Table 1.
Main Characteristics of the Included Studies Evaluating Association Between NAFLD and (A) Cardiovascular Diseases, (B) Stroke and Cerebrovascular Diseases, and (C) Extrahepatic Cancers.
| Author (Year) | Country/Region | Study Design | Study Period | Study Population | NAFLD Diagnosis Criteria | Exclusion Criteria | Outcomes Assessed |
|---|---|---|---|---|---|---|---|
| A. Cardiovascular Diseases | |||||||
| Golabi (2019)22 | USA | Cross-sectional study | Individuals from 2011 to 2016 | 3197 individuals (816 Asian American adults and 2381 Non-Hispanic Whites) | US-Fatty Liver Index ≥ 30 | Excessive alcohol consumption (≥20g/day in men and ≥10g/day in women), positive hepatitis C virus RNA, positive hepatitis B surface antigen, iron overload (defined as serum transferrin saturation ≥ 50%) | Prevalence of atherosclerotic cardiovascular disease using ASCVD: 10-year ASCVD risk score ≥7.5% |
| Viglino (2018)39 | France | Single-center prospective cohort study | Individuals from 2007 to 2012 | 111 individuals with chronic obstructive pulmonary disease | FibroMax algorithm, which incorporates 3 non-invasive tests: FibroTest, SteatoTest, and NashTest | Active pulmonary infection, chronic heart failure, left ventricular ejection fraction <45%, active smoking >10 cigarettes per day, neoplasia, prior antioxidant treatment, pregnancy, alcohol ≥20g for women and ≥30 g for men daily, viral hepatitis | Incidence of new cardiovascular events during follow-up, which include acute myocardial infarction, stroke, peripheral arterial disease or acute limb ischemia, venous thromboembolic disease and/or pulmonary embolism, and new-onset arrhythmias |
| Chinnadurai (2019)17 | England | Retrospective Cohort Study | Individuals from 1/2000–12/2014 with follow-up through 12/2015 | 149 individuals with diabetic kidney disease | Hepatic ultrasonography | Concurrent renal replacement therapy, excessive alcohol intake, hepatitis and other chronic liver diseases | Incidence of nonfatal cardiovascular events: myocardial infarctions, acute coronary syndromes, nonfatal cardiac arrest, congestive cardiac failure, peripheral vascular disease and cerebrovascular disease |
| Vanjiappan (2018)38 | India | Single-center prospective cohort study | Individuals from 4/2014–5/2016 | 300 individuals with type 2 diabetes mellitus | Hepatic ultrasonography | hepatitis B & C infections, chronic liver disease, those on hepatotoxic drugs | Incidence of cardiovascular disease |
| Allen (2018)14 | USA | Retrospective cohort study | Individuals from 1997 to 2014 with follow-up through 10/1/2016 | 19,078 individuals | Hospital International Classification of Diseases Adapted (HICDA) codes and ICD codes | Other liver diseases such as viral hepatitis, alcoholic liver disease, alcohol use, cholestatic liver disease, short follow-up time of less than 1 year | Incidence of clinical cardiovascular events (myocardial infarction, angina/ischemic heart disease, atrial fibrillation, cardiac arrest, congestive heart failure and stroke) |
| Mantovani (2016)28 | Italy | Single-center retrospective cohort study | Individuals enrolled from 1999 to 2001 | 286 individuals with type 1 diabetes | Findings on liver ultrasound which include diffuse hyperechogenicity of the liver relative to kidneys, beam attenuation, poor visualization of the intrahepatic vessel border and diaphragm | Missing liver ultrasound data, concurrent end-stage renal disease or malignancy, cirrhosis, and liver diseases due to secondary causes, such as excessive alcohol consumption, viral hepatitis, iron overload or use of steatogenic medications | Incidence of cardiovascular events: combined endpoint of nonfatal ischemic heart disease, nonfatal ischemic stroke, carotid endarterectomy, coronary or lower extremity artery revascularization |
| Zeb (2016)43 | USA | Multicenter retrospective cohort study | Not clearly stated, but MESA study started enrolling July 2000 | 4119 individuals | Computed tomography | Heavy alcohol intake (defined as >14 drinks per week for men and >7 drinks per week for women), use of oral steroids, cirrhosis | Incidence of nonfatal coronary heart disease: myocardial infarction, resuscitated cardiac arrest, angina with or without coronary revascularization |
| Stolic (2016)32 | Serbia | Single-center observational study | Not stated | 72 individuals over the age of 65 years on chronic hemodialysis | Findings on liver ultrasound demonstrating increased hepatic parenchyma echogenicity compared to right kidney cortex | Hemodialysis, patients hospitalized in the past 6 months, infection with hepatotropic virus, diabetes mellitus, hepatobiliary surgery, body mass index higher than 30 kg/m2, concurrent use of statins or glucocorticosteroids | Prevalence of cardiovascular disease |
