Controversies in anesthetic management of liver transplantation

Abstract

Controversies in the anesthetic management of liver transplantation (OLT) include the exclusion of patients with pulmonary hypertension, intracranial pressure monitoring in patients with fulminant hepatic failure, preoperative screening for coronary artery disease, perioperative use of anticoagulant medications, and early extubation after surgery. This manuscript discusses these controversies from the perspective of two institutions that possess a long tradition of providing liver transplantations for patients with end-stage liver disease.

Introduction

Although results of orthotopic liver transplantation (OLT) have steadily improved, many questions regarding best practices persist. A major objective of the Liver Anesthesiology and Critical Care Symposium of the Sixth World Congress of the International Hepato-Pancreato-Biliary Association (June, 2004) is to explore data and opinions regarding such questions. With that in mind, the authors will moderate a discussion at the Symposium with the intention of highlighting and illuminating some controversial areas in the realm of anesthetic management of OLT. Likely topics of discussion include anticoagulation, pulmonary hypertension, preoperative cardiac evaluation, early extubation and fulminant hepatic failure. In this paper, we intend to explore these topics and offer some data and opinions regarding practices at our institutions. Our discussion is primarily focused on perioperative management of adult OLT recipients.

Anticoagulation

Bleeding and thrombosis are two important complications of OLT. Acute thrombotic complications following OLT (e.g. hepatic artery thrombosis 9%, portal vein thrombosis 3%–13%) are associated with increased morbidity and mortality ¹,². Given the tendency toward bleeding, anticoagulation is not given routinely to these patients. While a hypercoagulable state is occasionally found following reperfusion of the liver (as demonstrated by thromboelastograph (TEG) or other point-of-care test of whole-blood coagulation) the role of prophylactic anticoagulation is not defined by current literature.

At Georgetown Transplant Institute (GTI) we rarely anticoagulate OLT recipients during the perioperative period. Generally, heparin is only given after an episode of thrombosis. Heparin is our preferred anticoagulant because of reversibility, short duration and ease of titration. Anticoagulation would only be recommended in the absence of significant bleeding and in the absence of laboratory evidence of coagulopathy. At Mayo Clinic low molecular weight heparin is sometimes given after surgery.

In most OLT recipients, we avoid giving any pro-coagulants (except FFP) prior to unclamping the hepatic artery. Even after unclamping, systemic pro-coagulants are often withheld unless the diagnosis of coagulopathy is supported by both laboratory evidence (usually TEG or platelet count) and clinical evidence (e.g. diffuse oozing and lack of clot in the field). The choice of pro-coagulants (usually FFP, platelet concentrate, aminocaproic acid or protamine) is guided by preoperative platelet count, estimated blood loss, and intraoperative TEG. Other centers have used alternate pro-coagulants such as desmopressin injection (DDAVP) and aprotinin (Trasylol).

In some cases of diffuse oozing not responsive to the therapy listed above, intravenous recombinant activated factor seven (rFVIIa, NovoSeven) has been an effective hemostatic agent ³.

All decisions regarding the use of anticoagulants and pro-coagulants are discussed with the surgeon beforehand (except FFP which we often give without discussion).

Pulmonary hypertension

Although definitions of pulmonary hypertension vary, portopulmonary hypertension (PPHTN) is an elevation in pulmonary artery (PA) pressures evolving in the presence of portal hypertension and in the absence of any other known causes of pulmonary hypertension. Severe PPHTN was considered a contraindication to OLT prior to the demonstration that a subset of these patients responded favorably to epoprostanol (Flolan) ⁴.

Our approach to PPHTN at GTI follows the algorithm of Kuo et al: assuming no other contraindications to OLT are present (and left ventricular function is preserved on dobutamine stress echocardiogram (DSE)), patients suspected of having severe PPHTN (usually based on echocardiographic estimate and respiratory alkalosis present on blood gas) undergo right heart catheterization (PAC). If mean PA pressure is >40 mmHg, epoprostanol is given by continuous infusion with an implantable pump. PAC is repeated at 3–6 month intervals. When mean PA pressure is < 40 mmHg, a DSE is obtained with the PAC in situ. If left and right ventricular function are adequate at rest and during stress, an intravenous fluid bolus is administered (1L normal saline over 10 minutes). If mean PA pressure remains <40 mmHg and biventricular function is preserved, the patient is considered a candidate for OLT. At Mayo Clinic the approach is similar, however, pulmonary vascular resistance and other factors influence timing of OLT if mean PA pressure is 35–50 mmHg.

Echocardiographic estimates of PA systolic pressure are based on the modified Bernoulli equation which allows one to calculate the pressure difference (during systole) between the right atrium and right ventricle from the velocity of the regurgitant jet across the tricuspid valve. This difference is added to the central venous pressure (whether measured or estimated). This method depends on the presence of tricuspid regurgitation and the ability to direct continuous wave Doppler across the valve parallel (or nearly so) to the direction of blood flow. The method also assumes that there is no gradient (during systole) between the RV and the PA. Use of estimated (rather than measured) CVP increases the degree of uncertainty in the result.

We place PA catheters in all adult OLT recipients. Transesophageal echocardiogram (TEE) is placed selectively. These monitors have special value in the setting of pulmonary hypertension.

