- Email: [email protected]
Selective double disconnection for cirrhotic portal hypertension
Accepted Manuscript Selective double disconnection for cirrhotic portal hypertension Guang-quan Zong , M.D. Yang Fei , M.D. Jian Chen , M.D. Ren-min Liu , M.D. PII:
S0022-4804(14)00532-0
DOI:
10.1016/j.jss.2014.05.065
Reference:
YJSRE 12765
To appear in:
Journal of Surgical Research
Received Date: 25 March 2014 Revised Date:
1 May 2014
Accepted Date: 21 May 2014
Please cite this article as: Zong G-q, Fei Y, Chen J, Liu R-m, Selective double disconnection for cirrhotic portal hypertension, Journal of Surgical Research (2014), doi: 10.1016/j.jss.2014.05.065. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Revised May 1st 2014
ACCEPTED MANUSCRIPT
Selective
double
disconnection
for
cirrhotic
portal
hypertension
RI PT
Guang-quan Zonga M.D., Yang Fei*a M.D., Jian Chen M.D. and Ren-min Liu M.D.
Department of General Surgery, the 81st Hospital of P.L.A., P.L.A. Cancer Center,
SC
Nanjing 210002, China.
* a
First co-authors
E-mail: [email protected] Telephone: +86-25-86648090
TE D
Fax: +86-25-80864499
M AN U
Corresponding author
EP
Contributors: Guang-quan Zong, Yang Fei wrote the paper, organized the figures and patient data and did the analysis; Ren-min Liu carried out the statistical analysis
AC C
and corrected the paper; Yang Fei supervised the writing and organization process.
ACCEPTED MANUSCRIPT
Abstract Object: To evaluate the effect of selective double portazygous disconnection with preserving vagus ( SDPDPV ) for patients with portal hypertension in the authors’ hospital.
RI PT
Methods: 453 patients with cirrhotic portal hypertension who underwent either SDPDPV or pericardial devascularization with splenectomy (PDS) for variceal bleeding from Feb 2007 to January 2013 were retrospectively reviewed. The
complications and clinical outcomes were analyzed.
SC
operation-relevant information, change of lavatory examination data, postoperative
Results: There were no significant difference between the SDPDPV group and the
M AN U
PDS group of mean operative time and intraoperative blood loss (P
0.05). The FPP
( free portal pressure) in the SDPDPV group was much lower than PDS group significantly after operation (P
0.05). The test of biochemical profile of hepatocyte
functions and Child-Pugh’s score at the end of the first postoperative year were significantly more altered in the SDPDPV group than in the PDS group (P
0.05).
TE D
Except encephalopathy, occurrences or development of postoperative complications including rebleeding, ascites and gastric stasis showed great difference between the two groups (P < 0.05). The operative mortality rate and the 3-year survival rates were
EP
great difference between the two groups too (P < 0.05). Conclusion: The SDPDPV not only controls recurrent bleeding from varices with
AC C
portal hypertension effectively but also maintains normal dynamics of stomach and physiological function of intestine and hepatobiliary.
Key words
portal hypertension; variceal hemorrhage; devascularization; vagus
Introduction
ACCEPTED MANUSCRIPT Portal hypertension (PHT) is defined as portal venous pressure exceeding 10 mmHg (14 cm of H2O). In China, the most common cause of PHT is cirrhosis[1]. This syndrome develops in the majority of patients with cirrhosis and is responsible for a frequent and severe complication of cirrhosis, massive bleeding from ruptured
RI PT
gastroesophageal varices is the main complication of PHT. Approximately 90% of patients with cirrhosis will develop gastroesophageal varices over 10 years and one-third of these will bleed from them. The risk of bleeding from gastroesophageal varices is about 25% within 2 years of diagnosis[2]. Catastrophic bleeding from
SC
gastroesphageal varices rupture is a life-threatening emergency. In many instances pharmacological therapy such as β- adrenergic blockers, and nitrates or endoscopic
M AN U
therapy such as band ligation cannot stop the bleeding and this area is also beyond the field of sclerotherapy[3,4]. Surgery seems the only definite therapy, which may offer chances for survival. Devascularization and shunting are the two widely accepted approaches
for
the
control
of
such
bleeding[5-8].
In
China,
pericardial
devascularization with splenectomy (PDS) has been widely used in patients with portal
TE D
hypertension as an effective haemostatic method, because it is less of a burden and is associated with a lower incidence of encephalopathy compared to shunt procedures[9]. But PDS truncated the diversion from portal vein to azygos vein, the reduction of portal vein
EP
pressure was not significant postoperatively[8,10]. It may promote the establishment of
collateral circulation rapidly, establish the path between portal vein and azygos vein again. Then, recurrence of gastroesophageal varices will appear and portal hypertensive gastric
AC C
mucosal lesions will be further aggravated. Thus PDS was shown to have a high rebleeding rate[11-13]. If patients with ascites undergone PDS, ascites was often difficult to be
controlled postoperatively[14]. On the other hand, once the vagus especially vagus trunk is interrupted during PDS procedure, it is easy for patients to suffer from delayed gastric emptying and gastric stasis[15]. In order to prevent these postoperative complications, a new procedure, selective pericardial vascular disconnection plus pericardial submembranous varices ligation by preserving vagus trunks (Selective double portazygous disconnection with preserving vagus, SDPDPV) has been performed in our department for the treatment of patients with cirrhotic portal
ACCEPTED MANUSCRIPT hypertension, since Feb 2007. The efficacy, safety, postoperative complications and survival were compared with those of single PDS performed during the same period.
Materials and methods
RI PT
General Information
The study population comprised 453 variceal bleeders with cirrhosis PHT
managed by surgery in the period between February 2007 and January 2013 at the 81st
SC
Hospital of P.L.A. in Nanjing, China. Consecutively 266 patients received SDPDPV and 187 PDS. The diagnoses of these patients were all confirmed by endoscopy,
M AN U
barium meal and Doppler ultrasonography. Our indications for operations include episodes of gastroesophageal variceal hemorrhage with a 1-month interval between the last attack and date of surgical procedure, which could not be controlled by medical means and endoscopic therapy, or thrombocytopenia (platelets count under 80×103/mm3) with gastroesophageal varices due to hypersplenism related to cirrhosis.
TE D
All of the patients accompanied with no serious cardiopulmonary diseases and concomitant chronic duodenal ulcers. The characteristics of the two surgical groups including age, gender, Child-Pugh’s score, preoperative variceal grade defined according to the criteria of Calès et al[16], follow-up time, preoperative biochemical
EP
tests were summarized in Table 1. Biochemical tests were examined within 1-3 days preoperatively, endoscopy or a barium meal examination was performed within
AC C
fourteen days before the operation to define the grade of varices.
Operative technique SDPDPV procedure was performed through an expanded left subcostal incision
of the left upper abdomen. The FPP (free portal pressure) is measured by inserting a catheter into the portal trunk through a branch of the right gastroepiploic vein. After conventional extreme splenectomy, The FPP is measured again. The proximal stomach was devascularized close to the gastric wall just above the crow’s foot along the lesser curvature from the incisura angularis up to the esophagus, we paid attention to avoid ligating the vagus nerve. Dissecting anterior Serosa layer and the left lateral
ACCEPTED MANUSCRIPT peritoneum covering the esophagus, and separating the muscle layer carefully, paraesophageal vein was exposed,. Following by lifting the stomach and dissecting gastropancreatic fold around the gastroesophageal junction, the left gastric artery and left gastric vein (coronary vein) were exposed. The gastric branch of left gastric vein
RI PT
and branches of left gastric artery were disconnected and suture ligated close to the gastric wall of lesser curvature in order to preserve the trunk of left gastric vein and artery. Then perforating branches from paraesophageal vein to the lower part of
esophagus were disconnected and suture ligated, tried to keep paraesophageal vein
SC
entirely. Dissection the lower part of esophagus was performed up to 7-10 cm above the cardia (the esophageal hiatus level). During the course of procedure, trunk of
M AN U
vagus nerve may be preserved. Whole layer discontinuous suture guided by a stomach tube with 3-0 prolene threads (Ethicon, Inc.) was performed around the lower part of esophagus at the area of 2-10cm above the cardia. The overlap length between the two needles was about two mm, and the direction of suture lines were not on the same horizontal level but showed zigzag-like to prevent esophageal stricture
TE D
postoperatively. The seromuscular layers of the lesser curvature were closed with interrupted silk sutures. After checking peritoneal cavity hemorrhage and circulation of free zone in the stomach and esophagus, we covered the wound of splenic fossa
EP
and retroperitoneal with omentum. Finally, the FPP was remeasured. A peritoneal cavity drainage tube was placed at left subphrenic location, before suturing the abdominal incision, a biopsy specimen of the liver was taken.
