EVALUATION OF NON-ALCOHOLIC FATTY LIVER DISEASE AND BONE MINERAL DENSITY IN POSTMENOPAUSAL WOMEN

Abstract

Background

In recent years, it has been claimed that non-alcoholoic fatty liver disease (NAFLD) has a negative effect on bone mineral density (BMD) and causes osteoporosis. This study was designed to investigate the relationship between non-alcoholic fatty liver disease and BMD in postmenopausal women.

Patients and Methods

This study was designed prospectively: 118 postmenopausal women were included in the study. The study participants were divided into two groups as 59 people with NAFLD and 59 people without NAFLD. The groups were compared in terms of bone mineral density.

Results

The mean age of the participants in the study was 59.09 ± 9.22 years. According to the degree of NAFLD adiposity of the participants, 59 people were non-NAFLD (50%), 21 people were grade 1 (17.8%), 34 people were grade 2 (28.8%), and 4 people were grade 3 (3.4%). According to the Femur T score of the participants, 2 (1.7%) were osteoporosis, 34 (28.8%) osteopenia, 82 (69.5%) were normal. According to the L1-L4 T score of the participants, 18 (15.3%) were osteoporosis, 44 (37.3%) osteopenia, 56 (47.5%) normal. There was no statistical difference between the two groups in terms of femoral T score and L1-L4 T score. Age, BMI, GGT negatively affect the femoral T score.

Conclusion

We found that there was no difference in BMD between groups with and without NAFLD ultrasonographically, but there was a negative correlation between GGT and Femur T score in the subjects.This data suggest that NAFLD may cause osteoporosis”.

INTRODUCTION

Non–alcoholic fatty liver disease (NAFLD), alongside of increasing the obesity, has started to become the most important cause of liver cirrhosis, leaving behind viral and alcoholic causes in developed and developing countries. NAFLD has a wide spectrum ranging from simple steatosis to non–alcoholic steatohepatitis, which can progress to liver failure and hepatocellular carcinoma (1-3). Although NAFLD pathogenesis has not been fully elucidated, many intra and extrahepatic causes have been described (4).

It is now recognised that NAFLD has extrahepatic effects in addition to liver damage. In a study of 102 morbidly obese patients with different degrees of NAFLD, extrahepatic effects of NAFLD were mentioned. Further investigations have linked NAFLD with cardiovascular diseases (CVD), chronic kidney disease (CKD), Type 2 diabetes mellitus (T2DM), thyroid dysfunction, polycystic ovary syndrome, and obstructive sleep apnea syndrome (5).

Some recent studies have suggested an association between NAFLD and decreased bone mineral density (BMD). One study involving 1720 participants revealed a negative impact of NAFLD on BMD (6). The exact mechanism underlying the purported relationship between NAFLD and low BMD remains unclear. It has been reported that increased mechanical loading due to excess body weight in patients with NAFLD leads to cortical bone loss and a decrease in BMD (7). In addition, it is thought that increased visceral adipose tissue in NAFLD may lead to osteopenia and osteoporosis by causing chronic inflammation and release of proinflammatory, procoagulant and fibrinogenic mediators (8). Tumour necrosis factor alpha (TNF-α), which occurs as a result of chronic inflammation in NAFLD, causes stimulation of osteoclasts and inhibition of osteoblast activity from precursor cells, preventing bone destruction and bone rebuilding (9). In NAFLD, osteopontin, which increases bone destruction, is released in high amounts (10).

In recent years, it has been claimed that NAFLD has a negative effect on BMD and causes osteoporosis, but the issue is still unclear. This study was designed to investigate the relationship between NAFLD and osteoporosis in postmenopausal women.

MATERIALS AND METHODS

A total of 118 postmenopausal women who applied to the internal medicine outpatient clinic of our hospital between 01/11/2021 and 01/04/2022 were included in the study. During participant selection, individuals with the following criteria were excluded: viral hepatitis, autoimmune hepatitis, toxic hepatitis, cirrhosis, biliary tract diseases, cancer, use of medications affecting lipid profiles and liver enzymes, alcohol consumption (more than 20 gr/day for women), hyperparathyroidism, hyperthyroidism, malabsorption, chronic kidney failure, celiac disease, psoriasis, use of medications affecting BMD (such as glucocorticoids, thiazide diuretics, anticoagulants, anticonvulsants, estrogens), inflammatory bowel disease, and patients undergoing osteoporosis treatment (bisphosphonates, calcium). The study was approved by local ethics committee (ethics code: 2021/306).

