Tacrolimus intrapatient variability (Tac IPV) has been considered a marker for post-graft risk. We investigated pre-transplant psychometric testing to predict Tac IPV after living kidney transplantation. Minnesota Multiphasic Personality Inventory-2 (MMPI-2) examined during pre-transplant evaluation by 102 recipients were analyzed. Subjects were divided into two groups, low IPV (L-IPV) and high IPV (H-IPV), by cutoffs of Tac IPV: median of 24 and value of 30. T-scores of MMPI-2 scales were used to analyze difference between L-IPV and H-IPV using independent t-tests. Stepwise multiple logistic regression was used to test whether MMPI-2 scales affected Tac IPV. Confusion matrix of logistic regression was used to explain statistical power. Cutoff values of significant scales for H-IPV were analyzed by constructing receiver operating characteristic curves. Hysteria (Hy) and depression (D) scale scores and Tac IPV were associated in IPV 24 (odds ratio [OR]: 1.08, p<0.01 for Hy; OR: 0.93, p<0.01 for D) and IPV 30 models (OR: 1.09, p<0.01 for Hy; OR: 0.92, p<0.01 for D). Paranoia (Pa) scale scores were associated with Tac IPV in IPV 24 model (OR=1.10, p<0.01) and were significantly higher in H-IPV 24 (p<0.01). F1 scores of confusion matrix in IPV 24 and 30 models were 0.70 and 0.71, respectively. Cutoffs of Hy, D, and Pa scales were 51, 57, and 47, respectively. MMPI-2 profile is suggested as a predictor for high Tac IPV after living kidney transplantation.
Read full abstract