| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Department of Radiology, Dartmouth-Hitchcock Medical Center, One Medical Center Dr., Lebanon, NH 03756.
Received February 11, 2002;
accepted after revision May 31, 2002.
Presented in part at the annual meeting of the Society of Cardiovascular
and Interventional Radiologists, San Antonio, TX, March 2001.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We followed up 103 patients who underwent TIPS for a mean of 20.10 ± 25.58 months (range, 1 day-92 months). Various statistical methods were used to analyze long-term survival, shunt patency, and correlation with Child-Pugh classification and indications for TIPS. The Rand 36-Item Health Survey 1.0 was used to assess quality of life over time.
RESULTS. The cumulative survival rate was 68%, 60%, 50%, 41%, and 41% at years 1-5, respectively. The cumulative survival rate was significantly higher for patients classified as Child-Pugh class A or B versus those classified as class C (p < 0.01), as well as for patients with the original indication of variceal bleeding versus refractory ascites or hydrothorax (p < 0.01). No significant difference in survival rates was found between patients with Child-Pugh A and those with Child-Pugh B. The cumulative primary patency rate was 50%, 34%, 21%, 13%, and 13% at years 1-5, respectively, with assisted patency rate of 80%, 61%, 46%, 42%, and 36%. Cumulative secondary patency rate was 85%, 64%, 55%, 55%, and 55% at years 1-5. Mean follow-up time in 33 patients who completed quality-of-life questionnaires with one follow-up was 17.46 months. Scores after TIPS in all nine of the health categories were higher than those preprocedure with statistically significant improvement in four categories. The second follow-up was completed by 21 patients at a mean of 30.58 months after TIPS creation. Both scores after TIPS were higher than those before TIPS, and scores in five categories were further improved at the second surveillance (p < 0.0, only for health change), whereas in four categories, the scores were slightly worse.
CONCLUSION. TIPS has positive efficacy both for controlling bleeding or ascites and for improving the quality of life. The precise impact of TIPS on long-term survival, however, requires further clarification.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In addition to the four patients in whom attempts at TIPS failed, several others were excluded from consideration for the TIPS procedure on clinical grounds because of severe right heart failure, portal venous system occlusion seen on CT, and protracted encephalopathy or hepatic failure.
Before the TIPS procedure, esophagogas-troduodenoscopy was performed more than once in 52 patients, with variceal banding performed at least once in all of these. Endoscopically confirmed recurrent variceal hemorrhage was the primary indication for TIPS in 73 patients (70.87%). In 27 (36.40%) of these patients, the procedure was performed on an emergency basis for active bleeding. Refractory ascites was the indication for TIPS in 27 patients (26.21%) and refractory hydrothorax, in the remaining three. The mean preprocedural laboratory values and standard deviations were the following: hematocrit, 29.79% ± 8.34%; prothrombin time, 16.22 ± 2.96 sec; platelet count, 118,800 ± 66,836/mL; blood urea nitrogen level, 29.10 ± 22.35 mg/dL; serum creatinine level, 1.29 ± 1.41 mg/dL; serum ammonia level, 90.94 ± 58.68 mg/dL; serum albumin level, 3.05 ± 1.70 g/dL; total serum bilirubin level, 2.63 ± 2.18 µg/dL; gamma-glutamyltransferase level, 142.58 ± 109.65 U/L; aspartate aminotransferase level, 78.08 ± 74.70 U/L; and alkaline phosphatase level, 157.26 ± 100.14 U/L.
TIPS Procedure
The Ring TIPS set (Cook, Bloomington, IN) was used in all patients. In the
first 15 patients, a 5-French catheter was introduced into the right portal
vein by sonographically guided percutaneous puncture of the left portal vein
to facilitate right hepatic to right portal vein puncture. In the remaining
patients, the standard TIPS technique was used
[4]. Wedge hepatic venography
was performed in 91 patients immediately before the TIPS creation. Carbon
dioxide (Northeast Airgas, Salem, NH) was used as the contrast agent in 78
patients and Conray 43% (Mallinckrodt, St. Louis, MO), in 13. The TIPS were
created with Wallstents (Schneider, Minneapolis, MN) in all patients in this
study. The length and number of stents were determined empirically, as needed
to provide optimal pressure gradient and flow. Stents were dilated with
standard balloons sufficiently to reduce the portosystemic gradient to 5-12 mm
Hg. If varices were still present after TIPS creation, they were embolized
using Gianturco coils (Cook).