| Fracanzani (2016)21 | Italy | Prospective cohort study | Individuals from 6/2002–12/2004 | 273 individuals | Liver ultrasound assessment of hepatorenal echo contrast, liver brightness, deep attenuation, and vascular blurring | Chronic viral hepatitis, autoimmune hepatitis, hereditary hemochromatosis, Wilson's disease, and drug-induced liver disease were excluded in NAFLD patients; Controls were negative for hepatitis B and C and had normal liver function tests | Incidence of major cardiovascular events |
| Wong (2016)41 | Hong Kong | Prospective cohort study | Individuals from 10/2007–11/2008 | 612 consecutive individuals who underwent coronary angiogram | Liver ultrasound findings including diffusely increased liver echogenicity compared to kidney or spleen, vascular blurring, and deep attenuation of the ultrasound signal | Excessive alcohol intake (>20 g/day in men and >10 g/day in women), secondary causes of fatty liver (e.g., systemic steroids or methotrexate), HBV or HCV, or antinuclear antibody titer >1/160 | Incidence of cardiovascular events: cardiovascular deaths, nonfatal myocardial infarction, heart failure, coronary interventions, and congestive heart failure |
| Mellinger (2015)29 | USA | Cross-sectional study from a large prospective longitudinal cohort study | Individuals from 2002 to 2005 | 3014 individuals | CT scan liver-phantom ratio with a liver-phantom ratio of 0.33 or lower representing the presence of thirty percent or more of hepatic steatosis | Pregnancy, weight >160 kg, CT scan results uninterpretable for hepatic steatosis, missing a complete profile | Prevalent cardiovascular disease: composite of nonfatal myocardial infarction, heart failure, stroke, transient ischemic attack, or peripheral arterial disease |
| Pisto (2014)30 | Finland | Population-based prospective cohort study | Individuals from 1991 to 2009 | 988 individuals | Liver ultrasound assessment of hepatic steatosis. | Previous hospital-diagnosed myocardial infarction or stroke | Incidence of cardiovascular disease: major coronary heart disease event and stroke based on ICD-10 coding. |
| Lai (2013)25 | Taiwan | Single-center retrospective cohort study | Individuals from 7/1998 to 10/2012 | 278 individuals undergoing hemodialysis | Liver ultrasound assessment that includes increased hepatic echogenicity compared to kidneys, blurring of the gallbladder wall, hepatic veins, or portal vein | Chronic drug or alcohol abuse, cirrhosis, history of CVD, malignancies, medications known to cause hepatic steatosis, insufficient medical records | Development of nonfatal cardiovascular events: angina pectoris, nonfatal myocardial infarction, acute pulmonary edema, congestive heart failure, and peripheral vascular disease |
| Choi (2013)18 | Korea | Single-center cross-sectional study | Individuals from 1/2009–6/27/2011 | 134 individuals undergoing elective coronary angiography | Liver ultrasound with characteristic echo patterns such as a diffuse increase in hepatic echogenicity compared to the kidney | Viral hepatitis, history of heavy alcohol ingestion, medications reported to affect hepatic steatosis within 3 months of enrollment, or other history of chronic liver disease. | Prevalence of coronary artery disease: at least 50% stenosis in at least one major coronary artery |
| Dunn (2013)19 | USA | Single-center retrospective cohort study | Individuals from 1/1/2002 to 12/31/2003 with follow-up through the end of 2008 | 2343 individuals with type 2 diabetes | CT scan assessment of liver and spleen attenuation difference of −10 or less (indicating ≥30% steatosis) | Patients with ICD-9 diagnostic codes for alcohol abuse, alcoholic liver disease, chronic hepatitis B and C, autoimmune hepatitis, biliary cirrhosis, Wilson disease, hemochromatosis, alpha-1 antitrypsin disease, or a prior liver transplant. | Development of cardiovascular outcomes: cardiovascular deaths, myocardial infarctions, strokes, angina, arrhythmias and congestive heart failure based on ICD-9 coding |