AC C
The technique of standard PDS was applied as described by Hassab[17].
Follow-up Studies
All patients had undergone follow-up from one to seven years (4.6+ 1.2 years). Clinical variables including episodes of recurrent hemorrhage, encephalopathy, ascites et al. were examined postoperatively, laboratory variables including biochemical tests were systematically examined at two weeks postoperatively, and then reperformed at six months and one year thereafter. In our studies, rebleeding was defined as at least 400ml blood transfusion
ACCEPTED MANUSCRIPT requirement and/or a decrease in the hematocrit to less than 30%, operative mortality was defined as death within thirty days of surgery[18].
Statistical Analysis All statistical analyses were performed using SPSS15.0 software (SPSS, Chicago,
RI PT
IL). Continuous data were expressed as mean values + standard deviation (SD). Significant differences between groups were determined by chi-squared analysis and unpaired Student's t test. Overall survival estimates were calculated by the Kaplan-Meier method and compared with similar estimates obtained using the
M AN U
SC
log-rank test. P-values <0.05 were considered statistically significant.
Results
Preoperative comparison of the two operational groups There were no significant difference between the SDPDPV group and the PDS group of the preoperative database including age, gender, Child-Pugh’s score,
0.05 )
Table 1
TE D
preoperative variceal grade, follow-up time, blood routine and biochemical tests (P . The two groups were well balanced in the distribution of
prognostic factors and other characteristics.
EP
operation-relevant information
All 453 operations were successful. The mean operative time of SDPDPV group
AC C
was 236.0 + 73.8 min with a range of 131 min to 238 min. In PDS group, it was 242.1 + 68.8 min with a range of 146 min to 242 min. The difference of mean operative time between the two groups was not significant (P = 0.387). (Table 2) The intraoperative blood loss of SDPDPV group was 572.2 + 71.5 ml, In PDS group, it was 630.7 + 102.1 ml. The difference of blood loss between the two groups was not significant too (P = 0.275). (Table 2)
Portal vein pressure As shown in Table 3, the difference of FPP between the SDPDPV group and the PDS group were not significant before operation and after splenectomy (P = 0.481, P
ACCEPTED MANUSCRIPT = 0.393 ). However, in contrast to the PDS group, the FPP in the SDPDPV group was much lower significantly after devascularization. (P = 0.013).
Biochemical data and hepatic functional reserve The test of biochemical profile of hepatocyte functions such as Albumin (Alb),
RI PT
alkaline phosphatase (ALP), alanine aminotransferase (ALT) and aspartate aminotransferase AST at two weeks, six months and one year after operations were all significantly more altered in the SDPDPV group than in the PDS group (P 0.05). Apart from these, both the total bilirubin (TB) and the prothrombin time (PT) in the
differences were not significant ( P
0.05 ) (Table 4).
SC
SDPDPV group were lower than that in the PDS group postoperatively, however the
M AN U
Child-Pugh’s classification is a well-validated method of assessing hepatic functional reserve in cirrhosis and is one of the best measures for prediction of long-term outcome[19]. In our studies, postoperative Child-Pugh’s score of SDPDPV group (at two weeks, six months and one year after operations) were all lower than that of PDS group significantly (P 0.05). (Table 4)
Recurrent hemorrhage
TE D
Postoperative complications
Recurrent hemorrhage of SDPDPV group and PDS group showed the following
Encephalopathy
EP
incidence of 5/266 (1.9%) and 21/187 (11.2%) respectively ( P= 0.001). (Table 2).
AC C
None of the studied patients experienced encephalopathy before operation. The incidence of postoperative encephalopathy was not significantly higher after SDPDPV (4.9%, 13/266) compared to PDS (5.3%, 10/187) ( P= 0.120) (Table 2). Ascites
Early postoperative (one month after operation) ascites developed in twelve patients (4.5%) after SDPDPV and in sixteen patients (8.6%) after PDS (P=0.032). Diuretic therapy controlled ascites in most of these surgical patients within the first year. At the time of a year after operation, ascites was detected in 7 SDPDPS (7.4%) and 4 PDS (5%) survivors (P=0.162). (Table 2).
ACCEPTED MANUSCRIPT Gastric stasis Gastric stasis developed in twenty-seven patients (14.4%) after PDS, but it didn’t develop in patients underwent SDPDPV, the difference was significant(P 0.001). (Table 2).
RI PT
Operative mortality The operative mortality rate was 0.4%(1/266) in the SDPDPV group and 3.7%
(7/187)in the PDS group. There was a significant difference among the groups (χ2 = 5.803; P=0.009). The reason of death in the SDPDPV group was acute portal vein
SC
thrombosis (one patient ), instead it were varicral bleeding (four patients ), hepatocyte function failure(two patients ) and portal vein thrombosis (one patient ) in the PDS
M AN U
group.
Long-term survival
After a mean follow-up of 4.6 + 1.2 years, forty-one patients (nineteen patients underwent SDPDPV, thirty-seven patients underwent PDS) had died. Thirteen
TE D
patients ( seven in SDPDPV group, six in PDS group) were lost to follow-up after one year. In SDPDPV group, recurrent bleeding was the cause of death in four patients,. Other causes of death included hepatomas in eight patients, liver failure in four patients, cecum carcinoma, portal vein thrombosis,and myocardial infarction in one
EP
patient each. Nevertheless in PDS group, the main reason of death were recurrent bleeding and liver failure. Seventeen patients died of recurrent bleeding and thirteen
AC C
patients died of liver failure respectively. Other causes of death included hepatomas in five patients, portal vein thrombosis in two patients. Less patients died of recurrent bleeding in SDPDPV group( 1.5% vs 9.4%), and the difference was significant (χ2 = 4.571; P=0.012). The data about liver failure were similar to that of recurrent bleeding compared between SDPDPV group and PDS group (1.5% vs 7.2%), the difference was significant too (χ2 = 2.937; P=0.0312 ). however, the difference about patients died of hepatomas between the two groups was not significant.( χ2 = 0.842; P=0.206) One-, two-, and three-year survival rates were 98.8%(256/259), 98.1% ( 254/259), and 95.4% (247/259), respectively, in SDPDPV group and 92.8%
ACCEPTED MANUSCRIPT ( 168/181), 90.1% ( 163/181), and 84.0% (152/181), respectively, in PDS group. The three-year survival rates of patients underwent SDPDPV was significantly greater than that underwent PDS (log-rank test, χ2 =4.263; P= 0.038). (Figure 1).
RI PT
Discussion PHT is associated with the most severe complications of cirrhosis, including bleeding from gastroesophageal varices, hepatic encephalopathy and ascites[2,20]. Many patients require surgical treatment each year. In China, PDS described by
SC
Hassab firstly in the 1964[21] is the most common elective procedure among all the surgical procedures performed. Though the initial control of bleeding is good and
M AN U
encephalopathy rate is low after PDS, the high rate of residual or recurrent varices and the exacerbated pathologic changes in the gastric mucosa were associated with a high rebleeding risk[12,13,22].
There are a few varicose rete between the outer longitudinal and inner annular tunica muscularis around cardia and lower esophagus (4-5cm above cardia) in patients
TE D
with PHT[23]. They connect paraesophageal venous rete wtih submucosal venous rete. Some studies showed that variceal hemorrhage located in the lamina propria, but not in the submucosa[24-26].
EP
PDS procedure only disconnected external vessels to serosa of perigastroesophagus, whereas neglected varicose vein plexus from submucosa and tunica muscularis. Thus, once the veins around the lower esophagus and fundus which
AC C
could be regarded as natural shunt channel was blocked after devascularization, portal vein pressure of splenic and gastric area was further increased, varicose vein plexus from perigastroesophageal submucosa and tunica muscularis was further aggravated, rerupture of vein could happen finally. In procedure of SDPDPV, we performed whole layer discontinuous suture with 3-0 prolene threads around the lower part of esophagus at the area of 2-10cm above the cardia, then both the extramural vessels and the intramural vessels were disconnected, venous pressure of area above the ligature would be much lower following. The direction of suture lines were not on the same horizontal level but showed zigzag-like to prevent esophageal stricture
ACCEPTED MANUSCRIPT postoperatively. Pericardial vascular disconnection plus pericardial submembranous varices ligation can get the significant efficiency of devascularization thoroughly without risk of anastomotic leakage which might happen after Sugiura’s operation[27]. Our studies showed that the incidence of postoperative recurrent hemorrhage was
RI PT
significantly lower after SDPDPV compared to PDS. In the present series, the operative mortality rate in the SDPDPV group was
lower than that in the PDS group. There were few patients die of varicral bleeding and hepatocyte function failure after SDPDPV. On the contrary, both of them were the
SC
main reason of death after PDS.
