Carbohydrate-Deficient Transferrin Evaluation Over Time Has Usefulness for Confirming Sobriety of Alcohol Rehabilitation Program Patients

Abstract

Background: For treating severe alcohol-related liver disease (ALD), abstinence from alcohol is essential. Because underreporting can occur, abstinence biomarkers are necessary during hepatology outpatient service. This study assesses whether serial changes of the serum carbohydrate-deficient transferrin to transferrin ratio (%CDT) reflect patient abstinence.

Methods: We retrospectively investigated serial %CDT measurements and laboratory data from two outpatient settings: patients in an alcohol rehabilitation program (A), and ALD patients at a hepatology clinic (H) who attempted abstinence with self-diary monitoring. Using group A data, correlation coefficients were determined between the number of days from the last drink (abstinence days) and laboratory data. Linear approximation was applied for significant coefficients. Using these results, %CDT was validated for group H.

Results: Group A data were 223 data samples from 55 cases with abstinence days. Four (7%) were diagnosed as having cirrhosis. Eighteen patients continued abstinence over 30 days. %CDT (Rho = -0.50, p < 0.001) and GGT (Rho = -0.34, p < 0.01) were negatively correlated with abstinence days. For %CDT, negative linear approximation was maintained with the number of days (y = 2.01-0.004 x).Group H data were 18 cases, of which 12 (66%) were diagnosed as cirrhosis cases. Seven patients reported continued abstinence for 8 weeks, during which time %CDT values dropped significantly. The group H %CDT was higher than the group A %CDT, perhaps because of severer liver disease.

Conclusion: The %CDT trend assessment is useful to support continued, multi-month self-reported abstinence from alcohol, but it must be interpreted carefully in severe liver disorder cases.

Keywords: alcohol use disorder; ambulatory care; liver cirrhosis.

FIGURE 1

Flowchart of data selection in Group A and Group H. Group A: Data from patients followed at Kure Midorigaoka Hospital. Group H: Data from patients followed at Kansai Medical University Medical Center. ALD: Alcohol‐related liver disease. CDT: Carbohydrate‐deficient transferrin.

FIGURE 2

Correlation between days from the last drink and laboratory values for group A patients are shown for abstinence cases and for drinkers. Each circle represents each measurement. The fitted line according to linear regression model is overlined: (a, b) %CDT, (c, d) GGT, and (e, f) MCV. Graphs (a), (c), and (e) are for abstinence cases and (b), (d), and (f) are for drinkers. The x‐axis shows the number of days since the last drink.

FIGURE 3

Line graphs for %CDT (a, b), GGT (c, d), and MCV (e, f) were drawn for each case in group A. The x‐axis shows the number of days since the last drink for abstinence cases (a, c, and e). Each dot in these graphs is identical to the respective circle in Figure 2a,c,e. The x‐axes for graphs b, d, and f show the visit date for drinking cases.

FIGURE 4

Change of average daily ethanol consumption (a, b), %CDT (c, d), GGT (e, f), MCV (g, h), and Albumin (i, j) for group H are presented: Graphs (a), (c), (e), (g), and (i) are for abstinence cases (n = 7) and (b), (d), (f), (h), and (j) are for drinkers (n = 11). The timings of calculations of the average daily ethanol consumption were the entry time, from entry to week 4 (Week 4), from week 4 to week 8 (Week 8), and at week 24 (Week 24). Only values found to be significant by paired t test are shown in the graph; *p < 0.05; **p < 0.01; asterisk not shown, not significant.

FIGURE 5

Change of average %CDT was graphed for abstainers (a) and for drinkers (b) according to their Child class. Cases with Child class C cirrhosis (n = 7) are shown as white boxes; other cases (n = 11) are shown as gray boxes.