| Feitosa (2013)20 | USA | Multicentered cross-sectional study | Individuals recruited from 1992 to 1996 | 2756 individuals in 510 extended random and high coronary heart disease risk families recruited from the Framingham Heart Study, the Utah Family Tree Study, and the North Carolina and Minnesota sites of the Atherosclerosis Risk in Communities Study. | CT scan liver attenuation of 40 Hounsfield units predicted hepatic steatosis, | Alcohol consumption over 21 drinks/week for men and >14 drinks/week) for women; Amiodarone use; HCV antibody positive. | Prevalence of coronary heart disease: coronary bypass, myocardial infarction, coronary angioplasty, balloon angioplasty, atherectomy, stent, percutaneous transluminal coronary angioplasty, or percutaneous coronary intervention |
| Wong (2011)42 | Hong Kong | Single-center prospective cohort study | Individuals from 10/2007 to 11/2008 | 612 individuals who underwent coronary angiogram | Liver ultrasound showing diffusely increased hepatic echogenicity compared to kidney or spleen, vascular blurring and deep attenuation of the ultrasound signal | Contraindications to coronary angiogram, excessive alcohol intake, and secondary causes of fatty liver (e.g., chronic use of systemic corticosteroids or methotrexate), positive hepatitis B surface antigen, antibody against hepatitis C virus and antinuclear antibody titer >1/160 | Incidence of coronary artery disease: presence of at least 50% stenosis at one or more coronary arteries |
| Hamaguchi (2007)24 | Japan | Prospective cohort study | Individuals from 1/1998 to 12/1998 with follow-up through 12/2004 | 1647 individuals who completed health checkups on an annual or biennial basis | Abdominal ultrasonography evidence of hepatorenal contrast and liver brightness | Previous myocardial infarction, angina pectoris, ischemic stroke, cerebral hemorrhage or cancer, alcohol intake of more than 20 g/d, concurrent HBV or HCV, other liver diseases | Incidence of cardiovascular disease via self-administered questionnaire: coronary heart disease, ischemic stroke, cerebral hemorrhage, unstable angina, acute MI, silent MI |
| Targher (2006)34 | Italy | Single-center retrospective cohort study | Not stated | 800 individuals with type 2 diabetes | Liver ultrasonography evidence of characteristic echo patterns such as increased echogenicity of the liver compared to kidneys | Heavy alcohol use, other known causes of chronic liver disease (e.g. viral hepatitis, autoimmune hepatitis, use of hepatotoxic medications such as glucocorticoids, antibiotics, amiodarone, methotrexate, tamoxifen or other anti-neoplastic drugs) | Prevalence of cardiovascular disease: coronary artery disease (myocardial infarction, angina pectoris, heart failure or revascularization procedures), cerebrovascular disease, or peripheral vascular disease |
| Targher (2005)36 | Italy | Prospective nested case-control study | Individuals from 1/1/2000 to 12/31/2000 with follow-up through 5/31/2005 | 248 cases (with nonfatal coronary heart disease) and 496 controls (without coronary heart disease) | Liver ultrasound findings of increased diffuse hyperechogenicity compared to kidneys, ultrasound beam attenuation, and poor visualization of intrahepatic structures | Alcohol abuse, other known causes of chronic liver disease (viral hepatitis, autoimmune hepatitis, use of hepatotoxic medications) | Prevalence of NAFLD in patients with and without any nonfatal coronary heart disease, ischemic stroke, or cardiovascular death. |
| Liu (2019)26 | China | Matched case-control study | Individuals from 3/2011 to 7/2016 who underwent coronary angiography for evaluation of angina-like chest pain and/or positive treadmill exercise test and/or significant stenosis by coronary computer tomography | 324 individuals with stable, new-onset coronary artery disease | Liver ultrasound evidence of diffusely increased liver echogenicity compared to kidney or spleen, vascular blurring, and deep attenuation of the ultrasound signal | Patients without abdominal ultrasound examination, HBV or HCV; autoimmune hepatitis; hereditary liver disease; excessive alcohol consumption; secondary causes of fatty liver (e.g., chronic use of systemic corticosteroids or methotrexate) or drug-induced liver disease | Incidence of cardiovascular events: all-cause death (death mainly cause by cardiovascular disease), nonfatal myocardial infarction and stroke |
| Hagstrom (2019)23 | Sweden | Retrospective cohort study | Individuals from 1971 to 2009 with biopsy proven NAFLD | 6872 individuals | Liver biopsy assessment of NAFLD activity score | Other causes of steatosis, low liver biopsy quality, CVD at baseline, liver outcome within 6 months | Incidence of cardiovascular outcomes: first event of either acute ischemic heart disease or stroke (ischemic or hemorrhagic) |