The therapeutic goals of surgical procedures for PHT are mainly achieved by
M AN U
selective reduction of the portal vein pressure[12,28]. Our studies illustrated that in contrast to the PDS group, the mean value of FPP in the SDPDPV group was much lower significantly after operation. We consider that the diversion from portal vein to azygos vein may be truncated after PDS, it will promote the establishment of collateral circulation rapidly[29,30]. and establish the path between portal vein and
TE D
azygos vein again, the FPP still maintain high level. Zhang et al. think the changes in hepatic hemodynamic status after PDS had a beneficial effect on hepatocyte functional reserve because these provided sufficient portal vein perfusion to liver[31],
EP
but the change can promote the formation of new collateral vessels in the periesophagogastric area, and increase the risk of rebleeding following. On the other hand, injury of paraesophageal vein and gastric coronary vein trunk were prevented
AC C
during the procedure of SDPDPV, then retention of these vicarious vessel channels may decrease FPP postoperatively. The similar results had been obtained by Xie et al[32]. But compared with PDS, the diminution of hepatic blood supply resulted from reduction of FPP was not significant. Perhaps the main reason was that some spontaneous shunts were retained intraoperatively, thus further reducing of portal blood flow could appear which led to reduced hepatic sinusoidal pressure and increased hepatic artery flow velocity[33,34]. Moreover, the vasodilator objects including nitric oxide were inactivated in liver due to shunts[35,36]. In summary, SDPDPV not only reduce the FPP and prevent variceal hemorrhage, but also maintain
ACCEPTED MANUSCRIPT hepatic perfusion and preserve hepatocyte functions. Our studies showed hepatocyte function and Child-Pugh’s score which is a well-validated method of assessing hepatocyte functional reserve in cirrhosis improved greater in SDPDPV group compared to that in PDS groups significantly after operations. Preservation of
RI PT
hepatocyte functions is a major determinant of long-term surgical outcome. In our studies, patients with PDS had a significantly worse prognosis than those with SDPDPV analyzed by 3-year survival rates.
Both SDPDPV and PDS belonged to devascularization, so encephalopathy was
SC
clinically detected with similar low incidence among survivors of both procedures.
Ascites was frequently observed in the devascularization, whereas it seldom occurred
M AN U
after shunting. It is already well known that shunts are effective in the prevention of ascites in patients with cirrhosis[37,38]. The mechanisms include a high hydrostatic pressure of portal hypertension leading to increased production, transudation of hepatic and splanchnic lymph into the peritoneal cavity, The ascitogenic effect of extensive dissection with lymphatic interruption may also explain the higher rate of
TE D
early postoperative ascites after PDS. Compared with PDS, a low portal hypertension pressure and limited dissection with lymphatic interruption occurred after the procedure of SDPDPV. Our studies showed incidence of early postoperative ascites
EP
after SDPDPV was lower than that after PDS. Furthermore, we found there were no difference between SDPDPS group and PDS group about incidence of ascites at the time of a year after operation, it should be considered diuretic therapy controlled
AC C
ascites in most of these surgical patients, and the natural history of portal hypertension with chronic liver disease. Complications of PDS included vagus injury[39]. Vagus injury tremendously enhances the incidence of gastric emptying delay, occasional bilious vomiting, and pyloric stricture; the injury possibly leads to gastric stasis, even make portal hypertensive gastropathy more severe[40]. However, in SDPDPV, the dissection was done close to the esophagus and the stomach to preserve the anterior and posterior vagus trunks and the nerves of Latarget. Thus, the function of antrum and pylorus could be retained, bile reflux might be reduced which effect gastric mucosal barrier.
ACCEPTED MANUSCRIPT The studies showed that Gastric stasis developed in 27 patients after SDPDPV was much lower than that
in 12 patients
after PDS significantly.
We found that there were no significant difference of mean operative time and the intraoperative blood loss between the two groups. It showed that SDPDPV as a
RI PT
new procedure was operable, viable and safe.
Conclusions
The SDPDPV not only controls recurrent bleeding from varices with portal
M AN U
physiological function of intestine and hepatobiliary.
Conflicts of interest
EP
TE D
There was no potential conflict of interest.
AC C
SC
hypertension effectively but also maintains normal dynamics of stomach and
RI PT
ACCEPTED MANUSCRIPT
Reference
SC
[1] Wang Y. Surgical treatment of portal hypertension. Hepatob Pancreat Dis 2002;1:211–214.
J Hepatol. 2000;32(1 Suppl):141-156.
M AN U
[2] Bosch J, García-Pagán JC. Complications of cirrhosis. I. Portal hypertension.
[3] Gournay J, Masliah C, Martin T, et al. Isosorbide mononitrate and propranolol compared with propranolol alone for the prevention of variceal rebleeding. Hepatology 2000;31:1239-1245.
TE D
[4] Larson AW, Cohen H, Zweiban B et al. Acute esophageal variceal sclerotherapy. Results of a prospective randomized controlled trial. JAMA 1986; 255: 497-500. [5] Jin G, Rikkers LF. Transabdominal esophagogastric devascularization as treatment
EP
for variceal hemorrhage. Surgery 1996;120:641.. [6] Yoshida H, Mamada Y, Taniai N, et al. New trends in surgical treatment for portal hypertension. Hepatol Res 2009;39:1044-1049.
AC C
[7] Vangeli M, Patch D, Burroughs AK. Salvage TIPS for uncontrolled variceal bleeding. J Hepatol 2002;37:703–704. [8] Henderson JM, Boyer TD, Kutner MH, et al. Distal splenorenal shunt versus transjugular intrahepatic portal systematic shunt for variceal bleeding: a randomized trial. Gastroenterology 2006;130:1643-1651 [9] da Silva LC, Strauss E, Gayotto LC, et al. A randomized trial for the study of the elective surgical treatment of portal hypertension in masonic schistosomiasis. Ann Surg 1986;204:148-156. [10] Bothe A, Stone MD, McDermott WV. Portoazygous disconnection for bleeding
ACCEPTED MANUSCRIPT esophageal varices. Am J Surg 1985; 149:546-550. [11] Gouge TH, Ranson JHC. Esophageal transection and paraesophagogastric devascularization for bleeding esophageal varices. Am J Surg 1986; 151:47-53. [12] El-Serag HB, Everhart JE. Improved survival after variceal hemorrhage
RI PT
over an 11-year period in the Department of Veterans Affairs. Am J Gastroenterol 2000;95:3566-3573.
[13] Carvalho DL, Capua A Jr, Leme PL. Portal flow and hepatic function after splenectomy and esophagogastric devascularization. Int Surg 2008;93:314.
SC
[14] Xu XB, Cai JX, Leng XS, et a1. Clinical analysis of surgical treatment of portal hypertension. World J Gastroenterol 2005;11:4552-4559.
M AN U
[15] Lu YF, Hou YD, Jia SR, et al. Experimental study of pylorus and pyloric vagus preserving gastrectomy. World J Surg 1993;17:525–529.
[16] Calès P, Braillon A, Lebrec D. Medical treatment of portal hypertension and oesophageal varices.Br Med J (Clin Res Ed) 1983;287:1144.
[17] Hassab MA. Gastroesophageal decongestion and splenectomy in the treatment of
TE D
esophageal varices in bilharzial cirrhosis: further studies with a report on 355 operations. Surgery 1967;61:169-176.
[18] Ezzat FA, Abu-Elmagd KM, Aly MA, et al. Selective shunt versus nonshunt
EP
surgery for management of both schistosomal and nonschistosomal variceal bleeders. Ann Surg 1990 ;212:97-108.
[19] Papatheodoridis GV, Goulis J, Leandro G, et al. Transjugular intrahepatic
AC C
portosystemic shunt compared with endoscopic treatment for prevention of variceal rebleeding: a metaanalysis. Hepatology 1999;30:612–622. [20] D’Amico G, de Franchis R, Cooperative Study Group. Upper digestive bleeding in cirrhosis. Post-therapeutic outcome and prognostic indicators. Hepatology 2003;38:599–612. [21] Hassab MA. 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 1964;41:232-238. [22] Estes NC, Pierce GE. Late results of extended devascularization procedure
ACCEPTED MANUSCRIPT for patients with bleeding esophageal varices. Am J Surg 1984; 50:381-385. [23] Wang ZH, Cai HZ, Zhao AX, et al. The effects of disconnection and shunt procedure on hepatic hemodynamics and function of active cirrhotic patients with portal hypertension. Chinese J of Clin Hepatol 2002;118:309–310.
