Prerpints.org logo
Preprint
Article

Smoking and Risk of Fatty Liver Disease: A Meta-Analysis of Cohort Studies

Altmetrics

Downloads

42

Views

33

Comments

0

Submitted:

30 October 2024

Posted:

31 October 2024

You are already at the latest version

Alerts
Abstract
It remains inconclusive whether smoking is associated with an increased risk of fatty liver dis-ease (FLD). We investigated the association between smoking and the risk of FLD by using a meta-analysis of cohort studies. PubMed and EMBASE were searched using keywords from in-ception to September 2023 to identify relevant studies. Out of 806 articles searched from data-bases, a total of 20 cohort studies were included in the final analysis. In the meta-analysis, smoking was significantly associated with an increased risk of FLD (odds ratio/relative risk/hazard ratio, 1.14; 95% confidence interval, 1.05 – 1.24; n = 20). Subgroup analyses showed a significant positive association between them in prospective cohort studies (odds ratio/relative risk/hazard ratio, 1.15; 95% confidence interval, 1.05 – 1.18; n = 5), but not in retrospective cohort studies and cross-sectional studies based on cohort studies. In the subgroup meta-analysis by gender in Asians, smoking significantly increased the risk of FLD in men, while there was no significant association between them in women. This meta-analysis showed that smoking in-creases the risk of FLD. In addition to well-known risk factors of FLD such as obesity and alcohol consumption, clinicians should recommend smoking cessation for the management of FLD.
Keywords: 
Subject: Medicine and Pharmacology  -   Gastroenterology and Hepatology

1. Introduction

Fatty liver disease (FLD), which is categorized into two major types such as nonalcoholic FLD (NAFLD) or alcoholic FLD (AFLD), indicates a morphological spectrum consisting of hepatic steatosis and steatohepatitis [1]. It is a persistent liver condition marked by macrovesicular steatosis in liver cells, which has the potential to advance to hepatic cirrhosis, liver failure, and possibly hepatocellular carcinoma [1]. Ethanol consumption serves as a pivotal determinant in distinguishing between NAFLD and AFLD in the guidelines from the European Association for the Study of the Liver [2]. NAFLD diagnosis involves considering ethanol intake of 20 g/d or less in females and 30 g/d or less in males, following the thorough exclusion of alternative causes such as hepatitis virus infection and the use of steatogenic drugs [3]. Beyond alcohol-related impacts, the pathogenesis of FLD is intricately linked to various contributors, including insulin resistance (IR), oxidative stress, mitochondrial dysfunction, immune system deregulation, and the release of adipokines [3,4]. These factors collectively underscore the complexity of FLD, emphasizing the importance of a comprehensive understanding to address its progression and associated risks such as hepatic cirrhosis, liver failure, and hepatocellular carcinoma [5].
The overall prevalence of NAFLD worldwide also has increased from 25.5% in or before 2005 to 37.8% in 2016 or later based on the report from a recent meta-analysis published in 2022 [6]. Also, FLD has become a predominant chronic liver disorder in developed Western countries [7]. Its risk factors include a higher body mass index (BMI), consumption of saturated fat and fructose, type 2 diabetes, and known single nucleotide polymorphisms [8].
However, it remains unclear whether smoking is associated with an increased risk of FLD. An animal study has reported that smoking increases lipid accumulation in hepatocytes by modulating an activity of critical molecules related with lipid synthesis [9]. Also, another animal study has shown that the histological severity of NAFLD in obese rats was exacerbated by tobacco exposure [10].
In the meantime, observational epidemiological studies [11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30] have reported inconsistent findings. Several cohort studies [14,17,18,20,23,24,27,28,29] have reported that smoking was associated with an increased risk of FLD, whereas other cohort studies [11,12,13,15,16,19,22,26] have reported no significant association between them.
In 2018, the only meta-analysis of observational studies has reported that smoking was significantly associated with the increased risk of NAFLD [31]. However, it included a small number of cohort studies to confirm the association, and subsequent cohort studies have been published since then. Furthermore, to our knowledge, no meta-analysis of cohort studies has been published regarding the association between smoking and the risk of FLD encompassing NAFLD and AFLD.
This study aimed to explore the associations between smoking and the risk of FLD by using a comprehensive meta-analysis of cohort studies and subgroup meta-analyses by important factors.

2. Materials and Methods

2.1. Search Strategy

PubMed and EMBASE were searched in September 2023 with terms of the National Library of Medicine (NLM) Medical Subject Headings (MeSH) and commonly used keywords. We used a PICO framework to combine search terms: P for population is any type of population; I for intervention is ‘smoking; C for comparison is ‘non-smoker’; and O for outcome is ‘fatty liver disease’. Also, the study type was confined to cohort studies. Thus, the final search terms were ‘smoking’, ‘fatty liver disease’, and ‘cohort study’.

2.2. Study Selection and Data Extraction

We included a cohort study that explored the associations between smoking and FLD (NAFLD or AFLD) and presented risk estimates such as odds ratio (OR), relative risk (RR), or hazard ratio (HR) with their corresponding 95% confidence intervals (CI). Two independent authors (MH. Lee and SH. Lee) conducted a selection of relevant studies by reviewing titles and abstracts. Discrepancies between them were resolved through discussion. The extracted information included the last name of the first author, publication year, study region, study design (prospective or retrospective cohort study), gender, study participants, comparison of exposure, risk estimates (OR, RR, and HR with corresponding 95% CIs), type of outcomes, and adjusted variables.