| Wild (2018)40 | Scotland | Retrospective cohort study | Individuals from 1/1/2004 to 12/31/2013 | 132,661 hospitalized individuals with type 2 diabetes | NAFLD ICD-9 and ICD-10 codes: 571.8, K76, K75.8 | Viral hepatitis, autoimmune hepatitis, hemochromatosis, and any cirrhosis, fibrosis, sclerosis, or portal hypertension with no mention of ALD or NAFLD | Incident/recurrent cardiovascular disease: ICD-9/10 codes for acute coronary syndrome, myocardial infarction, stroke, heart failure, coronary revascularization procedure and carotid revascularization procedure |
| Chan (2014)16 | Malaysia | Cross-sectional study | Individuals from 11/2011 to 4/2012 | 399 individuals with type 2 diabetes | Liver ultrasound showing evidence of increased echogenicity, posterior attenuation and loss of intrahepatic and architectural details | Heavy alcohol intake, other causes of chronic liver disease and use of drugs that could cause fatty liver. | Prevalence of ischemic heart disease: previous admission for acute coronary syndrome, previous coronary intervention, previous coronary angiography showing coronary artery disease, or under follow-up and treatment for ischemic heart disease |
| Stepanova (2012)31 | USA | Population-based cross-sectional study | Individuals from 1988 to 1994 | 11,613 individuals from the National Health and Nutrition Examination Survey III (NHANES III) | Abdominal ultrasonography evidence of moderate to severe hepatic steatosis | Excessive alcohol, iron overload, or positive for HBV or HCV | Prevalence of cardiovascular disease: self-reported history of congestive heart failure, stroke, angina, or myocardial infarction. |
| Targher (2012)37 | Italy | Single-center cross-sectional study | Individuals from 2008 to 2010 | 343 individuals with type 1 diabetes | Liver ultrasonography evidence of diffuse hyperechogenicity of the liver relative to the kidneys, ultrasound beam attenuation and poor visualization of the intrahepatic vessel borders and diaphragm | Unavailable liver ultrasound exam, history of end-stage renal disease, cirrhosis, malignancy, known causes of chronic liver disease (alcohol-induced or drug-induced liver disease, hemochromatosis, autoimmune or viral hepatitis) | Prevalence of cardiovascular disease composite endpoint including coronary heart disease, cerebrovascular disease, or peripheral vascular disease |
| Sun (2011)33 | China | Cross-sectional study | Individuals from 9/2008 to 9/2009 | 542 consecutive individuals with suspected coronary artery disease | Abdominal CT scan evidence of liver attenuation less than the spleen, blurred intrahepatic vessels, or markedly reduced attenuation of the liver with evident contrast between the liver and the intrahepatic vessels | Heavy alcohol consumption; other chronic liver diseases; medications known to induce fatty liver disease such as steroids, estrogens, amiodarone, tamoxifen, or other chemotherapeutic agents within the previous 6 months; creatinine >2 mg/dl; history of acute coronary syndrome or heart failure | Significant coronary artery disease: >70% stenosis of lumen diameter |
| Targher (2010)35 | Italy | Cross-sectional study | Individuals from 1/2009 to 12/2009 | 301 individuals with type 1 diabetes | Liver ultrasonography evidence of diffuse hyperechogenicity of liver relative to kidneys, ultrasound beam attenuation and poor visualization of intrahepatic vessel borders and diaphragm | Excessive alcohol consumption, other known causes of chronic liver disease, missing liver ultrasound data | Prevalence of cardiovascular disease composite endpoint including coronary heart disease, cerebrovascular disease, or peripheral vascular disease |
| Lu (2009)27 | China | Cross-sectional study | Individuals from 1/2002 to 1/2009 | 560 individuals with type 2 diabetes mellitus | Liver ultrasonography | Alcohol intake more than 20 g/day, no known etiologies of liver disease and negative tests for the presence of viral hepatitis | Prevalence of coronary heart disease |
| Arslan (2007)15 | Turkey | Cross-sectional study | Not stated | 92 consecutive individuals undergoing first coronary angiography | Liver ultrasound findings of diffuse increase in the echogenicity of the liver compared to kidney | coronary artery disease, or with a history of percutaneous surgical revascularization, prior acute coronary syndrome, chronic alcohol consumption (more than 20 g/day), HBV or HCV, systemic diseases that might cause fatty liver, use of drugs like statins and insulin-sensitizing agents (metformin and glitazones) | Prevalence of coronary artery disease: presence of at least 50% stenosis in at least one major coronary artery |
