Abstract
Nonalcoholic fatty liver disease (NAFLD) is considered an independent risk factor for the development of cardiovascular disease. However, the association between changes in NAFLD status and the risk of cardiovascular disease (CVD) remains uncertain. Starting January 1, 2013, participants were followed until the occurrence of CVD event, death, or December 31, 2020. This was a population-based cohort study that included data from adults aged ≥ 20, who underwent 2 consecutive health screenings from 2009 to 2012. NAFLD was defined as a Fatty Liver Index ≥ 60 at each screening. The primary endpoint was a CVD event, which encompassed ischemic heart disease and cerebrovascular disease. The association between changes in NAFLD status and the risk of CVD was determined using multivariable Cox proportional hazards regression. This cohort comprised 4656,305 adults with a median age of 53 years. During 36,396,968 person-years of follow-up, 238,933 (5.1%) CVD events were observed. Compared to patients with no NAFLD at both screenings, patients who developed NAFLD at the second screening exhibited an increased risk of CVD (adjusted hazard ratio, 1.15; 95% confidence interval, 1.13–1.17). In contrast, individuals who recovered from NAFLD at the second screening demonstrated a reduced CVD risk compared to those with persistent NAFLD (adjusted hazard ratio, 0.91; 95% confidence interval, 0.90–0.92). The reversal of NAFLD is associated with a reduced risk of CVD. Therefore, focusing on NAFLD treatment could serve as a clinical target for lowering CVD risk.
Keywords: cardiovascular disease, cerebrovascular disease, coronary heart disease, ischemic heart disease, nonalcoholic fatty liver disease, stroke
1. Introduction
Nonalcoholic fatty liver disease (NAFLD), characterized by the excessive accumulation of ectopic fat in the liver, has become the most prevalent liver disease worldwide.[1] The global prevalence of NAFLD has risen to approximately 25%, leading to significant health and economic burdens.[2] The majority of NAFLD patients also suffer from metabolic diseases, such as diabetes, obesity, or hyperlipidemia.[3] For this reason, NAFLD is sometimes perceived as a hepatic manifestation of metabolic disease.[4] Since both NAFLD and cardiovascular disease (CVD) are manifestations of end-organ damage from the metabolic syndrome, their association has been considered through various pathophysiological pathways.[5]
A number of previous studies have reported that NAFLD is a novel risk factor for CVD, independent of conventional risk factors, such as age, sex, obesity, smoking, family history, and the presence of metabolic diseases (hypertension, dyslipidemia, and diabetes).[6,7] Furthermore, a previous meta-analysis indicated that compared to patients without NAFLD, those with NAFLD had a higher risk of CVD,[6] and patients with more advanced NAFLD had increased risk of experiencing CVD events.[8]
However, the disease course of NAFLD does not always follow a unidirectional progressive worsening pattern. Previous studies have shown that the clinical course of NAFLD exhibited improvement or maintenance of NAFLD status in some patients.[9] For this reason, the disease status of NAFLD can be considered variable. While mounting evidence suggests that the presence of NAFLD and disease progression are risk factors for CVD, there is currently no study elucidating the association between changes in NAFLD status and the risk of CVD. If the risk of CVD is influenced by changes in NAFLD status, while considering various traditional cardiovascular risk factors, it could confirm that a strategy for preventing or managing NAFLD might provide a novel approach to preventing CVD. Therefore, the objectives of this study were to determine the association between changes in NAFLD status and the risk of CVD in a Korean nationwide cohort. This study aims to provide support for establishing strategies to manage NAFLD with the goal of preventing subsequent CVD.
2. Methods
2.1. Study population
The study population was obtained from the Korean National Health Insurance Service (NHIS). The NHIS offers mandatory health insurance coverage to all Korean citizens, encompassing nearly all types of health services.[10] Citizens aged at least 20 years are eligible for a biennial health screening, comprising blood and urine laboratory tests, anthropometric measurements, and self-reported lifestyle behavior questionnaires. Additionally, the NHIS gathers data on demographic characteristics, hospitalization and outpatient department visits, and drug prescriptions for research purposes. The NHIS has been utilized in numerous extensive epidemiology studies, and its validity has been previously demonstrated.[10–12]
The Institutional Review Board of Seoul National University approved this study (E-2011-005-1168). The requirement for informed consent was waived because the NHIS database provides anonymized data in accordance with the Personal Data Protection Act guidelines.