Quality of Life
Quality of life was assessed with the Rand 36-Item Health Survey 1.0
(Rand-36) developed for the study of medical outcomes by the Rand Corporation
[9]. It consists of 36 items,
including a single item indicating health change and eight scales related to
general health status; physical functioning (10 items), mental health (five
items), energy—vitality (four items), and general health perceptions
(five items); three related to general quality of life (limitations due to
physical problem [four items], social functioning [two items], and limitations
due to emotional problems [three items]); and one related to both health
status and quality of life (pain, two items). The Rand-36 includes the same
items as those in the 36-Item Short-Form Health Survey (SF-36), but scoring
differences exist for the pain and general scales
[9].
The scoring method is a 2-step process. First, each item is scored on a 0-100 range so that the lowest and highest possible scores are set at 0 and 100, respectively. Scores represent the percentage of total possible score achieved. In step 2, items in the same scale are averaged to create the eight scale scores. A high score defines a more favorable health state.
The survey of quality of life was performed in March 1996, September 1997, December 1998, and September 1999. The first forms were sent in March 1996 with follow-up questionnaires sent in September 1997 for new patients and second follow-ups. In December 1998, second follow-up questionnaires were sent to the first group and first follow-up questionnaires were sent to new patients. In September 1998, second follow-up questionnaires were sent only to patients who were sent first follow-up questionnaires in December 1998. We attempted to interview all patients who did not respond to the mailed survey by telephone 4 weeks after the mailing, but several patients were lost to follow-up.
Analysis Methods and Statistics
A Kaplan-Meier product-limit measure (SPSS, Chicago, IL) was used for
evaluating the following three factors regarding shunt patency: primary
patency (the duration of uninterrupted patency without re-intervention, which
ends at the first re-intervention), primary assisted patency (the duration of
continuous patency of the shunt with or without intervention until complete
occlusion), and secondary patency (the duration of patency with or without
prior occlusion until the next complete occlusion or complete loss of shunt
function). Doppler sonography confirmed by angiography or angiography alone
was used to determine shunt function. Doppler sonography was routinely
performed 1 month after TIPS placement or revision and subsequently at 3-month
intervals. The shunt was considered stenotic and was evaluated
angiographically if the maximum velocity that could be attained anywhere in
the shunt was less than 60 cm/sec. It was considered widely patent if flow
revealed by Doppler sonography was stable at 100-200 cm/sec. If velocity was
60-100 cm/sec, a decision was made on the basis of clinical status and change
from prior values. If patients presented with recurrent signs or symptoms,
angiographic evaluation was performed directly.
Correlation between survival time of the patients and the Child-Pugh classification or major clinical symptoms of gastrointestinal bleeding, ascites, or hydrothorax was analyzed using Kaplan-Meier product-limit analysis.
Quality-of-life scores before TIPS and after TIPS were compared using the Student's paired t test in the series with one survey. For the results in patients with two follow-up surveys after TIPS, one-way analysis of variance was used. Differences were considered significant if p was less than or equal to 0.05.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The initial mean (±SD) portosystemic pressure gradient was 24.60 ± 10.44 mm Hg (range, 11-66 mm Hg). After stent placement, the mean portosystemic gradient was reduced to 8.66 ± 4.15 mm Hg (range, 1-27 mm Hg). In only six patients did the gradient remain greater than 12 mm Hg. In all six, the shunt had been dilated to 12 mm, and no residual variceal filling was seen.