| B. Stroke and cerebrovascular diseases | |||||||
| Hagstrom (2019)23 | Sweden | Retrospective cohort study | Individuals from 1971 to 2009 with biopsy proven NAFLD | 6872 individuals | Liver biopsy assessment of NAFLD activity score | Other causes of steatosis, low liver biopsy quality, CVD at baseline, liver outcome within 6 months | Incidence of cardiovascular outcomes: first event of either acute ischemic heart disease or stroke (ischemic or hemorrhagic) |
| Alexander (2018)44 | USA | Retrospective case cohort study | Individuals from 2003 to 2007 with follow-up through 9/1/2011 | 1676 individuals (572 with incident ischemic stroke and a stratified stroke-free cohort random sample of 1017) | Fatty Liver Index > 60 | Medical conditions that precluded involvement, prior stroke, excessive alcohol consumption (14 drinks/week for men, 7 drinks/week for women) | Incidence of ischemic stroke |
| Mantovani (2016)28 | Italy | Single-center retrospective cohort study | Individuals enrolled from 1999 to 2001 | 286 individuals with type 1 diabetes | Liver ultrasound showing diffuse hyperechogenicity of the liver compared to kidneys, beam attenuation, poor visualization of the intrahepatic vessel border and diaphragm | Missing ultrasound data end-stage renal disease or malignancy, cirrhosis or other chronic liver diseases, excessive alcohol consumption, viral hepatitis, iron overload hepatitis or use of steatogenic medications | Incidence of cardiovascular events: combined endpoint of nonfatal ischemic heart disease, nonfatal ischemic stroke, carotid endarterectomy, coronary or lower extremity artery revascularization |
| Moshayedi (2014)45 | Iran | Cross-sectional study | Individuals from 5/2012 to11/2013 | 220 individuals (110 brain magnetic resonance imaging confirmed ischemic stroke patients and 110 age- and sex-matched controls) | Liver ultrasonography assessment of increased hepatic echogenicity and visualization of the diaphragm and intrahepatic vessel borders | Chronic HBV or HCV, heavy alcohol consumption more than 20 g/day and chronic hepatotoxic drug use | Prevalence of NAFLD in patients with and without imaging confirmed ischemic stroke |
| Pisto (2014)30 | Finland | Population-based prospective cohort study | Individuals from 1991 to 2009 | 988 individuals | Liver ultrasound assessment of steatosis | Previous hospital-diagnosed myocardial infarction or stroke | Incidence of cardiovascular disease: major coronary heart disease event and stroke based on ICD-10 coding. |
| Dunn (2013)19 | USA | Single-center retrospective cohort study | Individuals from 1/1/2002 to 12/31/2003 with follow-up through the end of 2008 | 2343 individuals with type 2 diabetes | Abdominal CT Scan assessment of liver and spleen attenuation difference of −10 or less (indicating ≥30% steatosis) | Patients with ICD-9 diagnostic codes for alcohol abuse, alcoholic liver disease, chronic hepatitis B and C, autoimmune hepatitis, biliary cirrhosis, Wilson disease, hemochromatosis, alpha-1 antitrypsin disease, or a prior liver transplant. | Development of stroke based on ICD-9 coding |
| Hamaguchi (2007)24 | Japan | Prospective cohort study | Individuals from 1/1998 to 12/1998 with follow-up through 12/2004 | 1647 individuals who completed health checkups on an annual or biennial basis | Abdominal ultrasonography evidence of hepatorenal contrast and liver brightness | Previous myocardial infarction, angina pectoris, ischemic stroke, cerebral hemorrhage or cancer, alcohol intake of more than 20 g/d, HBV or HCV positive, or other chronic liver diseases | Incidence of cardiovascular disease via self-administered questionnaire: coronary heart disease, ischemic stroke, cerebral hemorrhage, unstable angina, acute MI, silent MI |
| Targher (2005)36 | Italy | Prospective nested case-control study | Individuals from 1/1/2000 to12/31/2000 with follow-up through 5/31/2005 | 248 cases with cardiovascular outcomes and 496 controls (without cardiovascular outcomes) | Liver ultrasound findings of increased diffuse hyperechogenicity compared to kidneys, ultrasound beam attenuation, and poor visualization of intrahepatic structures | Alcohol abuse, other known causes of chronic liver disease (viral hepatitis, autoimmune hepatitis, use of hepatotoxic medications) | Prevalence of NAFLD in patients with and without cardiovascular outcomes (any nonfatal coronary heart disease, ischemic stroke, or cardiovascular death) |
