Secondary Logo

Journal Logo


S1051 Differences in Liver Elastography in Non-Diabetic, Prediabetic, and Known Diabetic Patients Based on HbA1c Levels in Community-Based Gastroenterology Practices

Lazas, Donald J. MD1; O'Rourke, Josh MBA1; Pontes, Alda BA1; Aldous, Mark MD2; Bachinski, Matthew MD3; Barish, Robert W. MD, FACG4; Brown, Michael MD, FACG5; Din, Raja MD6; Myers, Matthew MD2; Newman, Frederic MD7; Patel, Pankaj MD8; Patel, Vinay MD2; Andrady, Gerry BA1; Hoke, Colleen BS1; Wallace, Matthew BS1; Wiggins, Janice BS1

Author Information
The American Journal of Gastroenterology: October 2020 - Volume 115 - Issue - p S533-S534
doi: 10.14309/
  • Free


While the prevalence of non-alcoholic fatty liver disease (NAFLD) and liver fibrosis are higher in patients with Type 2 Diabetes (T2DM)1, the common biomarker Hemoglobin A1c (HbA1c) is not typically utilized by community clinicians in patient selection for transient elastography (TE) NAFLD screening. Our aim was to identify HbA1c trends that could assist in determining thresholds for TE screening protocols.


231 patients with suspected NAFLD were referred by GI practitioners across 8 US facilities. TE was performed and liver stiffness (kPa) and attenuation (CAP) were obtained. Most recent HbA1c values were collected from each patient’s medical record. A univariate logistic regression analysis was used to calculate odds ratios, risk ratios, sensitivity, specificity, PPV, and NPV to identify HbA1c thresholds for patients with kPa ≥ 8 and CAP ≥ 260. These values have been shown to correlate with moderate fibrosis and steatosis on liver biopsy.2


Of the 231 patients observed, 159 were female and 72 were male. 111 had a T2DM diagnosis. The mean TE values were 325 for CAP, 9.6 for kPa, and 6.8 for HbA1c. Patients with prediabetic HbA1c levels of ≥ 6.2 had increased risk for kPa ≥ 8 with an odds ratio of 2.19 (CI 1.3–3.7, P = 0.004) and for CAP ≥ 260 with an odds ratio of 2.28 (CI 1.1–4.9, P = 0.029) versus patients with HbA1c < 6.2. Relative risk ratios were also calculated at HbA1c levels in normal, prediabetic, and diabetic thresholds as defined by the American Diabetes Association. Risk for kPa ≥ 8 and CAP ≥ 260 was shown to increase as HbA1c values increased across groups. Patients with HbA1c values in prediabetic ranges (HbA1c ≥ 5.7) showed an increased risk of 1.14 for kPa ≥ 8 and 1.11 for CAP ≥ 260 and those with HbA1c values in diabetic ranges (HbA1c ≥ 6.5) showed in increased risk of 1.49 for kPa ≥ 8 and 1.16 for CAP ≥ 260 versus those with an HbA1c value in a normal range (HbA1c < 5.7). Additional results for risk ratios, sensitivity, specificity, PPV, and NPV can be found in Table 1, Table 2, and Figure 1.


Prediabetic and diabetic HbA1c values predicted higher TE measurements in ranges correlating with fibrosis and steatosis on liver biopsy. A greater emphasis should be placed on obtaining and reviewing HbA1c in assessing NAFLD risk. Future research should focus on comparing H1Ac and other predictive biomarkers in NAFLD.

Table 1
Table 2
Figure 1


1. Cusi, et al. Diabetes Care 2017;40:419-430.
2. Loomba, et al. Gastroenterology 2019;156:1264-81
© 2020 by The American College of Gastroenterology