS in clinical work [23]. In 1964, Hassab [24] proposed the procedure of gastroesophageal decongestion and splenectomy, which was based on angiography of portal hypertension. The surgical procedure includes removal of the spleen, total ligation of vessels in the first third of the proximal stomach and 6 cm
123
World J Surg Table 3 Long-term outcomes ([ 6 weeks) of the surgery Result
Group I
Group II
P value
Follow-up rate
152/167 (91.0 %)
81/89 (91.0 %)
0.999a
Average follow-up time(months)
67.9 ± 37.3
67.4 ± 44.6
0.930c
Esophageal stricture
–
4/81 (4.9 %)
0.014b*
Ascites
18/152 (11.8 %)
11/81 (13.6 %)
0.702a
Encephalopathy
10/152 (6.6 %)
10/81 (12.3 %)
0.135a
Incisional hernia
2/152 (1.3 %)
1/81 (1.2 %)
1.000b
Portal hypertensive gastropathy
30/152 (19.7 %)
23/81 (28.4 %)
0.133a
Rebleeding cases of 1 year
9/152 (5.9 %)
5/81 (6.2 %)
0.939a
Rebleeding cases of 3 year
26/152 (17.1 %)
11/81 (13.6 %)
0.483a
Rebleeding cases of 5 year Overall rebleeding cases
36/152 (23.7 %) 41/152 (27 %)
17/81 (21.0 %) 22/81 (27.2 %)
0.40a 0.976a
Variceal bleeding
35/152 (23.0 %)
17/81 (21.0 %)
0.722a
Gastropathy
6/152 (3.9 %)
5/81 (6.2 %)
0.521b
Mortality of 1 year
12/152 (7.9 %)
2/81 (2.5 %)
0.147b
Mortality of 3-year
25/152 (16.4 %)
10/81 (12.3 %)
0.404a
Mortality of 5-year
34/152 (22.4 %)
19/81 (23.5 %)
0.850a
Overall mortality
43/152 (28.3 %)
23/81 (28.4 %)
0.986a
Liver failure
20/152 (13.2 %)
11/81 (13.6 %)
0.928a
Variceal bleeding
8/152 (5.3 %)
6/81 (7.4 %)
0.512a
Hepatic carcinoma
6/152 (3.9 %)
3/81 (3.7 %)
0.616b
Overall esophageal transaction- related morbidity
–
8/98 (8.2 %)
\ 0.001b*
Long-term complications ([ 6 weeks)
Cause of rebleeding
Cause of death
a
Pearson’s Chi squared test
b
Fisher’s exact test
c
t-test
*Statistically significant
above the gastroesophageal junction. In the 1970s, Yang and Qiu [12] realized the importance of the upper esophageal branch, the heterotopic esophageal branch, and the posterior gastric vein in cases of variceal bleeding of portal hypertension, and proposed a more thorough paraesophagogastric vascular disconnection, including disconnection of the veins mentioned above. In 1984, Sugiura and Futagawa [7] proposed the Sugiura procedure, which consisted of esophageal transection through the chest, and paraesophagogastric devascularization through the chest and abdomen. However, because of the complexity of this procedure and the severe postoperative complications associated with it, some medical institutions improved the surgical procedure by adding paraesophagogastric devascularization with esophageal transection through the abdomen. They renamed it the modified Sugiura procedure [8]. The modified Sugiura procedure has proved to be effective in the management of acute variceal bleeding, but there is high rate of operative complications and a
123
remarkable rebleeding rate, up to 40 % in patients with liver cirrhosis [9, 10]. In the present study, whether esophageal transection was performed or not did not affect the rebleeding rate during the perioperative period (2.2 % in group I vs 3.1 % in group II), or during the first postoperative year (5.9 vs 6.2 %), the third postoperative year (17.1 vs 13.6 %), or the fifth postoperative year (23.7 vs 21.0 %), as well as overall (27 vs 27.2 %). The rebleeding rates show no statistical significance between the two groups (Table 3). These observations indicate that the esophageal transaction procedure may not be essential for devascularization. In the present study, the overall mortality is 28.3 % in group I and 28.4 % in group II. Mortality in the perioperative period was 7.2 % in group I versus 9.2 % in group II; during the first postoperative year, 7.9 versus 2.5 %; during the third postoperative year, 16.4 versus 12.3 %); and during the fifth postoperative year, 22.4 versus 23.5 %. The differences in the two groups show no statistical significance. The Kaplan–Meier survival test for the two
World J Surg
groups also shows that the differences in survival rates of these two groups have no statistical significance. In addition, the data indicate that the mortality of the two groups is not affected by the esophageal transection procedure. Mortality after emergency devascularization is 71.4 % (5/ 7) in group II, which is much higher than the 27.3 % (3/11) in group I, although the statistical analysis shows no significance. Several factors may contribute to rebleeding after devascularization, including the following: (1) failure to devascularize the paraesophagogastric veins completely, especially missing the upper esophageal branch, the heterotopic esophageal branch, and the posterior gastric vein [12]; (2) perfusion of the stomach is influenced by paraesophagogastric devascularization, which can cause portal hypertensive gastropathy [25]; (3) regeneration of esophageal and gastric varices. In