| Viglino (2018)39 | France | Single-center prospective cohort study | Individuals from 2007 to 2012 | 111 individuals with chronic obstructive pulmonary disease | FibroMax algorithm, which incorporates 3 non-invasive tests: FibroTest, SteatoTest, and NashTest | Active pulmonary infection, chronic heart failure, left ventricular ejection fraction <45%, active smoking >10 cigarettes per day, neoplasia, antioxidant treatment, pregnant women, a daily consumption of alcohol ≥20g for women and ≥30 g for men, viral hepatitis | Incidence of new cardiovascular events during follow-up: acute myocardial infarction; stroke; new diagnosis of peripheral arterial disease or acute limb ischemia; venous thromboembolic disease and/or pulmonary embolism and new-onset arrhythmias |
| Mellinger (2015)29 | USA | Cross-sectional study from a large prospective longitudinal cohort study | Individuals from 2002 to 2005 | 3014 individuals | CT scan liver-phantom ratio with a liver-phantom ratio of 0.33 or lower representing the presence of thirty percent or more of hepatic steatosis | Pregnancy, weight >160 kg, CT scan results uninterpretable for hepatic steatosis, missing a complete profile | Prevalent cardiovascular disease: composite of non-fatal myocardial infarction, heart failure, stroke, transient ischemic attack, or peripheral arterial disease |
| Targher (2006)34 | Italy | Single-center retrospective cohort study | Not stated | 800 individuals with type 2 diabetes | Liver ultrasonography evidence of characteristic echo patterns such as increased echogenicity of the liver compared to kidneys | Heavy alcohol use, other known causes of chronic liver disease (e.g. viral hepatitis, autoimmune hepatitis, use of hepatotoxic medications such as glucocorticoids, antibiotics, amiodarone, methotrexate, tamoxifen or other anti-neoplastic drugs) | Prevalence of cardiovascular disease: coronary artery disease, cerebrovascular disease,or peripheral vascular disease |
| Liu (2019)26 | China | Matched case-control study | Individuals from 3/2011 to7/2016 who underwent coronary angiography for evaluation of angina-like chest pain and/or positive treadmill exercise test and/or significant stenosis by coronary computer tomography | 324 individuals with stable, new-onset coronary artery disease | Liver ultrasound evidence of diffusely increased liver echogenicity compared to kidney or spleen, vascular blurring, and deep attenuation of the ultrasound signal | Patients without abdominal ultrasound examination, HBV or HCV; autoimmune hepatitis; hereditary liver disease; excessive alcohol consumption; secondary causes of fatty liver (e.g., chronic use of systemic corticosteroids or methotrexate) or drug-induced liver disease | Incidence of cardiovascular events: all-cause death (death mainly cause by cardiovascular disease), non-fatal myocardial infarction and stroke |
| Wild (2018)40 | Scotland | Retrospective cohort study | Individuals from 1/1/2004 to12/31/2013 | 132,661 hospitalized individuals with type 2 diabetes | NAFLD ICD-9 and ICD-10 codes: 571.8, K76, K75.8 | Viral hepatitis, autoimmune hepatitis, hemochromatosis, and any cirrhosis, fibrosis, sclerosis, or portal hypertension with no mention of ALD or NAFLD | Incident/recurrent cardiovascular disease: ICD-9/10 codes for acute coronary syndrome, myocardial infarction, stroke, heart failure, coronary revascularization procedure and carotid revascularization procedure |
| Stepanova (2012)31 | USA | Population-based cross-sectional study | Individuals from 1988 to 1994 | 11,613 individuals from the National Health and Nutrition Examination Survey III (NHANES III) | Abdominal ultrasonography evidence of moderate to severe hepatic steatosis | Excessive alcohol, iron overload, or positive for HBV or HCV | Prevalence of cardiovascular disease: self-reported history of congestive heart failure, stroke, angina, or myocardial infarction. |
| Targher (2012)37 | Italy | Single-center cross-sectional study | Individuals from 2008 to 2010 | 343 individuals with type 1 diabetes | Liver ultrasonography evidence of diffuse hyperechogenicity of the liver relative to the kidneys, ultrasound beam attenuation and poor visualization of the intrahepatic vessel borders and diaphragm | Unavailable liver ultrasound exam, known history of end-stage renal disease, cirrhosis, malignancy, known causes of chronic liver disease (alcohol-induced or drug-induced liver disease, hemochromatosis, autoimmune or viral hepatitis) | Prevalence of cardiovascular disease composite endpoint including coronary heart disease, cerebrovascular disease, or peripheral vascular disease |
