Clinical Review & Education

Special Communication | PACIFIC COAST SURGICAL ASSOCIATION

Fifty-three Years’ Experience With Randomized Clinical Trials of Emergency Portacaval Shunt for Bleeding Esophageal Varices in Cirrhosis 1958-2011 Marshall J. Orloff, MD

IMPORTANCE Emergency treatment of bleeding esophageal varices (BEV) consists mainly of

endoscopic and pharmacologic measures, with transjugular intrahepatic portal-systemic shunt (TIPS) performed when bleeding is not controlled. Surgical shunt has been relegated to salvage. At the University of California, San Diego, Medical Center, our group has conducted 10 studies of emergency portacaval shunt (EPCS) during 46 years. OBJECTIVE To describe 2 previously reported randomized clinical trials (RCTs) conducted from 1988 to 2011 in unselected consecutive patients who received emergency treatment for BEV. DESIGN, SETTING, AND PARTICIPANTS In RCT No. 1, a total of 211 unselected consecutive patients with cirrhosis and acute BEV were randomized to emergency endoscopic sclerotherapy (EEST) (n = 106) or EPCS (n = 105). In RCT No. 2, a total of 154 unselected consecutive patients with cirrhosis and acute BEV were randomized to TIPS (n = 78) or EPCS (n = 76). Diagnostic workup was completed within 6 hours of initial contact, and primary treatment was initiated within 8 to 12 hours. Regular follow-up for up to 10 years was accomplished in 100% of the patients. INTERVENTIONS In RCT No. 1, EEST or EPCS; in RCT No. 2, TIPS or EPCS. MAIN OUTCOMES AND MEASURES The 2 groups were compared with regard to survival, control of bleeding, portal-systemic encephalopathy, and direct cost of care. RESULTS Distribution in Child risk classes was almost identical. One-third of patients were in Child class C. Permanent control of bleeding was achieved by EEST in only 20% of the patients and by TIPS in only 22%. In contrast, EPCS permanently controlled bleeding in 97% and 100% of the patients in RCT No. 2 and RCT No. 1, respectively (P < .001). Survival was significantly greater following EPCS than after EEST and TIPS (P < .001). Median survival was more than 10 years following EPCS compared with 1.99 years after TIPS. Occlusion of TIPS was demonstrated in 84% of the patients, 63% of whom underwent TIPS revision, which failed in 80% of the cases. Recurrent portal-systemic encephalopathy developed in 35% of the patients who underwent EEST and 61% of those who received TIPS. In contrast, portal-systemic encephalopathy occurred in 15% of the patients who received EPCS in RCT No. 1 and 21% of those in RCT No. 2. Direct costs of care were 5 to 7 times greater in the EEST ($168 100) and TIPS ($264 800) groups than in the EPCS ($39 000) group (P < .001). CONCLUSIONS AND RELEVANCE Emergency portacaval shunt permanently stopped variceal bleeding, almost never became occluded, accomplished 5 times the long-term survival than EEST or TIPS, and was much less costly than EEST or TIPS. The widespread practice of using EPCS mainly as salvage for failure of endoscopic therapy or TIPS is not supported by the definitive results of these long-term RCTs in unselected patients with cirrhosis. TRIAL REGISTRATION clinicaltrials.gov Identifier: NCT00690027 and NCT00734227 JAMA Surg. 2014;149(2):155-169. doi:10.1001/jamasurg.2013.4045 Published online January 8, 2014.

Author Affiliation: Department of Surgery, School of Medicine, University of California, San Diego. Corresponding Author: Marshall J. Orloff, MD, Department of Surgery, University of California, San Diego, UCSD Medical Center, 200 W Arbor Dr, San Diego, CA 92103-8999 ([email protected]).

155

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

Clinical Review & Education Special Communication

T

his article summarizes 53 years’ experience with studies of emergency portacaval shunt (EPCS) for bleeding esophageal varices (BEV) in cirrhosis (1958-2011), with a focus on 2 previously published trials.1-3 Bleeding esophageal varices is responsible for high mortality.4,5 If untreated, 95% of the patients develop recurrent bleeding and die.6,7 Several modalities are used today for emergency treatment of BEV, including endoscopic therapy, pharmacologic measures, and transjugular intrahepatic portal-systemic shunt (TIPS). Surgical portal-systemic shunts (PSSs) are infrequently used because of the unsubstantiated belief that these shunts cause frequent portalsystemic encephalopathy (PSE) and liver failure. Surgical PSS has been used in 3 circumstances. First, prophylactic PSS has been advocated in patients with demonstrable varices that have never bled. Prophylactic PSS has been compared with medical therapy in 3 prospective studies8-11 that showed no influence on survival. There is no indication for prophylactic PSS. Second, elective PSS is the most widely used surgical treatment. It is performed in patients who have recovered from an episode of BEV because of the high likelihood that the varices will bleed again.12 From 1958 to 2001, elective PSS was performed in 1414 selected patients at the University of California, San Diego (UCSD), Medical Center who were referred to us after having recovered from an acute episode of BEV. In these highly selected referred patients, operative mortality was 1.6%, long-term shunt patency was 99.7%, recurrent PSE was 7%, and survival after 1, 5, 10, and 15 years was, respectively, 95%, 71%, 65%, and 61%. Emergency portacaval shunt is the third treatment used for control of BEV. Table 1 summarizes our 53 years’ experience (19582011) with PSS, which led us to conclude that EPCS is the most effective treatment of acute BEV. We conducted 10 prospective studies of EPCS involving 956 patients.1-3,13-19 The unique features of these studies are (1) EPCS was undertaken within 8 hours of initial contact, (2) all patients with BEV (all comers) were entered in the studies, (3) BEV was managed in all patients according to a well-defined protocol, and (4) patients were followed up monthly for the first year and every 3 months thereafter for life, such that the 1-, 5-, and 10year follow-up rates were 100%, 98%, and 97%, respectively. Our most recent studies have been 2 consecutive randomized clinical trials (RCTs) in unselected consecutive patients with acute BEV. Of these trials, RCT No. 1 compared emergency endoscopic sclerotherapy (EEST) with EPCS in 211 unselected consecutive patients,1,2 and RCT No. 2 compared TIPS with EPCS in 154 unselected consecutive patients.3 The second RCT was separate and distinct from the first. This report focuses on these 2 RCTs.

Methods

156

EPCS for Bleeding Esophageal Varices in Cirrhosis

17.75 years after the start of the study. Patient entry in RCT No. 2 lasted from July 25, 1996, until October 8, 2003, and follow-up continued until July 31, 2011, 15 years after the start of the study. Written informed consent was obtained by a physician coinvestigator before randomization. Participants did not receive financial compensation. The UCSD Human Subjects Committee and National Institutes of Health approved each study protocol and consent forms at regular intervals.

Eligibility Patients were enrolled in an RCT under 5 conditions: (1) requirement for 2 U or more of blood transfusion, (2) clinical and laboratory findings of cirrhosis, (3) upper endoscopy showing esophageal varices, (4) no other lesion that could reasonably account for bleeding of the observed magnitude, and (5) acute upper gastrointestinal bleeding that had been observed within 48 hours of study entry. Seventy referred patients were denied inclusion in an RCT because they did not have BEV or clinical or laboratory findings indicative of cirrhosis.

Randomization Patients were randomized by the investigators’ drawing a designation card from an opaque sealed envelope prepared by a statistician.

Diagnostic Workup The diagnostic workup was completed within 6 hours of presentation and included history and physical examination, blood studies,19 urinalysis, Doppler duplex ultrasonography, chest radiograph, electrocardiogram, and esophagogastroduodenoscopy, all performed at the patient’s bedside in the intensive care unit (ICU).

Liver Transplant Evaluation Bleeding esophageal varices in cirrhosis has been considered an indication for liver transplant.23-25 In both RCTs, patients were repeatedly evaluated by the UCSD Medical Center liver transplant program for indications for liver transplant. If patients exhibited progressive liver failure, they underwent extensive evaluation for liver transplant. As a supplement to the RCT data, we analyzed our results regarding liver transplant in 1300 unrandomized patients in whom we performed PCS beginning in 1978.26

Hepatocellular Carcinoma Hepatocellular carcinoma (HCC) is a frequent cause of death in patients with cirrhosis.27-29 Screening for HCC involved serial abdominal ultrasonography and measurements of serum α-fetoprotein at study entry and every 6 months thereafter. Computed tomography was performed as a supplement to ultrasonography.

Design of RCTs No. 1 and No. 2

Direct Cost of Care

Figure 1 is a Consolidated Standards for Reporting of Trials diagram that shows the overall design of the 2 RCTs.20-22 The objectives were to compare EPCS with EEST (RCT No. 1) and TIPS (RCT No. 2) with regard to survival, control of bleeding, PSE, and economic costs. Ninety-one physicians in 4 California counties agreed to promptly refer their patients with acute BEV to UCSD Medical Center for entry into an RCT. Patient entry in RCT No. 1 lasted from April 8, 1988, until July 25, 1996, and follow-up continued until December 31, 2005,

In each RCT, the 2 groups were compared with regard to direct costs of care. Complete hospital and outpatient charges and professional fee bills were obtained for every patient continuously for more than 10 years.

