Endocrine Abstracts

Introduction

Lipid accumulation product (LAP), an index based on waist circumference (WC) and serum triglyceride levels, is associated with the presence and severity of non-alcoholic fatty liver disease (NAFLD). However, its potential as a risk marker of NAFLD in subjects with type 1 diabetes (T1D) has not been studied yet.

Aims

This study evaluated the ability of LAP to estimate the presence of NAFLD in adults with T1D.

Methods

Four hundred and seven adult T1D patients (age 18–88, 56.8% males) were included. Anthropometric, biochemical and imaging data were acquired during the screening visit. LAP was calculated as [WC (cm) – 58] × triglycerides (mmol/l) in females; [WC – 65] × triglycerides (mmol/l) in males. NAFLD was assessed using ultrasound, controlled attenuation parameter (CAP) and fatty liver index (FLI), ultrasound was set as the reference method.

Results

NAFLD prevalence according to ultrasound was 20.4%. LAP was positively correlated with FLI (r = 0.728, P < 0.001), CAP (r = 0.433, P < 0.001), the number of elements of the metabolic syndrome (r = 0.659, P < 0.001), age (r = 0.115, P = 0.020), insulin dose/kg (r = 0.236, P < 0.001), liver stiffness (r = 0.110, P = 0.027), HbA1c (r = 0.188, P < 0.001) and uric acid (r = 0.324, P < 0.001). LAP was negatively correlated with estimated glucose disposal as a measure of insulin sensitivity (r = –0.478, P < 0.001) and HDL-c (r = –0.326, P < 0.001). LAP values could be used to diagnose NAFLD in the total cohort (AUC = 0.767 [0.708–0.827], P < 0.001). The AUC to indicate NALFD was 0.790 [0.715–0.865] in males and 0.732 [0.635–0.828] in females (P < 0.001 for both), indicating a slight difference between genders. The ideal gender-specific cut-off value was 32.6, with a sensitivity and specificity of 70% vs . 78% for males, and 27.0 (72% vs . 75%) for females. Dichotomised based on the optimal cut-off, the prevalence of NAFLD according to LAP was 32.9%. This was significantly higher compared to ultrasound (χ², P < 0.001). True positive rate was 58/133 subjects (44%), false negative rate was 25/271 (9%), false positive rate was 56%, while true negative rate was 91%. LAP ≥ 32.6 (males) or ≥ 27 (females) was associated with ultrasound-based NAFLD (OR = 4.120 [1.954–8.690], P < 0.001) in a multivariate regression model including BMI (OR = 1.146 [1.043–1.260], P = 0.005), insulin dose/kg (OR = 4.838 [1.672–13.995], P = 0.064) and liver stiffness (OR = 1.410 [1.114–1.784], P = 0.004).

Conclusion

We are the first to demonstrate that LAP is an easily available risk marker of NAFLD in T1D, with high accuracy to rule out. If LAP is ≥ 32.6 (males) or ≥ 27 (females), NAFLD should be suspected but other tools are needed to confirm the diagnosis.