2.2. Data availability
The dataset generated or used in this study is available from the NHIS. The authors have no right to share the data but researchers can submit their proposal to the NHIS to obtain data from the NHIS repository (https://nhiss.nhis.or.kr/).
2.3. Key variables
NAFLD was considered present when a participant had FLI ≥ 60, which was calculated using the following formula: (e0.953×loge(triglycerides) + 0.139×body mass index + 0.718×loge(γ-glutamyl transpeptidase) + 0.053×waistcircumference−15.745)/(1 + e0.953× loge(triglycerides) + 0.139×body mass index + 0.718×loge(γ-glutamyl transpeptidase) + 0.053×waistcircumference−15.745) × 100.[13] The FLI is considered as an acceptable alternative to imaging modality according to the European Clinical Practice Guidelines.[14] In Korean population, the FLI was validated to be accurate in a receiver operating characteristic curve (area under curve, 0.87).[15]
The International Classification of Disease 10th Revision (ICD-10) codes of K70-K77 and B18 were used for exclusion of chronic liver diseases and chronic hepatitis, respectively. The primary outcome was CVD (ICD-10, I20–I25 and I60–I69), and secondary outcomes were ischemic heart disease (I20–I25) and cerebrovascular disease (I60–I69), which were defined as at least 2 days of hospitalization due to the ICD-10 codes in accordance with the American Heart Association guidelines.[11,16]
The following variables were considered as key variables for multivariable analyses: age (continuous; years), sex (categorical; men and women), household income (categorical; quartiles), body mass index (kg/m2), cigarette smoking (categorical; never, past, and current), moderate-to-vigorous physical activity (categorical; 0, 1–2, 3–4, and ≥ 5 times/week), systolic blood pressure (continuous; mm Hg), fasting serum glucose (continuous; mg/dL), Charlson comorbidity index (categorical; 0, 1–2, and ≥ 3), antihypertensive medication (categorical; yes and no), antidiabetic medication (categorical; yes and no), antidyslipidemic medication (categorical; yes and no), aspirin (categorical; yes and no), acetaminophen (categorical; yes and no), nonsteroidal anti-inflammatory drugs (categorical; yes and no), and advanced fibrosis (categorical; yes and no).
2.4. Statistical analysis
Follow-up investigation was carried out from January 1, 2013 until the date of CVD, death, or December 31, 2020, whichever occurred earliest. The multivariable-adjusted hazard ratios (aHRs) and their corresponding 95% confidence intervals (CIs) for CVD, ischemic heart disease, and cerebrovascular disease were evaluated using Cox proportional hazards regression after adjustments for the key variables. The first model was conducted with adjustments for age and sex. The second model was calculated with additional adjustments for household income, body mass index, systolic blood pressure, fasting serum glucose, cigarette smoking, moderate-to-vigorous physical activity, and Charlson comorbidity index. The final model was further adjusted for mediation uses and advanced fibrosis. Starting from the date of January 1, 2009, all participants prescription records were checked for medication uses until 2012. Advanced fibrosis was considered present when a participant had the body mass index, aspartate aminotransferase/alanine aminotransferase ratio, and diabetes mellitus score ≥ 2.[17] The assumption for proportionality was not violated. Any participant with missing information on evaluation of NAFLD or covariates were excluded in the analytic cohort before the formal analysis. Sensitivity analyses were performed to assess the association between changes in NAFLD status and the risk of CVD by excluding CVD events that occurred within 1, 3, and 5 years of follow-up, respectively. A P value of < .05 was considered statistically significant. All data mining and statistical analyses were carried out using SAS version 9.4 (SAS Institute Inc.).
3. Results
The study population consisted of 9131,352 adults aged ≥ 20 years who underwent consecutive health screenings during both the first (2009–2010) and second (2011–2012) periods. Among them, 54,970 adults who died and 2524,322 adults who had CVD before the date of follow-up investigation (January 1, 2013) were excluded. In addition, 10,567 adults with missing information for the evaluation of NAFLD status, 52,577 adults with missing information for the covariates, and 1832,611 adults with chronic viral hepatitis B or C and other chronic liver diseases were excluded, respectively. The analytic cohort comprised 4656,305 adults (Fig. 1).
Figure 1.
Flow diagram showing the enrollment of the study population.