2.3. Assessment of Methodological Quality

We used the Newcastle-Ottawa Scale (NOS) in order to assess the methodological quality of the cohort studies included in the current meta-analysis [32]. The NOS comprises eight items and provides a scoring system ranging between 0 and 9. We classified individual cohort studies as having high or low quality based on the mean score.

2.4. Main and Subgroup Analyses

We investigated the association between smoking and the risks of FLD for the main analysis. We also conducted subgroup meta-analyses by type of study (prospective or retrospective cohort study), region (Europe, Asia, or US), type of FLD, gender (male or female), follow-up period (<5 years or >5 years), and study quality (high or low quality).

2.5. Statistical Analysis

A combined OR, RR, or HR with its corresponding 95% CIs was calculated utilizing the adjusted OR, RR, or HR and their respective 95% CIs from each study that reported the association between smoking and the risk of FLD. The DerSimonian and Laird method [33] was employed, opting for a random-effects model due to the diverse populations across studies. Heterogeneity was evaluated using Higgins I2 computed as follows:
I2 = 100% × (Q – df)/Q,
where Q is Cochran’s statistic for heterogeneity, and df is degrees of freedom [32]. I2 values range between 0% (no heterogeneity) and 100% (maximal heterogeneity) [32]. Publication bias was assessed using both Begg’s funnel plot and Egger’s test. We used the STATA SE version 15.1 software package (StataCorp, College Station, TX, USA) for statistical analyses.

3. Results

3.1. Study Selection

Figure 1 shows the diagram of identifying relevant studies. A total of 806 studies were identified by the initial search of PubMed and EMBASE databases by using keywords. After excluding duplicates, 622 articles were screened based on the review of each title and abstract. After excluding 546 articles according to the predetermined selection criteria, 76 full-text articles were reviewed and assessed for eligibility. Among them, 56 articles were excluded for the following reasons: not relevant (n = 46); not cohort studies (n = 4); identical population (n = 4); and insufficient data (n = 2). The final analysis included 20 cohort studies (Figure 1) [11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30].

3.2. General Characteristics of Included Studies

Table 1 shows general characteristics of the cohort studies included in the final analysis. Eight studies were conducted in Europe, 10 were conducted in Asia, and the remaining two studies were conducted in the US. Types of study designs are prospective cohort studies (n = 5), retrospective cohort studies (n = 6), and cross-sectional surveys based on cohort studies (n = 9). Types of outcomes are FLD (n = 6) and NAFLD (n = 14).

3.3. Methodological Quality of Studies

The average score of the 11 prospective and retrospective cohort studies assessed using the NOS was 7.5 (Table 2). Each study’s methodological quality was categorized as either high (a score of ≥8) or low (a score of <8). Out of the 11 studies, six were classified as high-quality studies, while the remaining 5 were categorized as low-quality studies.

3.4. Association Between Smoking and Risk of FLD

In the meta-analysis of all the included studies, smoking was associated with an increased risk of FLD (OR/RR/HR = 1.14; 95% CI, 1.05 – 1.24; n = 20) (Figure 2).

3.5. Subgroup Meta-Analyses

Table 3 shows the associations between smoking and the risk of FLD in the subgroup meta-analysis by various factors. In the subgroup meta-analysis by study design, smoking increased the risk of FLD in prospective cohort studies (HR = 1.15; 95% CI 1.05 – 1.18; n = 5), but not in retrospective cohort studies (OR/RR = 1.23; 95% CI 0.94 – 1.62) and cross-sectional surveys (OR/RR = 1.12; 95% CI 0.92-1.46; n = 9). In the subgroup meta-analysis by study region, smoking was significantly associated with an increased risk of FLD in in Europe (OR/RR/HR = 1.32; 95% CI 1.16 – 1.50; n = 8), but not in Asia (OR/RR/HR = 1.03; 95% CI 0.91 – 1.18; n = 10) and the US (OR/RR/HR = 0.75; 95% CI 0.28 – 2.06; n = 2). In the subgroup meta-analysis by gender, which data are available only for Asian studies, smoking significantly increased the risk of FLD in men (OR/RR/HR = 1.15; 95% CI 1.06 – 1.25; n = 4), while there was no significant association between them in women (OR/RR/HR = 1.12; 95% CI 0.94 – 1.34; n = 4).
Regardless of type of FLD, smoking consistently increased the risk of FLD. On the contrary, subgroup meta-analyses by follow-up period and study quality showed no significant association between smoking and the risk of FLD.

3.6. Publication Bias

Both the Begg’s funnel plots (Figure 3) and Egger’s test (P = 0.29) did not show publication bias.

