EXPERIMENTAL AND CLINICAL TRANSPLANTATION

Objectives: Recently, gamma-glutamyl transpeptidase has garnered increased attention as a diagnostic tool in the early identification of liver disease. However, its value in liver transplant is largely unknown, as the disease processes leading to abnormal gamma-glutamyl transpeptidase levels and the expected temporal trends in gamma-glutamyl transpeptidase levels during the period after liver transplant remain unclear.

Materials and Methods: Between January 2010 and August 2013, consecutive patients who underwent liver transplant at Vancouver General Hospital (Vancouver, Canada) were assessed longitudinally up to 1 year after liver transplant. A “gamma-glutamyl transpeptidase event” was defined as 2 abnormal gamma-glutamyl transpeptidase values (exceeding sex-specific limits of normal, at 55 U/L for female and 80 U/L for male patients) ≥ 1 week apart.

Results: Our study included 147 liver transplant recipients. The median gamma-glutamyl transpeptidase level on day 1 after liver transplant was 73 U/L, which peaked to 435 U/L during the first month after liver transplant and returned to within normal parameters by 1 year. In total, there were 282 gamma-glutamyl transpeptidase events, with biliary complications (22%), acute rejection (16%), and hepatitis C virus recurrence (10%) being the most common causes. In 39% of events, no cause was identified. When attempting to identify a disease-associated event, if gamma-glutamyl transpeptidase was the initial liver biochemistry test to double in value, it had 42% sensitivity and 40% specificity. Comparatively, if gamma-glutamyl transpeptidase was the initial liver biochemistry test to become abnormal, it had 3% sensitivity and 93% specificity.

Conclusions: Although gamma-glutamyl transpeptidase almost universally becomes abnormal after liver transplant, a specific pathologic cause was not commonly identified. Interpreting the characteristics of gamma-glutamyl transpeptidase elevation has limited use for identifying the underlying reason for its elevation.

Key words : Complications, Adverse events, Liver biochemistry

Liver transplant has rapidly evolved to become a widely accepted treatment for patients with end-stage liver disease. This has been facilitated by advances in organ preservation strategies, surgical techniques, and immunosuppression regimens, ultimately leading to improved survival among appropriate candidates.^1,2 However, despite these advances, it remains a complex procedure requiring a multidisciplinary approach to allow for effective care of patients after transplant, specifically, for diagnosis and treatment of the diverse array of complications. Unfortunately, treatment of complications is reliant on diagnostic tests such as magnetic resonance cholangiopancreatography, endoscopic retrograde cholangiopancreatography, and liver biopsy. These are expensive diagnostic approaches, which are not readily available at many medical centers and are not devoid of adverse events.^3,4 Therefore, a need exists for a simple, inexpensive, and readily available diagnostic modality to facilitate timely identification of posttransplant complications, with the hope that it would lead to improved patient outcomes and optimized use of resources.

Liver biochemical testing has historically been used to serially monitor a patient’s posttransplant course. However, there is currently no standardized approach among transplant centers regarding which liver biochemical tests should be used and the frequency at which they should be performed, most notably regarding gamma-glutamyl transpeptidase (GGT). Interestingly, GGT has recently begun to garner increased attention as a potentially useful diagnostic tool for complications after liver transplant such as acute rejection,^5,6 biliary complications,⁷ and primary liver disease recurrence.8 However, the temporal trends in GGT after transplant and the disease processes commonly leading to abnormal GGT values remain undefined, thus limiting its use at this time.

Given the current deficit in the literature regarding the overall use of GGT, we aimed to retrospectively delineate the causes of abnormal GGT values and characterize temporal trends in GGT after liver transplant among a large cohort of recipients. This is of importance, as an improved appreciation for the meaning of elevated GGT levels may allow for more timely identification of common complications after transplant, thereby improving patient outcomes, maximizing use of resources, and potentially altering current investigational algorithms.

