Patient Selection and the Effect of Transjugular Intrahepatic Portosystemic Shunt on Liver and Kidney Function

Chapter 11: Patient Selection and the Effect of Transjugular Intrahepatic Portosystemic Shunt on Liver and Kidney Function

Hector Ferral and George Behrens

Introduction

The management of portal hypertension (PHT) and its complications has changed drastically in the past 20 years.1,2 Changes in the recommendations for the management PHT are a result of expert consensus conferences that have taken place between 1986 and 2010.^1,3 The role of the transjugular intrahepatic porto-systemic shunt (TIPS) procedure in the management of patients with complications of PHT has changed as a result of these expert panel discussions.^4,5 The most common indications to perform a TIPS procedure have been control of esophageal variceal bleeding and management of refractory ascites.^4,6 The procedure is performed either in an emergency situation (active variceal bleeding) or in an elective fashion (recurrent bleed after failed medical and endoscopic therapy or for the management of ascites).

The purposes of this chapter are to convey an updated perspective of the role of TIPS in the management of patients with complicated PHT, emphasize the importance of careful patient selection before the procedure, and describe the hemodynamic effects of the shunt and its impact on liver and renal function.

The Role of Transjugular Intrahepatic Portosystemic Shunt in Patients with Variceal Bleeding

The first expert consensus conference was held in Baveno, Italy, in 1995.¹ The first Baveno conference in 1995 established that the first line of treatment for acute variceal hemorrhage should be based on endoscopic techniques; TIPS was recommended only in case of failure of endoscopic and pharmacologic therapy.¹ The concept of clinically significant PHT (hepatic venous pressure gradient [HVPG] >10 mm Hg) was introduced in the conference held in April 2000. There were no changes in recommendations regarding the role of TIPS in the management of variceal bleeding.¹ Endoscopic band ligation emerged as an alternative therapy for the management of bleeding esophageal varices during the conference held in 2005. Early TIPS was presented as an option in patients at high risk of rebleeding for the first time during the conference held in 2007,¹ and the recommendations for early TIPS in acute variceal bleeding were reinforced in the conference held in 2010.¹

Current recommendations in the management of hemorrhagic PHT depend on the clinical stage of the disease. The clinical stages include: (i) patients with PHT who have not developed esophageal varices; (ii) patients with PHT with esophageal varices who have never bled; (iii) patients with acute variceal hemorrhage; and (iv) patients who have survived a bleeding episode and require treatment to prevent rebleeding.

Patients with Portal Hypertension Who Have Not Developed Esophageal Varices

These patients require no therapy, and management is focused on treating the cause of cirrhosis.

Patients with Portal Hypertension with Esophageal Varices Who Have Never Bled

In this stage, patients are further classified into high risk and low risk. High-risk patients are those with advanced cirrhosis (Child-Pugh class C) and large varices. Low-risk patients are those who have less severe cirrhosis (Child-Pugh class A or B) and small varices. Treatment options for patients at high risk of bleeding include nonselective beta-blockers and endoscopic variceal ligation (EVL). Studies have shown that EVL and nonselective beta-blockers are equally effective in preventing first variceal bleed; therefore, management decisions depend on local expertise and resources.³ The downside of beta-blockers is that they are associated with side effects, and their use has to be discontinued in 15% to 20% of patients because of poor tolerance to the drug.³ The problem with EVL is that it requires technical expertise, and it may be associated with postprocedural complications, the worst of which is the development of ulcers that may be associated with severe bleeding.³ The focus in this stage is prophylaxis and the risks of beta-blocker side effects or EVL procedural complications need to be weighed against their potential clinical benefits.³ In patients with advanced liver failure and small, low-risk varices, the recommendation is to treat them with nonselective beta-blockers instead of using endoscopic options.³ Finally, in low-risk patients, the use of beta-blockers is optional. According to current recommendations, TIPS has no role in this prophylactic stage.