addition to paraesophagogastric devascularization, the modified Sugiura procedure adds esophageal transaction, which, theoretically at least, completely blocks the connection between esophageal and gastric varices. However, both the postoperative and long-term rebleeding rates show no statistical differences, whether esophageal transaction is performed or not. Possible reasons might be that the left common submucosal veins can hardly form varices because of disconnection between portal vein system and esophagogastric veins after complete paraesophagogastric devascularization. Under this circumstance, further esophageal transection seems meaningless in the prevention of rebleeding. Secondarily, the modified Sugiura procedure may aggravate submucosal congestion and ischemia of the stomach and eventually lead to gastropathy. The devascularization procedure without esophageal transection avoids transection-related morbidity, which is as high as 8.2 % (8/ 98 patients). The most common complications associated with esophageal transection are esophageal leak and bleeding from the esophageal stoma [26, 27]. The mortality of esophageal leak was reported to be as high as 60 % [27]. In our study, the rate of esophageal leak was 3.1 % in group II, compared to reports of 6–13 % [13, 26] from other studies. Our three patients with leakage from the transection line had undergone sclerotherapy. However, there were no statistical differences between patients who had prior sclerotherapy and those who had not in our group II (3/55 vs 0/43; p = 0.120), Nevertheless, it should be recognized that the use of prior sclerotherapy may lead to thickening around the distal esophagus as well as hypoperfusion at distal esophagus, which may eventually cause a leak. The rate of mortality related to leak was 1.0 % in our study. Furthermore, the incidence of esophageal strictures in our study was 4.9 %, and all were cured by endoscopic dilatation. The results of our study indicate that devascularization combined with esophageal transaction
can lead to a higher rate of postoperative transectionrelated morbidity [27–30]. Furthermore, it will not improve postoperative morbidity or reduce the rate of rebleeding below the levels achieved with devascularization without esophageal transection.
Conclusions In the management of variceal bleeding, paraesophagogastric devascularization without esophageal transection is as effective and safe as devascularization with esophageal transection, but with less esophageal transection-related morbidity. Acknowledgments This work was supported by the 5010 Clinical Research Project of SUN Yat-Sen University (2010009).
References 1. Infante-Rivard C, Esnaola S, Villeneuve JP (1989) Role of endoscopic variceal sclerotherapy in the long-term management of variceal bleeding: a meta-analysis. Gastroenterology 96: 1087–1092 2. D’Amico G, Pagliaro L, Bosch J (1995) The treatment of portal hypertension: a meta-analytic review. Hepatology 22:332–354 3. Bendtsen F, Krag A, Moller S (2008) Treatment of acute variceal bleeding. Dig Liver Dis 40:328–336 4. Vlachogiannakos J, Goulis J, Patch D et al (2000) Review article: primary prophylaxis for portal hypertensive bleeding in cirrhosis. Aliment Pharmacol Ther 14:851–860 5. Stanley AJ, Hayes PC (1997) Portal hypertension and variceal haemorrhage. Lancet 350:1235–1239 6. Van Stiegmann G, Goff JS, Sun JH et al (1990) Endoscopic ligation of esophageal varices. Am J Surg 159:21–26 7. Sugiura M, Futagawa S (1984) Esophageal transection with paraesophagogastric devascularizations (the Sugiura procedure) in the treatment of esophageal varices. World J Surg 8:673–679. doi:10.1007/BF01655762 8. Battaglia G, Ancona E, Patarnello E et al (1996) Modified Sujura operation: long-term results. World J Surg 20:319–325. doi:10. 1007/s002689900051 9. Comar KM, Sanyal AJ (2003) Portal hypertensive bleeding. Gastroenterol Clin North Am 32:1079–1105 10. Gouge TH, Ranson JH (1986) Esophageal transection and paraesophagogastric devascularization for bleeding esophageal varices. Am J Surg 151:47–54 11. Conn HO (1967) Ammonia tolerance in the diagnosis of esophageal varices. A comparison of endoscopic, radiologic, and biochemical techniques. J Lab Clin Med 70:442–451 12. Zhen Yang, Fazu Qiu (2001) The typical technique of pericardial devascularization and its effects on portal hypertension. Central China Med J 25:229–230 13. Mathur SK, Shah SR, Nagral SS et al (1999) Transabdominal extensive esophagogastric devascularization with gastroesophageal stapling for management of noncirrhotic portal hypertension: long-term results. World J Surg 23:1168–1175. doi:10.1007/ s002689900641 14. Johnson M, Rajendran S, Balachandar TG et al (2006) Transabdominal modified devascularization procedure with or without
123
World J Surg
15.
16.
17.
18. 19.
20.
21.
22.