Materials and Methods

Study population
Between January 1, 2010, and August 31, 2013, consecutive adult patients (≥ 19 y old) who underwent liver transplant at Vancouver General Hospital (Vancouver, Canada) were evaluated for study inclusion. Patients who underwent repeat transplants were excluded from analyses. Written ethics approval was obtained from our institution’s ethics committee with the research protocol conforming to the ethical guidelines of the Helsinki Declaration.

Data extraction and interpretation
Data were collected in a retrospective manner. The following variables were collected up to 1 year after liver transplant: (1) age; (2) sex; (3) indication for liver transplant; (4) donor type; (5) surgical methods; (6) initial immunosuppression regimen; (7) length of stay in hospital after transplant; and (8) liver biochemistry test results, including GGT, alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, and total bilirubin. When multiple values for any liver biochemistry test were available within a 24-hour period, the maximum value was recorded.

A “GGT event” was defined as at least 2 consecutive abnormal GGT values (exceeding sex-specific upper limits of normal, at 55 U/L for female and 80 U/L for male patients) ≥ 1 week apart. Similarly, resolution of a GGT event was defined as 2 consecutive GGT values within sex-specific normal limits ≥ 1 week apart. During persistent GGT elevation, if GGT prominently rose (≥ 2 times the new baseline GGT value), a new GGT event was defined as having occurred. Furthermore, in the setting of persistent GGT elevation without an attributable cause after ≥ 3 months, a new GGT event was defined as having occurred. For each GGT event, further data were collected, including (1) timing during the posttransplant period, (2) cause of GGT elevation, and (3) associated liver biochemistry measurements. During each GGT event, if multiple potential causes for GGT elevation were found, only the most prominent cause was recorded. Concerning liver biochemistry tests, values were recorded at time of diagnosis of the cause of GGT elevation. If no cause was identified, liver biochemistry values were recorded coinciding with the peak GGT value during the GGT event.

Statistical analyses
Numerical data are presented as means with standard error when normally distributed and nonparametric data as median with interquartile range. Categorical data are presented with their corresponding frequencies and percentages. Nonparametric data were compared using the Wilcoxon-rank sum test. Generalized estimated equations were used for this longitudinal study for the consideration that 1 patient could have multiple events to account for a possible unknown correlation between each event. The sensitivity and specificity of a GGT change when it was either the first liver biochemistry test (compared with alanine aminotransferase, aspartate aminotransferase, bilirubin and alkaline phosphatase) to become abnormal or to double also was calculated using all of the GGT events recorded. Only GGT events for which a cause was identified after thorough investigation were considered as true positives; GGT events with no identified diagnosis were considered as false positives. Patients with an identified complication during a GGT event but with no significant changes to GGT values (first liver biochemistry test to become abnormal or double) were defined as false negatives. All statistical analyses were performed using Stata/SE 11.2 (StataCorp LP, College Station, TX, USA).

Demographics of liver transplant recipients
Between January 1, 2010, and August 31, 2013, a total of 161 liver transplants were performed. Fourteen were excluded from our analyses as these were repeat transplants, leaving 147 liver transplant recipients for study inclusion (Table 1). The median age of recipients was 55 years, with 67% being male patients. The most common indications for liver transplant were hepatitis C virus (HCV) with or without concomitant hepatocellular carcinoma (41%), alcohol-related liver disease (15%), and primary sclerosing cholangitis (10%). The most common method of biliary anastomosis was duct-to-duct anastomosis (82%), with subsequent induction immunosuppression posttransplant being a combination of tacrolimus, mycophenolate mofetil, and prednisone (88%). The median length of hospital stay after liver transplant was 17 days.

Liver biochemistry and monthly gamma-glutamyl transpeptidase values
The median GGT value on day 1 after liver transplant was 73 U/L, in comparison to alkaline phosphatase of 78 U/L, total bilirubin of 6 mg/dL, alanine aminotransferase of 535 U/L, and aspartate aminotransferase of 685 U/L. When we analyzed peak monthly median GGT values, GGT peaked within the first month at 435 U/L and then gradually returned to within normal limits by 1 year posttransplant (Figure 1). When we compared recipients with and without HCV, those with HCV had significantly higher monthly median GGT values over the 1-year period after liver transplant (Figure 2).

