P2427 - Insulin Resistance is an Independent Predictor of Non-Alcoholic Fatty Liver Disease in Lean United States Adults

Monday, October 23, 2023

10:30 AM – 4:15 PM PT

Location: Exhibit Hall

Has Audio

Presenting Author(s)

Mouhand FH Mohamed, MD

Brown University
Providence, RI

Mouhand FH. Mohamed, MD¹, Manasik Abdu, MD², Basile Njei, MD, MPH, PhD³
¹Brown University, Providence, RI; ²University at Buffalo-Catholic Health System, Buffalo, NY; ³Yale School of Medicine, New Haven, CT

Introduction: The screening and early identification of Non-alcoholic Fatty Liver Disease (NAFLD) in lean individuals often present challenges due to typical presentations of normal liver enzymes and lack of diabetes. A strong underpinning of metabolic syndrome, Insulin Resistance (IR), is believed to be a significant pathophysiological component of NAFLD. Our study aims to investigate the predictive role of IR in discerning NAFLD amongst lean individuals.

Methods: We extracted data from the National Health and Nutrition Examination Surveys (NHANES) 2017 to 2020. Inclusion criteria encompassed lean adults > 18 years with valid transient elastography measurements, while we excluded individuals with heavy alcohol consumption, viral hepatitis, or human immunodeficiency virus. NAFLD was defined as a Controlled Attenuation Parameter (CAP) ≥302 dB/m as determined by Youden’s index. Subjects were stratified into NAFLD and non-NAFLD groups. We assessed IR using the Homeostasis Model Assessment of Insulin Resistance [HOMA-IR = insulin (μU/ml) × fasting glucose (mmol/L)/22.5]. The study's primary objective was to evaluate HOMA-IR's effectiveness in predicting NAFLD amongst lean individuals. Potential confounders were adjusted for through multivariate logistic regression analyses.

Results: Our dataset comprised 860 subjects, of which 783 (91.0%) were classified into the non-NAFLD group, while 77 (9.0%) were in the NAFLD group. The median ages for the non-NAFLD and NAFLD groups were 52 (IQR 36) and 62 (IQR 22) years, respectively. Female representation was numerically higher in the non-NAFLD group (48.9%) compared to the NAFLD group (42.9%), but this was not statistically significant. The percentage of current smokers was noticeably lower in the non-NAFLD group (51.4%) versus the NAFLD group (83.3%) (Table 1). From the univariate analysis, HOMA-IR was correlated with increased odds of NAFLD in lean individuals (OR 1.75, 95% CI 1.48-2.08, P< 0.001). This association remained significant in the multivariate regression analysis (OR 1.49, 95% CI 1.20-1.87, P< 0.001) (Figure 1).

Discussion: This study underscores the significance of HOMA-IR, a surrogate for IR, as a predictor of NAFLD in lean individuals. HOMA-IR can potentially serve as a tool to guide NAFLD screening in lean individuals. Further investigation is needed to ascertain the clinical utility of HOMA-IR as a diagnostic and therapeutic target for lean NAFLD patients.

Figure: Figure 1: Univariate and Multivariate Analysis of Homeostasis Model Assessment of Insulin Resistance (HOMA-IR) as a Predictor of NAFLD in Lean United States Adults

OR = Odds Ratio, using logistic regression; CI = Confidence interval.
aModel 1: univariate analysis
bModel 2: adjusted for age, sex, and race.
cModel 3: Model 2, further adjusted for Diabetes, BMI, Waist circumference, Hypertension, Triglycerides, HDL-cholesterol, LDL-cholesterol, ALT, GGT, RBC Folate, Uric Acid levels, and Liver fibrosis (LSM).

	*Lean Individuals (N=860)*
Variable	No NAFLD (N=783; 91.0 %)	NAFLD (N=77; 9.0%)	*P Value*
Age, median (IQR), years	52 (36)	62 (22)	< 0.001
Female	48.9%	42.9%	0.31
Race /Ethnicity
Non-Hispanic Asian	19.8%	27.3%	0.34
Non-Hispanic Black	24.4%	11.7%
Hispanic	7.5%	6.5%
Non-Hispanic White	36.0%	40.3%
Current smoker	51.4%	83.3%	0.02
Vigorous Exercise	71.5%	80.5%	0.10
Average Alcoholic Drinks per Day, median (IQR)	1(0.4)	1 (0.4)	0.43
Body Mass Index, median (IQR), Kg/m²	22.4 (3.2)	23.7 (2.2)	< 0.001
Waist Circumference, median (IQR), cm	82.5 (11.5)	90.1 (9.3)	< 0.001
Sedentary Activity, median (IQR), minutes/day	240 (240)	240 (300)	0.72
Income/poverty ratio, median (IQR)	2.1 (3.0)	2.0 (2.6)	0.66
Comorbidities
Metabolic Syndrome	8.9%	32.5%	< 0.001
Prediabetes	24.9%	33.8%	0.09
Diabetes	4.7%	18.2%	< 0.001
Hypertension	8.9%	31.2%	< 0.001
Hypertriglyceridemia	4.7%	18.2%	< 0.001
Low HDL Cholesterol	14.2%	37.7%	< 0.001
High ALT ( >40 U/L)	3.6%	9.1%	0.02
Laboratory Panel, median (IQR)
Fasting Glucose, mg/dL	99.0 (13)	113.5 (24)	< 0.001
HbA1c, %	5.5 (0.5)	5.8 (1.0)	< 0.001
Triglycerides, mg/dL	73 (43)	101 (66)	< 0.001
HDL-Cholesterol, mg/dL	57 (19)	47 (17)	< 0.001
LDL-Cholesterol, mg/dL	101 (51)	99 (42)
ALT, U/L	15 (8)	18 (10)	0.01
GGT, U/L	16 (12)	19 (14)	0.03
Platelets, X10⁹/L	227 (75)	227 (79)	0.74
Serum Albumin, g/dL	4.1 (0.4)	4.2 (0.4)	0.75
C-reactive Protein, g/dl	0.84 (1.6)	1.2 (2.0)	0.83
Ferritin, ng/mL	97 (131)	111 (149)	0.17
RBC Folate, ng/mL	444 (261)	490 (279)	0.03
Lactate Dehydrogenase, U/L	151 (40)	153 (42)	0.26
Uric Acid, mg/dL	4.8 (1.8)	4.9 (2.4)	0.02
Urine Albumin, g/dL	8.2 (12.4)	9.3 (15.7)	0.45
Transient Elastography
Steatosis, CAP, median (IQR), dB/m	213 (47)	311 (31)	< 0.001
Fibrosis, LSM, median (IQR), kPa	4.5 (1.8)	5.2 (2.0)	0.15
Clinically Significant Fibrosis (LSM ≥ 8.6)	3.7%	5.2%	< 0.001
HOMA-IR, median (IQR)	1.7 (0.9)	2.3 (1.4)	< 0.001

Disclosures:

Mouhand Mohamed indicated no relevant financial relationships.

Manasik Abdu indicated no relevant financial relationships.

Basile Njei indicated no relevant financial relationships.

Mouhand FH. Mohamed, MD¹, Manasik Abdu, MD², Basile Njei, MD, MPH, PhD³. P2427 - Insulin Resistance is an Independent Predictor of Non-Alcoholic Fatty Liver Disease in Lean United States Adults, ACG 2023 Annual Scientific Meeting Abstracts. Vancouver, BC, Canada: American College of Gastroenterology.