esophageal stapler transaction—an operation adequate for effective control of a variceal bleed. Is esophageal stapler transection necessary? World J Surg 30:1507–1519. doi:10.1007/s00268005-0754-x Jalan R, John TG, Redhead DN et al (1995) A comparative study of emergency transjugular intrahepatic portosystemic stent— shunt and esophageal transection in the management of uncontrolled variceal hemorrhage. Am J Gastroenterol 90:1932–1937 Luketic VA, Sanyal AJ (2000) Esophageal varices. II. TIPS (transjugular intrahepatic portosystemic shunt) and surgical therapy. Gastroenterol Clin North Am 29:387–421 Jalan R, Hayes PC (2000) UK guidelines on the management of variceal haemorrhage in cirrhotic patients. British Society of Gastroenterology. Gut 46(Suppl 3–4):I1–I15 De Franchis R (1996) Developing consensus in portal hypertension. J Hepatol 25:390–394 Krige JE, Bornman PC, Goldberg PA et al (2000) Variceal rebleeding and recurrence after endoscopic injection sclerotherapy: a prospective evaluation in 204 patients. Arch Surg 135:1315–1322 Sorbi D, Gostout CJ, Peura D et al (2003) An assessment of the management of acute bleeding varices: a multicenter prospective member-based study. Am J Gastroenterol 98:2424–2434 de Franchis R (2010) Revising consensus in portal hypertension: report of the Baveno V consensus workshop on methodology of diagnosis and therapy in portal hypertension. J Hepatol 53:762–768 Garcia-Tsao G, Bosch J (2010) Management of varices and variceal hemorrhage in cirrhosis. N Engl J Med 362:823–832
123
23. Courtney M. Townsend, R Daniel Beauchamp B, et al (2012) Portal hypertension. In: Sabiston textbook of surgery, 19th edn. W. B. Saunders Company, Philadelphia, pp 1433–1434 24. Hassab MA (1964) Gastroesophageal decongestion and splenectomy. A method of prevention and treatment of bleeding from esophageal varices associated with bilharzial hepatic fibrosis: preliminary report. J Int Coll Surg 41:232–248 25. Editorial (1991) Portal hypertensive gastropathy. Lancet 338: 1045-1046 26. Orozco H, Takahashi T, Mercado MA et al (1994) Surgical management of extrahepatic portal hypertension and variceal bleeding. World J Surg 18:246–250. doi:10.1007/BF00294409 27. Dagenais M, Langer B, Taylor BR et al (1994) Experience with radical esophagogastric devascularization procedures (Sugiura) for variceal bleeding outside Japan. World J Surg 18:222–228. doi:10.1007/BF00294405 28. Idezuki Y, Sanjyo K (1989) Twenty-five-year experiences with esophageal transection for esophageal varices. J Thorac Cardiovasc Surg 98:876–883 29. Han HS, Yi NJ, Kim YW et al (2004) New operative method for fundal variceal bleeding: fundectomy with periesophagogastric devascularization. World J Surg 28:406–410. doi:10.1007/ s00268-003-6923-x 30. Hassab MA (1998) Gastro-esophageal decongestion and splenectomy GEDS (Hassab), in the management of bleeding varices. Review of literature. Int Surg 83:38–41
Long-term Results of the Paraesophagogastric Devascularization with or without Esophageal Transection: Which Is More Suitable for Variceal Bleeding? He-yun Zhang • Wen-bin Li • Hua Ye • Zhi-yu Xiao • Yao-rong Peng • Jie Wang
Ó Socie´te´ Internationale de Chirurgie 2014
Abstract Background It has been reported that the paraesophagogastric devascularization with esophageal transection procedure, also known as the modified Sugiura procedure, was effective in the treatment of variceal bleeding. However, it was not widely accepted by other surgeons because of the high rate of rebleeding, complications, and mortality. To discover the effects of the paraesophagogastric devascularization procedure and the modified Sugiura procedure, we retrospectively analyzed the outcomes of these two procedures. Materials and methods During January 1990 and December 2009, 278 patients with variceal bleeding underwent devascularization after failed pharmacotherapy and endotherapy. In these 278 patients, 180 underwent paraesophagogastric devascularization without esophageal transection (group I), and the other 98 patients were subjected to the modified Sugiura procedure (group II). Results Postoperative mortality was 7.2 % in group I, and 9.2 % in group II (P = 0.563). The postoperative rebleeding rate in the two groups was 2.2 and 3.1 %, respectively (P = 0.474). After a mean follow-up of 67.9 ± 37.3 months and 67.4 ± 44.6 months, respectively, esophageal transaction-related morbidity (leak, bleeding,
He-yun Zhang and Wen-bin Li have contributed equally to this study. H. Zhang Department of Hepatobiliary Surgery, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, 510120 Guangzhou, China H. Zhang W. Li H. Ye Z. Xiao Y. Peng J. Wang (&) Guangdong Provincial Key Laboratory of Malignant Tumor Epigenetics and Gene Regulation, Sun Yat-sen Memorial Hospital of Sun Yat-Sen University, 510120 Guangzhou, China e-mail: [email protected]
and stricture) was 8.2 % (8/98) in group II and 0 % (0/180) in group I (P \ 0.001). The overall rebleeding rate was 27 % (41/152) in group I, and 27.2 % (22/81) in group II (P = 0.976). The overall mortality was 28.3 % (43/152) in group I, and 28.4 % (23/81) in group II (P = 0.986). Conclusions In the management of variceal bleeding, paraesophagogastric devascularization without esophageal transection is as effective and safe as devascularization with esophageal transaction, but with less esophageal transection-related morbidity.