Gamma-glutamyl transpeptidase event description
In total, 282 GGT events occurred among the 147 liver transplant recipients within the 1-year follow-up. The most commonly associated disease processes were biliary complications (22%), acute rejection (16%), and HCV recurrence (10%). However, in 39% of cases, no cause was identified. At time of diagnosis of the underlying reason for the GGT event (or at time of peak GGT value for GGT events without an identifiable cause), the median GGT value was 330 U/L. This was in comparison to alkaline phosphatase level of 192 U/L, total bilirubin level of 1 mg/dL, alanine aminotransferase level of 108 U/L, and aspartate aminotransferase level of 59 U/L. When specifically assessing GGT events attributed to biliary disease, the median GGT value was 341 U/L. Along with GGT values, alkaline phosphatase (233 U/L), total bilirubin (1 mg/dL), alanine aminotransferase (85 U/L), and aspartate aminotransferase (40 U/L) levels were relatively comparable to their values when assessing overall GGT events.

When stratifying GGT events by their frequency, 143 liver transplant recipients had ≥ 1 GGT event, 91 had ≥ 2 GGT events, 37 had ≥ 3 GGT events, and 11 had 4 GGT events (Table 2). Concerning the first GGT event for recipients, in 50% of cases, no cause was identified. However, as more GGT events were recorded, this became less likely, with only 9% of events having no identified source among the liver transplant recipients who had 4 GGT events. When assessing GGT events by timing 1 year after transplant, 59 occurred between days 1 and 30, 32 between days 31 and 90, 52 between days 91 and 180, and 139 between days 181 and 365. Among all 4 of these groups, the percentage of GGT events without an identifiable cause was relatively consistent (range, 35%-44%). Among earlier GGT events (occurring from days 1-30), there was a greater frequency of GGT events attributed to acute rejection compared with biliary complications (27% vs 12%). However, during the latter half of the 1-year follow-up, this reversed, with a greater proportion of biliary complications compared with acute rejection (27% vs 14%). Finally, when we stratified GGT events by the value of GGT at time of diagnosis of the attributable cause, 50 had a GGT value 1 to 2 times the upper limit of normal (ULN), 100 with 3 to 5 times ULN, 83 with 6 to 10 times ULN, 44 with 11 to 20 times ULN, and 5 with > 20 times ULN. Between 1 and 10 times ULN, there was a relatively consistent frequency of GGT events without an attributable cause (range, 32%-44%). However, among GGT events whose value rose to > 10 times ULN, we found that 30% to 40% of cases were due to acute rejection.

Use of gamma-glutamyl transpeptidase interpretation
When we assessed the use of GGT interpretation to identify a disease-associated GGT event, if GGT was the initial liver biochemistry test to double in value, it had a sensitivity of 42% and a specificity of 40%. Comparatively, if GGT was the initial liver biochemistry test to become abnormal, it had a sensitivity of 3% and a specificity of 93%.

Liver transplant has become an integral component to the treatment of end-stage liver disease and with it comes several posttransplant complications that require prompt recognition and treatment. Unfortunately, diagnosis is reliant on costly tests that carry with them notable morbidity risk. Therefore, there is currently a dependence on the initial interpretation of liver biochemistry to guide the direction of subsequent diagnostic evaluation. However, liver biochemistry is relatively nonspecific and can be quite difficult to interpret in the early period after liver transplant. This has hindered guidelines in providing firm recommendations regarding which liver biochemistry tests should be performed as well as their testing frequency.⁹ Interestingly, GGT has recently garnered increased attention pertaining to its use as a diagnostic tool for complications after liver transplant, most notably biliary complications, acute rejection, and HCV recurrence. Its interpretation, however, is currently limited as GGT commonly rises during the posttransplant period. Therefore, both the underlying causes and the temporal trends in GGT elevation after liver transplant are unclear. We therefore sought to evaluate these deficiencies, with our study showing that GGT almost universally rises after liver transplant and commonly normalizes within 1 year. Furthermore, although biliary tract complications, acute rejection, and HCV recurrence make up a large proportion of the underlying causes for GGT elevation, a large proportion of GGT events did not have a definitive cause identified.