4. Discussion

In this meta-analysis of cohort studies, we found that smoking was significantly associated with an increased risk of FLD. In the subgroup meta-analysis by study design, smoking increased the risk of FLD in prospective cohort studies, but not in retrospective cohort studies and cross-sectional surveys. Also, smoking was significantly associated with an increased risk of FLD in Europe, but not in Asia and the US. Interestingly, in the subgroup meta-analysis by gender in Asians, smoking significantly increased the risk of FLD in men, while there was no significant association between them in women.
There are several possible biological mechanisms that could explain the increased risk of FLD by smoking. First, Yuan et al.’s study demonstrated that smoking stimulated lipid accumulation in hepatocytes in mice and cultured hepatocytes [9]. When mice and cultured hepatocytes were exposed to sidestream whole smoke, lipid accumulation was increased by modulating the activity of AMP-activated protein kinase and sterol response element binding protein-1, which are critical molecules in lipid synthesis [9]. Second, smoking has the potential to induce insulin resistance, which leads to the development of NAFLD [34,35]. A study revealed that cigarette consumption is correlated with degree of insulin resistance in smokers [36]. Also, a study in biopsy-proven NAFLD patients and health control subjects showed that homeostasis model assessment of insulin resistance (HOMA-IR) levels were significantly higher in the NAFLD patients [35]. Insulin resistance induced by smoking could enhance hepatic fat accumulation through increasing free fatty acid delivery to the liver and through hyperinsulinemia to contribute to triacylglycerol accumulation in the liver [37]. Third, nicotine in tobacco smoke could increase the release of norepinephrine and epinephrine, which could affect thermogenesis in adipose tissue, leading to the increased lipolysis and the subsequent recycling of fatty acids into triglycerides [38]. Consequently, it may contribute to the development of NAFTD. Lastly, adiponectin could inhibit liver fat deposition, and gluthathione peroxidase (GPx) could reduce lipid and hydrogen peroxide [39]. Thus, it has been suggested that cigarette smoking in combination with single-nucleotide polymorphisms in the adiponectine gene and GPx1 gene mutation could contribute to the development of NAFLD [39].
Previously, a meta-analysis of 12 observational studies by Akhavan et al. has already reported that smoking moderately increased a risk NAFLD [31]. However, it included only two cohort studies, and the remaining studies were seven cross-sectional and three case-control studies. Although it is possible to combine different study designs such as cross-sectional, case-control, and cohort studies when conducting a meta-analysis, conducting subgroup meta-analyses by study designs is crucial and useful because there could be discrepancies in findings between different study designs. Also, based on the ‘levels of evidence pyramid’, cohort studies give us a higher level of evidence than cross-sectional and case-control studies [40]. Akhavan et al. performed subgroup meta-analyses by study designs and reported significant increased risks of NAFLD by smoking in all the subgroup meta-analyses by each study design such as cross-sectional, case-control, and cohort studies [31]. However, they included just two cohort studies, which are not enough to confirm the association between smoking and the risk of NAFLD. In the current meta-analysis, we included 11 cohort studies involving five prospective and six retrospective cohort studies and nine cross-sectional surveys based on cohort studies. Thus, our findings would provide more clear and convincing evidence on this topic.
A notable strength lies in the subgroup meta-analyses by various factors. We confirmed a significantly increased risk of FLD in prospective cohort studies, which are generally considered as having a higher level of evidence than retrospective cohort studies and cross-sectional studies although retrospective and cross-sectional studies in the current analysis showed no significant association between smoking and the risk of FLD.
Interestingly, a significantly increased risk of FLD by smoking was observed in the studies conducted only in Europe, but not Asia and the US. We included just two studies conducted in the US that are not enough to draw a definite conclusion. Thus, more studies are required to confirm the association for people in the US. On the contrary, the number of studies conducted in Asia is sufficient to draw a conclusion. We do not have exact reasons why there were discrepancies in findings on this topic between Europeans and Asians. However, we have a potential explanation. A misclassification of Asian women’s smoking status might lead to a non-significant association between smoking and the risk of FLD in the current analysis. Unlike adult men, smoking rates among adult women have been known to be very low (<10%) in South Korea, China, and Hong Kong [41,42]. One of the main reasons of the low smoking prevalence in Asian women is under-reporting. The accuracy of smoking rates self-reported by women in Asian countries has been doubted because social repression and disapproval of women’s smoking might make women reluctant to report their smoking status [43,44,45]. A study using the Korean National Health and Nutrition Examination Survey reported that among the cotinine-verified smokers, about 60% of women classified themselves as non-smokers in a self-report survey [42]. When we performed the subgroup analysis by gender, smoking significantly increased the risk of FLD in Asian men, although there was no significant association between them in Asian women. Thus, the discrepancy in findings between Europeans and Asians might be mainly attributable to a misclassification of Asian women’s smoking status, but not race or ethnicity.
Despite the strengths, there are several limitations in this study. First, we included only five prospective cohort studies. Out of 20 cohort studies included in the current analysis, most studies were retrospective cohort studies and cross-sectional surveys based on cohort studies. As mentioned above, because prospective cohort studies give us a higher level of evidence than retrospective cohort studies and cross-sectional studies, our findings should be confirmed by further prospective cohort studies. Second, most studies measured smoking status based on a self-report. As discussed earlier, self-reported measure of smoking status may lead to under-reporting and a misclassification of smoking status, which could result in biased conclusions. Thus, further studies with biochemical validation of smoking status such as urinary cotinine levels are warranted to confirm our findings. Last, we were unable to investigate the association between the number of cigarettes smoked and the risk of FLD due to a lack of data reported in each study.
In summary, we found that smoking increases the risk of FLD in the meta-analysis of cohort studies. In addition to well-known risk factors of FLD such as obesity and alcohol consumption, clinicians should recommend smoking cessation for the management of FLD.