Introduction Pharmacotherapy combined with endotherapy remains the first line treatment for the management of variceal bleeding, but repeated endotherapy procedures were required because of the high rebleeding rate, which ranged from 20 to 50 % after the first endotherapy [1–6], Patients who experienced rebleeding after the endotherapy, require surgical intervention as the second line salvage treatment for the emergent and long-term control of rebleeding before liver failure occurs. In 1984, Sugiura and Futagawa [7] proposed what has come to be called the Sugiura procedure, which included esophageal transection through the chest and paraesophagogastric devascularization through the chest and abdomen. However, because of the complexity of the procedure and the incidence of severe postoperative complications, some medical institutions improved the surgical procedure and renamed it the modified Sugiura procedure, in which both the paraesophagogastric devascularization and esophageal transection are performed in the abdomen [8]. The modified Sugiura procedure had been shown to be
123
World J Surg
n=12
n=4
n=38
Devascularization without stapler transection Devascularization with stapler transection Shunt
n=58
n=180
Splenectomy TIPS Liver transplantation
n=98
Fig. 1 The form of surgical intervention for patients whom had failed in endotherapy in our hospital (1990–2009)
effective for the management of variceal bleeding, but with a high rate of complications and rebleeding, ranging from 30 to 40 % in cirrhotic patients [9, 10]. The aim of the present study was to evaluate the safety and efficacy of paraesophagogastric devascularization with or without esophageal transection in the management of variceal rebleeding in patients who had accepted adequate endotherapy and pharmacotherapy. The results of these two procedures, paraesophagogastric devascularization without esophageal transection in one group (group I) esophageal transaction in a second group (group II), were analyzed in terms of their safety, complications, incidence of variceal rebleeding, and morbidity and mortality. Fig. 2 The main step of paraesophagogastric devascularization with or without esophageal transection [14]
Materials and methods This study was a retrospective analysis of patients with variceal bleeding (n = 708), who received endotherapy and pharmacotherapy as the first line treatment in our hospital during January 1990 and December 2009. Among them, 390 patients (55.1 %) had rebleeding after endotherapy and pharmacotherapy. Of the 278 patients who underwent devascularization procedures, 180 were subjected to devascularization without esophageal transection (group I), and the other 98 patients underwent devascularization with esophageal transection, a procedure known as the modified Sugiura procedure (group II) (Fig. 1). Devascularization with or without esophageal transection is the choice of second line salvage treatment for patients who experiences rebleeding after endotherapy and pharmacotherapy. Devascularization with esophageal transection was applied if it was technically practical and the esophagus was supple enough. In other cases, when it was difficult to perform transection, either for technical
123
reasons or when staplers were not suitable or not available, or the esophagus was edematous and thickened, friable or fibrosed, then the devascularization procedure alone was performed. Section of both vagus nerves and pyloroplasty were performed during devascularization in all patients. Clinical data of patients were studied from the patient records. Esophageal varices were graded by Conn’s classification through endoscopy [11], and the grades of varices at each endoscopy were noted. The cirrhotic grades of patients were catalogued according to the Child-Pugh classification. The number of bleeding episodes, the interval between endotherapy and rebleeding, and the operative procedures performed were documented. Postoperative data were retrieved, and included complications, encephalopathy, rebleeding episodes, and deaths. Follow-up details were documented at the end of the study, and included rebleeding episodes, complications of the surgery, liver dysfunction–related diseases, and deaths.
World J Surg
Operative procedures
Table 1 The comparisons and statistical analyzes of preoperative clinical details between the two groups
The paraesophagogastric devascularization procedure was described by Yang and Qiu [12], and the paraesophagogastric devascularization with stapler esophageal transection, also termed the modified Sugiura procedure, was described by Sugiura and Futagawa [7], and by Mathur et al. [13] through an abdominal incision. The main surgical procedures are shown in Fig. 2 [14]. The paraesophagogastric devascularization procedure included complete devascularization of at least half of the lesser curvature of the stomach and more than two-thirds of the greater curvature proximally, and 6–8 cm of the lower esophagus. The left gastric and short gastric vessels, coronary vein, retrogastric veins, as well as the periesophageal collaterals were dissected during this step. An EEA circular stapler (SDH 25–28 mm; Ethicon), was then used for transection and end-to-end anastomosis of the lower esophagus about 3 cm above gastroesophageal junction. Splenectomy was done by the common method. Pyloroplasty was required in the patients who underwent esophageal transection.
Total number of patients (n = 278)
Follow-up The schedule of the follow-up visits was every 3 months during the first year, every 6 months for the next 5 years, and once every year thereafter. All patients were scheduled for upper GI endoscopy once a year after surgery, according to their clinical situations.
Statistical analysis
Group I (n = 180)
Group II (n = 98)
5–77
8–70
47.4 ± 13.1
46.3 ± 13.1
Male
147
80
Female
33
18
Age (years) Range Mean ± SD Gender
Criteria used in the study We defined the study criteria as follows for the present study: (1) operative mortality was considered as death during hospitalization or within the first 6 weeks postoperatively [13]; (2) uncontrolled variceal bleeding referred to either continuing variceal bleeding, even after adequate pharmacotherapy and endotherapy, or major rebleeding
0.495a 0.994b
0.451b
Prior endotherapy EST
33
15
EST and EVL
60
40
EVL
87
43 0.095b
Time frame 1990–1999
66
46
2000–2009
114
52
Hepatitis B related
134
69
0.469b
Hepatitis C related
4
6
0.174c
Alcohol related
29
17
0.79b
Etiology
0.728b
Type of portal hypertension Cirrhotic
167
92
Non-cirrhotic
13
6 0.574b
Child-Pugh classification Grade A
93
47
Grade B
72
45
Grade C
15
6
Timing of surgery Emergency Elective
The statistical software SPSS 15.0 was used for the statistical analysis. The quantitative data of patients were described as the mean ± standard error and range. The qualitative data were described as frequencies. Complications, mortality, rebleeding rate, ascites, and encephalopathy between the two groups were analyzed by the v2 test and Fisher’s exact test. Comparisons of quantitative data were analyzed by t test. Statistical significance was noted when the P value \ 0.05.