RI PT
[24] Zhou GW, Yan DY, Li F, et al. The retrospective analysis of surgical outcome of portal hypertension. Zhonghua Wai Ke Za Zhi 2009;47:1532-1535.
[25] Chikamori F, Kuniyoshi N, Shibuya S, et al. Correlation between endoscopic and angiographic findings in patients with esophageal and isolated gastric varices.
SC
Digest Surg 2001;18:176-81.
[26] Stimec BV, Grønvold LA, Ignjatovic D, et al. Gastric varices without portal
M AN U
hypertension: role of left inferior vena cava? JBR-BTR 2012;95:302-305. [27] Dagenais M, Langer B, Taylor BR, et al. Experience with radical esophagogastric devascularization procedures (Sugiura) for variceal bleeding outside Japan. WorldJ Surg 1994;18:222-227.
[28] de Franchis R, Era Dell A, Iannuzzi F. Diagnosis and treatment of portal
TE D
hypertension. Digest Liver Dise 2004;36:787–798.
[29] Kitano S, Terblanche J, Kahn J, et a1. Venous anatomy of the lower oesophagus in portal hypertension: practical implications. Br J surg 1986;73:525-531.
EP
[30] Sato T, Yamazaki K, Toyota J, et a1. Perforating veins in recurrent esophageal varices evaluated by endoscopic color Doppler ultrasonography with a galactos-based contrast agent. J Gastroenterol 2004;39:422-428.
AC C
[31] Zhan Y, Wen T, Yan L, et al. The changes of hepatic hemodynamics and functional hepatic reserve after splenectomy with periesophagogastric devascularization. Hepatogastroenterology 2009;56:835-839. [32] Xie M, Huang J, Pan WN, et al. The clinical significance of retain traffic branch between portal vein and azygos in selective Devascularization. Zhonghua Pu tong Wai Ke Za Zhi. 2006;21:689-691. [33] Raia S, Meis S, Macedo AL. Surgical treatment of portal hypertension in schistosomiasis. World J Surg 1984; 8:738-752. [34] Ezzat FA, Abu-Elamgd KM, Aly IY, et al. Distal splenorenal shunt for
ACCEPTED MANUSCRIPT management of variceal bleeding in patients with schistosomal hepatic fibrosis. Ann Surg 1986; 204:566-573. [35] Huang HC, Chang CC, Wang SS, et al. The roles of angiotensin II receptors in the portosystemic collaterals of portal hypertensive and cirrhotic rats. J Vasc Res.
RI PT
2012;49:160-168. [36] Nagasaka H, Miida T, Yorifuji T, et al. Metabolic improvements in intrahepatic porto-systemic venous shunt presenting various metabolic abnormalities by 4-phenylacetate.Clin Chim Acta. 2013;419:52-56.
SC
[37] Orloff MJ. Pathogenesis and surgical treatment of intractable ascites associated with alcoholic cirrhosis. Ann N Y Acad Sci 1970;170:213-38.
M AN U
[38] Franco D, Vons C, Traynor O, et al. Should portal systemic shunt be reconsidered in the treatment of intractable ascites in cirrhosis? Arch Surg 1988; 123:987-91.
[39] Liu N, Liu B, Xu RY,et al. Splenectomy with endoscopic variceal ligation is superior to splenectomy with pericardial devascularization in treatment of portal
TE D
hypertension. World J Gastroenterol 2006;12:7375–7379.
[40] Elwood DR, Pomposelli JJ, Pomfret EA, et a1. Dislal splenorenal shunt: preferred treatment for recurrent varies hemorrhage in the patient with
AC C
EP
well-compensated cirrhosis. Arch Surg, 2006;141:385-388.
ACCEPTED MANUSCRIPT
Preoperative clinical characteristics of cirrhotic patients PDS (n=187)
t/χ2-value
P-value
47.2 + 7.5
45.1 + 8.2
0.711
0.160
81 123 62 4.0 + 1.6 14.8 + 2.7 33.2 + 8.1 25.5 + 4.1 94.0+ 11.6 26.3 + 7.2
AST (IU/L)
27.2 + 7.0
206 60 7.0 + 1.4
151 36 6.9 + 1.2 57 89 41 4.5 + 1.3 14.6 + 2.2 33.6 + 5.8 25.9+ 4.3 101.2+ 13.9 27.1 + 6.9 27.8 + 6.3
0.370
0.366
0.185
0.641
SC
age mean+SD (years) gender male female Child-Pugh’s score variceal grade Ⅰ Ⅰ Ⅰ follow-up time (years) PT(sec) Alb (g/L) TB(µmol/L) ALP (IU/L) ALT (IU/L)
RI PT
SDPDPV（n=266）
M AN U
Table 1
0.738
0.119
0.265 0.183 0.167 0.170 0.291 0.223
0.428 0.663 0.729 0.702 0.398 0.514
0.178
0.691
AC C
EP
TE D
SDPDPV, selective double portazygous disconnection with preserving vagus PDS, pericardial devascularization with splenectomy
ACCEPTED MANUSCRIPT
Table 2 Operation-relevant information and postoperative complications of SDPDPV group and PDS group PDS (n=187)
t/χ2-val ue
P-value
operative time (min) intraoperative blood loss (ml) recurrent hemorrhage encephalopathy early ascites
236.0 + 73.8 572.2 + 71.5 5 13 12
242.1 + 68.8 630.7 + 99.1 21 10 16
-0.365 -0.502 13.253 0.736 4.752
0.387 0.275 0.001 0.120 0.032
gastric stasis
0
27
8.852*
0.000
SC
RI PT
SDPDPV（n=266）
*is Fisher-value
AC C
EP
TE D
M AN U
SDPDPV, selective double portazygous disconnection with preserving vagus PDS, pericardial devascularization with splenectomy
ACCEPTED MANUSCRIPT
Table 3 Preoperative and postoperative FPP (cmH2O; mean+SD ) of SDPDPV group and PDS group SDPDPV（n=266）
PDS (n=187)
t-value
P-value
preoperative after splenectomy
38.7 + 3.6 26.9 + 3.2
39.2 + 4.1 27.7 + 2.9
-0.301 -0.392
0.481 0.393
postoperative
23.6 + 4.1
28.5 + 2.9
3.523
0.013
FPP, free portal pressure
RI PT
Group
AC C
EP
TE D
M AN U
SC
SDPDPV, selective double portazygous disconnection with preserving vagus PDS, pericardial devascularization with splenectomy
ACCEPTED MANUSCRIPT
Table 4 Postoperative biochemical Definitions of cirrhotic patients SDPDPV（n=266） Two weeks
Six months
PDS (n=187)
One year
Two weeks
Six months
P-value One year a
13.3 + 3.6
12.4 + 3.2
12.1 + 4.3
14.1 + 3.1
13.5 + 3.9
RI PT
PT(sec)
0.102
13.4 + 3.3
b
0.072
c
0.052
Alb (g/L)
33.1 + 9.0
36.3 + 8.3
39.7 + 8.4
30.2+9.0
32.9 + 8.4
35.1 + 8.9
a
0.027
b
0.021
c
ALT (IU/L)
AST (IU/L)
Child-Pugh’s score
93.5+ 11.9
24.9 + 5.8
26.1 + 6.3
6.1 + 1.7
23.0 + 4.7
86.2+ 11.4
21.9 + 6.3
22.4 + 6.0
19.6 + 4.2
5.3 + 2.0
25.8 + 3.6
25.2 + 4.3
M AN U
ALP (IU/L)
25.1 + 4.0
93.8+ 9.7
25.4 + 6.0
26.6 + 5.1
TE D
TB(µmol/L)
SC
0.016
5.9 + 1.8
119.1+15.9
29.6 + 7.2
31.9 + 6.6
102.7+10.6
27.7 +6.6
24.7 + 7.8
20.3 + 3.8
113.9+ 11.1
EP
0.275
b
0.097
c
0.192
a
0.008
b
0.007
c
0.008
28.3 + 6.3
a
0.036
b
0.032
c
0.045
a
30.6+ 6.3
0.031
b
0.092
c
0.037
6.6 + 1.2
6.2 + 1.5
6.5 + 1.7
a comparison of the two operational groups two weeks after operation b comparison of the two operational groups six months after operation c comparison of the two operational groups one year after operation SDPDPV, selective double portazygous disconnection with preserving vagus PDS, pericardial devascularization with splenectomy
AC C
a
a
0.011
b
0.004
c
0.009
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1 Kaplan-Meier curves of survival rates in two groups of patients with SDPDPV or PDS. The patients have better 3-year survival rate after SDPDPV (95.4%) compared to S&GD (84.0%; P=0.038).
