Underrecognition of NAFLD in Poorly Controlled Diabetes

In This Article

Abstract and Introduction

Abstract

Individuals with type 2 diabetes (T2DM) are at high risk for nonalcoholic fatty liver disease (NAFLD), and evidence suggests that poor glycemic control is linked to heightened risk of progressive NAFLD. We conducted an observational study based on data from a telehealth trial conducted in 2018–2020. Our objectives were to (1) characterize patterns of NAFLD testing/care in a cohort of individuals with poorly controlled T2DM; and (2) explore how laboratory based measures of NAFLD (eg, liver enzymes, fibrosis-4 [FIB-4]) vary by glycemic control. We included individuals with poorly controlled T2DM (n = 228), defined as hemoglobin A1c (HbA1c) ≥ 8.5% despite clinic-based care. Two groups of interest were (1) T2DM without known NAFLD; and (2) T2DM with known NAFLD. Demographics, medical history, medication use, glycemic control (HbA1c), and NAFLD testing/care patterns were obtained by chart review. Among those without known NAFLD (n = 213), most were male (78.4%) and self-identified as Black race (68.5%). Mean HbA1c was 9.8%. Most had liver enzymes (85.4%) and platelets (84.5%) ordered in the outpatient department over a 2-year period that would allow for FIB-4 calculation, yet only 2 individuals had FIB-4 documented in clinical notes. Approximately one-third had abnormal liver enzymes at least once over a 2-year period, yet only 7% had undergone liver ultrasound and 4.7% had referral to hepatology. Among those with known NAFLD (n = 15), mean HbA1c was 9.5%. Only 4 individuals had undergone transient elastography, half of whom had advanced fibrosis. NAFLD is underrecognized in poorly controlled T2DM, even though this is a high-risk group for NAFLD and its complications.

Introduction

Nonalcoholic fatty liver disease (NAFLD) disproportionately impacts individuals with type 2 diabetes (T2DM), with a prevalence of 60% to 70%.^[1–4] Importantly, the presence of T2DM is associated with higher odds of having progressive forms of NAFLD^[5,6] such as nonalcoholic steatohepatitis and advanced liver fibrosis, with a prevalence as high as 10% to 15% for advanced fibrosis in T2DM.^[3,4] T2DM also leads to a substantial increase in the risk of NAFLD complications, including a 2-fold increase in hepatocellular carcinoma and liver-related mortality risk.^[7,8] Recent evidence also suggests that beyond the presence of T2DM, glycemic control may be linked to greater severity of liver fibrosis in NAFLD.^[9] Therefore, individuals with poor glycemic control may be at particular risk of progressive NAFLD and poor liver outcomes.

Despite an increasing clinical burden of NAFLD, and calls to action across the medical community to prepare for this rising public health threat,^[10–14] NAFLD remains underrecognized in primary care settings.^[15] It is likely that this underrecognition extends to T2DM populations—even in those with poor glycemic control—though data specific to poorly controlled T2DM are lacking. The primary purpose of this study was to examine NAFLD testing and care patterns in a cohort of individuals with poorly controlled T2DM. Since laboratory-based measures (liver enzymes, noninvasive scores, eg, fibrosis-4 [FIB-4] index) constitute initial testing for NAFLD that may prompt additional work-up, we also aimed to explore how these measures vary by glycemic control.