P value
a
0.738b 11
7
169
91
t-test
b
Pearson’s Chi squared test
c
Fisher’s exact test
EVL endoscopic variceal ligation, EST endoscopic variceal ligation plus sclerotherapy
within 5 days after the first episode of variceal bleeding [15]; (3) failure of emergency endotherapy was defined as failure to control variceal bleeding, or as variceal rebleeding after two separate endotherapy sessions during the hospitalization or less than two weeks apart [15]; (4) failure of chronic endotherapy was defined as at least two episodes of life-threatening rebleeding occurring during the regular follow-up after chronic endotherapy; (5) variceal rebleeding referred to one new episode of upper gastrointestinal hemorrhage causing a hemoglobin drop of more than 2 g/L within 24 h or occurring within 2 weeks after the initial bleeding had been stabilized [16–18]; (6) early rebleeding
123
World J Surg Table 2 The early outcomes (B 6 weeks) of the surgery Result
Group I (n = 180)
Group II (n = 98)
P value
Bleeding from esophageal stoma
–
1/98 (1.0 %)
0.353b
Esophageal leak
–
3/98 (3.1 %)
0.043b*
Gastric retention
2/180 (1.1 %)
5/98 (5.1 %)
0.025b*
Respiratory infection
5/180 (2.8 %)
2/98 (2.0 %)
0.526b
Wound infection Ischemic necrosis of the stomach wall
4/180 (2.2 %) 1/180 (0.6 %)
2/98 (2.0 %) 0/98 (0 %)
0.643b 0.647b
Variceal rebleeding
4/180 (2.2 %)
3/98 (3.1 %)
0.474b
Intra-abdominal bleeding
6/180 (3.3 %)
2/98 (2.0 %)
0.420b
Portal vein thrombosis
4/180(2.2 %)
2/98(2.0 %)
0.643b
Acute pulmonary embolism
1/180(0.6 %)
1/98(1.0 %)
0.582b
Mortality for emergency surgery
3/11(27.3 %)
5/7(71.4 %)
0.145b
Overall mortality
13/180(7.2 %)
9/98(9.2 %)
0.563a
Liver failure
5/180(2.8 %)
3/98(3.1 %)
0.580b
Postoperative infection
1/180(0.6 %)
1/98(1.0 %)
0.582b
Intra-abdominal bleeding
4/180(2.2 %)
2/98(2.0 %)
0.643b
Variceal rebleeding
3/180(1.7 %)
2/98(2.0 %)
0.583b
Acute pulmonary embolism and DIC
1/180(0.6 %)
1/98(1.0 %)
0.582b
Esophageal transection-related
General complications
Cause of death
a
Pearson’s Chi squared test
b
Fisher’s exact test
*Statistically significant
was defined as rebleeding that occurred within 6 weeks of operation, and late rebleeding was that which occurred beyond 6 weeks after operation; (7) symptomatic hypersplenism referred to splenomegaly with malaise, anemia, thrombocytopenia (PLT \100 9 109/L), WBC \4 9 109/ L, as well as coagulation disorder [19, 20]; (8) emergency surgery referred to surgical procedures performed only if acute variceal bleeding could not be controlled by endotherapy combined with pharmacotherapy [8].
Results The comparisons and statistical analyses of preoperative clinical details between the two groups are shown in Table 1. Early outcomes of surgery (within 6 weeks of surgery) Four patients in group II had complications associated with esophageal transection. One of them bled from the esophagus anastomotic stoma and recovered after surgical hemostasis. The other three patients had esophageal leaks. Two of them recovered after conservative treatment and
123
B-ultrasound-guided percutaneous drainage, but the other died of intra-abdominal infection and septic shock. Four patients in group I and three patients in group II experienced upper gastrointestinal rebleeding after surgery. Three patents in group I and two patients in group II died because of rebleeding. The others recovered after endotherapy and pharmacotherapy. Six cases of intra-abdominal bleeding occurred in group I. One patient had bleeding from the vascular stump and recovered after reoperation. Five other patients experienced bleeding from splenic bed errhysis. Four of them died of hemorrhagic shock, and the fifth patient recovered after conservative treatment. In group II, two patients had errhysis from the splenic bed and both died of hemorrhagic shock. Acute pulmonary embolism occurred in one patient of each group, and both of them died. Four patients in group I and two patients in group II developed portal vein thrombosis, and all of them recovered after administration of anticoagulants. Five patients in group I and three patients in group II died of progressive liver failure. Ischemic necrosis of the stomach at the greater curvature occurred in 1 patient in group I who died of septic shock after surgical repair. Other complications, such as respiratory infection,
World J Surg
developed hepatocellular carcinoma and 6 of them died, whereas 5 patients in group II developed HCC and 3 of them died. The overall mortality in the two groups was 27 % in group I (41/152 patients) and 27.2 % in group II (22/81 patients). The Kaplan–Meier survival test analysis (P [ 0.05) of the two groups is shown in Fig. 3. Four patients in group II developed esophageal stricture with symptoms; 2 patients in group I and 1 patient in group II had incisional hernia repairs. In group I, 18 patients required pharmacotherapy for ascites, as did 7 patients in group II. None of the patients had post-splenectomy sepsis at the end of the study. The long-term outcomes of the two groups are shown in Table 3.
