Association of non-alcoholic fatty liver disease (NAFLD) and insulin resistance in type 2diabetes

Introduction

NAFLD is a potential pathologic condition encompassing a broad spectrum of disease severity ranging from isolated steatosis in its mildest form to a more serious condition called nonalcoholic steatohepatitis (NASH) with advanced fibrosis, cirrhosis and can lead to hepatocellular carcinoma.^1-3 NAFLD has now become the second leading cause of liver transplantation in the United States.⁴

Liver being the major organ of the metabolic system, the consequences of NAFLD are not restricted to the liver. Increasing reports on the association of NAFLD with cardiovascular disease, T2DM, insulin resistance (IR) and dyslipidaemia signifies the impact of NAFLD beyond the liver. Further, co-existence of NAFLD with T2DM may potentially worsen the risk of micro and macro vascular disease.^5,6

Studies have shown cause and consequence relation between NAFLD and T2DM. However, there are limited studies on association of NAFLD and IR in Diabetes. Though the association of NAFLD and IR is known, the causal link between these two is debatable. Both NAFLD and T2DM are having common risk factors like obesity, IR, dyslipidemia etc. IR being the common pathophysiological risk factor, the current study is an attempt to decipher the association of NAFLD and IR in T2DM individuals.

Therefore the present study is aimed to determine the association of NAFLD with IR in T2DM subjects and; compare between NAFLD and non-NALFD subgroups among T2DM subjects.

Materials and Methods:

Study population:

Total of 151 type 2 diabetic subjects in the age group of 18 to 60 years for the cross sectional study were selected, who visited endocrinology OPD of Tertiary care hospital attached to our institute during July 2016 to March 2018. Participants who consume alcohol (≥20g/day), pre-existing liver diseases other than NAFLD or those who are under medications which affect liver function were excluded. The study was approved by institutional ethics committee (EC REG ECR/134/Inst/KA/2013). A written informed consent was taken from all the participants before enrolment.

Data regarding age, gender, height, weight, waist and hip circumferences, blood pressure, family history of diabetes and systemic examination results were recorded.

Diagnosis of NAFLD:

NAFLD was diagnosed by abdominal ultrasonography using – 5.0 MHZ high frequency curvilinear transducer (C5-1) in Philips affinity 70 ultrasound machine (PHILIPS medical systems, Bothell, WA) by experienced radiologists in the department of Radio diagnosis.⁷

Biochemical measurements:

5 ml of fasting blood and 2 ml of postprandial blood were collected in plain vacutainer and allowed to clot for 20 min at room temperature. The serum was separated by centrifugation at 1500 rpm for 10 min. and serum aliquots were stored at -80 0C till further use. Another 2 ml of postprandial blood was collected in EDTA vacutainer for glycosylated haemoglobin (HbA1c) analysis is stored at -20 0C till further use.

Fasting and postprandial blood glucose, serum lipids (total cholesterol, triglycerides, HDL-c, LDL-c), SGOT, SGPT, bilirubin total and direct, total protein, albumin, globulin and ALP were measured in serum samples using Cobas C311 fully automated chemistry analyser (Roche Diagnostics).  HbA1c was determined in whole blood samples by turbidimetric immunoassay using Cobas C311 analyzer. Fasting serum insulin was determined using Cobas E411 automated immunoanalyser (Roche Diagnostics).

Insulin sensitivity was calculated using homeostasis model assessment - insulin resistance [HOMA-IR] formula:

HOMA-IR = [Fasting insulin (µU/ml) × Fasting glucose (mg/dl)] / 405

Definitions:

Diabetes was defined as fasting serum glucose ≥126 mg/dL, post prandial blood glucose ≥200 mg/dl, HbA1c ≥6.5% or use of oral anti-diabetic agents (after diagnosis of T2DM). HOMA-IR score ≥3 is considered as insulin resistant. Obesity was defined as BMI ≥25 kg/ m2 and central obesity was defined as waist circumference ≥90 cm for males and ≥80 cm for females according to the proposed cut-off for the diagnosis in Asians.⁸ Hypertension was defined as systolic blood pressure ≥140 mm Hg, diastolic blood pressure ≥90 mm Hg, or the use of antihypertensive medication. SGPT ≥34 U/L for females, ≥45 U/L for males and SGOT ≥31 for female, ≥35 for males are considered as elevated enzyme levels.9 Triglycerides ≥150 mg/dL and cholesterol ≥200 mg/dL, LDL-c ≥100 mg/dL, and HDL-c ≤50 mg/ dL for females and ≤40 mg/dL for males are considered as abnormal levels.¹⁰