3.1. Participant characteristics
Descriptive characteristics of the analytic cohort is shown in Table 1. The number of participants with NAFLD and no NAFLD at baseline were 1795,429 and 2860,876, respectively. Compared with no NAFLD individuals, those with NAFLD tended to be older, men, lower household income, higher body mass index, higher systolic blood pressure, higher diastolic blood pressure, higher triglyceride, higher high-density lipoprotein cholesterol, higher alanine aminotransferase, higher aspartate aminotransferase, higher γ-glutamyl transpeptidase, past or current smokers, and had more comorbidities.
Table 1.
Descriptive characteristics of the participants.
| Characteristic | Overall participant (n = 4656,305) | NAFLD (n = 1795,429) | No NAFLD (n = 2860,876) | P value |
|---|---|---|---|---|
| Age, yr | 53 (43 to 64) | 56 (47 to 65) | 51 (42 to 62) | <.001 |
| Sex, n (%) | <.001 | |||
| Men | 1355,278 (29.1) | 798,866 (44.5) | 556,412 (19.4) | |
| Women | 3301,027 (70.9) | 996,563 (55.5) | 2304,464 (80.6) | |
| Household income*, n (%) | <.001 | |||
| First quartile | 1507,758 (32.4) | 596,847 (33.2) | 910,911 (31.8) | |
| Second quartile | 1162,431 (25.0) | 456,043 (25.4) | 706,388 (24.7) | |
| Third quartile | 920,869 (19.8) | 340,875 (19.0) | 579,994 (20.3) | |
| Fourth quartile (highest) | 1065,247 (22.9) | 401,664 (22.4) | 663,583 (23.2) | |
| Body mass index, kg/m2 | 23.3 (21.3 to 25.5) | 25.7 (24.1 to 27.6) | 21.9 (20.4 to 23.5) | <.001 |
| Waist circumference, cm | 79 (72 to 85) | 86 (81 to 90) | 74 (70 to 79) | <.001 |
| Systolic blood pressure, mm Hg | 120 (110 to 130) | 128 (119 to 136) | 119 (110 to 130) | <.001 |
| Diastolic blood pressure, mm Hg | 76 (70 to 80) | 80 (71 to 84) | 73 (68 to 80) | <.001 |
| Triglyceride, mg/dL | 105 (73 to 151) | 146 (106 to 203) | 86 (64 to 117) | <.001 |
| HDL cholesterol, mg/dL | 55 (46 to 64) | 50 (43 to 58) | 58 (49 to 67) | <.001 |
| Alanine aminotransferase, IU/L | 18 (14 to 25) | 24 (18 to 34) | 16 (13 to 20) | <.001 |
| Aspartate aminotransferase, IU/L | 22 (19 to 26) | 24 (20 to 29) | 21 (18 to 25) | <.001 |
| γ-glutamyl transpeptidase, IU/L | 18 (14 to 27) | 28 (21 to 40) | 15 (12 to 19) | <.001 |
| Change in fatty liver index | 0.0 (−8.1−8.2) | 1.4 (−3.2−14.4) | −1.1 (−12.3−5.4) | <.001 |
| Cigarette smoking, n (%) | <.001 | |||
| Never smoker | 3751,651 (80.6) | 1266,870 (70.6) | 2484,781 (86.9) | |
| Former smoker | 373,963 (8.0) | 216,815 (12.1) | 157,148 (5.5) | |
| Current smoker | 530,691 (11.4) | 311,744 (17.4) | 218,947 (7.7) | |
| MVPA, n (%) | <.001 | |||
| 0 time/wk | 2317,746 (49.8) | 906,507 (50.5) | 1411,239 (49.3) | |
| 1–2 time/wk | 771,093 (16.6) | 282,422 (15.7) | 488,671 (17.1) | |
| 3–4 time/wk | 609,824 (13.1) | 230,184 (12.8) | 379,640 (13.3) | |
| ≥5 time/wk | 957,642 (20.6) | 376,316 (21.0) | 581,326 (20.3) | |
| Charlson comorbidity index, n (%) | <.001 | |||
| 0 | 2401,329 (51.6) | 826,227 (46.0) | 1575,102 (55.1) | |
| 1–2 | 1753,876 (37.7) | 711,613 (39.6) | 1042,263 (36.4) | |
| ≥3 | 501,100 (10.8) | 257,589 (14.3) | 243,511 (8.5) |
Data are presented as median (interquartile range) unless otherwise specified.