Discussion Fig. 3 The Kaplan–Meier survival test of the two groups
wound infection, and gastric retention occurred in both groups. The early outcomes of the two groups are shown in Table 2. Long-term outcomes of the surgery Of the 167 survivors in group I, 91 % (152/167 survivors) were followed from 8 to 216 months (mean: 67.9 ± 37.3 months), and 91 % of the survivors in group II (81/89 survivors) were followed from 7 to 204 months (mean: 67.4 ± 44.6 months). In group I, the rebleeding rate was 5.9 % (9/152) in 1 year, 17.1 % (26/152) in the first 3 years, and 23.7 % in the first 5 years (36/152). The overall rebleeding rate was 27 % (41/152), and 35 cases of rebleeding were caused by recurrent varices; 25 of those patients were cured by endotherapy and pharmacotherapy, 2 were cured by portosystemic shunt. The other 8 died of uncontrolled bleeding. The other 6 cases of rebleeding were the result of portal gastropathy or hemorrhagic gastritis, and all were successfully managed by pharmacotherapy. In group II, the rebleeding rate was 6.2 % (5/81 patients) in the first year, 13.6 % (11/81 patients) in the first 3 years, and 21 % in the first 5 years (17/81 patients). The overall rebleeding rate was 27.2 % (22/81 patients). In 17 of the 22 patients, the cause of rebleeding was recurrent varices; 11 were cured by endotherapy and pharmacotherapy but 6 died of uncontrolled bleeding. In the other 5 patients, the bleeding was due to portal gastropathy or hemorrhagic gastritis, and all of them were successfully managed by pharmacotherapy. Progressive liver failure occurred in 20 patients in group I and 11 patients in group II. Eight patients in group I
Therapy for variceal bleeding in patients with portal hypertension has evolved over time, and it now it encompasses a spectrum of treatment modalities among which sequential therapies are often necessary [21, 22]. In many centers, the first-line treatment is pharmacotherapy and endotherapy. Once variceal bleeding occurs, the risk of recurrence can be as high as 70 % [23]. The challenge of management for variceal bleeding in patients with portal hypertension is how to prevent rebleeding while maintaining satisfactory liver function. Liver transplantation is beneficial for patients with endstage liver failure, but it is not an ideal treatment for variceal bleeding. Also, because of the limited supply of donor organs, and for economic reasons, liver transplantation cannot be available for all patients. For these reasons, only if the liver is in its decompensated stage should liver transplantation be considered for patients with variceal bleeding secondary to portal hypertension. There is accumulating evidences that variceal bleeding with compensated liver function (Child-Pugh class A and B?) should initially be treated with non-transplantation therapy [23]. Transjugular intrahepatic portosystemic shunt (TIPS) is increasingly used in patients with variceal hemorrhage caused by portal hypertension. Because TIPS is a nonoperative approach, it appears to be a desirable treatment when only temporary portal decompression is required, such as in liver transplantation candidates who fail endotherapy and pharmacotherapy. However, the high rate (40–50 %) of shunt stenosis or shunt thrombosis within the first year will lead to variceal rebleeding, and this risk limits the use of TIPS in clinical work [23]. In 1964, Hassab [24] proposed the procedure of gastroesophageal decongestion and splenectomy, which was based on angiography of portal hypertension. The surgical procedure includes removal of the spleen, total ligation of vessels in the first third of the proximal stomach and 6 cm
123
World J Surg Table 3 Long-term outcomes ([ 6 weeks) of the surgery Result
Group I
Group II
P value
Follow-up rate
152/167 (91.0 %)
81/89 (91.0 %)
0.999a
Average follow-up time(months)
67.9 ± 37.3
67.4 ± 44.6
0.930c
Esophageal stricture
–
4/81 (4.9 %)
0.014b*
Ascites
18/152 (11.8 %)
11/81 (13.6 %)
0.702a
Encephalopathy
10/152 (6.6 %)
10/81 (12.3 %)
0.135a
Incisional hernia
2/152 (1.3 %)
1/81 (1.2 %)
1.000b
Portal hypertensive gastropathy
30/152 (19.7 %)
23/81 (28.4 %)
0.133a
Rebleeding cases of 1 year
9/152 (5.9 %)
5/81 (6.2 %)
0.939a
Rebleeding cases of 3 year
26/152 (17.1 %)
11/81 (13.6 %)
0.483a
Rebleeding cases of 5 year Overall rebleeding cases
36/152 (23.7 %) 41/152 (27 %)
17/81 (21.0 %) 22/81 (27.2 %)
0.40a 0.976a
Variceal bleeding
35/152 (23.0 %)
17/81 (21.0 %)
0.722a
Gastropathy
6/152 (3.9 %)
5/81 (6.2 %)
0.521b
Mortality of 1 year
12/152 (7.9 %)
2/81 (2.5 %)
0.147b
Mortality of 3-year
25/152 (16.4 %)
10/81 (12.3 %)
0.404a
Mortality of 5-year
34/152 (22.4 %)
19/81 (23.5 %)
0.850a
Overall mortality
43/152 (28.3 %)
23/81 (28.4 %)
0.986a
Liver failure
20/152 (13.2 %)
11/81 (13.6 %)
0.928a
Variceal bleeding
8/152 (5.3 %)
6/81 (7.4 %)
0.512a
Hepatic carcinoma
6/152 (3.9 %)
3/81 (3.7 %)
0.616b
Overall esophageal transaction- related morbidity
–
8/98 (8.2 %)
\ 0.001b*
Long-term complications ([ 6 weeks)
Cause of rebleeding
Cause of death
a
Pearson’s Chi squared test
b
Fisher’s exact test
c
t-test
*Statistically significant