Table 1. Characteristics of T2DM population without known NAFLD, stratified by clinical site
	Whole cohort (n = 212)	Durham VAMC (n = 117)	Richmond VAMC (n = 96)
Demographic
Age, mean (SD)	58.1 (8.1)	57.8 (7.8)	58.5 (8.6)
Male, n (%)	167 (78.4)	93 (79.5)	74 (77.1)
Race, n (%)
White	50 (23.5)	26 (22.2)	24 (25.0)
Black or African American	146 (68.5)	80 (68.4)	66 (68.8)
Other	17 (8.0)	11 (9.4)	6 (6.3)
Ethnicity, n (%)^a
Hispanic or Latino/a	9 (4.2)	3 (2.6)	6 (6.3)
≤40 miles from nearest VA clinic (%)	163 (76.5)	88 (75.2)	75 (78.1)
Medical history
Hypertension, n (%)	181 (85.0)	99 (84.6)	82 (85.4)
Hyperlipidemia, n (%)^a	178 (83.6)	99 (84.6)	79 (82.3)
BMI (kg/m²), mean (SD)	35.0 (6.2)	35.6 (6.4)	34.4 (5.9)
Diabetes duration (years), mean (SD)	12.5 (8.0)	12.6 (7.6)	12.5 (8.6)
T2DM medication use and control^a
Metformin, n (%)	168 (79.2)	98 (83.8)	70 (73.6)
Sulfonylurea, n (%)	86 (40.6)	49 (41.9)	37 (38.9)
Thiazolidinedione, n (%)	20 (9.4)	19 (16.2)	1 (1.1)
DPP4 inhibitor, n (%)	2 (0.9)	1 (0.8)	1 (1.1)
GLP-1 receptor agonist, n (%)	34 (16.0)	10 (8.5)	24 (25.3)
SGLT2 inhibitor, n (%)	26 (12.2)	15 (12.8)	11 (11.6)
Insulin, n (%)	151 (70.9)	87 (74.4)	64 (66.7)
HbA1c, %, mean (SD)^a	9.8 (1.5)	9.7 (1.4)	9.9 (1.7)
NAFLD testing/care patterns
Liver enzymes, n (%)	182 (85.4)	99 (84.6)	83 (86.4)
# of checks, mean (SD)	2.3 (1.8)	2.4 (2.0)	2.1 (1.6)
Platelets, n (%)	180 (84.5)	92 (78.6)	88 (91.7)
# of checks, mean (SD)	2.9 (2.6)	3.0 (3.3)	2.8 (1.7)
FIB-4 or other score documented, n(%)^b	2 (0.9)	0 (0)	2 (2.1)
Liver ultrasound, n (%)	15 (7.0)	8 (6.8)	7 (7.3)
Proportion with fatty liver, n (%)	12 (80.0)	6 (75.0)	6 (85.7)
CT/MRI with fatty liver, n (%)	6 (2.8)	3 (2.6)	3 (3.1)
Transient elastography, n (%)^c	5 (2.3)	2 (1.7)	3 (3.1)
F0–1 (%)	2 (40.0)	0 (0)	2 (40.0)
F2 (%)	1 (20.0)	1 (20.0)	0 (0)
F3 (%)	2 (40.0)	1 (20.0)	1 (20.0)
F4 (%)	0 (0)	0 (0)	0 (0)
Hepatology referral, n (%)	10 (4.7)	4 (3.4)	6 (6.3)

Abbreviations: BMI, body mass index; DPP4, dipeptidyl peptidase-4; GLP-1, glucagon-like pepetide-1; SGLT2, sodium glucose cotransporter-2; HbA1c, hemoglobin A1c; FIB-4, fibrosis-4; CT, computed tomography; MRI, magnetic resonance imaging.
^a1 Richmond participant was missing HbA1c and data on noninsulin medications; 1 Richmond participant responded "Don't Know" to Latino/Hispanic origin; 1 Richmond participant responded "Don't Know" to diagnosis of hyperlipidemia.
^b40 individuals (27 Durham 13 Richmond) were not included in this calculation because they did not have outpatient liver enzymes or platelets checked. We only included outpatient liver enzyme and platelet data in this table.
^cProportions and percentages of the individuals who underwent transient elastography (n = 5).

Table 2. Laboratory-based measures in cohort without known NAFLD, stratified by glycemic control
	Glycemic control groups^c
	Mean HbA1c <8.5% (n = 39)	Mean HbA1c 8.5–9.5% (n = 67)	Mean HbA1c >9.5% (n = 106)
Liver enzyme testing
AST, mean (SD)^d	22.9 (8.0)	23.0 (18.0)	19.3 (7.3)
ALT, mean (SD)^d	36.4 (12.2)	36.0 (21.9)	34.0 (11.9)
ALT >40 IU/L, n (%)^d	16 (44.4)	16 (29.1)	38 (42.2)
Platelet count, mean (SD)^e	236.2 (63.1)	258.9 (57.4)	260.3 (61.8)
Undiagnosed NAFLD^a, n (%)	6 (15.4)	8 (11.9)	22 (20.8)
Undiagnosed NAFLD^b, n (%)	15 (38.5)	19 (28.4)	43 (40.6)
FIB-4 score, mean (SD)^f	1.4 (1.4)	1.0 (0.4)	0.9 (0.5)
FIB-4 score risk categories
Low risk, n (%)	24 (72.7)	42 (85.7)	72 (93.5)
Indeterminate risk, n (%)^c	7 (21.2)	7 (14.3)	5 (6.5)
High risk, n (%)	2 (6.1)	0 (0)	0 (0)