Demographic data including age, gender, etc., were recorded for all participants. Also body mass index (BMI) was recorded.

Of the 118 participants included in the study, 59 were diagnosed ultrasonographically with hepatic steatosis, while the remaining 59 did not exhibit hepatic steatosis. These two groups were compared in terms of osteopenia and osteoporosis by assessing bone mineral density. Additionally, both groups were compared regarding BMI, Homeostasis Model Assessment – Insulin Resistance (HOMA – IR), aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma-glutamyl transferase (GGT), hemoglobin A1c (HbA1c), glucose, albumin, total bilirubin, total cholesterol, triglycerides, calcium, phosphorus, parathyroid hormone (PTH), (25-OH)vitamin D levels, and menopausal age. The results obtained were analysed statistically.

BMI (kg/m²): It was calculated with the formula of body weight (kg) / height (m²)(11).

HOMA – IR (mg/dl): It was calculated by the formula (glucose x insulin) / 405 (12).

Blood samples were collected between 08.00 and 10.00 in the morning after at least 8 hours of fasting. Blood samples were collected in SST II, LH PST II and EDTA tubes and analysed simultaneously.

Total cholesterol, calcium, phosphate and triglycerides were analysed by enzymatic colourimetric test.Parathormone was analysed by chemiluminescence method (Immulite, Bio-DPC, Los Angeles, CA, USA). (25-OH) vitamin D was analysed by radioimmunoassay (RIA kit, Diasorin Inc, Stillwater, MN, USA).

Fasting blood glucose and HBA1c levels were measured by Roche cobas "e 801 analyser. AST levels in sera were determined enzymatically-kinetically in an autoanalyser using Abbott brand kit (AST, 8L91-20 and 8L91-40 30- 4255/R3)). ALT values in sera were determined enzymatically-kinetically in an autoanalyzer using Abbott brand kit (ALT, 8L92-20 and 8L92-40 30-4106/R2). Roche commercial kit with catalogue number 03002721122 was used in the analysis.

Ultrasonographic (USG) measurements of fatty liver were performed with GE Healtcare brand VividTM S60N model ultrasound device C1-6-D convex probe and 11L-D linear probe. Fatty liver disease was determined by USG and grouped as grade 1, grade 2, grade 3 according to the degree of fatty liver disease.

BMD was measured by dual energy X-ray absorptiometry (DXA) and analysed for osteoporosis and osteopenia. DXA Hologic Horizon Wi (S/N 200894) device was used for this purpose.

BMD of the groups was determined by measuring femoral neck T score and lumbar spine T score. T score values lower than -2.5 were considered as osteoporosis. The presence of one of the lumbar L1-L4 vertebra or femur total T values at the osteoporotic limit was considered sufficient to diagnose osteoporosis. Values between -1 and -2.5 were considered osteopenic.

Statistical analysis

A power analysis was conducted using the G*Power (v3.1.9) software to determine the sample size, aiming for 80% power (1-β, where β represents the Type II error probability).

In statistical analysis, the data were analysed using Spss 25.0 package programme.The normality of the distribution of the data was checked by Kolmogorov Smirnov test.

In addition to descriptive statistical methods (mean, standard deviation, frequency), Student t test, One Way Anova, Pearson correlation analysis were used for parametric data. Chi-square and Mann Whitney U test, Kruskal Wallis test, Spearman correlation analysis were used for non-parametric data.

RESULTS

The study was carried out with 118 postmenopausal female patients who applied to the internal medicine outpatient clinic in our hospital. The study participants were 59 with NAFLD (50%) and 59 without NAFLD (50%). When the participants were analysed in terms of the degree of fatty liver disease, 59 were non-NAFLD (50%), 21 were grade 1 (17.8%), 34 were grade 2 (28.8%) and 4 were grade 3 fatty liver disease (3.4%).

