Prevalence and Related Factors of Non-alcoholic Fatty Liver Disease among the Elderly in Taiwan

European Geriatric Medicine 4 (2013) 78–81

Available online at

www.sciencedirect.com

Research paper

Prevalence and related factors of non-alcoholic fatty liver disease among

the elderly in Taiwan

S.-C. Hung

, S.-W. Lai

*,

M.-C. Chen

, P.-C. Li

, K.-C. Lin

Department of Emergency Medicine, Nantou Hospital, 504 Nantou, Taiwan Department of Surgery, Nantou Hospital, 504 Nantou, Taiwan

cDepartment of Public Health, China Medical University, 404 Taichung, Taiwan

dSchool of Medicine, China Medical University, 404 Taichung, Taiwan

eDepartment of Family Medicine, China Medical University Hospital, 2, Yuh-Der Road, 404 Taichung, Taiwan

b A R T I C L E I N F O A B S T R A C T Article history: Received 31 August 2012 Accepted 10 November 2012 Available online 21 December 2012 Keywords:

Alanine aminotransferase Hyperglycemia Hypertriglyceridemia Obesity

Purpose: The elderly with non-alcoholic fatty liver disease (NAFLD) are more at risk for cirrhosis, plus even higher mortality. However, few studies of NAFLD focused on the elderly. This study explores prevalence of NAFLD to determine related factors among the elderly in Taiwan.

Materials and methods: Hospital-based, retrospective cross-sectional study analyzed medical records of elderly patients receiving periodic health examination at a medical center in Taichung, Taiwan, during 2000 to 2004. Subjects with alcoholism, hepatitis B surface antigen positive, or hepatitis C virus antibody positive were excluded. Each subject received anthropometric measurement, biochemical testing, and abdominal ultrasonography; 521 underwent further analysis.

Results: There were 258 men (49.5%) and 263 women (50.5%), mean age 70.3 Æ 4.6 years (range 65–87). Overall NAFLD prevalence was 50.1%, no significant difference between men and women (46.1% vs. 54.0%, P > 0.05). Controlling for covariates, multivariate logistic regression pinpointed factors for NAFLD: generalized obesity (OR = 2.59, 95% CI = 1.50–4.48), central obesity (OR = 1.85, 95% CI = 1.20–2.87), hyperglycemia (OR = 1.50, 95% CI = 1.01–2.23), hypertriglyceridemia (OR = 2.00, 95% CI = 1.22–3.30), and elevated alanine aminotransferase (ALT) (OR = 6.43, 95% CI = 2.00–20.61). Smoking proved negatively correlated with NAFLD (OR = 0.55, 95% CI = 0.31–0.97).

Conclusion: This study discloses NAFLD prevalence among the elderly in Taiwan as above 50%. Related factors include obesity, hypertriglyceridemia, hyperglycemia, and elevated ALT. Despite the high prevalence in the elderly population, we identified no unique risk factors when comparing with other Oriental and European studies.

ß 2012 Elsevier Masson SAS and European Union Geriatric Medicine Society. All rights reserved.

1. Introduction

As with most developed countries, aging looms as an important medical issue in Taiwan. Nationwide statistics showed over 10% elderly among 23 million citizens at the end of 2009. Geriatrics, focused on health care and health promotion in the elderly, poses new medical challenges. Aging liver, with change of structure and function, constitutes a different threat from that in young patients

[1]. Fatty liver disease, once ignored as simple benign liver status, shows gradual impact on elderly health [2]. With similar gross and histological appearance [3], the disorder is usually classified as alcoholic fatty liver disease and non-alcoholic. As denominated, non-alcoholic fatty liver disease (NAFLD) is thought not related to alcohol consumption but to multiple causes, including metabolic factors [3]. Because NAFLD has the potential to progress to cirrhosis, hepatocellular carcinoma, and finally death, identifica-tion of individuals at risk of NAFLD is of greater concern [4]. In the past, few studies of NAFLD focused on the elderly; now, with the elderly as the major group for metabolic syndrome, and patients with NAFLD also thought at risk for cirrhosis and hepatocellular carcinoma, we are interested in prevalence of NAFLD in the elderly, as well as its related factors.

This study explores prevalence of NAFLD to determine related factors among the elderly in Taiwan.