S0022-4804(14)00532-0
DOI:
10.1016/j.jss.2014.05.065
Reference:
YJSRE 12765
To appear in:
Journal of Surgical Research
Received Date: 25 March 2014 Revised Date:
1 May 2014
Accepted Date: 21 May 2014
Please cite this article as: Zong G-q, Fei Y, Chen J, Liu R-m, Selective double disconnection for cirrhotic portal hypertension, Journal of Surgical Research (2014), doi: 10.1016/j.jss.2014.05.065. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Revised May 1st 2014
ACCEPTED MANUSCRIPT
Selective
double
disconnection
for
cirrhotic
portal
hypertension
RI PT
Guang-quan Zonga M.D., Yang Fei*a M.D., Jian Chen M.D. and Ren-min Liu M.D.
Department of General Surgery, the 81st Hospital of P.L.A., P.L.A. Cancer Center,
SC
Nanjing 210002, China.
* a
First co-authors
E-mail: [email protected] Telephone: +86-25-86648090
TE D
Fax: +86-25-80864499
M AN U
Corresponding author
EP
Contributors: Guang-quan Zong, Yang Fei wrote the paper, organized the figures and patient data and did the analysis; Ren-min Liu carried out the statistical analysis
AC C
and corrected the paper; Yang Fei supervised the writing and organization process.
ACCEPTED MANUSCRIPT
Abstract Object: To evaluate the effect of selective double portazygous disconnection with preserving vagus ( SDPDPV ) for patients with portal hypertension in the authors’ hospital.
RI PT
Methods: 453 patients with cirrhotic portal hypertension who underwent either SDPDPV or pericardial devascularization with splenectomy (PDS) for variceal bleeding from Feb 2007 to January 2013 were retrospectively reviewed. The
complications and clinical outcomes were analyzed.
SC
operation-relevant information, change of lavatory examination data, postoperative
Results: There were no significant difference between the SDPDPV group and the
M AN U
PDS group of mean operative time and intraoperative blood loss (P
0.05). The FPP
( free portal pressure) in the SDPDPV group was much lower than PDS group significantly after operation (P
0.05). The test of biochemical profile of hepatocyte
functions and Child-Pugh’s score at the end of the first postoperative year were significantly more altered in the SDPDPV group than in the PDS group (P
0.05).
TE D
Except encephalopathy, occurrences or development of postoperative complications including rebleeding, ascites and gastric stasis showed great difference between the two groups (P < 0.05). The operative mortality rate and the 3-year survival rates were
EP
great difference between the two groups too (P < 0.05). Conclusion: The SDPDPV not only controls recurrent bleeding from varices with
AC C
portal hypertension effectively but also maintains normal dynamics of stomach and physiological function of intestine and hepatobiliary.
Key words
portal hypertension; variceal hemorrhage; devascularization; vagus
Introduction
ACCEPTED MANUSCRIPT Portal hypertension (PHT) is defined as portal venous pressure exceeding 10 mmHg (14 cm of H2O). In China, the most common cause of PHT is cirrhosis[1]. This syndrome develops in the majority of patients with cirrhosis and is responsible for a frequent and severe complication of cirrhosis, massive bleeding from ruptured
RI PT
gastroesophageal varices is the main complication of PHT. Approximately 90% of patients with cirrhosis will develop gastroesophageal varices over 10 years and one-third of these will bleed from them. The risk of bleeding from gastroesophageal varices is about 25% within 2 years of diagnosis[2]. Catastrophic bleeding from
SC
gastroesphageal varices rupture is a life-threatening emergency. In many instances pharmacological therapy such as β- adrenergic blockers, and nitrates or endoscopic
M AN U
therapy such as band ligation cannot stop the bleeding and this area is also beyond the field of sclerotherapy[3,4]. Surgery seems the only definite therapy, which may offer chances for survival. Devascularization and shunting are the two widely accepted approaches
for
the
control
of
such
bleeding[5-8].
In
China,
pericardial
devascularization with splenectomy (PDS) has been widely used in patients with portal
TE D
hypertension as an effective haemostatic method, because it is less of a burden and is associated with a lower incidence of encephalopathy compared to shunt procedures[9]. But PDS truncated the diversion from portal vein to azygos vein, the reduction of portal vein
EP
pressure was not significant postoperatively[8,10]. It may promote the establishment of
collateral circulation rapidly, establish the path between portal vein and azygos vein again. Then, recurrence of gastroesophageal varices will appear and portal hypertensive gastric
AC C
mucosal lesions will be further aggravated. Thus PDS was shown to have a high rebleeding rate[11-13]. If patients with ascites undergone PDS, ascites was often difficult to be
controlled postoperatively[14]. On the other hand, once the vagus especially vagus trunk is interrupted during PDS procedure, it is easy for patients to suffer from delayed gastric emptying and gastric stasis[15]. In order to prevent these postoperative complications, a new procedure, selective pericardial vascular disconnection plus pericardial submembranous varices ligation by preserving vagus trunks (Selective double portazygous disconnection with preserving vagus, SDPDPV) has been performed in our department for the treatment of patients with cirrhotic portal
ACCEPTED MANUSCRIPT hypertension, since Feb 2007. The efficacy, safety, postoperative complications and survival were compared with those of single PDS performed during the same period.
Materials and methods
RI PT
General Information
The study population comprised 453 variceal bleeders with cirrhosis PHT
managed by surgery in the period between February 2007 and January 2013 at the 81st
SC
Hospital of P.L.A. in Nanjing, China. Consecutively 266 patients received SDPDPV and 187 PDS. The diagnoses of these patients were all confirmed by endoscopy,
M AN U
barium meal and Doppler ultrasonography. Our indications for operations include episodes of gastroesophageal variceal hemorrhage with a 1-month interval between the last attack and date of surgical procedure, which could not be controlled by medical means and endoscopic therapy, or thrombocytopenia (platelets count under 80×103/mm3) with gastroesophageal varices due to hypersplenism related to cirrhosis.
TE D
All of the patients accompanied with no serious cardiopulmonary diseases and concomitant chronic duodenal ulcers. The characteristics of the two surgical groups including age, gender, Child-Pugh’s score, preoperative variceal grade defined according to the criteria of Calès et al[16], follow-up time, preoperative biochemical
EP
tests were summarized in Table 1. Biochemical tests were examined within 1-3 days preoperatively, endoscopy or a barium meal examination was performed within
AC C
fourteen days before the operation to define the grade of varices.
Operative technique SDPDPV procedure was performed through an expanded left subcostal incision
of the left upper abdomen. The FPP (free portal pressure) is measured by inserting a catheter into the portal trunk through a branch of the right gastroepiploic vein. After conventional extreme splenectomy, The FPP is measured again. The proximal stomach was devascularized close to the gastric wall just above the crow’s foot along the lesser curvature from the incisura angularis up to the esophagus, we paid attention to avoid ligating the vagus nerve. Dissecting anterior Serosa layer and the left lateral
ACCEPTED MANUSCRIPT peritoneum covering the esophagus, and separating the muscle layer carefully, paraesophageal vein was exposed,. Following by lifting the stomach and dissecting gastropancreatic fold around the gastroesophageal junction, the left gastric artery and left gastric vein (coronary vein) were exposed. The gastric branch of left gastric vein
RI PT
and branches of left gastric artery were disconnected and suture ligated close to the gastric wall of lesser curvature in order to preserve the trunk of left gastric vein and artery. Then perforating branches from paraesophageal vein to the lower part of
esophagus were disconnected and suture ligated, tried to keep paraesophageal vein
SC
entirely. Dissection the lower part of esophagus was performed up to 7-10 cm above the cardia (the esophageal hiatus level). During the course of procedure, trunk of
M AN U
vagus nerve may be preserved. Whole layer discontinuous suture guided by a stomach tube with 3-0 prolene threads (Ethicon, Inc.) was performed around the lower part of esophagus at the area of 2-10cm above the cardia. The overlap length between the two needles was about two mm, and the direction of suture lines were not on the same horizontal level but showed zigzag-like to prevent esophageal stricture
TE D
postoperatively. The seromuscular layers of the lesser curvature were closed with interrupted silk sutures. After checking peritoneal cavity hemorrhage and circulation of free zone in the stomach and esophagus, we covered the wound of splenic fossa
EP
and retroperitoneal with omentum. Finally, the FPP was remeasured. A peritoneal cavity drainage tube was placed at left subphrenic location, before suturing the abdominal incision, a biopsy specimen of the liver was taken.