In the 103 patients who underwent successful TIPS, 26 of 27 patients with active bleeding during the procedure had immediate cessation of bleeding. Bleeding was controlled after 24 hr in the remaining patient. All patients were free of bleeding when they were discharged from the hospital, on average 4.3 days after TIPS. Rebleeding occurred 34 times in 30 (32.26%) of the 92 patients with a history of hemorrhage. The first rebleeding occurred 7.06 ± 8.98 months (range, 2 days—36 months) later. At the time of initial discharge, ascites or hydrothorax had decreased in 32 (58.18%) of 55 patients and increased in 10 patients (18.18%), with no obvious change in 13 patients (23.64%). Sixteen (53.33%) of 30 patients who had encephalopathy before TIPS showed subsequent improvement. The remaining patients did not change or deteriorate.
Twenty-three patients (22.33%) died during the first 30 days, and four patients, within the subsequent 2 months (3-month mortality rate, 26.21%). The primary causes of death in these patients were hepatorenal failure (n = 15), adult respiratory distress syndrome (n = 7), pneumonia (n = 2), HIV (n = 1), sepsis with hepatoma and bilateral pulmonary metastases (n = 1), and massive intraabdominal hemorrhage due to inadvertent injury to the gastroduodenal artery during the procedure (n = 1). Nineteen of these 27 patients were classified as Child-Pugh C, and the remaining eight were classified as Child-Pugh B. Only in the case of vascular injury was death attributed primarily to the procedure. In none of the others was the procedure thought to have directly caused death. Rather, TIPS was thought to have failed to halt a progressive downhill course.
Major potentially life-threatening procedure-related complications occurred in five patients (4.85%). These events included two fatalities (rupture of gastroduodenal artery and bleeding related to the positioning of a parallel TIPS stent across the major fissure of the liver during revision 20 months after the initial TIPS procedure). One patient had right atrial damage and recovered after pericardiotomy. Pulmonary edema with stent migration into the right atrium developed in one patient with a history of biventricular heart failure. In this case, two stents were extracted using a snare from a right femoral vein approach on day 2 with resolution of the pulmonary edema. The final patient showed worsening pulmonary edema, chest pain, and dyspnea. These conditions were thought to be due to right heart failure after enlargement of the shunt from 8 to 10 mm. The TIPS was selectively occluded on day 7 after the procedure by inflating an occlusion balloon catheter in the proximal shunt for 24 hr. The patient died of hepatorenal failure 2 months after initial TIPS. Additionally, encephalopathy as a complication of TIPS occurred in 18 patients who had no encephalopathy before TIPS. These 18 patients were treated with the usual medical regimen (dietary restriction and lactulose) with partial or complete return to baseline status in all.
Follow-Up
Average follow-up time was 20.10 ± 25.58 months (range, 1 day-92
months). Forty-six patients (44.66%) had died at the latest follow-up. The
average survival time among these patients was 12.64 ± 19.42 months
after the procedure. Eight patients were lost to follow-up immediately after
discharge. Two patients underwent liver transplantation, at 12 and 20 months
after the original TIPS. One patient underwent splenorenal shunt after two
revisions of TIPS and was alive at last follow-up 26 months after TIPS.
Another patient underwent a mesocaval shunt after five revisions of TIPS but
died 2 weeks later. No patients in the follow-up group had new or refractory
encephalopathy.
The Kaplan-Meier survival analysis for patients in this study is shown in Figure 1. The median survival time overall was 35.33 months (95% confidence interval [CI], 18.24-52.42 months). Figure 2 shows the overall survival time for the three Child-Pugh classes. Survival time was improved for patients with Child-Pugh class A or B versus patients with class C (p < 0.01, log-rank test). We found no significant difference between patients with Child-Pugh class A and Child-Pugh class B (p = 0.29, SPSS log-rank test). The survival time of the patients who presented with hemorrhage versus those with ascites was significantly improved for hemorrhage (p < 0.01, log-rank test) (Fig. 3).
|
|
|
In the 95 patients who were available for follow-up, acute shunt thrombosis occurred 15 times in 13 patients (13.67%) and restenosis with or without subacute occlusion, 93 times in 41 patients (43.16%). The distribution of the stenoses or occlusions was the following: hepatic vein and hepatic vein end of stent in 49, mid stent in 13, and diffuse in-stent stenosis or occlusion in 31. Proximal abnormalities occurred at a mean of 16.37 ± 18.29 months (median, 9.2 months), significantly later than mid stent abnormalities (6.50 ± 9.92 months; median, 2.8 months) or diffuse stenoses (7.53 ± 7.7 5 months; median, 5.6 months) (p < 0.05).