2. Materials and methods 2.1. Study population

This hospital-based, cross-sectional study analyzed medical records of older people undergoing self-referred health examina-tion at one medical center located at Taichung city in Taiwan

* Corresponding author. Tel.: +886 4 2206 2121x4507. E-mail address: wei@mail.cmuh.org.tw (S.-W. Lai).

1878-7649/$ – see front matter ß 2012 Elsevier Masson SAS and European Union Geriatric Medicine Society. All rights reserved.

http://dx.doi.org/10.1016/j.eurger.2012.11.004

Author's personal copy

during 2000 to 2004. The self-referred health examination was not a routine procedure. It was optional to all adults who concerned their own health status. Usually, people undergoing self-referred health examination were without specific medical complain. It was not covered by National Health Insurance in Taiwan, either. The institutional review board of this medical center approved this retrospective study. Subjects with previous malignant diseases, with habitually drinking alcohol, with hepatitis B surface antigen or with hepatitis C virus antibody were excluded. The laboratory test results were rechecked and life style interviews were completed by well-trained staff and physicians. In all, 521 subjects were included for further analysis.

2.2. Data collection

Table 1

Basic characteristics of the study population. Men

n (%)

Age (years) (mean Æ SD) Fatty liver No Yes Total * Women n (%) 70.6 Æ 4.9 121 (46.0) 142 (54.0)* 263 Total n (%) 70.3 Æ 4.6 260 (49.9) 261 (50.1) 521 70.0 Æ 4.4 139 (53.9) 119 (46.1)* 258

P = 0.073, by the Chi2 test.

3. Results

3.1. Characteristics of study population

Gastroenterologists using high resolution real-time machine performed abdominal ultrasonography (Toshiba Sonolayer SSA-270A, convex-type 3.5 MHz transducer, Tochigi-Ken, Japan). Fatty liver was diagnosed according to international criteria [5,6], blood pressure measured by a mercury sphygmomanometer with the subject in a sitting position. Weight and height were recorded, body mass index (BMI) measured as follows: weight (kg)/height (m2). Waist circumference (WC) was measured as the minimum

circumference with tape positioned between xyphoid process and umbilicus at the end of a normal expiration. Venous blood samples were obtained in the morning after 12-hour overnight fast. Biochemical markers like total cholesterol, triglyceride, low-density lipoprotein cholesterol (LDL), high-low-density lipoprotein cholesterol (HDL), fasting glucose, uric acid, aspartate amino-transferase (AST), alanine aminoamino-transferase (ALT), alkaline phosphatase (ALP), and gamma-glutamyl transpeptidase (GGT) were gauged by biochemical autoanalyser (Hitachi 736-15, Tokyo, Japan) at the Clinical Laboratory Department of this medical center, Hepatitis B surface antigen detected by Elisa test (Enzygnost, Dade Behring Marburg GmbH, Marburg, Germany), antibody to hepatitis C virus detected by EIA test (Abbott HCV EIA, third generation).

2.3. Diagnostic criteria

BMI ! 27 (kg/m2) was defined as generalized obesity. Central

obesity was defined as WC ! 90 cm for men, ! 80 cm for women; hypercholesterolemia, as fasting total cholesterol level ! 5.2 mmol/L, hypertriglyceridemia as fasting triglyceride level ! 1.7 m-mmol/L. High-level LDL was defined as fasting LDL ! 3.4 mmol/L; low-level HDL, as fasting HDL < 1.03 mmol/L for men

and < 1.3 mmol/L for women. The ratio of TC/HDL > 5 was defined as abnormal. Diabetes mellitus was defined as fasting glucose level ! 6.9 mmol/L or subjects with antidiabetic drug treatment. Subjects were considered to have hypertension if average of both arm readings exceeded 140 mmHg systolically and/or 90 mmHg diastolically or subjects with antihypertensive drug treatment. Hyperuricemia was defined as serum uric acid ! 420

m

without NAFLD by the t test (P = 0.114) (Table 2).

3.3. Related factors for non-alcoholic fatty liver disease by multivariate logistic regression

Only the significantly related factors identified in univariate analysis were further analyzed. After controlling for covariates, multivariate logistic regression proved significantly related factors for NAFLD as generalized obesity (OR = 2.59, 95% CI = 1.50–4.48, P < 0.001); central obesity (OR = 1.85, 95% CI = 1.20–2.87, P < 0.001); hyperglycemia (OR = 1.50, 95% CI = 1.01–2.23, P < 0.05); hypertriglyceridemia (OR = 2.00, 95% CI = 1.22–3.30, P < 0.001); elevated ALT (OR = 6.43, 95% CI = 2.00–20.61, P < 0.001). Smoking was negatively related to risk of NAFLD (OR = 0.55, 95% CI = 0.31–0.97, P < 0.05) (Table 3).