Initial Emergency Therapy During Workup All patients were housed in an ICU in which the nursing staff was expert in the care of patients with cirrhosis and acute upper

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

jamasurgery.com

EPCS for Bleeding Esophageal Varices in Cirrhosis

Special Communication Clinical Review & Education

Table 1. Clinical Characteristics at Time of Study Entry of Patients With Cirrhosis and Bleeding Esophageal Varicesa RCT No. 1 Characteristic UCSD admission, No. (%)

EEST (n = 106)

EPCS (n = 105)

Direct to UCSD emergency department

35 (33.0)

26 (24.8)

Transfer from outside hospital

71 (67.0)

79 (75.2)

RCT No. 2 P Value

.24

EPCS (n = 76)

TIPS (n = 78)

13 (17.1)

10 (12.8)

63 (82.9)

68 (87.2)

P Value

.50

Demographics Age, y Mean/median Range No. ≥70 y

47.8/45

49.8/47

.21

49.1/48

49.0/47

.89

23-75

28-82

.61

31-73

30-84

.62

7

Male sex, No. (%)

9

81 (76.4)

81 (77.1)

White

53 (50.0)

58 (55.2)

Hispanic

50 (47.2)

39 (37.1)

3 (2.8)

Alcoholism alone Hepatitis B or C alone Alcoholism and hepatitis

.61 .87

2

1

60 (78.9)

56 (71.8)

37 (48.7)

33 (42.3)

34 (44.7)

39 (50.0)

8 (7.6)

5 (6.6)

6 (7.7)

58 (54.7)

54 (51.4)

24 (31.6)

29 (37.2)

10 (9.4)

8 (7.6)

11 (14.5)

4 (5.1)

30 (28.3)

33 (31.4)

37 (48.7)

44 (56.4)

8 (7.5)

10 (9.5)

4 (5.3)

1 (1.3)

First

36 (34.0)

41 (39.0)

Second

38 (35.8)

32 (30.5)

≥Third

32 (30.2)

32 (30.5)

.35 .65

Race, No. (%)

Other

.09

.65

Cause of cirrhosis, No. (%)

Other

.73

.08

Bleeding episode, No. (%)

Chronic alcoholism, No. (%)

.76

28 (36.8)

3 (3.8)

16 (21.0)

23 (29.5)

32 (42.1)

25 (32.0)

.81

88 (83.0)

87 (82.8)

>.99

61 (80.3)

72 (92.3)

Years of alcoholism, mean (range)

22 (4-59)

25 (7-54)

.56

21.6 (8-41)

21.6 (5-45)

.66 .97

Recent alcohol ingestion, ≤7 d, No. (%)

55 (62.5)

57 (65.5)

.99

34 (44.7)

42 (58.3)

.44

Hematemesis, No. (%)

103 (97.2)

97 (92.4)

.13

74 (97.4)

73 (93.6)

.14

Hematochezia, No. (%)

88 (83.0)

93 (88.6)

.32

66 (86.8)

60 (76.9)

.14

History, No. (%) Jaundice

61 (57.5)

58 (55.2)

.78

42 (55.3)

32 (41.0)

.11

Ascites

70 (66.0)

48 (45.7)

.004b

53 (69.7)

40 (51.3)

.02b

Portal-systemic encephalopathy

20 (18.9)

30 (28.6)

.11

18 (23.7)

17 (21.8)

.85

Delirium tremens in alcoholics

23 (26.1)

28 (32.2)

.43

21 (34.4)

20 (27.8)

.45

Narcotic addiction

31 (29.2)

26 (24.8)

.44

15 (19.7)

19 (24.4)

.56

Type 2 diabetes mellitus

13 (12.3)

24 (22.8)

.047b

9 (11.8)

11 (14.1)

.81

Hypertension

13 (12.3)

16 (15.2)

.56

2 (2.6)

5 (6.4)

.44

Pulmonary disease

12 (11.3)

12 (11.4)

>.99

0

3 (3.8)

.25

>.99

2 (2.6)

0

.24

0

3 (3.8)

.25

4 (5.1)

.10

Other medical disorders, No. (%)

Coronary artery disease, atrial fibrillation

9 (8.5)

8 (7.6)

Renal disease

3 (32.8)

10 (9.5)

.049b

22 (20.8)

27 (25.7)

.42

Other

10 (13.2)

Physical examination, No. (%) Jaundice

45 (42.4)

38 (36.2)

.40

37 (48.7)

34 (43.6)

.63

Ascites

65 (61.3)

54 (51.4)

.17

46 (60.5)

37 (47.4)

.11

Portal-systemic encephalopathy

19 (17.9)

19 (18.1)

>.99

Severe muscle wasting, 2+ or 3+ on 0-3+ scale

50 (47.2)

67 (63.8)

Delirium tremens in alcoholics PSE index, median (IQR)

2 (1.9)

2 (1.9)

0 (0-0.15)

0 (0-0.15)

.02b >.99 .46

17 (22.4)

19 (24.4)

.80

49 (64.5)

40 (51.3)

.11

3 (4.9)

1 (1.3)

.33

0 (0-0.16)

0 (0-0.15)

.44

Child risk class, No. (%) A, 5-8 points

32 (30.2)

26 (24.8)

B, 9-11 points

46 (43.4)

49 (46.7)

C, 12-15 points

28 (26.4)

30 (28.6)

Child risk class points, mean/median

10.1/10

10.0/10

.71 .76

15 (19.7)

16 (20.5)

37 (48.7)

39 (50.0)

24 (31.6)

23 (29.5)

10.6/11

10.2/10

.98 .26

Abbreviations: EPCS, emergency portacaval shunt; EEST, emergency endoscopic sclerotherapy; IQR, interquartile range; PSE, portal-systemic encephalopathy; RCT, randomized clinical trial; UCSD, University of California, San Diego, Medical Center. a

Data are from Orloff et al.1,3

b

Statistically significant difference.

jamasurgery.com

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

157

Clinical Review & Education Special Communication

EPCS for Bleeding Esophageal Varices in Cirrhosis

Figure 1. Consolidated Standards for Reporting of Trials Flow Diagrams RCT No.1: EEST vs EPCS 258 Assessed for eligibility

211 Randomized

106 Allocated to intervention EEST 106 Received allocated intervention 0 Did not receive allocated intervention

RCT No. 2: TIPS vs EPCS

154 Randomized

23 Excluded 23 Did not have cirrhosis and BEV 0 Refused to participate

105 Allocated to intervention EPCS 105 Received allocated intervention 0 Did not receive allocated intervention

78 Allocated to intervention TIPS 77 Received allocated intervention 1 Did not receive allocated intervention (hepatocellular carcinoma)

76 Allocated to intervention EPCS 72 Received allocated intervention 4 Did not receive allocated intervention 3 Hepatocellular carcinoma 1 Patient withdrawal

0 Lost to follow-up 0 Discontinued intervention

0 Lost to follow-up 0 Discontinued intervention

0 Lost to follow-up

0 Lost to follow-up

106 Analyzed 0 Excluded from analysis

105 Analyzed 0 Excluded from analysis

78 Analyzed 0 Excluded from analysis

76 Analyzed 0 Excluded from analysis

The overall design and conduct of the 2 prospective randomized clinical trials (RCTs).20-22 Data are from Orloff et al.1,3 BEV indicates bleeding esophageal

varices; EPCS, emergency portacaval shunt; EEST, emergency endoscopic sclerotherapy; and TIPS, transjugular intrahepatic portal-systemic shunt.

gastrointestinal bleeding. Vasopressin or octreotide was administered by continuous intravenous infusion. Patients in both groups received broad-spectrum antibiotics before and for 3 days after primary therapy.

Long-term EEST (RCT No. 1) was performed for the purpose of obliterating esophageal varices and consisted of intravariceal injection of sodium tetradecyl sulfate, 1.5%. The schedule of EEST sessions after initial EEST was the second in 8 days, the third in 22 days, the fourth in 6 weeks, and thereafter, every 3 weeks until the esophageal varices were obliterated. Emergency portacaval shunt involved a direct anastomosis between the portal vein and inferior vena cava within 8 hours of initial contact in 97% of the patients and within 24 hours in the other 3%. Our unchanged technique of PCS has been described in detail repeatedly during the past 45 years.30 Side-to-side EPCS was performed in 94% of the patients and end-to-side EPCS was done in the remainder. Intraoperative pressure measurements were made before and after EPCS by direct needle puncture of the portal vein and inferior vena cava. A large-wedge liver biopsy was obtained and confirmed the diagnosis of cirrhosis. The study protocol required that patients in whom EPCS failed undergo crossover therapy. No EPCS patients experienced treatment failure.