Author Contributions

Moon Hyung: Conceptualization, data curation, formal analysis, investigation, methodology, visualization, writing-original draft, and writing-review and editing. Seung-Kwon Myung: Formal analysis, investigation, methodology, project administration, validation, writing-original draft, and writing-review and editing. Sang Hee Lee: Data curation, methodology, and writing-review and editing. Yoosoo Chang: Methodology, writing-original draft, and writing-review and editing.

Funding Support

The authors received no financial support to produce this manuscript.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflict of Interest

No potential conflicts of interest are disclosed.

Ethics Statement

We confirm that this work do not include any ethics issues because we used published data from individual studies.

References

  1. Adams, L.A.; Lymp, J.F.; St Sauver, J.; Sanderson, S.O.; Lindor, K.D.; Feldstein, A.; Angulo, P. The natural history of nonalcoholic fatty liver disease: a population-based cohort study. Gastroenterology 2005, 129, 113-121. [CrossRef]
  2. Nascimbeni, F.; Pais, R.; Bellentani, S.; Day, C.P.; Ratziu, V.; Loria, P.; Lonardo, A. From NAFLD in clinical practice to answers from guidelines. J. Hepatol. 2013, 59, 859-871. [CrossRef]
  3. Crabb, D.W.; Galli, A.; Fischer, M.; You, M. Molecular mechanisms of alcoholic fatty liver: role of peroxisome proliferator-activated receptor alpha. Alcohol 2004, 34, 35-38. [CrossRef]
  4. Machado, M.; Cortez-Pinto, H. Non-alcoholic steatohepatitis and metabolic syndrome. Curr. Opin. Clin. Nutr. Metab. Care 2006, 9, 637-642. [CrossRef]
  5. Machado, M.V.; Cortez-Pinto, H. Management of fatty liver disease with the metabolic syndrome. Expert Rev. Gastroenterol. Hepatol. 2014, 8, 487-500. [CrossRef]
  6. Riazi, K.; Azhari, H.; Charette, J.H.; Underwood, F.E.; King, J.A.; Afshar, E.E.; Swain, M.G.; Congly, S.E.; Kaplan, G.G.; Shaheen, A.A. The prevalence and incidence of NAFLD worldwide: a systematic review and meta-analysis. Lancet Gastroenterol. Hepatol. 2022, 7, 851-861. [CrossRef]
  7. Browning, J.D.; Szczepaniak, L.S.; Dobbins, R.; Nuremberg, P.; Horton, J.D.; Cohen, J.C.; Grundy, S.M.; Hobbs, H.H. Prevalence of hepatic steatosis in an urban population in the United States: impact of ethnicity. Hepatology 2004, 40, 1387-1395. [CrossRef]
  8. Ko, E.; Yoon, E.L.; Jun, D.W. Risk factors in nonalcoholic fatty liver disease. Clin. Mol. Hepatol. 2023, 29, S79-s85. [CrossRef]
  9. Yuan, H.; Shyy, J.Y.; Martins-Green, M. Second-hand smoke stimulates lipid accumulation in the liver by modulating AMPK and SREBP-1. J. Hepatol. 2009, 51, 535-547. [CrossRef]
  10. Azzalini, L.; Ferrer, E.; Ramalho, L.N.; Moreno, M.; Domínguez, M.; Colmenero, J.; Peinado, V.I.; Barberà, J.A.; Arroyo, V.; Ginès, P., et al. Cigarette smoking exacerbates nonalcoholic fatty liver disease in obese rats. Hepatology 2010, 51, 1567-1576. [CrossRef]
  11. Tsuneto, A.; Hida, A.; Sera, N.; Imaizumi, M.; Ichimaru, S.; Nakashima, E.; Seto, S.; Maemura, K.; Akahoshi, M. Fatty liver incidence and predictive variables. Hypertens. Res. 2010, 33, 638-643. [CrossRef]
  12. Hamabe, A.; Uto, H.; Imamura, Y.; Kusano, K.; Mawatari, S.; Kumagai, K.; Kure, T.; Tamai, T.; Moriuchi, A.; Sakiyama, T., et al. Impact of cigarette smoking on onset of nonalcoholic fatty liver disease over a 10-year period. J. Gastroenterol. 2011, 46, 769-778. [CrossRef]
  13. Koch, M.; Borggrefe, J.; Schlesinger, S.; Barbaresko, J.; Groth, G.; Jacobs, G.; Lieb, W.; Laudes, M.; Müller, M.J.; Bosy-Westphal, A., et al. Association of a lifestyle index with MRI-determined liver fat content in a general population study. J. Epidemiol. Community Health 2015, 69, 732-737. [CrossRef]
  14. Suomela, E.; Oikonen, M.; Virtanen, J.; Parkkola, R.; Jokinen, E.; Laitinen, T.; Hutri-Kähönen, N.; Kähönen, M.; Lehtimäki, T.; Taittonen, L., et al. Prevalence and determinants of fatty liver in normal-weight and overweight young adults. The Cardiovascular Risk in Young Finns Study. Ann. Med. 2015, 47, 40-46. [CrossRef]
  15. Zhang, T.; Zhang, C.; Zhang, Y.; Tang, F.; Li, H.; Zhang, Q.; Lin, H.; Wu, S.; Liu, Y.; Xue, F. Metabolic syndrome and its components as predictors of nonalcoholic fatty liver disease in a northern urban Han Chinese population: a prospective cohort study. Atherosclerosis 2015, 240, 144-148. [CrossRef]