Although PAC and TEE use are not mutually exclusive, it is worthwhile to briefly mention their relative merits. A PAC has some advantages over a TEE in perioperative management of pulmonary hypertension: The PAC will accurately measure PA pressures which must be measured in order to determine if the patient is a candidate for transplant. The PAC can be used as a continuous monitor in the operating room (OR) and intensive care unit (ICU) whereas the TEE can not be used as a continuous monitor. Determination of cardiac output and systemic vascular resistance are far easier with a PAC than with TEE. Also, the PAC can be used for determinations of mixed venous oxygen saturation and pulmonary vascular resistance.

A TEE should be inserted in the setting of unexpectedly high PA pressures both for determining the etiology of the pulmonary hypertension (e.g. pulmonary thromboembolus) and following the response to therapy (e.g. changes in left ventricular end-diastolic volume after furosemide). During OLT (and the immediate postoperative interval), TEE is especially useful when significant changes in hemodynamics are difficult to explain (e.g. severe hypotension refractory to pressors when filling pressures are roughly normal and labs and electrocardiogram are normal) ⁵. The presence of esophageal varices does not absolutely contraindicate use of TEE.

If mean PA pressure is < 40 mmHg (or greater than two thirds of mean arterial pressure (MAP)) when a liver becomes available, one must determine if the pulmonary hypertension can be reversed prior to proceeding with OLT; a fixed high PA pressure is associated with poor outcome and, in such cases, the available liver should be released to the next patient on the list. If a patient presents to the operating room for OLT with newly diagnosed or breakthrough pulmonary hypertension, the pulmonary hypertension ought to be treated before a decision is made whether to cancel the case. An attempt can be made to bring PA pressure down acutely with oxygen, hyperventilation and anesthesia (fentanyl/isoflurane). If PA occlusive pressure (wedge) is high, furosemide might be given especially if systemic pressure and cardiac output are normal or high. If the patient is on epoprostanol on arrival to the OR, this is continued. Milrinone or inamrinone may be useful in the setting of right or left ventricular failure.

Preoperative cardiac evaluation

Coronary artery disease (CAD) is found in 27% of patients over 50 years age referred for liver transplant evaluation ⁶. DSE has been shown to be sensitive and specific for diagnosing CAD in patients with multiple cardiac risk factors who have been referred for OLT ⁷.

Early extubation

The frequency of very early extubation (i.e. tracheal extubation at the conclusion of surgery prior to leaving the OR) in OLT recipients varies widely among transplant centers ⁸,⁹. Immediate extubation following OLT has several potential benefits. These include decreased cost, improved patient comfort and satisfaction, decreased ICU and hospital length of stay, early neurological evaluation and improved pulmonary toilet with decreased incidence of ventilator associated pneumonia. Early extubation also carries risks. These include postoperative ventilatory failure with impaired oxygen delivery to the newly grafted liver. Another possible disadvantage of early extubation is the need for re-intubation if the patient returns to the OR (for early complications such as bleeding, bile leak, thrombosis, or re-transplantation).

Several studies have explored these issues. For example, Mandell et al. studied 173 OLT recipients retrospectively: 41 were tracheally extubated at the end of surgery; 2 of these had to be re-intubated within 30 minutes due to respiratory failure. Both patients were successfully extubated after another 12–14 hours of ventilatory support. The authors found that early extubation was associated with an average savings of $2709. They concluded that immediate tracheal extubation following OLT is safe and cost-effective in selected patients. In this study, criteria associated with successful immediate tracheal extubation were lesser disease severity (i.e. old UNOS status 3 or 4), absence of encephalopathy or co-existing disease, age <50 years, good graft function, RBC transfusion <10 U, alveolararterial oxygen gradient < 150 mmHg and no vasoactive support at end of surgery. Although early extubation has been successful in many patients, optimal selection criteria and timing for early extubation are not defined by the current literature. Furthermore, results from a single center may not be generalizable due to differences in staffing and other resources.

Fulminant hepatic failure and coma

Fulminant hepatic failure is frequently associated with brain edema and concomitant intracranial hypertension ¹⁰. Sustained elevation of intracranial pressure (cerebral perfusion pressure < 40 mmHg for 2 hours) is associated with 100% mortality ¹¹. At GTI we follow an intracranial pressure (ICP) management protocol similar to that outlined by Hoofnagle et al.: an epidural ICP monitor is placed at the onset of stage III coma (marked confusion, sleeping but arousable) after baseline head CT scan and after coagulopathy is corrected (seebelow). Hyperventilation, diuretics, fluid restriction, hemodialysis, head elevation, intravenous anesthesia (usually propofol or fentanyl) and muscle relaxants are administered as needed to keep ICP below 20 mmHg. Hypothermia (33–35°C) is induced (with surface cooling) if ICP remains high. Cerebral perfusion pressure (CPP = MAP 7ICP) is maintained above 50 mmHg. If a sustained decrease in CPP (as defined above) occurs, the patient is not considered a suitable candidate for liver transplant. At Mayo Clinic an intraparenchymal ICP monitor is preferred.

As mentioned above, significant coagulopathy (INR >2 seconds or platelet count <100 k/mcL) is corrected prior to placement of the ICP monitor. If INR fails to correct with FFP, rFVIIa 80 mcg/kg is administered.

Author disclosures

Drs Manley and Yosaitis have received research support from NovoNordisk.