Additional Definitions Bleeding esophageal varices was defined as blood in the esophagus, stomach, or duodenum in a patient with cirrhosis who had hematemesis, melena, or both and shown by upper endoscopy to have actively BEV, had an adherent blood clot, or had red color signs on the varices with no other associated lesion that could reasonably account for bleeding of the observed magnitude. Unselected patients (all comers) encountered consecutively who had findings of cirrhosis and BEV, without exception and without selection, were entered in the RCTs. Failure of emergency primary therapy was declared when, after replacement of blood loss before and during primary treatment, continued or recurrent bleeding required transfusion of 6 U or more of packed red blood cells during the first 7 days after entry into the study. Failure of long-term therapy was declared when, after 7 days, recurrent upper gastrointestinal bleeding shown by endoscopy to be coming from esophagogastric varices required treatment with a total of 8 U or more of packed red blood cells during any 12-month period or, additionally, required blood transfusions after the attending faculty endoscopist had declared the esophageal varices obliterated or gone. Rescue therapy was considered the crossover treatment applied, whenever possible, to patients in either group in whom failure of primary therapy was declared. Emergency endoscopic sclerotherapy (RCT No. 1) was defined as EEST performed within 8 hours of the patient’s initial contact because of upper gastrointestinal bleeding. 158

177 Assessed for eligibility

47 Excluded 47 Did not have cirrhosis and BEV 0 Refused to participate

TIPS (RCT No. 2) TIPS

The TIPS procedure was undertaken within 24 hours after initial contact. A standard widely used TIPS procedure was performed.3 A portosystemic gradient of less than 10 mm Hg was the objective and sometimes required balloon dilatation, additional stents, and, occasionally, variceal embolization. TIPS-Specific Follow-up

Patients underwent color Doppler sonography within 24 hours before and 24 hours after the TIPS procedure, 1 month after TIPS, and then every 3 months. A color Doppler examination of the main por-

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

jamasurgery.com

EPCS for Bleeding Esophageal Varices in Cirrhosis

tal vein, intrahepatic portal vein branches, shunt, and hepatic veins was performed to assess vessel patency, regions of increased velocity suggesting stenosis, and direction of flow. Routine angiography of the stent was done at 1-year intervals following TIPS. Indications for transcatheter shunt recanalization included (1) shunt occlusion; (2) significant stenosis (>50% reduction in luminal diameter) within the portal vein, stent, or hepatic vein; (3) gradient from the portal vein to the inferior vena cava greater than 12 mm Hg; and (4) opacification of gastroesophageal varices on direct portogram. Shunt recanalization entailed thrombolysis, balloon angioplasty, and additional stent placement, alone or in combinations. There often were multiple attempts to correct shunt dysfunction. Recurrent bleeding following TIPS prompted evaluation for shunt dysfunction, and every effort was made to correct any dysfunction. Esophagogastroduodenoscopy was performed every 6 months looking for esophageal varices, gastric varices, and portal hypertensive gastropathy. Esophagogastroduodenoscopy was also done whenever there was other evidence of TIPS occlusion or upper gastrointestinal bleeding.

Posttreatment Therapy After primary therapy, all patients were returned to the same ICU. Patients and their families were given detailed dietary instructions and received repeated counseling about abstinence from alcohol. Patients were discharged from the hospital with a diet limited to 60 g of protein and 2 g of sodium salt per day.

Quantitation of PSE Portal-systemic encephalopathy was quantitated by grading 4 variously weighted components on a scale of 0 to 4: mental state, asterixis, number connection test, and arterial blood ammonia. A PSE index was calculated according to the method of Conn and Liberthal,31 in which the PSE sum was divided by the maximum possible PSE sum. To increase objectivity, a blinded senior faculty hepatologist evaluated the patients for PSE during clinic visits. Details of our assessment of PSE were described in recent publications.32,33 The definition of PSE was based on any of the following criteria: (1) classic signs of altered mental status on physical examination; (2) classic signs of altered mental status described by outside physicians, close relatives, or the patient himself; and (3) a high PSE index (ⱖ0.33). Patients were classified as having recurrent PSE when they had 2 or more episodes of PSE after primary therapy.

Life-long Follow-up All patients were followed up in a designated portal hypertension clinic biweekly for the first 4 weeks, monthly for the remaining 11 months of the first year, and every 3 months thereafter for life. At each clinic visit, clinical status was evaluated and measurements were made of blood cell count, liver function, renal function, psychomotor function, and fluid and electrolyte balance. A dietitian counseled the patients and their families at each clinic visit. Biochemical serum markers for HCC were measured at regular intervals. Patients who received EPCS underwent Doppler duplex ultrasonography yearly to determine PCS patency and function. Whenever upper gastrointestinal bleeding developed, esophagogastroduodenoscopy was performed to document the source. In the EEST group, upper endoscopy was performed routinely every

Special Communication Clinical Review & Education

6 months. If a patient missed 2 consecutive follow-up visits, a research nurse coordinator visited the patient at his or her place of residence to urge subsequent clinic visits. Patients who moved from the referral area were put in contact with a physician who agreed to return completed study data forms to us. There were no dropouts or withdrawals from either RCT, and follow-up was 100%. In RCT No. 1, 203 of the 211 patients (96.2%) underwent follow-up for 10.0 or more years or until death, and the remaining 8 patients underwent 9.4 to 9.9 years of follow-up. Of the 154 participants in RCT No. 2, 131 patients (85.1%) underwent complete follow-up for 5 to 10 years or until death. The remaining 23 patients underwent 3.0 to 4.5 years of follow-up.

Statistical Analysis Variables were summarized by medians and ranges. Statistical comparisons between treatment groups used the Wilcoxon rank sum test for continuous data or Fisher exact test for count data. Survival distributions were obtained by the Kaplan-Meier method and compared using the log-rank test.

Results Patient Characteristics Clinical characteristics are summarized in Table 1. The groups in each RCT were similar in every important characteristic of cirrhosis and BEV. In more than 75% of the patients, the cause of cirrhosis was chronic alcoholism, often in combination with hepatitis. Distribution of patients in Child risk classes, determined by the Campbell et al34 quantitative modification of the criteria originally proposed by Child and Turcotte,35 was almost identical in the 2 groups. Overall Child risk class points in the 2 groups also were essentially identical. Findings on liver biopsy, bedside upper endoscopy, and admission laboratory blood tests are summarized in Table 2. Liver biopsies demonstrated cirrhosis in all patients. Upper endoscopy demonstrated sizable esophageal varices in all patients, with active bleeding or evidence of recent acute bleeding in more than 90% of the cases. Almost two-thirds of the patients had serologic evidence of hepatitis B or C.

Rapidity of Therapy Table 3 provides data on the rapidity of therapy. Mean and median times from onset of bleeding to entry in RCT No. 1 were less than 20 hours in all groups of patients. Mean and median times from onset of bleeding to start of treatment were less than 24 hours. After initial contact, primary therapy was started in less than 8 hours in most patients and always in less than 24 hours, clearly a reflection of the rapidity of the diagnostic workup. Mean time from initial contact to primary therapy was always less than 24 hours. Before a patient’s entry into a study, active bleeding had been observed within 4 hours in 84% of the participants. Without doubt, each study involved evaluation of emergency treatment of acute BEV.

Control of Variceal Bleeding Data on control of variceal bleeding are reported in Table 4. There was a striking and highly significant difference between EEST and EPCS in RCT No. 1 and between TIPS and EPCS in RCT No. 2 (P < .001).

jamasurgery.com

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

159

Clinical Review & Education Special Communication

EPCS for Bleeding Esophageal Varices in Cirrhosis

Table 2. Findings on Liver Biopsy, Endoscopy, and Laboratory Blood Tests at Time of Study Entrya RCT No. 1 Finding

EEST (n = 106)

RCT No. 2

EPCS (n = 105)

P Value

EPCS (n = 76)

TIPS (n = 78)

P Value

Cirrhosis on liver biopsy, No. (%)

106 (100.0)

105 (100.0)

>.99

76 (100.0)

78 (100.0)

>.99

Esophageal varices on endoscopy, No. (%)

106 (100.0)

105 (100.0)

>.99

76 (100.0)

78 (100.0)

>.99

102 (96.2)

105 (100.0)

.68

71 (93.4)

69 (88.5)

.40

Size 2+ to 4+ (scale, 0-4+), No. (%) Active bleeding, No. (%)

41 (38.7)

29 (27.6)

.11

37 (48.7)

50 (52.0)

.50

Clot on varices, No. (%)

53 (50.0)

51 (48.6)

.89

45 (63.0)

40 (52.0)

.18

Red color signs on varices, No. (%)

67 (63.2)

66 (62.9)

>.99

34 (47.0)

35 (44.9)

>.99

Gastric varices on endoscopy, No. (%)

19 (17.9)

17 (16.2)

.86

25 (32.9)

28 (35.9)

.74

Portal hypertensive gastropathy, No. (%)

23 (21.7)

22 (21.0)

>.99

46 (60.5)

38 (48.7)

.15

Laboratory blood tests, median/mean (range) Lowest hemoglobin, g/dL

8.0/8.4 (2.0-8.1)

8.0/8.0 (2.6-11.9)

.40

7.9/7.9 (2.8-22.5)

.17

Lowest hematocrit, %

23.4/22.8 (6.0-39.4)

23.5/23.7 (11.0-35.4)

.46

22.3/21.85 (2-36.8)

7.5/7.4 (2.3-13.7)

23.0/22.6 (7.5-32.8)

.32

Lowest platelet count, ×103/μL

87.0/97.6 (25-340)

76.0/95.4 (15-268)

.36

78.0/88.1 (3-328)

77.5/92.7 (25-296)

.56

.89

Serum bilirubin, mg/dL Direct

0.5/1.1 (0.1-12.0)

0.4/0.7 (0.1-3.7)

.12

0.75/0.91 (0.1-3.7)

0.65/1.29 (0.1-14.7)

Total

2.4/3.5 (0.6-15.2)

1.9/27 (0.4-148.0)

.25

2.25/2.66 (0.9-7.5)

2.15/3.18 (0.5-25.4)

.89

Albumin, g/dL

2.6/2.6 (1.4-3.9)

2.5/2.6 (1.4-4.5)

.49

2.5/2.78 (1.2-24.0)

2.4/2.45 (1.6-3.5)

.43

AST, IU/L

73/116 (16-1359)

66/88 (16-582)

.37

65/83 (24-351)

67/178 (16-4010)

.41

ALT, IU/L

35/54 (11-432)

36/47 (9-582)

.56

32/43 (13-230)

34/72 (9-645)

.52

Alkaline phosphatase, IU/L

86/95 (1.3-367)

79/88 (29-218)

.20

75/90 (12-476)

75/88 (32-299)

