Return to Alcohol Use After Liver Transplant: Patterns and Surveillance

Watch a video presentation of this article

Watch the interview with the author

Answer questions and earn CME

Abbreviations

AALD

alcohol‐associated liver disease

CDT

carbohydrate‐deficient transferrin

EtG

ethyl glucuronide

EtS

ethyl sulfate

GGT

gamma‐glutamyltransferase

LT

liver transplant

LTX

liver transplantation

MCV

mean corpuscular volume

PEth

phosphatidylethanol

TLFB

Timeline Follow‐Back

Clinic Consult

You are providing a clinic consult for a 50‐year‐old woman status post–liver transplant (LT) for alcohol cirrhosis and follow‐up with the transplant team. What is the recommended approach to alcohol relapse prevention and monitoring?

Introduction

Alcohol‐associated liver disease (AALD) represents 40% to 50% of liver deaths in the United States. Since 2016, AALD has become the most common indication for LT in the United States.1 Although LT treats the hepatic complications of alcohol use disorder, it does not treat the addiction to alcohol, a chronic illness with relapsing and remitting periods of disease.2 This review discusses the patterns of alcohol use after LT, its negative consequences, and the tools available to transplant providers to monitor alcohol use.

Monitoring Alcohol Use after LT: Tools Exist, but There is Lack of Consensus

No controlled studies have directly linked vigilant monitoring of alcohol use to improved outcomes; however, there is strong evidence that sustained alcohol use leads to worse outcomes, including death, which has not been shown for brief periods of limited alcohol use after LT.3, 4 Thus, the purpose of monitoring for alcohol use in the post‐LT setting is to allow for early detection of alcohol use and to prevent a slip from becoming sustained alcohol use.

Nonbiochemical Testing: Self‐Report

Monitoring for alcohol use typically includes patient interview, with direct questioning of quantity, type, and frequency of alcohol use. Independent collateral from a family member or caregiver is helpful to confirm or add to the patient’s self‐report. With regard to formal self‐report tools of alcohol use, the most widely used is the Timeline Follow‐Back (TLFB), a calendar‐based questionnaire readily available online, which has been validated as a highly reproducible and detailed measure of quantity and frequency of alcohol use.5, 6 The TLFB is designed to improve recall by allowing the patient to use “anchor points” (i.e., birthdays, holidays) and to identify triggers of alcohol use.5, 6 Reliance on tools of self‐report, although efficient,7 are at risk for underreporting, particularly in the LT setting where a patient’s embarrassment or fear of the consequences of discovery by LT providers can limit patient candor.2

Biochemical Testing: Indirect and Direct Markers of Ethanol

Biochemical testing (Table 1) has been shown to increase sensitivity for detection of alcohol use beyond self‐report methods.8 For example, in the ACCELERATE‐AH cohort, which addressed posttransplant outcomes after LT for severe alcoholic hepatitis, 11 of 40 (28%) patients with evidence of alcohol use posttransplant were first discovered by biochemical testing.9

Table 1.

Selected Biomarkers of Alcohol Use

Biomarkers	Type of Test	Window of Alcohol Detection	Sensitivity	Specificity	Reasons for Inaccuracy
Indirect biomarkers
GGT	Serum	2‐4 weeks	35%‐85%	63%‐85%
CDT	Serum	Weeks to months	25%‐90%	90%
Direct biomarkers
EtG/EtS	Urine (most common), serum, hair (EtG only)	4 days (urine) Months (hair)	62%‐89%	93%‐99%	False negatives: very low alcohol consumption, delayed testing False positives: yeast ingestion, urinary bactiuria Positive results from inadvertent alcohol exposure*: hundreds of products (e.g., “alcohol‐free” beer/wine, ripe fruits, mouthwash, skin contact with cleaning products)
PEth	Whole blood	28 days	90%‐99%	100%	False negatives: low alcohol consumption (<2 drinks per day) False positives: none

Adapted with permission from Alcohol Abuse and Liver Disease.19 Copyright 2015, John Wiley & Sons, Ltd.; and Hepatology. 10 Copyright 2011, American Association for the Study of Liver Diseases.

^*Levels of urine EtG/EtS can be unpredictably high, particularly with renal dysfunction, and can be affected by a number of other factors, including age, sex, cannabis, among others.

As with any screening test, both sensitivity and specificity are important; however, in the context of LT, decisions misinformed by false‐positive results can have imminently life‐threatening repercussions by withholding LT listing in the pre‐LT period or withholding retransplantation in the post‐LT period.

Indirect markers of ethanol consumption, including gamma‐glutamyltransferase (GGT), mean corpuscular volume (MCV), aspartate aminotransferase, and carbohydrate‐deficient transferrin (CDT), have unacceptable specificity, are understudied in the LT population, and cannot be considered standard of care for monitoring of alcohol use in the LT setting.10

Direct markers of ethanol use include ethyl glucuronide (EtG), ethyl sulfate (EtS), and phosphatidylethanol (PEth).11, 12 These direct markers have the benefits of high specificity and long windows of detection after last alcohol use; these tests, however, are not perfect, and all transplant providers must understand the nuances to interpret the results correctly.11, 12