4. Discussion

It is difficult to estimate prevalence of NAFLD accurately, since histology remains the gold standard of diagnosis [8]. Ultrasonog-raphy and liver function test now serve as tools to evaluate the presence of NAFLD, precluding risk of liver biopsy. At the same time, it is difficult to figure simple steatosis from steatohepatitis in this situation. Therefore, prevalence of NAFLD is still discussed in recent decades. Prevalence of 15.9% in Northern China adults was reported [9]. A study conducted in Chinese taxi driver in Taiwan reported high NAFLD prevalence in the elderly [10]. There were also NAFLD studies in Europe. Spain reported the prevalence of 33.4% in men and 20.3% in women in general adult populations

[11]. For elderly population, prevalence of 35.1% was reported from Rotterdam [12]. Our figures show prevalence of NAFLD as 50.1% without statistically significant difference between genders, far higher than other studies with general adult and other population groups. Some studies also found different prevalence between genders, and increasing prevalence after menopause is also

Author's personal copy

S.-C. Hung et al. / European Geriatric Medicine 4 (2013) 78–81 Table 3

Odds ratio of related factors for non-alcoholic fatty liver disease by multivariate logistic regression. Variable 0.114 < 0.001 234 (90.0) 26 (10.0) 183 (70.1) 78 (29.9) < 0.001 131 (50.4) 129 (49.6) 59 (22.6) 202 (77.4) 0.025 130 (50.0) 130 (50.0) 105 (40.2) 156 (59.8) < 0.001 139 (53.5) 121 (46.5) 94 (36.0) 167 (64.0) 0.047 121 (46.5) 139 (53.5) 99 (37.9) 162 (62.1) < 0.001 219 (84.2) 41 (15.8) 169 (64.8) 92 (35.2) 0.105 142 (54.6) 118 (45.4) 124 (47.5) 137 (52.5) < 0.001 158 (60.8) 102 (39.2) 115 (44.1) 146 (55.9) 0.001 190 (73.1) 70 (26.9) 155 (59.4) 106 (40.6) 0.001 186 (71.5) 74 (28.5) 149 (57.1) 112 (42.9) 0.004 232 (89.2) 28 (10.8) 209 (80.1) 52 (19.9) < 0.001 255 (98.1) 5 (1.9) 22 5 (86.2) 36 (13.8) 0.100 255 (98.1) 5 (1.9) 260 (99.6) 1 (0.4) 0.011 238 (91.5) 22 (8.5) 219 (84.2) 41 (15.8) 0.009 206 (79.2) 54 (20.8) 229 (87.7) 32 (12.3) Generalized obesity Central obesity Hyperglycemia Hypertriglyceridemia Elevated ALT Smoke use

ALT: alanine aminotransferase. *P < 0.05.

**P < 0.001. OR 2.59 1.85

Yes Hypertriglyceridemia No Yes High level of LDL No Yes Low level of HDL No Yes TC/HDL ratio > 5 No Yes Hyperuricemia No Yes Elevated AST No Yes Elevated ALT No Yes Elevated ALP (%) No Yes Elevated GGT No Yes Smoke use No Yes Normal n (%) 70.6 Æ 4.8 Fatty liver n (%) 70.0 Æ 4.5 P value

LDL: low-density lipoprotein cholesterol; HDL: high-density lipoprotein cholester-ol; TC/HDL: ratio of total cholesterol to high-density lipoprotein cholestercholester-ol; AST: aspartate aminotransferase; ALT: alanine aminotransferase; ALP: alkaline phos-phatase; GGT: gamma-glutamyl transpeptidase.

discussed in women [13]. In pediatric population, male sex is an independent predictor of fatty liver. In general adult population, prevalence in both genders seems equal [3,14]. Our study focused on the elderly population, with prevalence of NAFLD similar in both genders.