When the femoral neck T score of the study participants was analysed, 2 (1.7%) had osteoporosis, 34 (28.8%) had osteopenia, and 82 (69.5%) had normal bone density. Similarly, when evaluating the lumbar spine T scores, 18 (15.3%) were identified as osteoporosis, 44 (37.3%) as osteopenia, and 56 (47.5%) as normal BMD (Table 1).

Table 1.

Frequency table of individuals participating in the study

		N	%
Group	Non – NAFLD With NAFLD	59 59	50.0% 50.0%
Liposis Degree	No NAFLD Grade 1 Grade 2 Grade 3	59 21 34 4	50.0% 17.8% 28.8% 3.4%
Femur T – Score	Osteoporosis Osteopenia Normal	2 34 82	1.7% 28.8% 69.5%
L1- L4 T – Score	Osteoporosis Osteopenia Normal	18 44 56	15.3% 37.3% 47.5%

The average age of the participants was 59.09 ± 9.22.

When both groups were analysed in terms of demographic characteristics and laboratory parameters, there was no statistically significant difference between the two groups in terms of age and age at menopause.

The group with NAFLD exhibited higher levels of BMI, ALT, AST, triglycerides, GGT, glucose, HbA1c, and HOMA-IR compared to the non-NAFLD group. Conversely, there were no statistically significant differences between the two groups in total cholesterol, LDL, albumin, calcium, phosphorus, PTH, total bilirubin, (25-OH) vitamin D, femoral neck T score, and lumbar spine T score (Table 2).

Table 2.

Demographic and Laboratory data of the patients

	Non – NAFLD	NAFLD	Total
	Ort±SS	Ort±SS	Ort±SS	P
Age	59.22±9.35	58.97±9.17	59.09±9.22	0.882*
Age of Menopause	45.22±4.73	45.73±6.02	45.47±5.40	0.611*
BMI	28.89±5.25	32.37±4.84	30.63±5.32	0.000*
ALT	20.08±13.34	28.68±18.86	24.38±16.82	0.001**
AST	18.86±6.10	23.34±11.45	21.10±9.41	0.044**
Total Cholesterol	211.85±36.85	215.10±37.14	213.47±36.87	0.634*
LDL – C	126.32±37.56	132.12±33.62	129.22±35.61	0.379*
Triglyceride	116.49±47.15	181.44±89.89	148.97±78.56	0.000**
Albumin	4.43±0.37	4.48±0.24	4.45±0.32	0.379**
Calcium	9.34±0.43	9.49±0.50	9.42±0.47	0.119**
Phosphorus	3.70±0.50	3.53±0.52	3.61±0.52	0.085*
PTH	49.23±19.80	49.31±22.88	49.27±21.30	0.929**
GGT	26.58±29.69	38.03±39.34	32.31±35.17	0.005**
Total Bilirubin	0.37±0.19	0.41±0.22	0.39±0.21	0.133**
Glucose	100.08±29.57	113.15±34.68	106.62±32.75	0.008**
HbA1C	5.77±0.73	6.33±1.30	6.05±1.09	0.005**
HOMA	2.59±1.73	5.02±5.38	3.81±4.16	0.000**
25-OHVitamin D	17.75±9.18	17.12±8.92	17.43±9.02	0.704*
Femur T-Score	-0.38±1.23	-0.18±1.05	-0.28±1.14	0.348
L 1- 4 T- Score	-0.96±1.39	-0.95±1.36	-0.95±1.37	0.947*

*Independent t test, **Mann-Whitney U. BMI: Body mass index, GGT: Gamma-glutamyl transferase, AST: Aspartate aminotransferase, ALT: Alanine aminotransferase, PTH: parathyroid hormone, LDL-C: LDL cholesterol.

When the participants in the study were divided into groups according to their degree of adiposity and compared in terms of femoral neck T scores and lumbar spine T scores, it was determined that there was no statistically significant difference (Table 3).

Table 3.

Relationship between Liposis Degree and L1-L4 T-Score

	Liposis Degree
	Non - NAFLD		GRADE 1		GRADE 2		GRADE 3
	N	%	N	%	N	%	N	%	X2	P
Osteoporosis	10	16.9	4	19.0	4	11.8	0	0.0	3.801	0.704
Osteopenia	22	37.3	10	47.6	11	32.4	1	25.0
Normal	27	45.8	7	33.3	19	55.9	3	75.0

*Ki-Kare.