EtG and EtS are metabolites formed during the elimination of ethanol.11 They are nonvolatile, water‐soluble, and widely used by LT programs as urine tests, although serum testing is available.10, 11 EtG and EtS can be detectable in urine up to 90 hours after alcohol ingestion, and liver disease has been shown to have no influence on the validity of this testing.11, 13 However, in vitro formation of EtG has been described, and given the high sensitivity, small amounts of yeast ingestion and urinary bacteriuria can lead to false‐positive results.11, 12 Further, impaired renal function can increase EtG and EtS concentrations in the urine and lead to prolonged elimination; although this has the benefit of a longer detection window, the EtG/EtS level can be unpredictable and poorly correlated to quantity of alcohol consumption.13, 14 The urine level can also be affected by a number of other factors, including age, sex, and cannabis consumption, as well as inadvertent alcohol consumption from use of hand sanitizer, ripe fruits, soft drinks, and hundreds of other products.11, 12, 13 The typical thresholds for EtG and EtS are 250 and 50 ng/mL, respectively, but can be variable across centers.12 Even though EtG and EtS are currently widely used in US LT centers, the risks of false‐positive or positive results from inadvertent alcohol consumption warrant caution and consideration of other biomarkers of alcohol use.

Hair and nail analyses with EtG and fatty acid ethyl ester also exist and have been reported to accurately identify alcohol use within the last several months, with higher specificity than urine testing, but these tests are not yet routinely used in LT centers.13, 15

PEth is an attractive alternative to EtG and EtS; PEth is a phospholipid formed only in the presence of alcohol and can be identified from a whole blood sample.12, 16 PEth can detect alcohol consumption in the last 28 days, with more than 90% sensitivity in patients consuming two or more drinks per day, and the level is linearly correlated to the quantity of alcohol consumption.12, 16 The test has been validated in patients with liver disease and does not seem to be affected by age, sex, renal function, or other factors.8 Thus, the test is thought to be 100% specific for alcohol consumption and, with a threshold of 8 ng/mL, can exclude any positive results from unintentional low‐level alcohol exposure.12, 16 Unlike EtG and EtS, PEth is typically a send‐out laboratory test for most LT centers, and it costs approximately $75.8 A recent prospective study examined the utility of PEth to monitor alcohol use after LT8; the study involved 213 LT recipients at two US centers and found that 24% with a history of AALD had at least one positive PEth with negative self‐report of alcohol use.8

Patterns of Alcohol Use after LT

Reported rates of alcohol use after LT for AALD range from 8% to 22% and 30% to 40% at 1 and 5 years after LT, respectively, and surpass 50% with longer‐term follow‐up.6, 8, 17 Comparing between studies is difficult because there is no standard definition for alcohol relapse, and methods to capture alcohol use are variable.2, 6, 18 In 2010, the University of Pittsburgh published a prospective study to describe the trajectories of alcohol use after LT in 208 patients who underwent LT for AALD.6 Patients were assessed for post‐LT alcohol use at scheduled intervals with the TLFB questionnaire (a validated tool to capture quantity and frequency of recent alcohol use by patient report)5 and routine blood alcohol level tests.6 LT recipients were stratified into five “trajectories” of post‐LT alcohol use (Fig. 1): (1) no evidence of alcohol use (54%); (2) low amounts infrequently (26%); (3) early‐onset moderate use (average of 3.5 drinks per week at peak amount) that diminished over time (6%); (4) later‐onset moderate use that increased over time (7%); or (5) early‐onset, heavy (average of 4 drinks per day at peak amount), increasing pattern of use (6%).6 Five deaths were secondary to recurrent AALD, all of whom had either early‐onset moderate or heavy alcohol use after LT (groups 3 and 5).6 This study demonstrated the infrequency of clinically relevant alcohol use and the importance in distinguishing patterns of drinking6; other studies are congruent in identifying the early post‐LT period (within 1 year of LT) as the highest risk period for return to alcohol use and in associating continuous harmful use of alcohol with rapid allograft fibrosis and death. This highlights the importance of monitoring and treating alcohol use after LT, particularly in the first year after LT.3, 4

Figure 1.

Trajectories of alcohol use after LT for AALD. In 2010, the University of Pittsburgh published a prospective study to describe the trajectories of alcohol use after LT in 208 patients who underwent LT for AALD.6 LT recipients were stratified into five “trajectories” of post‐LT alcohol use: (1) no evidence of alcohol use (54%); (2) low amounts infrequently (26%); (3) early‐onset moderate use (average of 3.5 drinks per week at peak amount) that diminished over time (6%); (4) later‐onset moderate use that increased over time (7%); or (5) early‐onset, heavy (average of 4 drinks per day at peak amount), increasing pattern of use (6%).6 Five deaths were secondary to recurrent AALD, all of whom had either early‐onset moderate or heavy alcohol use after LT (groups 3 and 5).6 LTX, liver transplantation. Reproduced with permission from author and publisher.6

Conclusion

Although there is agreement regarding the need to address the consequences of alcohol use after LT for AALD, there is no consensus as to how to monitor alcohol use.8, 10 Given the association of harmful drinking posttransplant with poor outcomes, transplant providers should query all LT recipients for alcohol use at every posttransplant visit, preferably supplemented by random biochemical testing to detect covert drinking. PEth, with 100% specificity and its long detection window for recent alcohol consumption (28 days), seems to be the most promising biomarker for the LT population. The first year posttransplant is the highest risk period for alcohol use, but use can occur thereafter, and thus vigilance should be maintained at all periods.6 Although it would appear that only harmful levels of alcohol use are clinically significant, in the context of an unclear trajectory of alcohol use, the goals should be complete abstinence and preventing slips from becoming sustained use.2 Further study into the implications of covert versus overt drinking and tools to prevent and treat alcohol use after LT are warranted to better inform providers how to act on the results from alcohol use monitoring.

Potential conflict of interest

Nothing to report.