above the gastroesophageal junction. In the 1970s, Yang and Qiu [12] realized the importance of the upper esophageal branch, the heterotopic esophageal branch, and the posterior gastric vein in cases of variceal bleeding of portal hypertension, and proposed a more thorough paraesophagogastric vascular disconnection, including disconnection of the veins mentioned above. In 1984, Sugiura and Futagawa [7] proposed the Sugiura procedure, which consisted of esophageal transection through the chest, and paraesophagogastric devascularization through the chest and abdomen. However, because of the complexity of this procedure and the severe postoperative complications associated with it, some medical institutions improved the surgical procedure by adding paraesophagogastric devascularization with esophageal transection through the abdomen. They renamed it the modified Sugiura procedure [8]. The modified Sugiura procedure has proved to be effective in the management of acute variceal bleeding, but there is high rate of operative complications and a
123
remarkable rebleeding rate, up to 40 % in patients with liver cirrhosis [9, 10]. In the present study, whether esophageal transection was performed or not did not affect the rebleeding rate during the perioperative period (2.2 % in group I vs 3.1 % in group II), or during the first postoperative year (5.9 vs 6.2 %), the third postoperative year (17.1 vs 13.6 %), or the fifth postoperative year (23.7 vs 21.0 %), as well as overall (27 vs 27.2 %). The rebleeding rates show no statistical significance between the two groups (Table 3). These observations indicate that the esophageal transaction procedure may not be essential for devascularization. In the present study, the overall mortality is 28.3 % in group I and 28.4 % in group II. Mortality in the perioperative period was 7.2 % in group I versus 9.2 % in group II; during the first postoperative year, 7.9 versus 2.5 %; during the third postoperative year, 16.4 versus 12.3 %); and during the fifth postoperative year, 22.4 versus 23.5 %. The differences in the two groups show no statistical significance. The Kaplan–Meier survival test for the two
World J Surg
groups also shows that the differences in survival rates of these two groups have no statistical significance. In addition, the data indicate that the mortality of the two groups is not affected by the esophageal transection procedure. Mortality after emergency devascularization is 71.4 % (5/ 7) in group II, which is much higher than the 27.3 % (3/11) in group I, although the statistical analysis shows no significance. Several factors may contribute to rebleeding after devascularization, including the following: (1) failure to devascularize the paraesophagogastric veins completely, especially missing the upper esophageal branch, the heterotopic esophageal branch, and the posterior gastric vein [12]; (2) perfusion of the stomach is influenced by paraesophagogastric devascularization, which can cause portal hypertensive gastropathy [25]; (3) regeneration of esophageal and gastric varices. In addition to paraesophagogastric devascularization, the modified Sugiura procedure adds esophageal transaction, which, theoretically at least, completely blocks the connection between esophageal and gastric varices. However, both the postoperative and long-term rebleeding rates show no statistical differences, whether esophageal transaction is performed or not. Possible reasons might be that the left common submucosal veins can hardly form varices because of disconnection between portal vein system and esophagogastric veins after complete paraesophagogastric devascularization. Under this circumstance, further esophageal transection seems meaningless in the prevention of rebleeding. Secondarily, the modified Sugiura procedure may aggravate submucosal congestion and ischemia of the stomach and eventually lead to gastropathy. The devascularization procedure without esophageal transection avoids transection-related morbidity, which is as high as 8.2 % (8/ 98 patients). The most common complications associated with esophageal transection are esophageal leak and bleeding from the esophageal stoma [26, 27]. The mortality of esophageal leak was reported to be as high as 60 % [27]. In our study, the rate of esophageal leak was 3.1 % in group II, compared to reports of 6–13 % [13, 26] from other studies. Our three patients with leakage from the transection line had undergone sclerotherapy. However, there were no statistical differences between patients who had prior sclerotherapy and those who had not in our group II (3/55 vs 0/43; p = 0.120), Nevertheless, it should be recognized that the use of prior sclerotherapy may lead to thickening around the distal esophagus as well as hypoperfusion at distal esophagus, which may eventually cause a leak. The rate of mortality related to leak was 1.0 % in our study. Furthermore, the incidence of esophageal strictures in our study was 4.9 %, and all were cured by endoscopic dilatation. The results of our study indicate that devascularization combined with esophageal transaction
can lead to a higher rate of postoperative transectionrelated morbidity [27–30]. Furthermore, it will not improve postoperative morbidity or reduce the rate of rebleeding below the levels achieved with devascularization without esophageal transection.
Conclusions In the management of variceal bleeding, paraesophagogastric devascularization without esophageal transection is as effective and safe as devascularization with esophageal transection, but with less esophageal transection-related morbidity. Acknowledgments This work was supported by the 5010 Clinical Research Project of SUN Yat-Sen University (2010009).