  16. Kim, T.J.; Sinn, D.H.; Min, Y.W.; Son, H.J.; Kim, J.J.; Chang, Y.; Baek, S.Y.; Ahn, S.H.; Lee, H.; Ryu, S. A cohort study on Helicobacter pylori infection associated with nonalcoholic fatty liver disease. J. Gastroenterol. 2017, 52, 1201-1210. [CrossRef]
  17. Liu, P.; Xu, Y.; Tang, Y.; Du, M.; Yu, X.; Sun, J.; Xiao, L.; He, M.; Wei, S.; Yuan, J., et al. Independent and joint effects of moderate alcohol consumption and smoking on the risks of non-alcoholic fatty liver disease in elderly Chinese men. PLoS One 2017, 12, e0181497. [CrossRef]
  18. van den Berg, E.H.; Amini, M.; Schreuder, T.C.; Dullaart, R.P.; Faber, K.N.; Alizadeh, B.Z.; Blokzijl, H. Prevalence and determinants of non-alcoholic fatty liver disease in lifelines: A large Dutch population cohort. PLoS One 2017, 12, e0171502. [CrossRef]
  19. Bayerl, C.; Lorbeer, R.; Heier, M.; Meisinger, C.; Rospleszcz, S.; Schafnitzel, A.; Patscheider, H.; Auweter, S.; Peters, A.; Ertl-Wagner, B., et al. Alcohol consumption, but not smoking is associated with higher MR-derived liver fat in an asymptomatic study population. PLoS One 2018, 13, e0192448. [CrossRef]
  20. Okamoto, M.; Miyake, T.; Kitai, K.; Furukawa, S.; Yamamoto, S.; Senba, H.; Kanzaki, S.; Deguchi, A.; Koizumi, M.; Ishihara, T., et al. Cigarette smoking is a risk factor for the onset of fatty liver disease in nondrinkers: A longitudinal cohort study. PLoS One 2018, 13, e0195147. [CrossRef]
  21. Okamura, T.; Hashimoto, Y.; Hamaguchi, M.; Obora, A.; Kojima, T.; Fukui, M. Low urine pH is a risk for non-alcoholic fatty liver disease: A population-based longitudinal study. Clin. Res. Hepatol. Gastroenterol. 2018, 42, 570-576. [CrossRef]
  22. Wang, L.; Li, M.; Zhao, Z.; Xu, M.; Lu, J.; Wang, T.; Chen, Y.; Wang, S.; Dai, M.; Hou, Y., et al. Ideal Cardiovascular Health Is Inversely Associated with Nonalcoholic Fatty Liver Disease: A Prospective Analysis. Am. J. Med. 2018, 131, 1515.e1511-1515.e1510. [CrossRef]
  23. Jung, H.S.; Chang, Y.; Kwon, M.J.; Sung, E.; Yun, K.E.; Cho, Y.K.; Shin, H.; Ryu, S. Smoking and the Risk of Non-Alcoholic Fatty Liver Disease: A Cohort Study. Am. J. Gastroenterol. 2019, 114, 453-463. [CrossRef]
  24. van den Berg, E.H.; Gruppen, E.G.; Blokzijl, H.; Bakker, S.J.L.; Dullaart, R.P.F. Higher Sodium Intake Assessed by 24 Hour Urinary Sodium Excretion Is Associated with Non-Alcoholic Fatty Liver Disease: The PREVEND Cohort Study. J. Clin. Med. 2019, 8, 2157. [CrossRef]
  25. Chen, X.; Ma, T.; Yip, R.; Perumalswami, P.V.; Branch, A.D.; Lewis, S.; Crane, M.; Yankelevitz, D.F.; Henschke, C.I. Elevated prevalence of moderate-to-severe hepatic steatosis in World Trade Center General Responder Cohort in a program of CT lung screening. Clin. Imaging 2020, 60, 237-243. [CrossRef]
  26. Okamura, T.; Hashimoto, Y.; Hamaguchi, M.; Obora, A.; Kojima, T.; Fukui, M. Creatinine-to-bodyweight ratio is a predictor of incident non-alcoholic fatty liver disease: A population-based longitudinal study. Hepatol. Res. 2020, 50, 57-66. [CrossRef]
  27. Takenaka, H.; Fujita, T.; Masuda, A.; Yano, Y.; Watanabe, A.; Kodama, Y. Non-Alcoholic Fatty Liver Disease Is Strongly Associated with Smoking Status and Is Improved by Smoking Cessation in Japanese Males: A Retrospective Study. Kobe J. Med. Sci. 2020, 66, E102-e112.
  28. Zhang, Q.; Ma, X.; Xing, J.; Shi, H.; Yang, R.; Jiao, Y.; Chen, S.; Wu, S.; Zhang, S.; Sun, X. Serum Uric Acid Is a Mediator of the Association Between Obesity and Incident Nonalcoholic Fatty Liver Disease: A Prospective Cohort Study. Front. Endocrinol. (Lausanne) 2021, 12, 657856. [CrossRef]
  29. Jeong, S.; Oh, Y.H.; Choi, S.; Chang, J.; Kim, S.M.; Park, S.J.; Cho, Y.; Son, J.S.; Lee, G.; Park, S.M. Association of Change in Smoking Status and Subsequent Weight Change with Risk of Nonalcoholic Fatty Liver Disease. Gut Liver 2023, 17, 150-158. [CrossRef]
  30. Sadeghianpour, Z.; Cheraghian, B.; Farshchi, H.R.; Asadi-Lari, M. Non-alcoholic fatty liver disease and socioeconomic determinants in an Iranian cohort study. BMC Gastroenterol. 2023, 23, 350. [CrossRef]