Excluding two patients who underwent liver transplantation, 106 re-interventions were performed in these 52 patients to maintain patency: once in 24 patients, twice in 14 patients, three times in eight patients, four times in two patients, five times in three patients, and seven times in one patient. Treatments were the following: single dilatation of the shunt with a balloon catheter in 64, dilatation and placement of additional stents in 26, thrombolysis with urokinase (Abbokinase; Abbott Laboratories, North Chicago, IL) or recombinant tissue plasminogen activator (Genentech, South San Francisco, CA) followed by mechanical percutaneous thrombectomy with a catheter in nine, thrombolysis or thrombectomy and re-stenting in five, parallel TIPS creation in one, and recanalization of the shunt with a Colapinto needle (Cook) because of fibrotic occlusion in one.
The initial indications for re-intervention were rebleeding in 26 (24.53%), increase in ascites in 10 (9.43%), decreased flow in the shunt by routine Doppler sonography in 65 (61.32%), and routine angiographic follow-up in five (4.72%). The average interval between re-interventions was 9.88 ± 13.48 months (range, 1 day-61 months).
The cumulative patency by Kaplan-Meier analysis is shown in Figure 4. The mean duration of patency was 20.18 months (95% CI, 14.64-25.71 months) for primary patency, 41.60 months (95% CI, 32.14-51.07 months) for assisted primary patency, and 55.95 months (95% CI, 44.64-67.25 months) for secondary patency.
|
Quality of Life
Completed quality-of-life questionnaires were obtained in 33 of 39 eligible
patients. Of the remaining six, answers in four patients were excluded because
of inadequate recall of their status before TIPS. The two other patients were
excluded because of liver transplantation after TIPS.
Mean follow-up time in the 33 patients with one follow-up was 17.46 ± 11.54 months (range, 6-48 months). The second follow-up was completed in 21 of 23 patients who were eligible, with a mean elapsed time of 15.43 ± 4.96 months from the first surveillance.
The quality-of-life scores in 33 patients (one follow-up in 12 patients and the first follow-up in the 21 patients who underwent two follow-ups) are shown in Table 1. The scores in all of the nine categories of health status after TIPS are higher than those before TIPS. There was statistically significant improvement in the status of limitations due to emotional problems, energy and fatigue, emotional well-being, and health change. There were no statistically significant differences in general health, physical function, limitation due to physical health, social functioning, and pain. In the 21 patients with two measures of surveillance (Table 2), both summary scores after TIPS were higher than those before TIPS. In five categories, further improvement was seen at the second surveillance, whereas four categories showed a slight nonsignificant deterioration in scores. In only one category, health change, was there significant improvement over pre-TIPS and the first post-TIPS surveillance (p < 0.05). Improvement occurred but was not significant for general health, physical function, limitation due to physical health, and pain.