Gamma-glutamyl transpeptidase’s use in diagnosing complications after liver transplant previously has been established, in part by comparing liver biochemistry levels taken immediately after transplant with its value at the time of definitive diagnostic evaluation.⁷ This is potentially flawed given the multiple confounders for baseline liver biochemistry posttransplant, including baseline hepatic dysfunction before liver transplant and the extent of preservation-reperfusion injury. Moreover, studies^5,10 assessing the correlation of liver biochemistry taken at the time of definitive diagnostic testing, such as histopathologic evaluation, do not fully portray the role of liver biochemistry in this setting; which is also to guide subsequent diagnostic evaluation at the time of initial concern for a potential complication. Our findings are unique in that we attempted to delineate the underlying causes for GGT elevation and to identify early characteristics of GGT elevation to help delineate the presence of an underlying cause. What we found is that there appears to be limited use in the interpretation of the early characteristics (first liver biochemical test to become abnormal or to double in value) of GGT elevation.

An interesting and unique feature of our study was the establishment of 1-year trends in GGT values posttransplant, with previous studies largely focusing on the immediate period after liver transplant.6 We found that GGT values commonly peak within the first month after liver transplant and subsequently return to within normal parameters by 1 year. The peak in GGT within the first month is likely secondary to preservation-reperfusion injury. Interestingly, only 10% of GGT events within the first 30 days posttransplant were felt to be secondary to preservation-reperfusion injury. However, we believe this is likely an underestimate, with a large proportion of the GGT events not yet diagnosed representing preservation-reperfusion injury during this period, with misclassification secondary to a lack of confirmatory liver biopsy as it is not a structured component of care at our center. This is supported by Naik and associates,⁶ who showed elevations in GGT among liver transplant recipients who did not have acute rejection during the immediate period after transplant. Of note, there was a significant discrepancy in GGT trends between liver transplant recipients with HCV versus those without HCV. Although it has been shown that GGT elevation may be a positive prognostic sign after liver transplant,^11-13 this has been directly questioned for the population of liver transplant patients with HCV.^8,14 Cholestasis has been associated with both HCV recurrence and, on rare occasions, fibrosing cholestatic hepatitis.^15-17 A recent study by Ueda and associates demonstrated that cholestasis is a common (69% cases) feature of HCV recurrence.⁸ Furthermore, it is associated with liver fibrosis progression after liver transplant, with cholestasis improving with HCV treatment. Although this was not a specific focus of our study, given the almost universal recurrence of HCV after transplant, these findings in conjunction with our own suggest that GGT has the potential to act as an early marker for HCV recurrence and warrants further evaluation.

Our study has several limitations. First, it is retrospective in study design. Second, for each GGT event, only the single most prominent cause was coded for the purpose of this study, thus potentially underscoring the multifactorial nature of the underlying causes of GGT elevation. Finally, at our center, there is no standardized work-up for GGT elevation, which helps to some extent explain the large proportion of GGT events with no underlying cause identified.

In conclusion, the interpretation of liver biochemistry is clearly an integral component of care after liver transplant, with GGT almost universally rising after transplant and returning to normal by 1 year. Moreover, although GGT is likely commonly elevated secondary to preservation-reperfusion injury, biliary complications, HCV recurrence, and acute rejection, no cause was commonly identified. However, in the population of liver transplant patients with HCV, this elevation may be an early indicator of HCV recurrence; therefore, further research in this specific area is needed.

Volume: 14 Issue: 3 June 2016

FULL TEXT