.90

International normalized ratio

1/2.8 (0.6-5.8)

1.3/1.4 (1.1-2.6)

.39

1.4/1.5 (1.1-4.1)

1.4/1.5 (1.1-3.0)

.94

Arterial ammonia, μmol/L

76/93 (33-253)

75/92 (26-497)

.83

85/122 (32-2002)

88/97 (24-251)

.65

Creatinine, mg/dL

0.8/1.4 (0.4-10.0)

Serum urea nitrogen, mg/dL Glucose, mg/dL

.89

0.8/0.92 (0.4-2.9)

0.8/0.9 (0.4-2.2)

.94

21/23 (6-69)

.22

20/23 (5-91)

19/21 (4-68)

.43

133/152 (74-548)

135/157 (60-368)

.44

128/146 (11-389)

143/154 (89-367)

.17

Blood alcohol ≥0.01%, No. (%)

9 (8)

13 (12)

.29

7 (9)

6 (8)

.78

Blood α-fetoprotein, ng/mL, mean (range)

4 (0-18)

.84

4 (0-118)

4 (0-181)

.98

Positive HBV or HCV serology, No. (%)

160

0.8/1.0 (0.5-2.7)

22/27 (7-85)

37 (35)

4 (0-10) 44 (42)

.32

47 (62)

49 (63)

>.99

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; EEST, emergency endoscopic sclerotherapy; EPCS, emergency portacaval shunt; HBV, hepatitis B virus; HCV, hepatitis C virus; RCT, randomized clinical trial; TIPS, transjugular intrahepatic portal-systemic shunt.

micromoles per liter, 17.104; creatinine to micromoles per liter, 88.4; α-fetoprotein to micrograms per liter, 1; glucose to millimoles per liter, 0.0555; hematocrit to proportion of 1.0, 0.01; hemoglobin to grams per liter, 10; platelets to ×109 per liter, 1; and urea nitrogen to millimoles per liter, 0.357.

SI conversion factors: to convert albumin to grams per liter, multiply by 10; alkaline phosphatase, ALT, and AST to microkatals per liter, 0.0167; bilirubin to

a

Excluding indeterminate deaths within the first 3 weeks from causes other than bleeding, EEST achieved long-term control of bleeding in only 20% of the patients, and TIPS achieved long-term control of bleeding in only 22%. In contrast, EPCS promptly and permanently controlled bleeding in 100% of the patients in RCT No. 1 and in 97% of those in RCT No. 2.

and TIPS in RCT No. 2, were essentially identical. Subsequently, however, there were highly significant differences at all time intervals in each RCT (P < .001). Median survival following TIPS was 1.99 years, and median survival following EPCS was longer than 10 years (P < .001). Survival rates were related to effectiveness in control of bleeding as well as to the severity of liver disease.

Survival

Operative EPCS Data

Table 4 summarizes data on survival, and Figure 2 and Figure 3 show long-term Kaplan-Meier estimated survival plots in each RCT. Thirtyday survival rates for EEST and EPCS in RCT No. 1, as well as for EPCS

Portal vein–inferior vena cava pressure gradient before shunt insertion averaged 244 mm of saline in RCT No. 1 and 241 mm of saline in RCT No. 2. The EPCS procedure reduced the portal vein–inferior vena

Data are from Orloff et al.1,3

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

jamasurgery.com

EPCS for Bleeding Esophageal Varices in Cirrhosis

Special Communication Clinical Review & Education

Table 3. Rapidity of Therapy in RCT No. 1 and RCT No. 2a RCT No. 1

RCT No. 2

EEST (n = 106)

EPCS (n = 105)

EPCS (n = 76)

TIPS (n = 78)

Median/ Mean

Range

Median/ Mean

Range

P Value

Median/ Mean

Bleeding onset to study entry, h

12/20

0-144

16/19

0-95

.30

20/25

0-112

17/24

0.5-84

.56

Bleeding onset to primary therapy, h

15/23

3-147

19/22

2.6-100

.056

28/36

8-131.7

34/39

11.1-99.4

.08

3/4

1.4-24.3

.99

%>8h

0

2.9

3.9

68 (87)

.50

.17

63 (83)

68 (87)

Transfer patients, No. (%)

71 (67)

Bleeding onset to entry in referring hospital, h

4/10

Entry into referring hospital to study entry, h

7/12

Last observation of bleeding to study entry, h

0/2

80 (76) 0-127 1.5-53 0-32

.50

4/10

0-83.6

.92

6/12

0-120

6/10

0-62

.72

8/12

0-53

.56

13/18

2.2-110

10/16

2.6-65.3

.42

0/2

0-30

.76

0/2

0-32

0/4

0-84

≤4 h, % of group

84

84

.94

67 (88)

63 (81)

.35

>4 h, % of group

16

16

>.99

9 (12)

15 (19)

.27

Abbreviations: EPCS, emergency portacaval shunt; EEST, emergency endoscopic sclerotherapy; RCT, randomized clinical trial; TIPS, transjugular intrahepatic portal-systemic shunt. a

Data are from Orloff et al.1,3

b

Statistically significant difference.

cava gradient to a mean 17 mm in RCT No. 1 and 20 mm in RCT No. 2, and EPCS promptly eliminated portal hypertension in all patients and stoppedthebleeding.ThePCSremainedpatentthroughoutlifein98% of the patients in RCT No. 1 and 97% of those in RCT No. 2.

Operative TIPS Data Median portal vein–inferior vena cava pressure gradient before TIPS was 22 mm Hg (range, 10-42 mm Hg), and after TIPS it was reduced to 7 mm Hg (range, 0-17 mm Hg). At the conclusion of the procedure, TIPS was considered a technical success in 87% of patients, a failure in 11%, and indeterminate in 2%.

TIPS Technical Success or Failure During Follow-up Specific follow-up of TIPS by color Doppler sonography, angiography, esophagogastroduodenoscopy, and regular clinical examinations was conducted rigorously. Stenosis or occlusion associated with TIPS was demonstrated in 38 of the patients (49%). They developed a mean (SD) of 2.1 (1.8) episodes of TIPS stenosis or occlusion (range, 0-9), which was demonstrated 46 times in the first year after entry into the study, 26 times in the second year after entry, and 26 times in patients who survived for 3 or more years. Twenty-four of these 38 patients (63.2%) with TIPS malfunction underwent TIPS revision by balloon angioplasty or insertion of 1 or more additional stents. Revisions failed in 80% of the patients, and success was indeterminate in an additional 8%. The durability of TIPS was disappointing.

episodes of PSE after primary therapy in patients who survived 30 days and were discharged from the hospital. In RCT No. 1, recurrent PSE developed in 35% of the EEST group and 15% of the EPCS group, a highly significant difference (P = .001). In RCT No. 2, recurrent PSE developed in 61% of the TIPS patients compared with 21% of the EPCS patients, demonstrating a highly significant 3-fold greater incidence in the TIPS group (P < .001).

Liver Transplant Of the 211 participants in RCT No. 1, only 13 patients (6.2%) were ultimately referred for liver transplant, only 7 (3.3%) were approved for liver transplant, and only 4 (1.9%) underwent liver transplant. The 1- and 5-year liver transplant survival rates were 0.68% and 0%, respectively, compared with 81% and 73% after EPCS. In the 1300 unrandomized PCS patients, 50 individuals (3.8%) were referred and 19 (1.5%) underwent liver transplant. The 5-year survival rate was 53% compared with 72% for all 1300 patients. We concluded that if bleeding is permanently controlled, patients with cirrhosis and BEV seldom require liver transplant. Should liver transplant be required in patients with PCS, numerous studies have shown that PCS does not increase mortality or complications. Our experience confirmed that conclusion.

Hepatocellular Carcinoma Portal-Systemic Encephalopathy Portal-systemic encephalopathy was reported in detail in our recent publications.32,33 Recurrent PSE was defined as 2 or more

Hepatocellular carcinoma developed in 10 of the 106 patients randomized to EEST and 5 of the 105 patients randomized to EPCS. No significant differences distinguished the patients who

jamasurgery.com

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

161

Clinical Review & Education Special Communication

EPCS for Bleeding Esophageal Varices in Cirrhosis

Table 4. Control of Bleeding and Survival in RCT No. 1 and RCT No. 2a RCT No. 1

RCT No. 2

EEST (n = 106)

EPCS (n = 105)

P Value

TIPS (n = 78)

P Value

Received vasopressin or octreotide intravenously

95 (89.6)

80 (76.2)

.01b

60 (83)

71 (92)

.13

Bleeding at start of vasopressin or octreotide infusion

47 (49.5)

55 (68.8)

.01b

60 (83)

44 (71)

.59

Bleeding decreased or stopped with vasopressin or octreotide

36 (76.6)

38 (69.1)

.51

19 (48)

24 (58)

.66

4 (3.8)

12 (11.4)

.04b

16 (21.1)

14

22 (21)

93 (100)

5.44)

b

>9.80 (5.09->9.80)

2.12 (1.03-3.00)

9.66)

0.08 (0.04-1.99)

.02b

2.80)

1.25 (0.04->9.79)

1.03 (0.49->7.13)

Transfer

2.45 (1.59-3.76)

6.25 (5.81-11.03)

>10 (5.18->10)

1.99 (0.51-3.0)

Mean survival (95% CI), y

Rescue portacaval shunt in EEST or TIPS effect on median survival (95% CI), y Successful EEST (n = 21) or TIPS (n = 15)

2.76 (1.52-7.35)

.48

3.75 (1.15->7.18)

.10

>8.06 (3.93->8.06)