  31. Akhavan Rezayat, A.; Dadgar Moghadam, M.; Ghasemi Nour, M.; Shirazinia, M.; Ghodsi, H.; Rouhbakhsh Zahmatkesh, M.R.; Tavakolizadeh Noghabi, M.; Hoseini, B.; Akhavan Rezayat, K. Association between smoking and non-alcoholic fatty liver disease: A systematic review and meta-analysis. SAGE Open Med. 2018, 6, 2050312117745223. [CrossRef]
  32. Higgins, J.P.; Thompson, S.G. Quantifying heterogeneity in a meta-analysis. Stat. Med. 2002, 21, 1539-1558. [CrossRef]
  33. Borenstein, M.; Hedges, L.V.; Higgins, J.P.; Rothstein, H.R. A basic introduction to fixed-effect and random-effects models for meta-analysis. Res. Synth. Methods 2010, 1, 97-111. [CrossRef]
  34. Artese, A.; Stamford, B.A.; Moffatt, R.J. Cigarette Smoking: An Accessory to the Development of Insulin Resistance. Am. J. Lifestyle Med. 2017, 13, 602-605. [CrossRef]
  35. Cetin, E.G.; Demir, N.; Sen, I. The Relationship between Insulin Resistance and Liver Damage in non-alcoholic Fatty Liver Patients. Sisli Etfal Hastan. Tıp Bul. 2020, 54, 411-415. [CrossRef]
  36. Eliasson, B.; Attvall, S.; Taskinen, M.R.; Smith, U. The insulin resistance syndrome in smokers is related to smoking habits. Arterioscler. Thromb. 1994, 14, 1946-1950. [CrossRef]
  37. Utzschneider, K.M.; Kahn, S.E. Review: The role of insulin resistance in nonalcoholic fatty liver disease. J. Clin. Endocrinol. Metab. 2006, 91, 4753-4761. [CrossRef]
  38. Jia, W.P. The impact of cigarette smoking on metabolic syndrome. Biomed. Environ. Sci. 2013, 26, 947-952. [CrossRef]
  39. Zhang, C.X.; Guo, L.K.; Qin, Y.M.; Li, G.Y. Association of polymorphisms of adiponectin gene promoter-11377C/G, glutathione peroxidase-1 gene C594T, and cigarette smoking in nonalcoholic fatty liver disease. J. Chin. Med. Assoc. 2016, 79, 195-204. [CrossRef]
  40. Myung, S.K. How to review and assess a systematic review and meta-analysis article: a methodological study (secondary publication). J. Educ. Eval. Health Prof. 2023, 20, 24. [CrossRef]
  41. Mackay, J.; Eriksen, M.P.; Shafey, O.; American Cancer Society. The tobacco atlas. 2nd ed.; American Cancer Society: Atlanta, Ga., 2006; pp 1 atlas (128 pages) : color illustrations, color maps ; 125 cm.
  42. Jung-Choi, K.H.; Khang, Y.H.; Cho, H.J. Hidden female smokers in Asia: a comparison of self-reported with cotinine-verified smoking prevalence rates in representative national data from an Asian population. Tob. Control 2012, 21, 536-542. [CrossRef]
  43. Hwang, J.-e.; Choi, Y.; Yang, Y.-s.; Oh, Y. Gender differences in the perceived effectiveness of female-focused graphic health warnings against smoking in South Korea. Health Educ. J. 2020, 79, 58-72. [CrossRef]
  44. Seo, D.C.; Torabi, M.R.; Kim, N.; Lee, C.G.; Choe, S. Smoking among East Asian college students: prevalence and correlates. Am. J. Health Behav. 2013, 37, 199-207. [CrossRef]
  45. Nakhaee, N.; Divsalar, K.; Bahreinifar, S. Prevalence of and factors associated with cigarette smoking among university students: a study from Iran. Asia Pac. J. Public Health 2011, 23, 151-156. [CrossRef]
Preprints 137982 g001
Figure 1. Flow diagram of identifying relevant Studies.
Figure 1. Flow diagram of identifying relevant Studies.
Preprints 137982 g001
Preprints 137982 g002
Figure 2. Smoking and risk of fatty liver disease in the meta-analysis (n = 20). OR, odds ratio; RR, relative risk; HR, hazard ratio; CI, confidence interval.
Figure 2. Smoking and risk of fatty liver disease in the meta-analysis (n = 20). OR, odds ratio; RR, relative risk; HR, hazard ratio; CI, confidence interval.
Preprints 137982 g002
Preprints 137982 g003
Figure 3. Begg’s funnel plot and Egger’s test to test publication bias (n = 20). OR, odds ratio; RR, relative risk; HR, hazard ratio; S.E, standard error.
Figure 3. Begg’s funnel plot and Egger’s test to test publication bias (n = 20). OR, odds ratio; RR, relative risk; HR, hazard ratio; S.E, standard error.
Preprints 137982 g003
Table 1. General characteristics of the studies Included in the final analysis (n = 20).
Table 1. General characteristics of the studies Included in the final analysis (n = 20).
Study Region Type of study Gender Study participants (% of men) Comparison Odds ratio, relative Risk, or hazard ratio, and 95% confidence interval Outcomes Adjusted variables