Patients with Acute Variceal Hemorrhage

This stage is a medical emergency. Patients are treated in intensive care units (ICUs) and require aggressive medical management, including airway control, transfusion of blood products, prophylactic antibiotics, and vasoactive drugs. A diagnostic endoscopy is mandatory and should be performed within 12 hours of admission.¹ EVL is the treatment of choice if varices are confirmed as the source of bleeding.¹ Sclerotherapy may be used if EVL proves to be technically difficult. TIPS procedure is indicated in patients who fail endoscopic and vasoactive treatment.¹ Recent reports have determined that approximately 10% to 20% of patients with acute variceal bleeding fail standard therapy,^3,7 and those are the patients who need to be evaluated for an emergency TIPS procedure. Clinical studies have reported the effectiveness of emergency TIPS procedure to treat acute esophageal and gastric variceal bleeding in patients who have failed medical and endoscopic therapy.⁸

Patients with a Child-Pugh class C, HPVG greater than 20 mm Hg and bleeding varices during diagnostic endoscopy are considered to be at high risk of failing standard therapy,¹ and this is the subgroup of patients that may benefit from an early TIPS procedure.^1,3,9 A recent prospective, randomized, multicenter trial conducted by Garcia-Pagan and coworkers⁹ demonstrated that the application of early TIPS (within 72 hours of admission) resulted in a significant improvement in patient survival and a significant decrease in rebleeding rates in a group of patients with cirrhosis at high risk of conventional treatment failure. Careful patient evaluation before the procedure is crucial in selecting patients who may benefit from early TIPS. The clinical history, laboratory data, and diagnostic imaging studies must be reviewed.6,9 The role of the interventional radiologist as a true clinical consultant is to determine if a patient is a suitable candidate to undergo a TIPS procedure after the patient’s evaluation has been completed. The operator must also be able and willing to recommend alternative forms of endovascular therapy (i.e., balloon-occluded retrograde transvenous obliteration [BRTO] or direct embolization of varices) for patients who are not suitable candidates to undergo a TIPS procedure.^10,11

It is important to point out that in this recent study by Garcia-Pagan, early TIPS was not performed in patients who had a Child-Pugh score greater than 13 points, who had isolated bleeding gastric varices, who were older than 75 years of age, who had hepatocellular carcinoma, who had portal vein thrombosis, and who had renal failure. Furthermore, a total of 22 patients in this study were excluded for nonspecified reasons. Therapeutic options offered to the patients who were excluded were not specified.⁹

Clinical studies have shown that patients who are Child-Pugh class C and have an APACHE II (Acute Physiology and Chronic Health Evaluation II) score greater than 18 have a very poor prognosis if they undergo an emergency TIPS procedure to control variceal bleed. The 30-day mortality rate for such patients is 98% to 100%.^12,13 Multiple variables have been associated with a poor prognosis in patients undergoing an emergency TIPS, including delayed admission to the ICU; presence of ascites; emergent requirement for mechanical ventilation; elevation of serum creatinine; elevated international normalized ratio (INR)¹⁴; and elevated serum bilirubin, especially if it reaches levels higher than 6 mg/dL.¹²

Chalasani and coworkers¹⁵ studied 129 patients who underwent a TIPS procedure in a single center in the United States to evaluate the variables associated with early death. Only 16% of the patients in this series underwent an emergency TIPS; however, these authors identified four factors associated with 30-day mortality after TIPS: active variceal bleeding, emergency TIPS, prolonged prothrombin time (>17 sec), and bilirubin level greater than 3 mg/dL.¹⁵

Patch and coworkers¹⁶ developed the prognostic index, a model specifically designed to predict mortality for patients undergoing an emergency TIPS. This predictive model was developed after the evaluation of 54 patients undergoing emergency TIPS for variceal bleeding.¹⁶ In this study, five factors were associated with a poor prognosis: ascites (moderate or severe), need for emergent mechanical ventilation, white blood cell count, serum creatinine, and activated partial thromboplastin time (aPTT). Based on statistical methods, these authors developed a formula to calculate the prognostic score:

PI = 1.54 (ascites) + 1.27 (ventilation) + 1.38 log e (white blood cell count) + 2.48 log e (aPTT) + 1.55 log e (creatinine) – 1.05 log e (platelet count).