Other reported risk factors of NAFLD included gene [15],

obesity, hypertension, hyperuricemia, higher AST, higher ALT, hypertriglyceridemia, higher fasting plasma glucose and diabetes in Chinese and Taiwan population [9,10], as well as male sex, age, metabolic syndrome, insulin resistance, ALT, total physical activity level, pack years of smoking, WC, fasting glucose, elevated blood pressure in Europe Studies [11,12].

The elderly with NAFLD are more at risk for cirrhosis, plus even higher mortality [16]. Metabolic syndrome, mainly resulting from insulin resistance, is thought strongly related to NAFLD in adults and children [13,17]. For elderly with NAFLD, hypertriglyceridemia seems to play a more important role then hypercholesterolemia and high-level LDL. Some studies focused on adults and children also figure effect of hypertriglyceridemia on NAFLD [13,18]. On the other hand, we paid more attention to central than generalized obesity when discussing metabolic syndrome, since we found both related to NAFLD in the elderly. Metabolic syndrome is discussed in relation to cardiovascular disease, and patients with NAFLD now show higher prevalence of coronary heart disease [19,20]. We then correlated NAFLD with hypertension, a crucial risk factor for cardiovascular disease; no obvious relationship emerged. Smoking was negatively correlated with risk of NAFLD in the elderly in our study. Lipid metabolic condition, important risk factors for NAFLD improved, even with weight gain when smoking was discontinued [21]. Future study can clarify correlation among smoking, NAFLD, metabolic syndrome, and cardiovascular disease. Serum uric acid is mentioned frequently when discussing metabolic syndrome and cardiovascular disease. Some studies correlate hyperuricemia with NAFLD [22]. In our study, hyperuri-cemia seems not an independent related factor to NAFLD by multivariate logistic regression. We also figured elevated ALT as more closely related to NAFLD than other hepatobiliary biomark-ers: e.g., elevated AST, ALP and GGT in the elderly. It may be because of ALT is a much more specific indicator than AST in hepatitis [23]; patients diagnosed with alcoholic hepatitis exhibit higher AST than ALT. NAFLD thus appears to elevate ALT, even with a histological picture quite similar [23–25] to that of alcoholic hepatitis. Yet some studies cite AST/ALT ratio above 1 portending severe liver fibrosis in non-alcoholic steatohepatitis [26]; elevated ALT level could not stand for the extent of liver damage [3]. ALT cutoff values are discussed too [3,27]. Studies of hepatic biomarkers in NAFLD in the elderly, as well as elevated and reduced hepatic biomarkers throughout the course of NAFLD, from simple steatosis to severe steatohepatitis and end-stage liver disease, are also needed in the future.

Except for hepatitis viruses’ infection, alcohol is usually cited as the major cause leading to cirrhosis and hepatocellular carcinoma, even though some environmental substances may likewise show hepatic toxicity [23]. Hence, we educate people to quit alcohol, to receive vaccination and decrease prevalence of life-threatening conditions. We also promote essential lifestyle changes for cholesterol control and reducing CHD risk, life style modification, Dietary Approaches to Stop Hypertension (DASH) eating plan to prevent and manage hypertension. So far, there is no approved pharmacologic therapy for NAFLD in children and adults [26,28].

We may need lifestyle modification for NAFLD [28,29], along with more education about NAFLD for the general population [30].

Author's personal copy

Age is an independent risk factor for severe fibrosis in NAFLD. To understand the related factors of NAFLD in the elderly is important. Then we could have the chance to decrease the prevalence of fibrosis in the elderly.

There were some limitations in our study. Subjects are all older people undergoing self-referred health examination at one medical center located at Taichung City, Taiwan. There are diversity of lifestyles and dietary habits in different areas. Hospital-based study can thus not be extrapolated to the general population. Whereas histology remains the gold standard for pathologic diagnosis of NAFLD, imaging modalities like ultrasonography with reasonably high sensitivity can identify NAFLD. This study diagnosed NAFLD via ultrasonography. Ethical reasons preclude biopsy in epidemiological study, nor is influence of exercise taken into account. Larger studies with a more representative sample are required for more accurate description of related factors of NAFLD. 5. Conclusion

Prevalence of NAFLD is high in the elderly, both in male and female, far higher than other studies with general adult and other population groups. The elderly with NAFLD are under the risk of cirrhosis and probably hepatocellular carcinoma, and should not be ignored anymore. Although no unique risk factors of NAFLD were identified among the elderly population in Taiwan, the study confirmed factors, including generalized obesity, central obesity, hyperglycemia, hypertriglyceridemia are related to NAFLD. Patients with hypertriglyceridemia were more risky then hyper-cholesterolemia and high-level LDL. Physicians should practice to control risk factors. Elevated ALT was a better following indicator for NAFLD then AST.