2010 Tsuneto [11] Asia Prospective Both 1,635 atomic bomb
survivors who underwent biennial examinations in Nagasaki without NAFLD at baseline		 Ex-smoker or current smoker vs. none 0.92 (0.64-13.4) FLD Age, sex, BMI, DM, HTN, dyslipidemia, drinking habits, and atomic radiation dose
2011 Hamabe [12] Asia Retrospective Both 1,560 subjects without NAFLD who underwent a complete medical health checkup at the Kagoshima Kouseiren Medical Healthcare Center Cigarette smoking vs. no smoking 1.44 (0.86-2.42) NAFLD Age, sex, obesity, HTN, dyslipidemia, dysglycemia, and alcohol intake
2015 Koch [13] Europe Cross-sectional Both 747 official population registeries in Kiel Cigarette smoking vs. no smoking 1.14 (0.71-1.82) FLD Age, sex, years of education, total energy intake, physical activity, and waist circumference
2015 Suomela [14] Europe Cross-sectional Both 3,592 Young Finns Current smoker vs. none 2.56 (1.18-5.52) FLD Age, sex, BMI, and waist circumference
2015 Zhang [15] Asia Prospective Both 15,791 health check-up participants at the Center for Health Management of Shandong Provincial
Qianfoshan Hospital and Shandong Provincial Hospital		 Current smoker vs. none 1.03 (0.95-1.11) NAFLD Baseline Mets status, sex, age, diet, smoking status, and regular exercise
2017 Kim [16] Asia Retrospective Both 17,028 health-screening exam participants at the Center for Health Promotion of the Samsung Medical Center, South Korea Current smoker vs. none 0.96 (0.81-1.15) NAFLD Age, sex, body mass index, year of screening exam, alcohol
intake, regular exercise, and education level
2017 Liu [17] Asia Cross-sectional Male 9,432 DFTJ cohort study among retirees of Dong feng Motor corporation Current smoker vs. none 1.52 (1.22-1.88) NAFLD Age, body mass index, waist circumference alcohol
intake, DM, HTN, dyslipidemia, and past history of CHD
2017 van den Berg [18] Europe Cross-sectional Both 37,496 Framework of the Lifelines Cohort Study Current smoker vs. none 1.32 (1.21-1.43) FLD Age, sex, Hemoglobin, ALT/ALP/Albumin, HBA1c, Type 2 DM, dyslipidemia, and past history of CHD
2018 Bayerl [19] Europe Cross-sectional Both 1,282 persons from Cooperative Health Research in German region Ex-smoker or current smoker vs. none 0.56 (0.27-1.17) FLD Age, sex, DM, and alcohol intake
2018 Okamoto [20] Asia Retrospective Both 7,905 persons who underwent health checkup at Ehime General Health Care Association Current smoker vs. none 2.25 (1.10-4.38) FLD Age, sex, BMI, DM, HTN, CVD, dyslipidemia, and snacking habit
2018 Okamura [21] Asia Retrospective Both 29,555 medical
examination program at Murakami Memorial Hospital using the NAGALA
(NAFLD in the Gifu Area, Longitudinal Analysis) database		 Current smoker vs. none 0.88 (0.78-0.99) NAFLD Age, sex, BMI, ALT, triglycerides, exercise habit, alcohol consumption, systolic blood pressure, fasting
plasma glucose, and uric acid
2018 Wang [22] Asia Prospective Both 10,375 participants from community residents in the Jiading District of Shangia Current smoker or quit<12mo vs. none 1.11 (0.78-1.56) NAFLD Age, sex, alcohol consumption, education, and HOMA-IR
2019 Jung [23] Asia Prospective Both 199,468 persons who underwent health checkup at Kangbuk Samsung Health Study Ex-smoker or current smoker vs. none Men: 1.15
(1.12-1.18)
Women: 1.14
(1.03-1.27)		 FLD Age, sex, BMI, DM, HTN, dyslipidemia, alcohol drinking, education level, physical activity, waist circumference, and laboratory test
2019 van den Berg [24] Europe Cross-sectional Both 6,132 participants of the prevention of Renal and Vascular End-stage Disease cohort study Current smoker vs. none 1.24 (1.05-1.46) NAFLD Age, sex, BMI, DM, HTN, dyslipidemia, alcohol drinking, estimated GFR, urine albumin excretion, use of antihypertensive medication, glucose lowering drugs, lipid lowering drugs, and HOMA-IR
2020 Chen [25] US Cross-sectional Both 154 World Trade Center participants in NIOSH Current smoker vs. none 0.41 (0.17-0.99) FLD Age, sex , Ethnicity, BMI, DM,HTN,COPD, and membership in the WTC
2020 Okamura [26] Asia Retrospective Both 13,728 population-based longitudinal study
of participants in a medical checkup program at Asahi