.95

Failed EEST (n = 81) or TIPS (n = 55)

2.71 (1.65-3.92)

Rescue shunt–failed EEST (n = 50) or rescue shunt–failed TIPS (n = 8)

3.01 (1.65-4.34)

2.04 (0.92-2.88)

No rescue shunt–failed EEST (n = 31) or rescue shunt–failed TIPS (n = 39)

2.36 (0.72-4.34)

Overall survival: rescue shunt (n = 50) vs primary EPCS (n = 105) (RCT No. 1), rescue shunt (n = 8) vs primary EPCS (RCT No. 2)

3.01 (1.65-4.34)

6.18 (5.61-10.38)

10 (5.1->10)

>8.06 (3.93->8.06)

.98

Postoperative survival: rescue shunt (n = 50) vs primary EPCS (n = 105) (RCT No. 1), rescue shunt (n = 8) vs primary EPCS (n = 76) (RCT No. 2)

1.99 (1.34-3.73)

6.18 (5.61-10.38)

10 (5.1->10)

>8.04 (3.08->8.04)

.67

.001b

1.03 (0.50-2.67)

Abbreviations: EPCS, emergency portacaval shunt; EEST, emergency endoscopic sclerotherapy; PRBC, packed red blood cell; RCT, randomized clinical trial; TIPS, transjugular intrahepatic portal-systemic shunt; UCSD, University of California, San Diego. a

Data are from Orloff et al.1,3

b

Statistically significant difference.

a consequence of differences in the effectiveness of emergency treatment of BEV.

Direct Cost of Care Total postindex charges were significantly greater in patients who underwent EEST or TIPS compared with those who underwent EPCS (P < .001), and total overall charges for emergency and long-term care required for several years were greater in patients who received emergency followed by long-term repetitive EEST or by TIPS related to length of survival and, therefore, days or years during which care was required. Emergency portacaval shunt was significantly less expensive than EEST or TIPS in every aspect of care except for the index admission. Charges for postindex care per year in the EEST and EPCS groups, respectively, were a mean $108 500 vs $25 100 (P < .001). Mean total overall charges for care of patients who entered RCT No. 1 were $168 100 per year in the EEST group and $39 400 per year in the EPCS group (P < .001). Direct costs of care were 5 to 7 times greater in the EEST ($168 100) and TIPS ($264 800) groups than in the EPCS ($39 000) group (P < .001). The reasons why EPCS was less costly than EEST and TIPS are likely 164

Discussion The 2 RCTs that are the substance of this report were unique in several important respects. They involved a large number of unselected consecutive patients (all comers) with acute BEV: 211 in RCT No. 1 and 154 in RCT No. 2. No comparable studies of similar magnitude have been reported. Unlike patients in other studies, nearly 30% of the patients in these RCTs were in Child class C with advanced cirrhosis. Both RCTs were community-wide endeavors in which, by prior agreement, 72% to 82% of the patients were rapidly referred to UCSD from hospitals in 4 area California counties. The entire diagnostic workup was completed at the bedside in the ICU within less than 8 hours of study entry. Definitive therapy was undertaken within 15 hours of initial contact with the UCSD staff. EPCS was performed by 2 senior faculty surgeons with extensive op-

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

jamasurgery.com

EPCS for Bleeding Esophageal Varices in Cirrhosis

Special Communication Clinical Review & Education

Figure 2. Overall Survival After Emergency Endoscopic Sclerotherapy (EEST) and Emergency Portacaval Shunt (EPCS)

Figure 3. Overall Survival After Emergency Portacaval Shunt (EPCS) and Transjugular Intrahepatic Portal-Systemic Shunt (TIPS) 1.0

1.0

0.8 EPCS

0.6

Probability of Survival

Probability of Survival

0.8

EPCS

0.4

0.6

0.4 TIPS

0.2 EEST 0.2 0 0 0

No. at Risk EPCS 105 EEST 106

2

5

80 57

76 22

Years

10

15

40 8

3 1

Kaplan-Meier estimates of overall survival after EEST (n = 106) and EPCS (n = 105); 95% CIs are represented by shaded areas; P < .001. Data are from Orloff et al.1

erative experience with portacaval shunt. EEST was performed by board-certified attending faculty gastroenterologists with long experience in endoscopic therapy. Transjugular intrahepatic portalsystemic shunt was performed by senior attending faculty interventional radiologists with vast experience in TIPS. The type of EPCS was a direct portacaval shunt, side-to-side in almost all patients and endto-side in a few. Cirrhosis was confirmed by liver biopsy in 100% of patients. Regular follow-up was accomplished in 100% of patients and lasted for 10 or more years or until death in RCT No. 1 and for 5 to 10 years or until death in RCT No. 2. No comparable studies with similar length of follow-up have been reported. To date, there have been no reports of RCTs in which EEST or TIPS has been compared with emergency PSS in a broad spectrum of patients with cirrhosis and acute BEV. All reported studies36-58 have involved elective treatment of BEV and are clearly not applicable to emergency treatment. These studies have been thoroughly reviewed in a recent publication.3 Three RCTs36-38 have been reported in which direct PCS was compared with EEST in elective treatment directed at prevention of rebleeding in patients who had recovered from a variceal hemorrhage. It is difficult to draw conclusions from these studies because they involved a small number of highly selected patients who were monitored for short periods of time. Most important, the results of these studies are clearly not applicable to the emergency treatment of BEV. Unlike the random clinical trials reported herein, there have been no reports of the use of TIPS as first-line, primary emergency treatment of acute BEV in a sizable group of patients with long-term follow-up. Several reports49-57 have described studies of the use of TIPS as rescue treatment following failure to control bleeding by endoscopic and pharmacologic therapy. Seven of these reports50-56 pub-

No. at Risk EPCS 76 TIPS 78

1

3

56 43

49 28

Years

5

7

37 13

25 6

Kaplan-Meier estimates of overall survival after EPCS (n = 76) and TIPS (n = 78); 95% CIs are represented by shaded areas; P < .001. Data are from Orloff et al.3

lished between 1994 and 2001 generally involved few patients and short follow-up. Bleeding was controlled initially in 90% to 100% of patients, but the mortality rate remained at 27% to 50% despite control of bleeding. Four RCTs39-43 have been reported in which distal splenorenal shunt (DSRS), a so-called selective shunt, was compared with EEST in elective treatment aimed at prevention of rebleeding in patients who had recovered from 1 or more episodes of BEV. All 4 trials excluded patients with Child risk class C and were highly selective, involving fewer than one-third of the patients with cirrhosis who received treatment for variceal hemorrhage at the 4 institutions. In all 4 studies, recurrent variceal bleeding was much more frequent in the EEST groups than in the DSRS groups. Nevertheless, there was no difference in survival rates between the 2 groups of patients in any of the 4 studies. In addition, 2 of the studies40-42 observed a significant incidence of shunt thrombosis, amounting to 10% and 14%, respectively, after DSRS. Although these randomized trials of elective therapy in selected patients are interesting, they have no direct bearing on the comparative effectiveness of PCS and EEST in emergency treatment of BEV. Four studies44-48 have been reported in which surgical PSSs and TIPS were compared in selected patients with BEV due to cirrhosis. Three of these studies were RCTs and 1 was a retrospective case-control study.45 Three of the 4 studies were restricted to lowrisk patients, excluding the most seriously ill patients in Child risk class C, and had short-term follow-up lasting, respectively, 1.6 to 1.9, 2.4, and 3.75 years. The surgical shunt in 2 of the studies was a DSRS. In a third study, a DSRS was used in 12 patients or a “small stoma” portacaval shunt was used in 28 patients. In the fourth trial, a small-diameter prosthetic H-graft PCS was used. Unlike our RCT, none of these trials involved a standard direct side-to-side portacaval shunt and none addressed the important issue of emergency treatment of acute BEV. The results of these 4 trials

