Identification of Myocardial Insulin Resistance by Using Liver Tests: A Simple Approach for Clinical Practice

Figures

• 

  Figure 1. Differences between HIS (A), NAFLD-LFS (B) and FIB-4 (C) indexes between both phenotypes of T2D patients (ISm in blue and IRm in orange). Data are shown as mean ± SD.

• 

  Figure 2. ROC curves of variables displaying the optimal threshold for splitting the two groups (mIR vs. mIS) together with their threshold and AUC values. The corresponding ROC metrics (sensitivity and specificity) are as follows: ALT (0.88, 0.68), AST (0.88, 0.64), GGT (0.69, 0.77), Protein (0.68, 0.75), FIB-4 (0.31, 0.95), and HOMA-IR (0.56, 0.77).

• 

  Figure 3. ROC curves of each of the presented functions as a combination of variables displaying the AUC values. The corresponding ROC metrics (sensitivity and specificity) are as follows: f(Protein, AST, ALT, GGT, FIB-4, HOMA-IR) (0.81,0.95), f(AL, AST, GGT, HOMA-IR, FIB-4) (1.00,0.73), f(Protein, ALT, AST, GGT, FIB-4) (0.82,0.96), f(ALT, AST, GGT, FIB-4) (0.88,0.80), f(ALT, AST, GGT, HOMA-IR) (0.88,0.77), f(Protein, AST, ALT, GGT, HOMA-IR) (0.69,0.86), f(ALT, AST, GGT) (0.53,0.88). The best function obtained for splitting both groups with AUC = 0.95 was: f = 27.444 + (−2.9636 × [protein]) + 0.25195 × [AST] + (−0.36118 × [ALT]) + 0.014958 × [GGT] + (−3.6456 × [FIB-4]) + (−0.057339 × [HOMA-IR]). The p value of this function was 0.000043 with a threshold to split both cohorts of 1.37.

• 

  Figure 4. Results of liver assessment of fibrosis by fibroscan of both groups patients with T2D (mIR and mIS). Data are shown as median ± SD.

• 

  Figure 5. ΔSUV in liver after performing an HEC procedure of both groups (mIR and mIS). Data are shown as median ± SD.

• 

  Figure 6. Outline of the clinical trial to obtain PET/CT, biochemical and anthropometrical data in T2D patients.