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In our study, the procedural mortality rate was only 0.97%. The cumulative survival rates at 1, 2, 3, 4, and 5 years in the patients with variceal hemorrhage were 79%, 70%, 58%, 52%, and 52%, respectively, rates similar to those previously reported for 1- to 4-year survival rates [1, 7]. These rates are comparable to those of patients who have undergone sclerotherapy or surgery with, clearly, a much lower procedural mortality rate than that with surgery [11]. Direct comparison may not be valid, however, because in surgical series, exact survival rate is usually used [11], whereas the Kaplan-Meier cumulative survival rate is commonly used in TIPS series. Also, more seriously ill patients were likely excluded from consideration for surgery. Patients classified as Child-Pugh class B in our study, as indirect support of this hypothesis, had a 51% survival rate at 5 years, significantly better than that of patients classified as class C (29%, p < 0.01). Also, many patients were treated with TIPS after failed sclerotherapy and while acutely or subacutely bleeding. Interestingly, two randomized controlled trials comparing survival rates for TIPS versus those for endoscopic sclerotherapy had contradictory results: Cello et al. [4] showed a trend toward improved long-term survival in the TIPS group, whereas Sanyal et al. [10] showed the opposite. Neither series, however, reached statistical significance. Long-term survival, then, is a complex issue. As prospective studies of sclerotherapy versus medical therapy [11] and surgery [14] have suggested, long-term survival may be more a function of the disease than of the treatment. The long-term efficacy may not be resolved in the absence of large-scale prospective controlled trials.
Our study supports the finding of Fillmore et al. [15] that survival in patients who underwent TIPS for ascites is significantly worse than that for those with variceal hemorrhage (p < 0.01); these findings may be a reflection of the Child-Pugh class in these two groups. The cumulative survival at 1-5 years in patients with refractory ascites in our series was 51%, 45%, 37%, 19%, and 19%, respectively, which is similar to the results reported previously with TIPS, large-volume paracentesis [16], and side-to-side portacaval shunts [17]. This surgical approach, however, which would be thought to have physiologic effects similar to those of TIPS, has been essentially replaced by peritoneovenous shunts, which are more effective than medical management [11]. A single randomized controlled trial that compared TIPS with peritoneovenous shunt for the treatment of intractable ascites showed comparable results for the control of ascites and similar overall mortality rates (Zervos EE et al., presented at the American Gastroenterological Association meeting, May 1996). In a larger prospective randomized trial by Rossle et al. [16], TIPS was compared with repeated large-volume paracentesis in the treatment of refractory or recurrent ascites. TIPS improved the rate of survival without raising the incidence of hepatic encephalopathy. The superior efficacy of TIPS in patients with ascites may be due to a decrease in the risk of sepsis [16] and improvement of the nutritional states of patients with cirrhosis [18].
Long-Term Patency
Probably the major concern with TIPS is stenosis or occlusion of the shunt
[19], reported to require
revision or a new shunt due to recurrent bleeding in up to 26% of patients at
1 year and 32% at 2 years after TIPS
[1]. Stenosis or occlusion has
been shown in animals and in humans to be due to myofibroblastic cell
proliferation [20], similar to
the process seen with all vascular stents. This problem may at least in part
be addressed by the use of covered stents
[21,
22].
In retrospective studies, cumulative primary patencies (unassisted) at 1 and 2 years have been reported as 5-23% [13] and 32-50% [23]. In a prospective trial, the patencies at 1 year and 2 years were, respectively, 66% and 42% [1]. The results from our study are similar regarding primary patency, but we encountered a 64% incidence of stenosis within the first year. At 5 years, with surveillance and redilatation when indicated, overall assisted primary patency in our series was 36%, and secondary patency (re-opening after complete occlusion) was 55%. On the one hand, these rates support the long-term efficacy of TIPS, with monitoring and re-intervention. On the other hand, however, the high stenosis rate highlights the need for new tools and techniques.
Quality of Life
Quality of life has now become firmly established as an important end point
in medical care [24]. The aim
of evaluation of any therapy with a quality-of-life tool is improved
understanding of patients with chronic or incurable diseases. The efficacy of
a therapy must be assessed not only in terms of prolongation of life but also
in terms of patient function and sense of well-being. Orozco et al.
[25] assessed quality of life
after surgical shunt creation for portal hypertension but used only a 3-grade
scale (good, regular, and bad). To our knowledge, there are, to date, no
published studies on quality of life after endoscopic therapy. Two other
studies have evaluated quality of life in patients with TIPS, both using the
Karnofsky performance status scale
[26,
27]. This is a rating scale
scored by an observer or interviewer, consisting of three definitions and 10
criteria, originally designed as an outcome measure in patients with cancer.