jamasurgery.com

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

165

Clinical Review & Education Special Communication

favored surgical shunts in all outcome measures. Specifically, recurrent BEV developed in 25% of TIPS patients but in only 4% of surgical patients. Shunt stenosis occurred in 65% of the TIPS group compared with 10% of the surgical shunt group. In the only RCT with long-term follow-upthat involved 132 patients,46,47 survival following prosthetic HGPCS was significantly longer than survival following TIPS in patients in Child class A (91 vs 19 months; P = .09) and in patients in Child class B (63 vs 21 months; P = .02). On the other hand, survival of patients in Child class C was significantly longer after TIPS (45 vs 22 months; P = .04). Overall survival after 2 years was 68% following HGPCS vs 53% after TIPS and after 5 years was 47% after HGPCS vs 31% after TIPS. The definition of shunt failure involved death, rebleeding, liver transplant, irreversible shunt occlusion, and technical inability to accomplish TIPS or surgical shunt. In the final 7 years of the RCT, the important matter of surgical shunt patency was not examined. In a retrospective study by Tzeng and colleagues58 in 107 Asian patients, two-thirds (68%) of whom had cirrhosis due to viral hepatitis, rescue TIPS performed after failure to control bleeding by endoscopicandpharmacologictherapywasfollowedbyhighmortalityrates of 28% within 30 days, 35% within 60 days, and 50% within 1 year. A recent report by García-Pagán and colleagues59 of an RCT comparing early TIPS with standard pharmacologic and endoscopic therapy is noteworthy. The study was conducted during a period of 2.75 years among 9 European centers and involved randomization of 63 highly selected patients with Child class B and C cirrhosis. A total of 296 patients were excluded from the RCT, including all patients in Child class A, as well as patients in Child class C with scores greater than 13 points. The 32 patients randomized to TIPS received the expanded polytetrafluoroethylene (PTFE)–covered stent. Transjugular intrahepatic portal-systemic shunt was performed within 72 hours after diagnostic endoscopy, and vasoactive drugs were administered until TIPS was performed. Median follow-up was 16 months and maximum follow-up was 2 years. Outcomes strongly favored early TIPS compared with standard treatment with regard to initial control of bleeding (97% vs 55%), 1-year actuarial probability of control of BEV (97% vs 50%), survival for 6 weeks (97% vs 61%), and 1-year actuarial survival (86% vs 61%). Rescue TIPS was performed in 7 patients in the endoscopic-pharmacotherapy group, 4 of whom died. As for PSE, 1-year actuarial probability was 28% following early TIPS versus 40% in the endoscopic-pharmacotherapy patients. On the other hand, after discharge, 1-year probability of additional PSE was higher in the early TIPS group than in the endoscopic-pharmacotherapy group (19% vs 10%). These results are contrary to those of previous studies that showed reduction in the rebleeding rate by TIPS but an increase in PSE without improvement in survival.60-62 In an editorial that followed the publication of García-Pagán et al,59 Afdhal and Curry63 summarized with the statement that “additional clinical trials of adequate size should be performed to confirm these findings and to examine the effect of a rapid reduction in portal pressure on disease progression in patients with cirrhosis of other causes” (ie, other than alcoholism). Currently, TIPS is being performed most often with the PTFEcovered stent. A potential limitation of RCT No. 2 is that it was initiated before the covered stent was introduced and was conducted in part before the covered stent was in widespread use. Several studies of TIPS have been undertaken in which covered stents were compared with bare stents, but none of these studies pro166

EPCS for Bleeding Esophageal Varices in Cirrhosis

vided scientifically acceptable or conclusive evidence of the superiority of the covered stent. These studies were summarized by Clark and colleagues64,65 in 2010 and 2011 under the direction of Rosemurgy, a group with more than a decade of experience with studies of TIPS and surgical PSSs. Clark et al concluded that “the early data with covered stents indicate no differences in rates of encephalopathy or survival between covered and noncovered TIPS.” Contrary to the conclusions of Clark et al, a meta-analysis by Yang and colleagues66 stated that “this meta-analysis of 6 studies shows that the use of PTFE-covered stent grafts clearly improves shunt patency without increasing the risk or hepatic encephalopathy and with a trend towards better survival.” However, the authors observed appropriately that their meta-analysis was limited by the retrospective design of all the selected observational studies (except for one RCT) because all patients with bare stents and those with PTFEcovered stents were studied at different times and because the overall care of the patients had improved from the period of bare stents to the period of PTFE-covered stents, a fact that undermined the validity of the comparison. In a critique of the meta-analysis of Yang et al, also in 2010, Amarapurkar and Sanyal67 cataloged some of the flaws in the study, which included (1) only 1 of 6 studies was a randomized comparison, (2) there was no systematic assessment of shunt patency, (3) the criteria used to identify the need for angiographic assessment of the TIPS were not explicit, (4) the individual studies were not appropriately designed to rigorously answer the issue of survival benefit, and (5) the individual studies were not designed to dissect the issue of PSE. In summary of their critique, Amarapurkar and Sanyal concluded that PTFE-coated TIPS have a higher patency rate and require less interventions compared with TIPS using bare stents, but “while overall survival and encephalopathy appear to be improved, these findings are not definitive and require further rigorous analysis.” In summary of the issue of a covered vs a bare stent in TIPS, we propose that there is no basis for concluding that the results of TIPS in RCT No. 2 would have been better if covered stents had been used for this procedure. The EPCS between the portal vein and inferior vena cava permanently controlled BEV in 95% or more of the patients in our current studyandresultedinlong-termsurvivalthatwasmarkedlygreaterthan that obtained with TIPS or EEST. These results are similar to our recent and past experience with EPCS.1-3,13-19 Moreover, the survival rate is higher than that reported by other investigators29-36,68-73 who used PSSs as emergency treatment or to prevent bleeding. We believe that several factors were responsible for the consistent results of EPCS that we have obtained during a period of 4 decades. We reviewed these factors in a recent publication.1 Briefly, the first factor is simplification of the diagnostic workup by elimination of unnecessary studies as a routine practice prior to definitive therapy, which made it possible to accomplish the entire diagnostic workup at the bedside in less than 8 hours without moving the patients out of the ICU. A second factor is adoption of a specific protocol for an organized system of care before and after EPCS—something that is difficult to accomplish with uniformityinmulti-institutionalstudies.Patientsinall4armsofourRCTswere admitted directly to the same ICU where the personnel had specific training and long experience in the care of patients with cirrhosis of the liver, and they were returned to that same ICU postoperatively. Careofallpatientsinall4armsoftheRCTswassupervisedbyonegroup of attending physicians throughout the study. Third, a rigorous, lifelong program of follow-up was conducted in which there was an

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

jamasurgery.com

EPCS for Bleeding Esophageal Varices in Cirrhosis

Special Communication Clinical Review & Education

intensification of efforts to obtain dietary protein control and abstinence from alcohol. Visits to the portal hypertension clinic were scheduledmonthlyforthefirstpostoperativeyearandevery3monthsthereafter. A dietitian who was employed by the RCT grant and was trained in the care of patients with cirrhosis after shunt, EEST, and TIPS procedures was stationed in the clinic to counsel the patient at each visit on the restriction of dietary protein intake to 60 g/d. Serious efforts were made to enroll all patients with alcoholism into an alcohol rehabilitation program, such as Alcoholics Anonymous, or a similar programatourowninstitution.Ourpatientswithportacavalshuntsproved to be more receptive to rehabilitation therapy compared with many alcoholic individuals, perhaps because of their almost lethal experience with massive bleeding and EPCS. The frequency of permanent abstinence, as a result, was 85% in the EPCS group. Finally, we had a low incidence of early and long-term shunt thrombosis. The PCS remained patent in 97.4% of the patients, which is consistent with our past and recent experience with direct side-to-side PCS.1 Shunt occlusion represents a serious technical failure of surgical therapy and is usually followed by recurrent BEV and often death. As documented in our 2009 publication,1 high rates of occlusion of various types of shunts have been reported from centers known for experience in the treatment of portal hypertension. A frequent criticism of PSSs is the observation or assumption that control of bleeding is achieved at the cost of a high rate of PSE. There is a widespread belief that long-term use of EEST of BEV is associated with a substantially lower rate of PSE than is treatment with a PSS.74,75 The results of our RCTs are contrary to both of these conclusions. On initial contact before entry in the trial, 19% of patients in each group had PSE, most likely from upper gastrointestinal bleeding. A history of PSE was noted in 18% of the EEST patients and in 29% of the EPCS patients. Chronic, recurrent PSE that required treatment and diminished quality of life developed in 35% of patients treated by EEST and 15% of patients who received EPCS in RCT No. 1 and in 61% of patients treated by TIPS compared with 21% of those treated with EPCS in RCT No. 2, a significant difference (P = .001). Forty percent of the patients who experienced posttherapy PSE had PSE pretherapy. This relatively low incidence of PSE in patients with PCS is consistent with our past experience.13-19 Experience with the

ARTICLE INFORMATION Accepted for Publication: April 18, 2013. Published Online: January 8, 2014. doi:10.1001/jamasurg.2013.4045. Conflict of Interest Disclosures: None reported. Funding/Support: The RCTs were supported by grants 1R01 DK41920 and AM17103 from the National Institutes of Health and grants from the Surgical Education and Research Foundation, a 501(c)(3) organization. This work was supported in part by Health Resources and Services Administration contract 234-2005-370011C. Role of the Sponsors: The funding sources had no role in the design and conduct of the study; collection, management, analysis, or interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Disclaimer: The content is the responsibility of the author alone and does not necessarily reflect the views or policies of the Department of Health and

patients in this study and the results in our other studies of EPCS13-19 demonstrate that a low incidence of PSE is possible when, with the help of rigorous follow-up, patients abstain from alcohol and adhere to a diet of moderate protein restriction.

Conclusions In these 2 RCTs of emergency treatment of BEV in 365 unselected consecutive patients, EPCS was found to be superior in every respect to emergency and long-term use of EEST and to TIPS, even when failure of EEST and TIPS was treated by crossover rescue PCS as salvage therapy. The EPCS procedure promptly and permanently stopped variceal bleeding in virtually all patients, rarely became occluded or lost its effectiveness, was accomplished with a relatively low incidence of PSE, and produced long-term survival rates that were significantly greater than those resulting from EEST or TIPS. In addition, EPCS resulted in significantly lower charges than EEST and TIPS. In regard to HCC, regular long-term screening in our 2 RCTs by routine serum α-fetoprotein measurement and ultrasonography plus supplementary computed tomography failed to identify HCC at a stage when curative treatment by resection or liver transplant was possible. The detection of HCC in patients with cirrhosis following control of BEV remains a serious unsolved problem. Finally, in regard to liver transplant, our 2 RCTs and prospective data from our large series of unrandomized patients in whom we performed PCS indicate clearly that BEV alone should not be considered an indication for liver transplant. Portacaval shunt produced long-term survival rates superior to those produced by liver transplant. The importance of these results is underscored by the severe shortage of donor organs, the much higher cost of liver transplant compared with PCS, and the substantial difficulties in obtaining insurance coverage for transplant. Finally, it should be clear that the widespread practice of using surgical procedures mainly or only as salvage for failure of endoscopic therapy or TIPS is not supported by the results of these trials and by our 53 years of experience with treatment of BEV.