Femural Neck T Score Regression Analysis Application:

Multivariate regression analysis showed that age, BMI and GGT were effective on femural neck T score. BMI value had the most significant effect on the model, followed by GGT value. While the effects of BMI and GGT were found to be significant at p<0.05 level, age was found to be effective in the model at p<0.001 level (Table 4).

Table 4.

Relationship between age, BMI and GGT value with Femur T score

Model	Unstandardized Coefficients		95.0% Confidence Interval for B
	B	P	Lower Bound	Upper Bound
(Constant)	-2.298	0.002	-3.712	-0.885
Age	-.027	0.004	-0.046	-0.009
BMI	.112	0.000	0.080	0.144
GGT	.006	0.017	0.001	0.011

BMI: Body mass index,GGT: Gamma-glutamyl transferase.

DISCUSSION

In this study, we investigated the relationship between bone mineral density (BMD) and NAFLD in postmenopausal women. We found that there was no difference in BMD between the groups with and without NAFLD ultrasonographically, but there was a negative association between GGT, a marker of fatty liver, and femoral neck T score in the study paticipants.

NAFLD is the most common liver disease in the world. With the increase in obesity, NAFLD has started to become the most important cause of liver cirrhosis, leaving behind the causes of alcohol and viral infections in developed and developing countries (1,2).

NAFLD has a wide spectrum ranging from simple steatosis to non–alcoholic steatohepatitis, which can progress to liver failure and hepatocellular carcinoma (3).

Obesity, which has become a pandemic in recent years, is also a risk factor for osteoporosis as well as NAFLD. In a large cohort study conducted in China by Hsu et al., they found a negative correlation between BMI and body fat mass and BMD. They found that the risk of osteoporosis, osteopenia, and non-vertebral fractures was significantly higher in obese people, that is, people with a higher body fat percentage (13). A meta-analysis conducted in Japan found that there was no significant difference in BMD between those with NAFDL and the control group, but BMI was the factor with the strongest and most significant effect on BMD (14). Similarly, in our study, we determined that there was a significant negative correlation between BMI and femoral neck and lumber spine T scores.

There is a lot of evidence in the literature that NAFLD is not only a liver disease but also a multisystemic pathology affecting many organs other than the liver. It has been found that NAFLD is frequently associated with many diseases such as T2DM, CVD, CKD, thyroid diseases, polycystic ovary syndrome, obstructive sleep apnea syndrome (15, 16). In recent years, osteoporosis is thought to be among the diseases accompanying NAFLD and various studies have been conducted on this subject.

In a cross-sectional study of 1659 men and postmenopausal women in China, Xia et al. found that BMD was significantly lower in the whole body and lumbar spine in people with NAFLD (17). Moon et al. examined whether BMD was associated with NAFLD in 481 premenopausal and postmenopausal women and determined that postmenopausal women with NAFLD had lower lumbar BMD than those without NAFLD (18). In a cross-sectional study conducted by Lee et al. in 6634 patients, they suggested that there was a significant negative relationship between NAFLD and femoral neck BMD in men and a positive relationship between NAFLD and lumber spine BMD in postmenopausal women. However, the use of hormone replacement therapy, corticosteroids, calcium and vitamin D, which affect bone metabolism, were not taken into account in this study by Lee et al. (19,20).In our study, we did not detect a significant difference in lumber spine BMD in postmenopausal women with and without NAFLD. However, patients taking corticosteroids, hormone replacement therapy, calcium and (25-OH)vitamin D, which may affect BMD, were not included in our study.

In a study by Kaya et al. found that patients with NAFLD had a higher lumber spine BMD than the control group. However, in this study, those in the NAFLD group had higher serum (25-OH)vitamin D levels, which explains the higher BMD in NAFLD patients. In our study, there was no significant difference between the lumber spine BMD and (25-OH)vitamin D levels of the subjects with and without NAFLD (21).