HDL = high-density lipoprotein, MVPA = moderate-to-vigorous physical activity, NAFLD = nonalcoholic fatty liver disease.
Proxy for socioeconomic status based on the insurance premium of the National Health Insurance Service.
3.2. Association of change in NAFLD status with risk of CVD
CVD outcomes, the primary end point of this study, is shown in Table 2. During 36,396,968 person-years of follow-up investigation, 238,933 (5.1%) CVD events were identified. Compared with continual no NAFLD, newly developed NAFLD at second health screening had elevated risk of CVD (crude rate, 4.6 vs 7.0 per 1000 person-years; aHR, 1.15; 95% CI, 1.13–1.17; P < .001). In contrast, those who had no NAFLD at second health screening had lower risk of CVD compared to participants with continual NAFLD (crude rate, 9.4 vs 8.5 per 1000 person-years; aHR, 0.91; 95% CI, 0.90–0.92; P < .001). Risks of ischemic heart disease and cerebrovascular disease were consistent with CVD, and variations of aHR were relatively wider in ischemic heart disease compared to cerebrovascular disease. The sensitivity analyses that were performed after excluding CVD events occurred within 1, 3, and 5 years since the date of follow-up began, risk of CVD supported the primary finding (Table 3).
Table 2.
Association of change in fatty liver index with risk of cardiovascular disease according to the fatty liver index cutoff criteria for NAFLD.
| At 1st health screening | No NAFLD | P value | NAFLD | P value | ||
|---|---|---|---|---|---|---|
| At 2nd health screening | No NAFLD | NAFLD | NAFLD | No NAFLD | ||
| No. of participants | 2495,321 | 356,827 | 1438,602 | 365,555 | ||
| Cardiovascular disease | ||||||
| PY | 19,646,508 | 2784,680 | 11,127,840 | 2837,940 | ||
| Event | 90,174 (3.6) | 19,619 (5.5) | 105,000 (7.3) | 24,140 (6.6) | ||
| Crude rate/1000 PY | 4.6 | 7.0 | 9.4 | 8.5 | ||
| HR (95% CI) | 1.00 (reference) | 1.54 (1.51–1.56) | <.001 | 1.00 (reference) | 0.90 (0.89–0.91) | <.001 |
| aHR (95% CI)* | 1.00 (reference) | 1.29 (1.28–1.32) | <.001 | 1.00 (reference) | 0.85 (0.84–0.86) | <.001 |
| aHR (95% CI)† | 1.00 (reference) | 1.16 (1.15––1.18) | <.001 | 1.00 (reference) | 0.91 (0.89–0.92) | <.001 |
| aHR (95% CI)‡ | 1.00 (reference) | 1.15 (1.13–1.17) | <.001 | 1.00 (reference) | 0.91 (0.90–0.92) | <.001 |
| Ischemic heart disease | ||||||
| PY | 19,844,591 | 2824,301 | 11,322,981 | 2887,756 | ||
| Event | 31,182 (1.2) | 8101 (2.3) | 49,885 (3.5) | 9870 (2.7) | ||
| Crude rate/1000 PY | 1.6 | 2.9 | 4.4 | 3.4 | ||
| HR (95% CI) | 1.00 (reference) | 1.83 (1.78–1.87) | <.001 | 1.00 (reference) | 0.78 (0.76–0.79) | <.001 |
| aHR (95% CI)* | 1.00 (reference) | 1.51 (1.47–1.55) | <.001 | 1.00 (reference) | 0.77 (0.75–0.79) | <.001 |
| aHR (95% CI)† | 1.00 (reference) | 1.30 (1.26–1.33) | <.001 | 1.00 (reference) | 0.84 (0.82–0.86) | <.001 |
| aHR (95% CI)‡ | 1.00 (reference) | 1.26 (1.23–1.30) | <.001 | 1.00 (reference) | 0.86 (0.84–0.88) | <.001 |
| Cerebrovascular disease | ||||||
| PY | 19,751,844 | 2812,353 | 11,299,778 | 2871,618 | ||
| Event | 61,374 (2.5) | 12,158 (3.4) | 59,017 (4.1) | 15,067 (4.1) | ||
| Crude rate/1000 PY | 3.1 | 4.3 | 5.2 | 5.2 | ||
| HR (95% CI) | 1.00 (reference) | 1.39 (1.37–1.42) | <.001 | 1.00 (reference) | 1.01 (0.99–1.02) | <.001 |
| aHR (95% CI)* | 1.00 (reference) | 1.18 (1.16–1.21) | <.001 | 1.00 (reference) | 0.91 (0.90–0.93) | <.001 |
| aHR (95% CI)† | 1.00 (reference) | 1.10 (1.08–1.13) | <.001 | 1.00 (reference) | 0.95 (0.93–0.97) | <.001 |
| aHR (95% CI)‡ | 1.00 (reference) | 1.10 (1.08–1.12) | <.001 | 1.00 (reference) | 0.95 (0.93–0.96) | <.001 |
aHRs calculated using the Cox proportional hazards model. NAFLD defined as fatty liver index ≥ 60.