The patients with a higher score had higher mortality rates. Overall, the 6-week mortality rate in this series was 48%.¹⁶ Using this model, these investigators found that the 6-week mortality rate was 100% in patients with a prognostic index greater than 18.52.¹⁶ The model was prospectively tested in an additional group of 31 patients undergoing emergency TIPS. From this group, there were 11 deaths (35%) within the 6-week period. All patients who had an early death had a prognostic index greater than 17.1.^6,16 In essence, these authors demonstrated that the 6-week mortality rate was 100% for any patient with a prognostic score greater than 18.52.

Scores to predict early mortality have been applied and tested clinically in patients with acute variceal bleeding; however, most of these scores have not been validated with further studies.⁶ The scores are useful guides and allow the consulting interventional radiologist to give referring physicians, patients, and family members a realistic opinion on what to expect after the procedure if it is at all performed.⁶ In our practice, we apply the APACHE II score to patients who are being considered for an emergency TIPS. If the APACHE II score is greater than 18 or the patient has more than one of the previously mentioned high-risk factors, we recommend against the TIPS and offer alternative options, including BRTO or even direct variceal embolization.

Patients Who Have Survived a Bleeding Episode and Require Treatment to Prevent Rebleeding

The risk of rebleeding in a patient who has survived an episode of acute variceal bleed is high, ranging between 50% to 60%.^1,17 Two groups of patients are identified within this clinical category: (i) patients who underwent a TIPS procedure and (ii) patients who responded to medical and endoscopic therapy. Patients who underwent a TIPS procedure require surveillance of the shunt but do not require drugs to prevent rebleeding.^1,18 Patients who responded to medical therapy require continued prophylactic therapy with nonselective beta-blockers.¹ In this latter group, either TIPS or surgical shunts are indicated for those who rebleed despite medical and endoscopic treatment.^1,3 TIPS procedures performed in this setting are considered to be either emergent or even elective; thus, these patients in general have a better prognosis. Evaluation of the clinical history, laboratory examinations, and imaging studies is still very important.

At least five prognostic scores have been described to try to determine the survival prognosis of patients undergoing a TIPS procedure.^6,19 The scores that have been more commonly used in clinical practice for this purpose include the Child-Pugh score and the model for end-stage liver disease (MELD) score.⁶

The Child-Pugh score ( Table 11.1) has been used to predict outcomes in patients undergoing portal hypertensive surgery since 1973 and was adopted to predict outcomes in patients undergoing TIPS since 1989.²⁰ The application of the Child-Pugh score divides the patients in three groups: A (score, 5–6), B (score, 7–9), and C (score, 10–15). The higher the score, the worse the liver function.²¹ Patients with high scores (10–15) are considered to be poor operative risks.²² The Child-Pugh score has some disadvantages; the calculation includes two subjective variables: ascites and degree of encephalopathy.²³ In addition, it is subject to the so-called ceiling effect.⁶ This is considered to be a major disadvantage of the Child-Pugh score.²⁴ Despite these theoretical disadvantages, the Child-Pugh scoring system has withstood the test of time and is still very effective in predicting outcomes in patients with end-stage liver disease undergoing TIPS procedures^25–28; it has been compared with other recently described predictive scoring systems and has performed very well in patients undergoing both emergency²⁵ and elective TIPS.²⁶ In general, most authors coincide in the opinion that a Child-Pugh class C patient with a score of 12 or higher is at a very high risk of having an early death after a TIPS procedure.^13,29,30