AC C
The technique of standard PDS was applied as described by Hassab[17].
Follow-up Studies
All patients had undergone follow-up from one to seven years (4.6+ 1.2 years). Clinical variables including episodes of recurrent hemorrhage, encephalopathy, ascites et al. were examined postoperatively, laboratory variables including biochemical tests were systematically examined at two weeks postoperatively, and then reperformed at six months and one year thereafter. In our studies, rebleeding was defined as at least 400ml blood transfusion
ACCEPTED MANUSCRIPT requirement and/or a decrease in the hematocrit to less than 30%, operative mortality was defined as death within thirty days of surgery[18].
Statistical Analysis All statistical analyses were performed using SPSS15.0 software (SPSS, Chicago,
RI PT
IL). Continuous data were expressed as mean values + standard deviation (SD). Significant differences between groups were determined by chi-squared analysis and unpaired Student's t test. Overall survival estimates were calculated by the Kaplan-Meier method and compared with similar estimates obtained using the
M AN U
SC
log-rank test. P-values <0.05 were considered statistically significant.
Results
Preoperative comparison of the two operational groups There were no significant difference between the SDPDPV group and the PDS group of the preoperative database including age, gender, Child-Pugh’s score,
0.05 )
Table 1
TE D
preoperative variceal grade, follow-up time, blood routine and biochemical tests (P . The two groups were well balanced in the distribution of
prognostic factors and other characteristics.
EP
operation-relevant information
All 453 operations were successful. The mean operative time of SDPDPV group
AC C
was 236.0 + 73.8 min with a range of 131 min to 238 min. In PDS group, it was 242.1 + 68.8 min with a range of 146 min to 242 min. The difference of mean operative time between the two groups was not significant (P = 0.387). (Table 2) The intraoperative blood loss of SDPDPV group was 572.2 + 71.5 ml, In PDS group, it was 630.7 + 102.1 ml. The difference of blood loss between the two groups was not significant too (P = 0.275). (Table 2)
Portal vein pressure As shown in Table 3, the difference of FPP between the SDPDPV group and the PDS group were not significant before operation and after splenectomy (P = 0.481, P
ACCEPTED MANUSCRIPT = 0.393 ). However, in contrast to the PDS group, the FPP in the SDPDPV group was much lower significantly after devascularization. (P = 0.013).
Biochemical data and hepatic functional reserve The test of biochemical profile of hepatocyte functions such as Albumin (Alb),
RI PT
alkaline phosphatase (ALP), alanine aminotransferase (ALT) and aspartate aminotransferase AST at two weeks, six months and one year after operations were all significantly more altered in the SDPDPV group than in the PDS group (P 0.05). Apart from these, both the total bilirubin (TB) and the prothrombin time (PT) in the
differences were not significant ( P
0.05 ) (Table 4).
SC
SDPDPV group were lower than that in the PDS group postoperatively, however the
M AN U
Child-Pugh’s classification is a well-validated method of assessing hepatic functional reserve in cirrhosis and is one of the best measures for prediction of long-term outcome[19]. In our studies, postoperative Child-Pugh’s score of SDPDPV group (at two weeks, six months and one year after operations) were all lower than that of PDS group significantly (P 0.05). (Table 4)
Recurrent hemorrhage
TE D
Postoperative complications
Recurrent hemorrhage of SDPDPV group and PDS group showed the following
Encephalopathy
EP
incidence of 5/266 (1.9%) and 21/187 (11.2%) respectively ( P= 0.001). (Table 2).
AC C
None of the studied patients experienced encephalopathy before operation. The incidence of postoperative encephalopathy was not significantly higher after SDPDPV (4.9%, 13/266) compared to PDS (5.3%, 10/187) ( P= 0.120) (Table 2). Ascites
Early postoperative (one month after operation) ascites developed in twelve patients (4.5%) after SDPDPV and in sixteen patients (8.6%) after PDS (P=0.032). Diuretic therapy controlled ascites in most of these surgical patients within the first year. At the time of a year after operation, ascites was detected in 7 SDPDPS (7.4%) and 4 PDS (5%) survivors (P=0.162). (Table 2).
ACCEPTED MANUSCRIPT Gastric stasis Gastric stasis developed in twenty-seven patients (14.4%) after PDS, but it didn’t develop in patients underwent SDPDPV, the difference was significant(P 0.001). (Table 2).
RI PT
Operative mortality The operative mortality rate was 0.4%(1/266) in the SDPDPV group and 3.7%
(7/187)in the PDS group. There was a significant difference among the groups (χ2 = 5.803; P=0.009). The reason of death in the SDPDPV group was acute portal vein
SC
thrombosis (one patient ), instead it were varicral bleeding (four patients ), hepatocyte function failure(two patients ) and portal vein thrombosis (one patient ) in the PDS
M AN U
group.
Long-term survival
After a mean follow-up of 4.6 + 1.2 years, forty-one patients (nineteen patients underwent SDPDPV, thirty-seven patients underwent PDS) had died. Thirteen
TE D
patients ( seven in SDPDPV group, six in PDS group) were lost to follow-up after one year. In SDPDPV group, recurrent bleeding was the cause of death in four patients,. Other causes of death included hepatomas in eight patients, liver failure in four patients, cecum carcinoma, portal vein thrombosis,and myocardial infarction in one
EP
patient each. Nevertheless in PDS group, the main reason of death were recurrent bleeding and liver failure. Seventeen patients died of recurrent bleeding and thirteen
AC C
patients died of liver failure respectively. Other causes of death included hepatomas in five patients, portal vein thrombosis in two patients. Less patients died of recurrent bleeding in SDPDPV group( 1.5% vs 9.4%), and the difference was significant (χ2 = 4.571; P=0.012). The data about liver failure were similar to that of recurrent bleeding compared between SDPDPV group and PDS group (1.5% vs 7.2%), the difference was significant too (χ2 = 2.937; P=0.0312 ). however, the difference about patients died of hepatomas between the two groups was not significant.( χ2 = 0.842; P=0.206) One-, two-, and three-year survival rates were 98.8%(256/259), 98.1% ( 254/259), and 95.4% (247/259), respectively, in SDPDPV group and 92.8%
ACCEPTED MANUSCRIPT ( 168/181), 90.1% ( 163/181), and 84.0% (152/181), respectively, in PDS group. The three-year survival rates of patients underwent SDPDPV was significantly greater than that underwent PDS (log-rank test, χ2 =4.263; P= 0.038). (Figure 1).
RI PT
Discussion PHT is associated with the most severe complications of cirrhosis, including bleeding from gastroesophageal varices, hepatic encephalopathy and ascites[2,20]. Many patients require surgical treatment each year. In China, PDS described by
SC
Hassab firstly in the 1964[21] is the most common elective procedure among all the surgical procedures performed. Though the initial control of bleeding is good and
M AN U
encephalopathy rate is low after PDS, the high rate of residual or recurrent varices and the exacerbated pathologic changes in the gastric mucosa were associated with a high rebleeding risk[12,13,22].
There are a few varicose rete between the outer longitudinal and inner annular tunica muscularis around cardia and lower esophagus (4-5cm above cardia) in patients
TE D
with PHT[23]. They connect paraesophageal venous rete wtih submucosal venous rete. Some studies showed that variceal hemorrhage located in the lamina propria, but not in the submucosa[24-26].
EP
PDS procedure only disconnected external vessels to serosa of perigastroesophagus, whereas neglected varicose vein plexus from submucosa and tunica muscularis. Thus, once the veins around the lower esophagus and fundus which
AC C
could be regarded as natural shunt channel was blocked after devascularization, portal vein pressure of splenic and gastric area was further increased, varicose vein plexus from perigastroesophageal submucosa and tunica muscularis was further aggravated, rerupture of vein could happen finally. In procedure of SDPDPV, we performed whole layer discontinuous suture with 3-0 prolene threads around the lower part of esophagus at the area of 2-10cm above the cardia, then both the extramural vessels and the intramural vessels were disconnected, venous pressure of area above the ligature would be much lower following. The direction of suture lines were not on the same horizontal level but showed zigzag-like to prevent esophageal stricture
ACCEPTED MANUSCRIPT postoperatively. Pericardial vascular disconnection plus pericardial submembranous varices ligation can get the significant efficiency of devascularization thoroughly without risk of anastomotic leakage which might happen after Sugiura’s operation[27]. Our studies showed that the incidence of postoperative recurrent hemorrhage was
RI PT
significantly lower after SDPDPV compared to PDS. In the present series, the operative mortality rate in the SDPDPV group was
lower than that in the PDS group. There were few patients die of varicral bleeding and hepatocyte function failure after SDPDPV. On the contrary, both of them were the
SC
main reason of death after PDS.