The ratings are narrowly based on physical disabilities and need for care. The
disadvantages of this scale are lack of a standardized observational procedure
and its single-item format that limits usefulness in detecting
small-to-moderate differences between groups and even large differences
between individual patients. It is thus helpful when dealing with illnesses
such as liver failure and treatments such as TIPS to use a multiitem scale,
including a subset of items shown to best reproduce a full-length measurement
scale of proven validity
[28].
Rand-36 is a self-administered questionnaire used to measure general health status and quality of life. It has applications both as a measure of outcome and in assessment of the impact of a disease treatment on a specific population. The multiitem scale of Rand-36 is designed to elucidate two well-accepted comprehensive standards: representation of multidimensional health concepts and measurement of the full range of health status, including levels of well-being and personal evaluation of health. Most of the items of the questionnaire are adapted from instruments that have been well validated and used for 20-40 years in large numbers of patients [28]. It is easily completed within 5-10 min. Its reliability and validity have been verified by worldwide application in such diverse situations as cardiovascular disease [29]; dialysis [30]; hypertension, diabetes, heart disease, and depressive symptoms [31]; lung transplantation [32]; cancer [33]; neuropathy [34]; and migraine. This instrument has been used both prospectively and retrospectively and is readily available on the Internet.
To date, the Rand-36 has not been used for the evaluation of treatment of portal hypertension. The results in our study show that all nine concepts related to quality of life had higher scores at a mean time of 17.46 months after TIPS than before the procedure. Among the concepts, four (limitation due to emotional problems, energy and fatigue, emotional well-being, and health change) showed a statistically significant improvement compared with those before TIPS. The other concepts related to quality of life (general health, physical function, limitation due to physical health, social functioning, and pain) all showed improved scores after TIPS, but the changes were not statistically significant. Although scores after TIPS were in general higher than before TIPS with both one and two follow-up surveys, in the group surveyed twice after TIPS, there was a nonsignificant decrease in four categories at the time of the second evaluation at a mean of 15.43 months after the first evaluation. These categories related primarily to emotional rather than physical parameters—limitation due to emotional problems, social functioning, energy and fatigue, and emotional well-being. This outcome is perhaps not surprising because TIPS would not be expected to affect the progression of chronic liver disease that causes portal hypertension.
For validation of the use of the Rand-36, baseline scores from its application in patients with portal hypertension must be close to the baseline scores of 2471 patients [9], as in our cohort of patients. The reliability and validity of the questionnaire for use in assessing quality of life in patients with TIPS, then, are comparable with the results in other disease states. As we already noted, this comparison suggests that this instrument is more useful than others used previously.
Potential limitations to the quality-of-life assessment exist in our study. First, the survey was retrospective, so patients' recall of their status before TIPS may not have been accurate. The similarity of baseline scores to those in the Medical Outcome Study [9], however, suggests that this application of the Rand-36 is valid. Second, direct comparison with patients undergoing endoscopic sclerotherapy or surgery would be useful. This comparison probably should be performed either prospectively or using a case-matching approach.
Overall our study suggests that in comparison with other treatment options, TIPS offers comparable long-term survival and probably decreased procedural morbidity rates. Perhaps more importantly, TIPS has a positive efficacy for improving the quality of life, a factor that remains to be compared to that in other treatment options in a large randomized controlled studies.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Rossi, F. M. Salvatori, F. Fanelli, M. Bezzi, M. Rossi, G. Marcelli, D. Pepino, O. Riggio, and R. Passariello Polytetrafluoroethylene-covered Nitinol Stent-Graft for Transjugular Intrahepatic Portosystemic Shunt Creation: 3-year Experience Radiology, June 1, 2004; 231(3): 820 - 830. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. C. J. ter Borg, M. Hollemans, H. R. van Buuren, F. P. Vleggaar, M. Groeneweg, W. C. J. Hop, and J. S. Lameris Transjugular Intrahepatic Portosystemic Shunts: Long-term Patency and Clinical Results in a Patient Cohort Observed for 3-9 Years Radiology, May 1, 2004; 231(2): 537 - 545. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