University Hospital		 Current smoker vs. none 1.16 (0.88-1.52) NAFLD Age, aspartate aminotransferase, fasting plasma glucose, triglyceride to high-density lipoprotein cholesterol ratio, systolic blood pressure, alcohol consumption, and exercise.
2020 Takenaka [27] Asia Cross-sectional Both 8,297 health check-up participants at
Yodogawa Christian Hospital		 Current smoker vs. none 1.31 (1.17-1.47) NAFLD Age, sex, presence of metabolic syndrome, and light alcohol consumption
2021 Zhang [28] Asia Prospective Both 16,839 participants who received the Kailuan Group’s detailed
and thorough medical examination at Tangshan City, China		 Current smoker vs. none 1.15 (1.06-1.25) NAFLD Age, sex, marital status, working type, education level, physical activity, systolic blood pressure , lipid profile, CRP, and Cr
2023 Jeong [29] Asia Retrospective Both 296,033 in NHIS of Korea Current smoker vs. none 1.64 (1.39-1.94) FLD Age, sex, household income, BMI, HTN, DM, HL, physical activity, and Charlson comorbidity index
2023 Sadeghianpour [30] Asia Cross-sectional Both 180,000 Iranian adults in Hoveyzeh Cohort Study Current smoker vs. none 0.63 (0.50-0.79) FLD Age, sex, area, physical activity, Energy intake Household income, DM, HL, Education level, wealth status, and skill level
Abbreviations: NAFLD, non-alcoholic fatty liver disease; FLD, fatty liver disease; DMC, Dong feng Motor Corporation; BMI, body mass index; DM, Diabetes Mellitus; HTN, Hypertension; HL, Hyperlipidemia; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; NIOSH, National Institute of Occupational Safety and Health ; NHIS, National Health Insurance Service ; HOMA-IR, homeostasis model assessment of insulin resistance ; GFR, glomerular filtration rate.
Table 2. Methodological quality of studies assessed by the Newcastle-Ottawa Scale (n = 11).*.
Table 2. Methodological quality of studies assessed by the Newcastle-Ottawa Scale (n = 11).*.
Studies Selection Comparability Exposure Total score
1 2 3 4 1 1 2 3
Representativeness of the exposed cohort Selection of the non-exposed cohort Ascertainment of exposure Outcome of interest not present at start of the study Comparability of cohorts Assessment of outcome Adequate follow-up period for outcome of interest Adequacy of follow-up of cohorts
2010 Tsuneto 0 1 1 1 2 1 1 0 7
2011 Hamabe 1 1 0 1 2 1 1 0 7
2015 Zhang 1 1 0 1 2 1 1 1 8
2017 Kim 1 1 0 1 2 1 1 1 8
2018 Okamoto 1 1 0 1 2 1 1 1 8
2018 Okamura 1 1 0 1 2 1 1 0 7
2018 Wang 1 1 1 1 2 1 1 0 8
2019 Jung 1 1 0 1 2 1 1 0 7
2020 Okamura 1 1 0 1 2 1 1 1 8
2021 Zhang 1 1 0 1 2 1 1 0 7
2023 Jeong 1 1 0 1 2 1 1 1 8
*Among the 20 cohort studies included in the final analysis, cross-sectional surveys were excluded for the assessment of methodological quality. The average score of all the prospective and retrospective cohort studies is 7.5.
Table 3. Association between smoking and fatty liver disease in subgroup meta-analysis by various factors.
Table 3. Association between smoking and fatty liver disease in subgroup meta-analysis by various factors.
Factor No. of studies RR (95%CI) Heterogeneity I2 (%)
All studies 20 1.14 (1.05-1.24)* 83.7
Type of cohort study
   Prospective 5 1.15 (1.05-1.18)* 51.7
   Retrospective 6 1.23 (0.94-1.62) 88.4
   Cross-sectional 9 1.12 (0.92-1.36) 85.2
Region
   Europe 8 1.32 (1.16-1.50)* 82.1
   Asia 10 1.03 (0.91-1.18) 83.5
   US 2 0.75 (0.28-2.06) 79.5
Type of fatty liver disease
   Fatty liver disease 6 1.27 (1.01-1.59)* 72.7
   Non-alcoholic fatty liver disease 14 1.09 (1.00-1.19)* 83.2
Gender (All from Asia)
   Men 4 1.15 (1.06-1.25)* 71.2
   Women 4 1.12 (0.94-1.34) 47.8
Follow-up period
   <5 years 3 1.44 (0.95-2.13) 93.0
   >5 years 7 1.08 (0.98-1.19) 70.9
Quality of study†
   High 6 1.16 (0.94-1.42) 85.3
   Low 5 1.07 (0.95-1.20) 67.1
*Indicates a significant association. RR, relative risk; CI, confident interval. † Study quality was assessed based on the Newcastle-Ottawa Scale.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
Prerpints.org logo

Preprints.org is a free preprint server supported by MDPI in Basel, Switzerland.

facebook logotwitter logolinkedin logo
MDPI Initiatives
Important Links
Subscribe

Disclaimer

Terms of Use

Privacy Policy

© 2024 MDPI (Basel, Switzerland) unless otherwise stated