The therapeutic goals of surgical procedures for PHT are mainly achieved by
M AN U
selective reduction of the portal vein pressure[12,28]. Our studies illustrated that in contrast to the PDS group, the mean value of FPP in the SDPDPV group was much lower significantly after operation. We consider that the diversion from portal vein to azygos vein may be truncated after PDS, it will promote the establishment of collateral circulation rapidly[29,30]. and establish the path between portal vein and
TE D
azygos vein again, the FPP still maintain high level. Zhang et al. think the changes in hepatic hemodynamic status after PDS had a beneficial effect on hepatocyte functional reserve because these provided sufficient portal vein perfusion to liver[31],
EP
but the change can promote the formation of new collateral vessels in the periesophagogastric area, and increase the risk of rebleeding following. On the other hand, injury of paraesophageal vein and gastric coronary vein trunk were prevented
AC C
during the procedure of SDPDPV, then retention of these vicarious vessel channels may decrease FPP postoperatively. The similar results had been obtained by Xie et al[32]. But compared with PDS, the diminution of hepatic blood supply resulted from reduction of FPP was not significant. Perhaps the main reason was that some spontaneous shunts were retained intraoperatively, thus further reducing of portal blood flow could appear which led to reduced hepatic sinusoidal pressure and increased hepatic artery flow velocity[33,34]. Moreover, the vasodilator objects including nitric oxide were inactivated in liver due to shunts[35,36]. In summary, SDPDPV not only reduce the FPP and prevent variceal hemorrhage, but also maintain
ACCEPTED MANUSCRIPT hepatic perfusion and preserve hepatocyte functions. Our studies showed hepatocyte function and Child-Pugh’s score which is a well-validated method of assessing hepatocyte functional reserve in cirrhosis improved greater in SDPDPV group compared to that in PDS groups significantly after operations. Preservation of
RI PT
hepatocyte functions is a major determinant of long-term surgical outcome. In our studies, patients with PDS had a significantly worse prognosis than those with SDPDPV analyzed by 3-year survival rates.
Both SDPDPV and PDS belonged to devascularization, so encephalopathy was
SC
clinically detected with similar low incidence among survivors of both procedures.
Ascites was frequently observed in the devascularization, whereas it seldom occurred
M AN U
after shunting. It is already well known that shunts are effective in the prevention of ascites in patients with cirrhosis[37,38]. The mechanisms include a high hydrostatic pressure of portal hypertension leading to increased production, transudation of hepatic and splanchnic lymph into the peritoneal cavity, The ascitogenic effect of extensive dissection with lymphatic interruption may also explain the higher rate of
TE D
early postoperative ascites after PDS. Compared with PDS, a low portal hypertension pressure and limited dissection with lymphatic interruption occurred after the procedure of SDPDPV. Our studies showed incidence of early postoperative ascites
EP
after SDPDPV was lower than that after PDS. Furthermore, we found there were no difference between SDPDPS group and PDS group about incidence of ascites at the time of a year after operation, it should be considered diuretic therapy controlled
AC C
ascites in most of these surgical patients, and the natural history of portal hypertension with chronic liver disease. Complications of PDS included vagus injury[39]. Vagus injury tremendously enhances the incidence of gastric emptying delay, occasional bilious vomiting, and pyloric stricture; the injury possibly leads to gastric stasis, even make portal hypertensive gastropathy more severe[40]. However, in SDPDPV, the dissection was done close to the esophagus and the stomach to preserve the anterior and posterior vagus trunks and the nerves of Latarget. Thus, the function of antrum and pylorus could be retained, bile reflux might be reduced which effect gastric mucosal barrier.
ACCEPTED MANUSCRIPT The studies showed that Gastric stasis developed in 27 patients after SDPDPV was much lower than that
in 12 patients
after PDS significantly.
We found that there were no significant difference of mean operative time and the intraoperative blood loss between the two groups. It showed that SDPDPV as a
RI PT
new procedure was operable, viable and safe.
Conclusions
The SDPDPV not only controls recurrent bleeding from varices with portal
M AN U
physiological function of intestine and hepatobiliary.
Conflicts of interest
EP
TE D
There was no potential conflict of interest.
AC C
SC
hypertension effectively but also maintains normal dynamics of stomach and
RI PT
ACCEPTED MANUSCRIPT
Reference
SC
[1] Wang Y. Surgical treatment of portal hypertension. Hepatob Pancreat Dis 2002;1:211–214.
J Hepatol. 2000;32(1 Suppl):141-156.
M AN U
[2] Bosch J, García-Pagán JC. Complications of cirrhosis. I. Portal hypertension.
[3] Gournay J, Masliah C, Martin T, et al. Isosorbide mononitrate and propranolol compared with propranolol alone for the prevention of variceal rebleeding. Hepatology 2000;31:1239-1245.
TE D
[4] Larson AW, Cohen H, Zweiban B et al. Acute esophageal variceal sclerotherapy. Results of a prospective randomized controlled trial. JAMA 1986; 255: 497-500. [5] Jin G, Rikkers LF. Transabdominal esophagogastric devascularization as treatment
EP
for variceal hemorrhage. Surgery 1996;120:641.. [6] Yoshida H, Mamada Y, Taniai N, et al. New trends in surgical treatment for portal hypertension. Hepatol Res 2009;39:1044-1049.
AC C
[7] Vangeli M, Patch D, Burroughs AK. Salvage TIPS for uncontrolled variceal bleeding. J Hepatol 2002;37:703–704. [8] Henderson JM, Boyer TD, Kutner MH, et al. Distal splenorenal shunt versus transjugular intrahepatic portal systematic shunt for variceal bleeding: a randomized trial. Gastroenterology 2006;130:1643-1651 [9] da Silva LC, Strauss E, Gayotto LC, et al. A randomized trial for the study of the elective surgical treatment of portal hypertension in masonic schistosomiasis. Ann Surg 1986;204:148-156. [10] Bothe A, Stone MD, McDermott WV. Portoazygous disconnection for bleeding
ACCEPTED MANUSCRIPT esophageal varices. Am J Surg 1985; 149:546-550. [11] Gouge TH, Ranson JHC. Esophageal transection and paraesophagogastric devascularization for bleeding esophageal varices. Am J Surg 1986; 151:47-53. [12] El-Serag HB, Everhart JE. Improved survival after variceal hemorrhage
RI PT
over an 11-year period in the Department of Veterans Affairs. Am J Gastroenterol 2000;95:3566-3573.
[13] Carvalho DL, Capua A Jr, Leme PL. Portal flow and hepatic function after splenectomy and esophagogastric devascularization. Int Surg 2008;93:314.
SC
[14] Xu XB, Cai JX, Leng XS, et a1. Clinical analysis of surgical treatment of portal hypertension. World J Gastroenterol 2005;11:4552-4559.
M AN U
[15] Lu YF, Hou YD, Jia SR, et al. Experimental study of pylorus and pyloric vagus preserving gastrectomy. World J Surg 1993;17:525–529.
[16] Calès P, Braillon A, Lebrec D. Medical treatment of portal hypertension and oesophageal varices.Br Med J (Clin Res Ed) 1983;287:1144.
[17] Hassab MA. Gastroesophageal decongestion and splenectomy in the treatment of
TE D
esophageal varices in bilharzial cirrhosis: further studies with a report on 355 operations. Surgery 1967;61:169-176.
[18] Ezzat FA, Abu-Elmagd KM, Aly MA, et al. Selective shunt versus nonshunt
EP
surgery for management of both schistosomal and nonschistosomal variceal bleeders. Ann Surg 1990 ;212:97-108.
[19] Papatheodoridis GV, Goulis J, Leandro G, et al. Transjugular intrahepatic
AC C
portosystemic shunt compared with endoscopic treatment for prevention of variceal rebleeding: a metaanalysis. Hepatology 1999;30:612–622. [20] D’Amico G, de Franchis R, Cooperative Study Group. Upper digestive bleeding in cirrhosis. Post-therapeutic outcome and prognostic indicators. Hepatology 2003;38:599–612. [21] Hassab MA. 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 1964;41:232-238. [22] Estes NC, Pierce GE. Late results of extended devascularization procedure
ACCEPTED MANUSCRIPT for patients with bleeding esophageal varices. Am J Surg 1984; 50:381-385. [23] Wang ZH, Cai HZ, Zhao AX, et al. The effects of disconnection and shunt procedure on hepatic hemodynamics and function of active cirrhotic patients with portal hypertension. Chinese J of Clin Hepatol 2002;118:309–310.