Abbreviations: AST, aspartate aminotransferase; ALT, alanine aminotransferase; NAFLD, nonalcoholic fatty liver disease; FIB-4, fibrosis-4.
^aBased on Kanwal et al. criteria (ALT ≥40 IU/mL in males, ALT ≥31 IU/mL in females at least 6 months apart) [19].
^bBased on Prati et al. criteria for healthy ALT ranges (ALT ≥30 IU/mL in males, ALT ≥19 IU/mL in females at least 6 months apart) [20].
^c1 individual did not have a measurable A1c collected thus is excluded from this table.
^d31 individuals had no AST or ALT values measured (3 with HbA1c <8.5%, 12 with HbA1c 8.5–9.5%, and 16 with HbA1c >9.5%).
^e47 individuals had no platelets measured (5 with HbA1c <8.5%, 18 with HbA1c 8.5–9.5% and 24 in with HbA1c >9.5%).
^f53 individuals had no FIB-4 score calculated due to missing values (6 with HbA1c <8.5%, 18 with HbA1c 8.5–9.5%, and 29 with HbA1c >9.5%). We used liver enzymes and platelets collected within 3 months of each other for FIB-4 calculation.

Table 3. Characteristics of population with T2DM and known NAFLD
Demographic	Whole cohort (n = 15)
Age, mean (SD)	57.9 (9.3)
Male, n (%)	12 (80.0)
Race, n (%)
White	5 (33.3)
Black or African American	9 (60.0)
Mixed race	1 (6.7)
Ethnicity, n (%)
Hispanic or Latino/a	2 (13.3)
≤40 miles from nearest VA clinic (%)	9 (60.0)
Medical history
Hypertension, n (%)	15 (100.0)
Hyperlipidemia, n (%)	15 (100.0)
Diabetes duration (years) mean (SD)	15.1 (5.7)
BMI (kg/m²), mean (SD)	37.8 (6.8)
T2DM medication use and control
Metformin, n (%)	13 (86.7)
Sulfonylurea, n (%)	5 (33.3)
Thiazolidinedione, n (%)	2 (13.3)
DPP4 inhibitor, n (%)	0 (0)
GLP-1 receptor agonist, n (%)	3 (20.0)
SGLT2 inhibitor, n (%)	2 (13.3)
Insulin, n (%)	12 (80.0)
HbA1c, %, mean (SD)	9.5 (1.8)
NAFLD testing/care patterns
Liver enzymes, n (%)	15 (100)
# of checks, mean (SD)	3.3 (1.8)
Platelets, n (%)	15 (100)
# of checks, mean (SD)	3.2 (2.1)
FIB-4 or other score documented, n (%)	0 (0)
FIB-4 score, mean (SD)^a	1.65 (0.94)
Liver ultrasound, n (%)	6 (40.0)
Transient elastography, n (%)	4 (26.7)
F0–1 (%)	1 (25.0)
F2 (%)	1 (25.0)
F3 (%)	2(50.0)
F4 (%)	0 (0)
Following with hepatology, n (%)	5 (33.3%)
Following with endocrinology, n (%)^b	13 (86.7)

Abbreviations: BMI, body mass index; DPP4, dipeptidyl peptidase-4; GLP-1, glucagon-like pepetide-1; SGLT2, sodium glucose cotransporter-2; HbA1c, hemoglobin A1c; FIB-4=fibrosis-4.
^a1 individual had missing data due to lack of liver enzymes and platelets done within 3 months of each other. Only outpatient laboratory data were included in this table.
^bHad to be following endocrinology for T2DM care.

My Alerts

Add to Email Alerts

Your Name is required.

Subject is required.

Please enter a Recipient Address and/or check the Send me a copy checkbox.

Your email has been sent.

is an Invalid Email Address.

Underrecognition of Nonalcoholic Fatty Liver Disease in Poorly Controlled Diabetes

A Call to Action in Diabetes Care

Abstract and Introduction

Abstract

Introduction

Underrecognition of Nonalcoholic Fatty Liver Disease in Poorly Controlled Diabetes

A Call to Action in Diabetes Care

Abstract and Introduction

Abstract

Introduction

Tables

Tables

Tables

References

Authors and Disclosures

Authors and Disclosures

Email This

Feedback