University, New Haven, Connecticut), Haile T. Debas, MD (University of California, San Francisco), Peter B. Gregory, MD (Stanford University, Stanford, California), and Alexander R. Margulis, MD (University of California, San Francisco).

Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government. Previous Presentation: This study was presented by invitation at the 84th Annual Meeting of the Pacific Coast Surgical Association, February 18, 2013; Kauai, Hawaii. Additional Information: Coinvestigators at UCSD: Department of Surgery: Marshall J. Orloff, MD, Robert J. Hye, MD, Mark S. Orloff, MD, Susan L. Orloff, MD, and Karen J. Orloff, MSW. Department of Medicine–Gastroenterology and Hepatology: Henry O. Wheeler, MD (deceased), Jon I. Isenberg, MD (deceased), Kevin S. Haynes, MD, Roderick Rapier, MD, Kenneth R. McQuaid, MD, and Horacio Jinich-Brook, MD. Department of Radiology: Karim Valji, MD, Anne C. Roberts, MD, and Steven C. Rose, MD. Department of Pathology: Katsumi Miyai, MD, PhD, and Cynthia Behling, MD. Department of Biostatistics: Florin Vaida, PhD. External Advisory, Data-Monitoring, and Safety Committee, 1988-2011: Harold O. Conn, MD (deceased) (Yale

jamasurgery.com

Additional Contributions: We thank the many residents in the departments of Surgery, Radiology, and Medicine at UCSD who played a major role in the care of patients in this study. We are particularly grateful to Karim Valji, MD, and his colleagues in the Department of Radiology who played an important role in the conduct of the radiologic procedures such as TIPS. We thank Harold O. Conn, MD, of Yale University, and Haile T. Debas, MD, and Alexander R. Margulis, MD, of the University of California, San Francisco, who served voluntarily as an external advisory data safety and monitoring committee. We thank the many physicians practicing in the California counties of San Diego, Imperial, Orange, and Riverside who helped with patient recruitment, referral, and long-term follow-up.

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

167

Clinical Review & Education Special Communication

Correction: This article was corrected on May 9, 2014, to revise its classification and to indicate the two previously reported randomized clinical trials that this article summarizes. REFERENCES

patients with alcoholic cirrhosis. Ann Surg. 1980; 192(3):325-340. 17. Orloff MJ, Orloff MS, Rambotti M, Girard B. Is portal-systemic shunt worthwhile in Child’s class C cirrhosis? Ann Surg. 1992;216(3):256-268.

1. Orloff MJ, Isenberg JI, Wheeler HO, et al. Randomized trial of emergency endoscopic sclerotherapy versus emergency portacaval shunt for acutely bleeding esophageal varices in cirrhosis. J Am Coll Surg. 2009;209(1):25-40.

18. Orloff MJ, Bell RH Jr, Orloff MS, Hardison WGM, Greenburg AG. Prospective randomized trial of emergency portacaval shunt and emergency medical therapy in unselected cirrhotic patients with bleeding varices. Hepatology. 1994;20(4, pt 1):863-872.

2. Orloff MJ, Isenberg JI, Wheeler HO, et al. Direct costs of care in a randomized controlled trial of endoscopic sclerotherapy versus emergency portacaval shunt treatment of acutely bleeding esophageal varices in cirrhosis—part 4. J Gastrointest Surg. 2011;15:38-47.

19. Orloff MJ, Orloff MS, Orloff SL, Rambotti M, Girard B. Three decades of experience with emergency portacaval shunt for acutely bleeding esophageal varices in 400 unselected patients with cirrhosis of the liver. J Am Coll Surg. 1995;180(3): 257-272.

3. Orloff MJ, Vaida F, Haynes KS, Hye RJ, Isenberg JI, Jinich-Brook H. Randomized controlled trial of emergency transjugular intrahepatic portosystemic shunt versus emergency portacaval shunt treatment of acute bleeding esophageal varices in cirrhosis. J Gastrointest Surg. 2012;16(11):2094-2111.

20. Moher D, Schulz KF, Altman D; CONSORT Group (Consolidated Standards of Reporting Trials). The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomized trials. JAMA. 2001;285(15):1987-1991.

4. Graham DY, Smith JL. The course of patients after variceal hemorrhage. Gastroenterology. 1981; 80(4):800-809.

21. Altman DG, Schulz KF, Moher D, et al; CONSORT GROUP (Consolidated Standards of Reporting Trials). The revised CONSORT statement for reporting randomized trials. Ann Intern Med. 2001; 134(8):663-694.

5. Smith JL, Graham DY. Variceal hemorrhage: a critical evaluation of survival analysis. Gastroenterology. 1982;82(5, pt 1):968-973. 6. Orloff MJ, Duguay LR, Kosta LD. Criteria for selection of patients for emergency portacaval shunt. Am J Surg. 1977;134(1):146-152. 7. Mikkelsen WP. Therapeutic portacaval shunt: preliminary data on controlled trial and morbid effects of acute hyaline necrosis. Arch Surg. 1974; 108(3):302-305. 8. Conn HO. Prophylactic portacaval shunts. Ann Intern Med. 1969;70(4):859-864. 9. Conn HO, Lindenmuth WW. Prophylactic portacaval anastomosis in cirrhotic patients with esophageal varices and ascites. Am J Surg. 1969;117 (5):656-661. 10. Jackson FC, Perrin EB, Smith AG, Dagradi AE, Nadal HM. A clinical investigation of the portacaval shunt, II: survival analysis of the prophylactic operation. Am J Surg. 1968;115(1):22-42. 11. Resnick RH, Chalmers TC, Ishihara AM, et al; Boston Inter-Hospital Liver Group. A controlled study of the prophylactic portacaval shunt: a final report. Ann Intern Med. 1969;70(4):675-688. 12. Orloff MJ. Elective therapeutic porta-caval shunt. In: Orloff MJ, Stipa S, Ziparo V, eds. Medical and Surgical Problems of Portal Hypertension. New York, NY: Academic Press Inc; 1980:127-136. 13. Orloff MJ. Emergency portacaval shunt. Ann Surg. 1967;166(3):456-478. 14. Orloff MJ, Chandler JG, Charters AC III, et al. Emergency portacaval shunt treatment for bleeding esophageal varices. Arch Surg. 1974;108 (3):293-299. 15. Orloff MJ, Charters AC III, Chandler JG, et al. Portacaval shunt as emergency procedure in unselected patients with alcoholic cirrhosis. Surg Gynecol Obstet. 1975;141(1):59-68. 16. Orloff MJ, Bell RH Jr, Hyde PV, Skivolocki WP. Long-term results of emergency portacaval shunt for bleeding esophageal varices in unselected

168

EPCS for Bleeding Esophageal Varices in Cirrhosis

22. Schulz KF, Altman DG, Moher D; CONSORT Group. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c332. doi:10.1136/bmj.c332. 23. Iwatsuki S, Starzl TE, Todo S, et al. Liver transplantation in the treatment of bleeding esophageal varices. Surgery. 1988;104(4):697-705. 24. Reyes J, Iwatsuki S. Current management of portal hypertension with liver transplantation. Adv Surg. 1992;25:189-208. 25. Bismuth H, Adam R, Mathur S, Sherlock D. Options for elective treatment of portal hypertension in cirrhotic patients in the transplantation era. Am J Surg. 1990;160(1):105110. 26. Orloff MJ, Isenberg JI, Wheeler HO, et al. Liver transplantation in a randomized controlled trial of emergency treatment of acutely bleeding esophageal varices in cirrhosis. Transplant Proc. 2010;42(10):4101-4108. 27. Benvegnù L, Gios M, Boccato S, Alberti A. Natural history of compensated viral cirrhosis. Gut. 2004;53(5):744-749. 28. Degos F, Christidis C, Ganne-Carrie N, et al. Hepatitis C virus related cirrhosis. Gut. 2000;47(1): 131-136. 29. Sangiovanni A, Del Ninno E, Fasani P, et al. Increased survival of cirrhotic patients with a hepatocellular carcinoma detected during surveillance. Gastroenterology. 2004;126(4):10051014. 30. Orloff MJ, Orloff SL, Orloff MS. Portacaval shunts: side-to-side and end-to-side. In: Clavien P-A, Saar MG, Fong Y, eds. Atlas of Upper Gastrointestinal and Hepato-Pancreato-Biliary Surgery. Berlin, Germany: Springer-Verlag; 2007: 687-702. 31. Conn HO, Liberthal MM. The Hepatic Coma Syndromes and Lactulose. Baltimore, MD: Williams & Wilkins; 1978.