RI PT
[24] Zhou GW, Yan DY, Li F, et al. The retrospective analysis of surgical outcome of portal hypertension. Zhonghua Wai Ke Za Zhi 2009;47:1532-1535.
[25] Chikamori F, Kuniyoshi N, Shibuya S, et al. Correlation between endoscopic and angiographic findings in patients with esophageal and isolated gastric varices.
SC
Digest Surg 2001;18:176-81.
[26] Stimec BV, Grønvold LA, Ignjatovic D, et al. Gastric varices without portal
M AN U
hypertension: role of left inferior vena cava? JBR-BTR 2012;95:302-305. [27] Dagenais M, Langer B, Taylor BR, et al. Experience with radical esophagogastric devascularization procedures (Sugiura) for variceal bleeding outside Japan. WorldJ Surg 1994;18:222-227.
[28] de Franchis R, Era Dell A, Iannuzzi F. Diagnosis and treatment of portal
TE D
hypertension. Digest Liver Dise 2004;36:787–798.
[29] Kitano S, Terblanche J, Kahn J, et a1. Venous anatomy of the lower oesophagus in portal hypertension: practical implications. Br J surg 1986;73:525-531.
EP
[30] Sato T, Yamazaki K, Toyota J, et a1. Perforating veins in recurrent esophageal varices evaluated by endoscopic color Doppler ultrasonography with a galactos-based contrast agent. J Gastroenterol 2004;39:422-428.
AC C
[31] Zhan Y, Wen T, Yan L, et al. The changes of hepatic hemodynamics and functional hepatic reserve after splenectomy with periesophagogastric devascularization. Hepatogastroenterology 2009;56:835-839. [32] Xie M, Huang J, Pan WN, et al. The clinical significance of retain traffic branch between portal vein and azygos in selective Devascularization. Zhonghua Pu tong Wai Ke Za Zhi. 2006;21:689-691. [33] Raia S, Meis S, Macedo AL. Surgical treatment of portal hypertension in schistosomiasis. World J Surg 1984; 8:738-752. [34] Ezzat FA, Abu-Elamgd KM, Aly IY, et al. Distal splenorenal shunt for
ACCEPTED MANUSCRIPT management of variceal bleeding in patients with schistosomal hepatic fibrosis. Ann Surg 1986; 204:566-573. [35] Huang HC, Chang CC, Wang SS, et al. The roles of angiotensin II receptors in the portosystemic collaterals of portal hypertensive and cirrhotic rats. J Vasc Res.
RI PT
2012;49:160-168. [36] Nagasaka H, Miida T, Yorifuji T, et al. Metabolic improvements in intrahepatic porto-systemic venous shunt presenting various metabolic abnormalities by 4-phenylacetate.Clin Chim Acta. 2013;419:52-56.
SC
[37] Orloff MJ. Pathogenesis and surgical treatment of intractable ascites associated with alcoholic cirrhosis. Ann N Y Acad Sci 1970;170:213-38.
M AN U
[38] Franco D, Vons C, Traynor O, et al. Should portal systemic shunt be reconsidered in the treatment of intractable ascites in cirrhosis? Arch Surg 1988; 123:987-91.
[39] Liu N, Liu B, Xu RY,et al. Splenectomy with endoscopic variceal ligation is superior to splenectomy with pericardial devascularization in treatment of portal
TE D
hypertension. World J Gastroenterol 2006;12:7375–7379.
[40] Elwood DR, Pomposelli JJ, Pomfret EA, et a1. Dislal splenorenal shunt: preferred treatment for recurrent varies hemorrhage in the patient with
AC C
EP
well-compensated cirrhosis. Arch Surg, 2006;141:385-388.
ACCEPTED MANUSCRIPT
Preoperative clinical characteristics of cirrhotic patients PDS (n=187)
t/χ2-value
P-value
47.2 + 7.5
45.1 + 8.2
0.711
0.160
81 123 62 4.0 + 1.6 14.8 + 2.7 33.2 + 8.1 25.5 + 4.1 94.0+ 11.6 26.3 + 7.2
AST (IU/L)
27.2 + 7.0
206 60 7.0 + 1.4
151 36 6.9 + 1.2 57 89 41 4.5 + 1.3 14.6 + 2.2 33.6 + 5.8 25.9+ 4.3 101.2+ 13.9 27.1 + 6.9 27.8 + 6.3
0.370
0.366
0.185
0.641
SC
age mean+SD (years) gender male female Child-Pugh’s score variceal grade Ⅰ Ⅰ Ⅰ follow-up time (years) PT(sec) Alb (g/L) TB(µmol/L) ALP (IU/L) ALT (IU/L)
RI PT
SDPDPV（n=266）
M AN U
Table 1
0.738
0.119
0.265 0.183 0.167 0.170 0.291 0.223
0.428 0.663 0.729 0.702 0.398 0.514
0.178
0.691
AC C
EP
TE D
SDPDPV, selective double portazygous disconnection with preserving vagus PDS, pericardial devascularization with splenectomy
ACCEPTED MANUSCRIPT
Table 2 Operation-relevant information and postoperative complications of SDPDPV group and PDS group PDS (n=187)
t/χ2-val ue
P-value
operative time (min) intraoperative blood loss (ml) recurrent hemorrhage encephalopathy early ascites
236.0 + 73.8 572.2 + 71.5 5 13 12
242.1 + 68.8 630.7 + 99.1 21 10 16
-0.365 -0.502 13.253 0.736 4.752
0.387 0.275 0.001 0.120 0.032
gastric stasis
0
27
8.852*
0.000
SC
RI PT
SDPDPV（n=266）
*is Fisher-value
AC C
EP
TE D
M AN U
SDPDPV, selective double portazygous disconnection with preserving vagus PDS, pericardial devascularization with splenectomy
ACCEPTED MANUSCRIPT
Table 3 Preoperative and postoperative FPP (cmH2O; mean+SD ) of SDPDPV group and PDS group SDPDPV（n=266）
PDS (n=187)
t-value
P-value
preoperative after splenectomy
38.7 + 3.6 26.9 + 3.2
39.2 + 4.1 27.7 + 2.9
-0.301 -0.392
0.481 0.393
postoperative
23.6 + 4.1
28.5 + 2.9
3.523
0.013
FPP, free portal pressure
RI PT
Group
AC C
EP
TE D
M AN U
SC
SDPDPV, selective double portazygous disconnection with preserving vagus PDS, pericardial devascularization with splenectomy
ACCEPTED MANUSCRIPT
Table 4 Postoperative biochemical Definitions of cirrhotic patients SDPDPV（n=266） Two weeks
Six months
PDS (n=187)
One year
Two weeks
Six months
P-value One year a
13.3 + 3.6
12.4 + 3.2
12.1 + 4.3
14.1 + 3.1
13.5 + 3.9
RI PT
PT(sec)
0.102
13.4 + 3.3
b
0.072
c
0.052
Alb (g/L)
33.1 + 9.0
36.3 + 8.3
39.7 + 8.4
30.2+9.0
32.9 + 8.4
35.1 + 8.9
a
0.027
b
0.021
c
ALT (IU/L)
AST (IU/L)
Child-Pugh’s score
93.5+ 11.9
24.9 + 5.8
26.1 + 6.3
6.1 + 1.7
23.0 + 4.7
86.2+ 11.4
21.9 + 6.3
22.4 + 6.0
19.6 + 4.2
5.3 + 2.0
25.8 + 3.6
25.2 + 4.3
M AN U
ALP (IU/L)
25.1 + 4.0
93.8+ 9.7
25.4 + 6.0
26.6 + 5.1
TE D
TB(µmol/L)
SC
0.016
5.9 + 1.8
119.1+15.9
29.6 + 7.2
31.9 + 6.6
102.7+10.6
27.7 +6.6
24.7 + 7.8
20.3 + 3.8
113.9+ 11.1
EP
0.275
b
0.097
c
0.192
a
0.008
b
0.007
c
0.008
28.3 + 6.3
a
0.036
b
0.032
c
0.045
a
30.6+ 6.3
0.031
b
0.092
c
0.037
6.6 + 1.2
6.2 + 1.5
6.5 + 1.7
a comparison of the two operational groups two weeks after operation b comparison of the two operational groups six months after operation c comparison of the two operational groups one year after operation SDPDPV, selective double portazygous disconnection with preserving vagus PDS, pericardial devascularization with splenectomy
AC C
a
a
0.011
b
0.004
c
0.009
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1 Kaplan-Meier curves of survival rates in two groups of patients with SDPDPV or PDS. The patients have better 3-year survival rate after SDPDPV (95.4%) compared to S&GD (84.0%; P=0.038).