References 1. Infante-Rivard C, Esnaola S, Villeneuve JP (1989) Role of endoscopic variceal sclerotherapy in the long-term management of variceal bleeding: a meta-analysis. Gastroenterology 96: 1087–1092 2. D’Amico G, Pagliaro L, Bosch J (1995) The treatment of portal hypertension: a meta-analytic review. Hepatology 22:332–354 3. Bendtsen F, Krag A, Moller S (2008) Treatment of acute variceal bleeding. Dig Liver Dis 40:328–336 4. Vlachogiannakos J, Goulis J, Patch D et al (2000) Review article: primary prophylaxis for portal hypertensive bleeding in cirrhosis. Aliment Pharmacol Ther 14:851–860 5. Stanley AJ, Hayes PC (1997) Portal hypertension and variceal haemorrhage. Lancet 350:1235–1239 6. Van Stiegmann G, Goff JS, Sun JH et al (1990) Endoscopic ligation of esophageal varices. Am J Surg 159:21–26 7. Sugiura M, Futagawa S (1984) Esophageal transection with paraesophagogastric devascularizations (the Sugiura procedure) in the treatment of esophageal varices. World J Surg 8:673–679. doi:10.1007/BF01655762 8. Battaglia G, Ancona E, Patarnello E et al (1996) Modified Sujura operation: long-term results. World J Surg 20:319–325. doi:10. 1007/s002689900051 9. Comar KM, Sanyal AJ (2003) Portal hypertensive bleeding. Gastroenterol Clin North Am 32:1079–1105 10. Gouge TH, Ranson JH (1986) Esophageal transection and paraesophagogastric devascularization for bleeding esophageal varices. Am J Surg 151:47–54 11. Conn HO (1967) Ammonia tolerance in the diagnosis of esophageal varices. A comparison of endoscopic, radiologic, and biochemical techniques. J Lab Clin Med 70:442–451 12. Zhen Yang, Fazu Qiu (2001) The typical technique of pericardial devascularization and its effects on portal hypertension. Central China Med J 25:229–230 13. Mathur SK, Shah SR, Nagral SS et al (1999) Transabdominal extensive esophagogastric devascularization with gastroesophageal stapling for management of noncirrhotic portal hypertension: long-term results. World J Surg 23:1168–1175. doi:10.1007/ s002689900641 14. Johnson M, Rajendran S, Balachandar TG et al (2006) Transabdominal modified devascularization procedure with or without
123
World J Surg
15.
16.
17.
18. 19.
20.
21.
22.
esophageal stapler transaction—an operation adequate for effective control of a variceal bleed. Is esophageal stapler transection necessary? World J Surg 30:1507–1519. doi:10.1007/s00268005-0754-x Jalan R, John TG, Redhead DN et al (1995) A comparative study of emergency transjugular intrahepatic portosystemic stent— shunt and esophageal transection in the management of uncontrolled variceal hemorrhage. Am J Gastroenterol 90:1932–1937 Luketic VA, Sanyal AJ (2000) Esophageal varices. II. TIPS (transjugular intrahepatic portosystemic shunt) and surgical therapy. Gastroenterol Clin North Am 29:387–421 Jalan R, Hayes PC (2000) UK guidelines on the management of variceal haemorrhage in cirrhotic patients. British Society of Gastroenterology. Gut 46(Suppl 3–4):I1–I15 De Franchis R (1996) Developing consensus in portal hypertension. J Hepatol 25:390–394 Krige JE, Bornman PC, Goldberg PA et al (2000) Variceal rebleeding and recurrence after endoscopic injection sclerotherapy: a prospective evaluation in 204 patients. Arch Surg 135:1315–1322 Sorbi D, Gostout CJ, Peura D et al (2003) An assessment of the management of acute bleeding varices: a multicenter prospective member-based study. Am J Gastroenterol 98:2424–2434 de Franchis R (2010) Revising consensus in portal hypertension: report of the Baveno V consensus workshop on methodology of diagnosis and therapy in portal hypertension. J Hepatol 53:762–768 Garcia-Tsao G, Bosch J (2010) Management of varices and variceal hemorrhage in cirrhosis. N Engl J Med 362:823–832
123
23. Courtney M. Townsend, R Daniel Beauchamp B, et al (2012) Portal hypertension. In: Sabiston textbook of surgery, 19th edn. W. B. Saunders Company, Philadelphia, pp 1433–1434 24. Hassab MA (1964) Gastroesophageal decongestion and splenectomy. A method of prevention and treatment of bleeding from esophageal varices associated with bilharzial hepatic fibrosis: preliminary report. J Int Coll Surg 41:232–248 25. Editorial (1991) Portal hypertensive gastropathy. Lancet 338: 1045-1046 26. Orozco H, Takahashi T, Mercado MA et al (1994) Surgical management of extrahepatic portal hypertension and variceal bleeding. World J Surg 18:246–250. doi:10.1007/BF00294409 27. Dagenais M, Langer B, Taylor BR et al (1994) Experience with radical esophagogastric devascularization procedures (Sugiura) for variceal bleeding outside Japan. World J Surg 18:222–228. doi:10.1007/BF00294405 28. Idezuki Y, Sanjyo K (1989) Twenty-five-year experiences with esophageal transection for esophageal varices. J Thorac Cardiovasc Surg 98:876–883 29. Han HS, Yi NJ, Kim YW et al (2004) New operative method for fundal variceal bleeding: fundectomy with periesophagogastric devascularization. World J Surg 28:406–410. doi:10.1007/ s00268-003-6923-x 30. Hassab MA (1998) Gastro-esophageal decongestion and splenectomy GEDS (Hassab), in the management of bleeding varices. Review of literature. Int Surg 83:38–41