Disclosure of interest

The authors declare that they have no conflicts of interest concerning this article.

References

[1] Floreani A. Liver disorders in the elderly. Best Pract Res Clin Gastroenterol 2009;23:909–17.

[2] O’Connor BJ, Kathamna B, Tavill AS. Nonalcoholic fatty liver (NASH syndrome). Gastroenterologist 1997;5:316–29.

[3] Choudhury J, Sanyal AJ. Clinical aspects of fatty liver disease. Semin Liver Dis 2004;24:349–62.

[4] Sugimoto K, Takei Y. Clinicopathological features of non-alcoholic fatty liver disease. Hepatol Res 2011;41:911–20.

[5] Chang WY, Chen CJ, Lu SN, You SL, Chuang WL, Chen SC, et al. Relationship between fatty liver, alanine aminotransferase, HBsAg and hepatitis C virus. J Gastroenterol Hepatol 1992;7:455–8.

[6] Lin DY, Sheen IS, Chiu CT. Clinical significance of ultrasonographic fatty liver in asymtomatics: analysis of 1040 check-up subjects. J Med Ultrasound 1993;1:165–71.

[7] Lai SW, Liao KF. Related factors for rectosigmoid hyperplastic polyps: a hospital-based study. Kuwait Med J 2011;43:208–13.

[8] Clark JM, Diehl AM. Nonalcoholic fatty liver disease: an underrecognized cause of cryptogenic cirrhosis. JAMA 2003;289:3000–4.

[9] He S, Bao W, Shao M, Wang W, Wang C, Sun J, et al. Risk factors for alcoholic fatty liver disease in a Chinese population. Acta Gastroenterol Belg 2011;74:503–8.

[10] Tung TH, Chang TH, Chiu WH, Lin TH, Shih HC, Chang MH, et al. Clinical correlation of nonalcoholic fatty liver disease in a Chinese taxi drivers lation in Taiwan: Experience at a teaching hospital. BMC Res Notes 2011;4:315.

[11] Caballeria L, Pera G, Auladell MA, Toran P, Munoz L, Miranda D, et al. lence and factors associated with the presence of nonalcoholic fatty liver disease in an adult population in Spain. Eur J Gastroenterol Hepatol 2010;22:24–32.

[12] Koehler EM, Schouten JN, Hansen BE, van Rooij FJ, Hofman A, Stricker BH, et al. Prevalence and risk factors of non-alcoholic fatty liver disease in the elderly: Results from the Rotterdam study. J Hepatol 2012;57:1305–11.

[13] Clark JM. The epidemiology of nonalcoholic fatty liver disease in adults. J Clin Gastroenterol 2006;40:S5–10.

[14] Adams LA, Lymp JF, St Sauver J, Sanderson SO, Lindor KD, Feldstein A, et al. The natural history of nonalcoholic fatty liver disease: a population-based cohort study. Gastroenterology 2005;129:113–21.

[15] Lin YC, Chang PF, Hu FC, Yang WS, Chang MH, Ni YH. A common variant in the PNPLA3 gene is a risk factor for non-alcoholic fatty liver disease in obese Taiwanese children. J Pediatr 2011;158:740–4.

[16] Frith J, Day CP, Henderson E, Burt AD, Newton JL. Non-alcoholic fatty liver disease in older people. Gerontology 2009;55:607–13.

[17] Tominaga K, Fujimoto E, Suzuki K, Hayashi M, Ichikawa M, Inaba Y. Prevalence of non-alcoholic fatty liver disease in children and relationship to metabolic syndrome, insulin resistance, and waist circumference. Environ Health Prev Med 2009;14:142–9.

[18] Tessari P, Coracina A, Cosma A, Tiengo A. Hepatic lipid metabolism and alcoholic fatty liver disease. Nutr Metab Cardiovasc Dis 2009;19:291–302. [19] Chen CH, Nien CK, Yang CC, Yeh YH. Association between nonalcoholic fatty liver disease and coronary artery calcification. Dig Dis Sci 2010;55:1752–60.