Cui et al. investigated the effects of NAFLD on BMD in men with NAFLD and postmenopausal women. They found that BMD was significantly lower in the NAFLD group compared to the control group in both genders. Although most clinical studies have shown a negative relationship between NAFLD and lumbar BMD, Cui’s study found no significant difference in lumbar spine BMD between NAFLD and control group. However, lower BMD was found in the right hip and femoral neck in men and in the right hip in women in the NAFLD group (22). Similarly, we did not find a difference in lumber spine BMD between the NAFLD and non-NAFLD groups in our study, but we found a negative correlation between fatty liver disease and femoral neck T scores.

Montovani et al. examined the relationship between the degree of fatty liver and BMD in postmenopausal female patients diagnosed with Type 2 DM in Italy. NAFLD diagnosis and staging were made by evaluating the patients with USG and Fibroscan. Femoral neck and lumber spine T scores were examined by grouping the patients according to the degree of fatty liver. They found no correlation between the degree of fatty liver disease and femoral neck and lumber spine BMD. (23). In our study, we also divided people with NAFLD into groups according to the degree of adiposity and compared their BMD, but we did not find a relationship between the degree of fatty liver disease and BMD.

In a cross-sectional study conducted by Umehara in the USA, no significant association was found between NAFLD and BMD, but in the same study, it was found that NAFLD patients with higher ALT levels had lower BMD levels (24). In our study, we found that ALT was significantly higher in patients with NAFLD, but we did not find a relationship between ALT and BMD.

Mild to moderate elevation of serum ALT, AST is the most common and sometimes the only laboratory abnormality found in NAFLD (25). Primarily elevated ALT, triglycerides and GGT appear to be the most sensitive biochemical indicators of the presence of hepatic steatosis (26). Elevated liver enzymes (ALT, AST, GGT) indicate increase in grade of NAFLD and progression towards end-stage liver disease (27). GGT is an important indicator of liver dysfunction, but its specificity is intermediate as its value is also increased in conditions that cause liver dysfunction (28). The fatty liver index (FLI) can be used as a simple indicator of fatty liver in the general population. The FLI is an algorithm based on BMI, waist circumference, triglycerides and GGT. The GGT value is included in the diagnostic algorithm FLI to detect the presence of steatosis. Its specificity was found to be high together with other markers. Haring et al. found high GGT levels in male patients with hepatic steatosis in their study (29). Additionally, Dongyeop et al. found relationship between GGT and osteoprosis in their study in Korea (30). Ho Jeong Do et al. found in their studies that there is a relationship between GGT and BMD.Thet found that this was especially evident in postmenopausal women and elderly men (31). In our study, we also found that there is a statistically significant negative correlation between GGT elevation and femoral neck T score.

The relationship between NAFLD and BMD is not entirely clear. It is thought that increased visceral adipose tissue and chronic inflammation play an important role in this regard and proinflammatory, procoagulant and profibrogenic cytokines secreted by visceral adipose tissue cause systemic abnormalities (16). Various proinflammatory cytokines, hormones and molecules that cause insulin resistance are released from increased visceral adipose tissue.

NAFLD causes chronic inflammation and triggers the release of TNF-α from the liver. TNF-α affects BMD by both stimulating osteocalsts and inhibiting osteoblast activation. Osteopontin is also released in excessive amounts in NAFLD. Osteoprotegerin normally prevents bone apoptosis by inhibiting osteoclast activation as a decoy for receptor activator of nuclear factor kappa-B (RANKL) ligand. Osteoprotegerin levels were found to be decreased in patients with NAFLD. Osteocalcin is synthesized by osteoblasts. Osteocalcin levels were also found to be decreased in patients with NAFLD (32). All of these mechanisms may contribute to the development of osteoporosis in NAFLD.

Our study has some limitations.It is a cross-sectional study. Since it was not prospective randomized, a definite cause - effect relationship could not be established. A prospective further study is needed for this.

In conclusion, with the rapid increase in obesity in the world, NAFLD and its accompanying diseases are increasing day by day. The fact that osteoporosis is among the diseases accompanying NAFLD has started to attract attention and in recent years several studies have been published this subject. In this study, we investigated the relationship between NAFLD and osteoporosis in postmenopausal women. As a result, we found that there was no difference in BMD between the groups with and without NAFLD ultrasonographically, but there was a negative correlation between NAFLD and femoral neck T score in the subjects included in the study.

Conflict of Interest

The authors declare that they have no conflict of interest.