| Targher (2010)35 | Italy | Cross-sectional study | Individuals from 1/2009 to12/2009 | 301 individuals with type 1 diabetes | Liver ultrasonography evidence of diffuse hyperechogenicity of liver relative to kidneys, ultrasound beam attenuation and poor visualization of intrahepatic vessel borders and diaphragm | Excessive alcohol consumption, other known causes of chronic liver disease, missing liver ultrasound data | Prevalence of cardiovascular disease composite endpoint including coronary heart disease, cerebrovascular disease, or peripheral vascular disease |
| C. Extrahepatic Cancers | |||||||
| Kim (2017)49 | Korea | Single-center retrospective cohort study | Individuals from 9/1/2004 to 12/31/2005 with follow-up through 12/31/2015 | 25,947 individuals who underwent a comprehensive health checkup | Liver ultrasound evidence of hepatic steatosis | Excessive alcohol consumption; HBV or HCV; HIV; cancer or diagnosis of cancer at baseline; prior organ transplantation; cirrhosis; chronic kidney disease; subjects who had not visited hospital for > 2 years but later returned after cancer diagnosis; | Incidence of all cancers |
| Chang (2018)47 | Taiwan | Single-center cross-sectional study | Individuals diagnosed with pancreatic cancer between 1/2009 and 12/2013 | 143 individuals with pancreatic cancer and 414 randomly selected control patients without pancreatic cancer | Unenhanced CT scanning evidence of hepatic steatosis | HBV, HCV, or cirrhosis; history of alcohol drinking; prior major operation involving abdominal organ; history of medications that may have contribute to hepatotoxicity | Prevalence of NAFLD in cases and controls |
| Wild (2018)40 | Scotland | Retrospective cohort study of national population-based registry of patients with type 2 diabetes mellitus | Individuals diagnosed with type 2 diabetes between 1/1/2004 and 12/31/2013 | 132,661 adults age 40–89 with type 2 diabetes mellitus | Presence of the ICD-9 and ICD-10 codes for NAFLD: 571.8, K76, K75.8 | ICD-9 and ICD-10 codes for viral hepatitis, autoimmune hepatitis, hemochromatosis, and any cirrhosis, fibrosis, sclerosis, or portal hypertension with no mention of ALD or NAFLD | Incident or recurrent cancer, excluding HCC |
| Nseir (2017)53 | Israel | Single-center cross-sectional study | Individuals referred for a mammography screening exam from 1/2008 to 12/2011 | 73 patients with malignant breast cancer with abdominal CT imaging data within one month of diagnosis and 73 controls with normal screening mammography and breast ultrasonography who had abdominal CT imaging within 3 months of screening examinations | Presence of hepatic steatosis on abdominal CT exam | For NAFLD diagnosis: alcohol consumption > 20 g per day, positive serology for hepatitis B, hepatitis C, positive autoimmune hepatitis antibodies, or any history of another known liver disease | Prevalence of NAFLD in cases and controls |
| Seko (2015)55 | Japan | Single-center retrospective cohort study | Individuals from 1/1999 to 4/2013 | 312 individuals with liver biopsy confirmed NAFLD | Liver biopsy findings of steatosis in 5% or more of hepatocytes | Presence of viral hepatitis, autoimmune hepatitis, drug-induced liver disease, primary biliary cirrhosis, biliary obstruction, hemochromatosis, Wilson's disease and α-1-antitrypsin-deficiency-associated liver disease, heavy alcohol use; decompensated cirrhosis or HCC | Incidence of all cancers |
| Kwak (2019)50 | Korea | Single-center case-control study | Individuals from 1/2008 to 5/2017 | 270 patients with breast cancer diagnosed by screening and 270 controls with normal breast mammography and/or ultrasonography | Hepatic steatosis by ultrasonography | Other chronic liver disease, including HBV or HCV, significant alcohol consumption (defined as >20 g/d for women) | Prevalence of NAFLD in cases and controls |
| Cho (2019)48 | Korea | Single-center cross-sectional study | Individuals from 1/2013 to 11/2018 | 476 individuals with biopsy proven NAFLD | Liver biopsy in which at least 5% of hepatocytes displayed macrovesicular steatosis | Hepatitis B or C virus infection; autoimmune hepatitis or primary biliary cholangitis; drug-induced liver injury or steatosis; Wilson disease or hemochromatosis; excessive alcohol consumption; diagnosis of malignancy within the prior year; family history of CRC in first-degree relatives; having an inherited syndrome (e.g. Lynch syndrome, Peutz-Jeghers syndrome, MYH-associated polyposis or familial adenomatous polyposis); past medical history of colorectal neoplasm; inflammatory bowel disease; bowel symptoms (e.g. hematochezia, melena, or bowel habit change); patients who underwent polypectomy within the last 5 years; patients who declined to undergo colonoscopy | Prevalence of advanced colorectal neoplasm (defined as adenomatous polyp 10 mm or larger and/or with a feature of villous adenoma, and/or high grade dysplasia or adenocarcinoma |