Statistical Analysis:

Mean differences were analysed by student’s t-test and ANOVA. Chi-square and odds ratio were used to analyse the association between NAFLD and IR. Statistical tests were performed using SPSS software. Z-test in EpiTools epidemiological calculators was used to analyse the difference in proportions (http://epitools.ausvet.com.au/ content.php?page=z-test-2).

Results:

The study population involved 151 diabetic subjects and majority of them were females (65%). Mean age was 51±7. NAFLD was diagnosed in 37 out of 53 males and 73 out of 98 females. The overall prevalence of NAFLD and IR were 73% and 81% respectively (Fig 1). Hypertensive subjects were also high (80%) among the study participants. But, there was no statistically significant difference between male and female in the prevalence of NAFLD, IR and hypertension at P<0.05. However, females showed comparatively higher frequency of NAFLD (74% vs 70%). On the other hand obesity data revealed that females are having significantly high prevalence of overall obesity (70% vs 51%) and central obesity (85% vs 58%) when compared to males.

Females showed higher prevalence of obesity (70% vs 51%) and central obesity (85% vs 58%) when compared to males in T2DM subjects. Other parameters (NAFLD, IR and hypertension) didn’t shows significant difference across gender.

The comparison between NAFLD and non-NAFLD subjects showed significant high prevalence of IR, obesity, central obesity, hypertension, elevated SGPT, elevated TG and hepatomegaly among NAFLD subjects as revealed in Fig. 2. However, there was no significant difference in the prevalence of elevated SGOT, cholesterol, LDL-c and reduced HDL-c, between NAFLD and nonNAFLD subgroups at P<0.05.

*P<0.05, **P<0.001, ns no significant difference at P<0.05.

The serum biochemical data revealed that mean values of FBS, PPBS, HbA1c, fasting insulin, TG, HDL-c and LDL-c were at clinically abnormal level in both NAFLD and non-NAFLD subgroups. However, there was no significant difference in mean values of serum biochemical parameters except TG, VLDL and SGPT, between NAFLD and non-NAFLD subjects (Table 1).

Also, the IR groups made based on HOMA-IR score revealed that the odds of having NAFLD was increased from 1.8 to 13.2 with increasing severity of IR (Fig-3).

Prevalence of NAFLD increased significantly as the severity of IR increses. The values in the table indicate the increase in odds of having NAFLD as the HOMA-IR score increases.

Discussion

In spite of substantial information available on the prevalence of NAFLD in patients with T2DM, where most of the information are largely from highly selected populations of T2DM with predominantly elevated liver transaminase,11 the assessment has been achieved through biochemical tests that may not have given the exact prevalence owing to its limitations compared to ultrasound diagnosis of NAFLD. To support this, the earlier reports on prevalence of NAFLD have shown varied data depending on the method of diagnosis such as elevated transaminase level to USG. Histopathology, though being the gold standard to diagnose NAFLD, cannot be used for routine screening of NAFLD.

Both NAFLD and T2DM share common risk factors such as obesity, IR and dyslipidemia. IR being the common pathophysiological link between NAFLD and T2DM, the study focused on exploring the association of NAFLD with IR in diabetic condition.

In the present study, we found 73% of NAFLD and around 81% of IR in T2DM subjects. Also, we found that IR is strongly associated with NAFLD as evidenced by increase in odds of having NAFLD as the severity of IR increases among diabetic subjects as shown in Fig.3. Some of the earlier reports have shown high prevalence and early occurrence of NAFLD in males than female.¹² However, in the current study, though majority of the participants were females (65%), there were no significant difference found in the prevalence of NAFLD and IR between male and female. Whereas, prevalence of obesity (70% vs 51%) and central obesity (85% vs 58%) were significantly high among female T2DM subjects.