32. Orloff MJ, Isenberg JI, Wheeler HO, et al. Portal-systemic encephalopathy in a randomized controlled trial of endoscopic sclerotherapy versus emergency portacaval shunt treatment of acutely bleeding esophageal varices in cirrhosis. Ann Surg. 2009;250(4):598-610. 33. Orloff MD, Vaida F, Haynes KS, Hye RJ, Iseberg JI, Jinich-Brook H. Portal-systemic encephalopathy in a randomized controlled trial of emergency transjugular intrahepatic portosystemic shunt versus emergency portacaval shunt treatment of acute bleeding esophageal varices in cirrhosis. Am J Surg. in press. 34. Campbell DP, Parker DE, Anagnostopoulos CE. Survival prediction in portacaval shunts. Am J Surg. 1973;126(6):748-751. 35. Child CG III, Turcotte JG. Surgery and portal hypertension. In: Child CG III, ed. The Liver and Portal Hypertension. Philadelphia, PA: WB Saunders; 1964:1-85. 36. Korula J, Yelin A, Yamada S, Weiner J, Cohen H, Reynolds TB. A prospective randomized controlled comparison of chronic endoscopic variceal sclerotherapy and portal-systemic shunt for variceal hemorrhage in Child’s class A cirrhotics [abstract]. Hepatology. 1988;8:1242. 37. Planas R, Boix J, Broggi M, et al. Portacaval shunt versus endoscopic sclerotherapy in the elective treatment of variceal hemorrhage. Gastroenterology. 1991;100(4):1078-1086. 38. Isaksson B, Jeppsson B, Bengtsson F, Hannesson P, Herlin P, Bengmark S. Mesocaval shunt or repeated sclerotherapy. Surgery. 1995;117 (5):498-504. 39. Henderson JM, Kutner MH, Millikan WJ Jr, et al. Endoscopic variceal sclerosis compared with distal splenorenal shunt to prevent recurrent variceal bleeding in cirrhosis. Ann Intern Med. 1990;112(4): 262-269. 40. Rikkers LF, Burnett DA, Volentine GD, Buchi KN, Cormier RA. Shunt surgery versus endoscopic sclerotherapy for long-term treatment of variceal bleeding: early results of a randomized trial. Ann Surg. 1987;206(3):261-271. 41. Rikkers LF, Jin G, Burnett DA, Buchi KN, Cormier RA. Shunt surgery versus endoscopic sclerotherapy for variceal hemorrhage: late results of a randomized trial. Am J Surg. 1993;165(1):27-33. 42. Terés J, Bordas JM, Bravo D, et al. Sclerotherapy vs. distal splenorenal shunt in the elective treatment of variceal hemorrhage. Hepatology. 1987;7(3):430-436. 43. Spina GP, Santambrogio R, Opocher E, et al. Distal splenorenal shunt versus endoscopic sclerotherapy in the prevention of variceal rebleeding. Ann Surg. 1990;211(2):178-186. 44. Khaitiyar JS, Luthra SK, Prasad N, Ratnakar N, Daruwala DK. Transjugular intrahepatic portosystemic shunt versus distal splenorenal shunt. Hepatogastroenterology. 2000;47(32):492497. 45. Helton WS, Maves R, Wicks K, Johansen K. Transjugular intrahepatic portasystemic shunt vs surgical shunt in good-risk cirrhotic patients. Arch Surg. 2001;136(1):17-20. 46. Rosemurgy AS, Bloomston M, Clark WC, Thometz DP, Zervos EE. H-graft portacaval shunts versus TIPS: ten-year follow-up of a randomized

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

jamasurgery.com

EPCS for Bleeding Esophageal Varices in Cirrhosis

Special Communication Clinical Review & Education

trial with comparison to predicted survivals. Ann Surg. 2005;241(2):238-246.

shunt (TIPS) for variceal bleeding in portal hypertensionDig Dis Sci. 1998;43(11):2463-2469.

47. Rosemurgy AS, Frohman HA, Teta AF, Luberice K, Ross SB. Prosthetic H-graft portacaval shunts vs transjugular intrahepatic portasystemic stent shunts: 18-year follow-up of a randomized trial. J Am Coll Surg. 2012;214(4):445-455.

56. Chau TN, Patch D, Chan YW, Nagral A, Dick R, Burroughs AK. “Salvage” transjugular intrahepatic portosystemic shunts. Gastroenterology. 1998;114 (5):981-987.

48. Henderson JM, Boyer TD, Kutner MH, et al; DIVERT Study Group. Distal splenorenal shunt versus transjugular intrahepatic portal systematic shunt for variceal bleeding. Gastroenterology. 2006;130(6):1643-1651. 49. LaBerge JM. Transjugular intrahepatic portosystemic shunt—role in treating intractable variceal bleeding, ascites, and hepatic hydrothorax. Clin Liver Dis. 2006;10(3):583-598, ix.

65. Clark W, Golkar F, Luberice K, et al. Uncovering the truth about covered stents: is there a difference between covered versus uncovered stents with transjugular intrahepatic portosystemic shunts? Am J Surg. 2011;202(5):561-564. 66. Yang Z, Han G, Wu Q, et al. Patency and clinical outcomes of transjugular intrahepatic portosystemic shunts with polytetrafluoroethylene-covered stents versus bare stents: a meta-analysis. J Gastroenterol Hepatol. 2010;25(11):1718-1725.

57. Azoulay D, Castaing D, Majno P, et al. Salvage transjugular intrahepatic portosystemic shunt for uncontrolled variceal bleeding in patients with decompensated cirrhosis. J Hepatol. 2001;35(5): 590-597.

67. Amarapurkar D, Sanyal A. Improving survival in patients with decompensated cirrhosis. J Gastroenterol Hepatol. 2010;25(11):1716-1717.

58. Tzeng W-S, Wu R-H, Lin C-Y, et al. Prediction of mortality after emergent transjugular intrahepatic portosystemic shunt placement. Korean J Radiol. 2009;10(5):481-489.

68. Gralnek IM, Jensen DM, Kovacs TO, et al. The economic impact of esophageal variceal hemorrhage. Hepatology. 1999;29(1):44-50.

50. Kalva SP, Salazar GM, Walker TG. Transjugular intrahepatic portosystemic shunt for acute variceal hemorrhage. Tech Vasc Interv Radiol. 2009;12(2): 92-101.

59. García-Pagán JC, Caca K, Bureau C, et al; Early TIPS (Transjugular Intrahepatic Portosystemic Shunt) Cooperative Study Group. Early use of TIPS in patients with cirrhosis and variceal bleeding. N Engl J Med. 2010;362(25):2370-2379.

51. McCormick PA, Dick R, Panagou EB, et al. Emergency transjugular intrahepatic portasystemic stent shunting as salvage treatment for uncontrolled variceal bleeding. Br J Surg. 1994;81 (9):1324-1327.

60. Papatheodoridis GV, Goulis J, Leandro G, Patch D, Burroughs AK. Transjugular intrahepatic portosystemic shunt compared with endoscopic treatment for prevention of variceal rebleeding. Hepatology. 1999;30(3):612-622.

52. Jalan R, John TG, Redhead DN, et al. A comparative study of emergency transjugular intrahepatic portosystemic stent-shunt and esophageal transection in the management of uncontrolled variceal hemorrhage. Am J Gastroenterol. 1995;90(11):1932-1937.

61. Luca A, D’Amico G, La Galla R, Midiri M, Morabito A, Pagliaro L. TIPS for prevention of recurrent bleeding in patients with cirrhosis. Radiology. 1999;212(2):411-421.

53. Sanyal AJ, Freedman AM, Luketic VA, et al. Transjugular intrahepatic portosystemic shunts for patients with active variceal hemorrhage unresponsive to sclerotherapy. Gastroenterology. 1996;111(1):138-146. 54. Bañares R, Casado M, Rodríguez-Láiz JM, et al. Urgent transjugular intrahepatic portosystemic shunt for control of acute variceal bleeding. Am J Gastroenterol. 1998;93(1):75-79. 55. Gerbes AL, Gülberg V, Waggershauser T, Holl J, Reiser M. Transjugular intrahepatic portosystemic

69. Villeneuve J-P, Pomier-Layrargues G, Duguay L, et al. Emergency portacaval shunt for variceal hemorrhage. Ann Surg. 1987;206(1):48-52.

71. Spina GP, Santambrogio R, Opocher E, et al. Emergency portosystemic shunt in patients with variceal bleeding. Surg Gynecol Obstet. 1990;171(6): 456-464. 72. Garcia N Jr, Sanyal AJ. Portal hypertension. Clin Liver Dis. 2001;5(2):509-540.

62. Escorsell A, Bañares R, García-Pagán JC, et al. TIPS versus drug therapy in preventing variceal rebleeding in advanced cirrhosis. Hepatology. 2002;35(2):385-392.

73. Klempnaue J, Schrem H. Review: surgical shunts and encephalopathy. Metab Brain Dis. 2001; 16(1-2):21-25.

63. Afdhal NH, Curry MP. Early TIPS to improve survival in acute variceal bleeding. N Engl J Med. 2010;362(25):2421-2422.

74. Garcia N Jr, Sanyal AJ. Portal hypertension. Clin Liver Dis. 2001;5(2):509-540.

64. Clark W, Hernandez J, McKeon B, et al. Surgical shunting versus transjugular intrahepatic portasystemic shunting for bleeding varices resulting from portal hypertension and cirrhosis. Am Surg. 2010;76(8):857-864.

jamasurgery.com

70. Soutter DI, Langer B, Taylor BR, Greig P. Emergency portasystemic shunting in cirrhotics with bleeding varices—a comparison of portacaval and mesocaval shunts. HPB Surg. 1989;1(2):107-118.

75. Klempnaue J, Schrem H. Review: surgical shunts and encephalopathy. Metab Brain Dis. 2001; 16(1-2):21-25.

JAMA Surgery February 2014 Volume 149, Number 2

Copyright 2014 American Medical Association. All rights reserved.

Downloaded From: http://archsurg.jamanetwork.com/ by a Tulane University User on 05/18/2015

169

Fifty-three years' experience with randomized clinical trials of emergency portacaval shunt for bleeding esophageal varices in Cirrhosis: 1958-2011.

Emergency treatment of bleeding esophageal varices (BEV) consists mainly of endoscopic and pharmacologic measures, with transjugular intrahepatic port...
323KB Sizes 0 Downloads 0 Views