| Yang (2017)56 | Korea | Single-center retrospective cohort study | Individuals from 1/2009 to 12/2013 who underwent surveillance colonoscopy after index colonoscopy | 441 patients with NAFLD and 441 patients without NAFLD – propensity score matched cohorts | Liver ultrasound or CT scan showing diffuse increased hepatic echogenicity compared to kidney, blurring of vascular structure, lower hepatic parenchymal attenuation than that of the spleen. | Prior history of colorectal surgery or colorectal disease; inflammatory bowel disease; an incomplete index colonoscopy; any colonoscopy within the previous 3 years of the index colonoscopy; chronic liver disease, including HBV, HCV or cirrhosis; significant alcohol consumption; no diagnostic examination for NAFLD within 3 months of the index colonoscopy; incomplete clinical information (missing variables ≥2) | Incidence of advanced colorectal neoplasm (defined as cancer or adenomatous polyp 10 mm or larger, any adenoma with tubulovillous or villous histology, high grade dysplasia) |
| Pan (2017)54 | China | Single-center cross-sectional study | Individuals from 1/2011–to 11/2015 | 1793 individuals undergoing colonoscopy as part of routine health checkup (27 with colorectal cancer and 1767 without colorectal cancer) | Hepatic ultrasonography with features including hepatomegaly, diffusely increased echogenicity, and blurring of vasculature | viral hepatitis; cirrhosis; liver cancer or other liver disease; excess alcohol consumption | Prevalence of NAFLD in cases and controls |
| Ahn (2017)46 | Korea | Single-center cross-sectional study | Individuals from 1/2003 to 12/2012 | 26,540 adults undergoing routine comprehensive health check up | Liver ultrasound showing increased parenchymal brightness, liver-to-kidney contrast, deep beam attenuation and bright vessel walls during examination | Cirrhosis, HBV or HCV, excess alcohol consumption >30g/day in men or 20g/day in women | Incidence of advanced colorectal neoplasia, defined as an invasive cancer or adenoma that was at least 10 mm in diameter, had high grade dysplasia, villous or tubulovillous histology or any combination thereof |
| Lin (2014)52 | China | Single-center cross-sectional study | Individuals enrolled from 10/2007–12/2011 | 2315 individuals undergoing routine screening colonoscopy | Ultrasonography features including hepatomegaly, diffusely increased echogenicity of liver parenchyma, and blurring of vasculature | History of colorectal cancer, adenoma and polyp; history of other extraintestinal malignancies; contraindications to colonoscopy; viral hepatitis, cirrhosis, liver cancer or other liver diseases; heavy alcohol consumption | Prevalence of colorectal malignant neoplasm (adenocarcinoma) |
| Lee (2012)51 | Korea | Single-center retrospective cohort study | Individuals enrolled from 7/1/2002–6/30/2006 with follow-up through 12/31/2008 | 5517 women aged 35–80 years undergoing life insurance health examinations | Abdominal ultrasonography assessment of hepatorenal contrast, liver brightness, deep attenuation, and blurring of the vasculature | HBV or HCV positive, alcohol consumption of more than 40 g/week; history of chronic liver disease, including viral, toxic, and autoimmune liver diseases; history of receiving previous medical insurance benefits | Incidence of colorectal cancer |
| Wong (2011)42 | Hong Kong | Multicenter cross-sectional study | Individuals enrolled from 1/2008–7/2010 (two cohorts: community subjects and hospital patients) | 380 individuals | Liver biopsy (hospital cohort) and proton-magnetic resonance spectroscopy (community cohort) | HBV or HCV positive; excess alcohol consumption (men >30g/day and women >20g/day; history of colorectal cancers or polyps; inflammatory bowel disease; bowel symptoms including per rectal bleeding and altered bowel habit; prior colorectal cancer screening; contraindications to colonoscopy | Prevalence of advanced colorectal neoplasms (cancer or adenoma with high grade dysplasia or villous architecture) |