Comparison of biochemical parameters between NAFLD and non-NAFLD subgroups showed hyperglycemia, hyperinsulinimia, and dyslipidemia in diabetic subjects irrespective of their fatty liver condition (Table1). The data showed clinically abnormal levels of glycemic status (FBS, PPBS, HbA1c) and lipid profile except cholesterol and VLDL. But the statistically significant difference between NAFLD and non NAFLD subgroups was found only with respect to TG and VLDL. The liver function tests showed clinically normal range of protein, bilirubin and aminotransferase levels in both the groups (Table 1).

It is assumed that low level of liver aminotransferase in diabetes have the less chance of developing complications related to NAFLD. However, in the current study, more than 85% of diabetic subjects were having normal levels of serum aminotransferase (SGPT) level even in presence of NAFLD. This warrants the need for routine screening of diabetic patients with USG for early diagnosis of NAFLD instead of depending on liver function tests.

The co-existence of NAFLD with diabetes will have high probability of developing diabetes associated complications like NASH, cirrhosis and hepatocellular carcinoma.¹³ The early diagnosis and treatment of NAFLD may help in reducing these complications in diabetes.

Significantly high prevalence of IR in NAFLD subpopulation (88%) and increasing odds of having NAFLD above 13 with increase in severity of IR (Fig. 3) implies that NAFLD is strongly associated with IR in T2DM subjects. Number of previous studies explored the association of NAFLD with several insulin-dependent metabolic impairments regardless of the presence of diabetes.[14] Reduction of hepatic TG with proportional improvement of hepatic insulin sensitivity in diabetic individuals with low calorie, low fat diet also supported the association of hepatic fat with IR.¹⁵ Few studies also showed that it is the liver fat content rather than intramyocellular and visceral fat content associated with insulin resistance.16 Findings of the current study also showed a strong association of fatty liver with IR (Fig 3). A limitation of the current study is we could not quantify the liver fat content by USG unlike MRS method. Also, we couldn’t correlate the severity of fatty liver with increased IR which requires liver histopathology data.

The increase in fasting insulin in NAFLD group is evidenced by proportionately increased beta cell function (Table 1). This data supports the previous observations on hyperinsulinimia in fatty liver individuals without diabetes.14 The increased insulin may also attribute to impaired insulin clearance.16 Further studies proposed that, diacyl glycerol in the fatty liver may reduce hepatic insulin signalling.17 Hence, increase in beta cell function and hyperinsulinimia observed in the current study in NAFLD group are in response to reduced insulin sensitivity due to hepatic fat content. 

In the current study, though there was no significant difference in mean HbA1c level between NAFLD and non-NAFLD groups, we found that individuals with HbA1c level of >6.4% were more in NAFLD than non-NAFLD group (85% vs 75%). This suggests that NAFLD is one of the factors associated with difficult to manage hyperglycemia. The same was observed by Ryysy et al., where the authors suggests hepatic steatosis in patients with T2DM is associated with more difficult-to-control hyperglycemia, worse insulin resistance, and the need for higher insulin doses.¹⁸

Although muscle is the tissue i.e. mainly affected by IR, impairment of insulin action in liver and adipose tissue among NAFLD individuals may occur even before the onset of diabetes. The loss of insulin sensitivity in adipose tissue results in decreased suppression of lipolysis leading to increased free fatty acids influx into the liver. Peripheral IR requires higher insulin secretion. The combination of higher insulin level and high plasma lipids promotes hepatic free fatty acid uptake and triglyceride synthesis.¹⁹ The impairment in insulin action at the level of liver and adipose tissue is proportional to the amount of hepatic and visceral fat.^20,21 On the other hand, there are some reports showing liver fat causing IR. Increase in diacylglycerol (DAG) content and inhibition of insulin signalling were found in lipid-infused rodents and rodents fed with highfat diets.²² Thus in subjects with IR, more calorie intake may favour development of fatty liver.^19,23 These reports suggest a link between IR and NAFLD though the cause and effect conundrum still exists.

Conclusion

The current study showed the high prevalence of NAFLD and its strong association with insulin resistance among T2DM subjects. The study suggests early diagnosis of NAFLD and; consider NAFLD as the therapeutic target for prevention and management of diabetes, as clinicians concentrate more on micro and macro vascular complications while managing diabetes conditions.