aHR = adjusted hazard ratio, CI = confidence interval, NAFLD = nonalcoholic fatty liver disease, PY = person-year.
Adjusted for age and sex.
Further adjusted for household income, body mass index, systolic blood pressure, fasting serum glucose, cigarette smoking, moderate-to-vigorous physical activity, and Charlson comorbidity index on the basis of Model A.
Further adjusted for antihypertensive medication, antidiabetic medication, antidyslipidemic medication, aspirin, acetaminophen, nonsteroidal anti-inflammatory drugs, and advanced fibrosis on the basis of Model B.
Table 3.
Sensitivity analysis on association of change in fatty liver status with risk of cardiovascular disease.
| At 1st health screening | No NAFLD | P value | NAFLD | P value | ||
|---|---|---|---|---|---|---|
| At 2nd health screening | No NAFLD | NAFLD | NAFLD | No NAFLD | ||
| aHR (95% CI)* | 1.00 (reference) | 1.15 (1.13–1.17) | <.001 | 1.00 (reference) | 0.92 (0.90–0.93) | <.001 |
| aHR (95% CI)† | 1.00 (reference) | 1.13 (1.11–1.16) | <.001 | 1.00 (reference) | 0.92 (0.91–0.94) | <.001 |
| aHR (95% CI)‡ | 1.00 (reference) | 1.13 (1.10–1.15) | <.001 | 1.00 (reference) | 0.93 (0.91–0.96) | <.001 |
aHRs calculated using the Cox proportional hazards model after adjustments for baseline fatty liver index, age, sex, household income, body mass index, systolic blood pressure, fasting serum glucose, cigarette smoking, moderate-to-vigorous physical activity, Charlson comorbidity index, antihypertensive medication, antidiabetic medication, antidyslipidemic medication, aspirin, acetaminophen, nonsteroidal anti-inflammatory drugs, and advanced fibrosis.
aHR = adjusted hazard ratio, CI = confidence interval, NAFLD = nonalcoholic fatty liver disease.
Calculated after excluding cardiovascular disease cases within 1 year.
Calculated after excluding cardiovascular disease cases within 3 years.
Calculated after excluding cardiovascular disease cases within 5 years.
3.3. Stratified analysis
Stratified analyses of the association between NAFLD and the risk of CVD among initially NAFLD-free adults is shown in Figure 2. All participants were stratified according to age, sex, obesity, advanced fibrosis, Charlson comorbidity index, and medication uses. aHRs for NAFLD were in a range of 1.09 to 1.18 among different subgroups in participants without NAFLD at first health screening, and all subgroups showed higher risk of CVD in NAFLD group along with significant interaction. Stratified analyses comparing NAFLD and non-NAFLD among initially NAFLD participants are shown in Figure 3. All subgroups showed lower risk of CVD in No NAFLD at second health screening group (P < .001), and aHRs were in a range between 0.86 and 0.94. In addition, significant interaction was found for age, sex, Charlson comorbidity index, antihypertensive medication, antidyslipidemic medication, aspirin, acetaminophen, and nonsteroidal anti-inflammatory drugs.
Figure 2.
Stratified analysis of the association of NAFLD with CVD risk among participants without NAFLD at baseline. CVD = cardiovascular disease, NAFLD = Nonalcoholic fatty liver disease.
Figure 3.
Stratified analysis of the association of no NAFLD with risk of CVD among participants with NAFLD at baseline. CVD = cardiovascular disease, NAFLD = Nonalcoholic fatty liver disease.
4. Discussion
In this nationwide large-scale cohort study, we discovered a significant association between changes in NAFLD status and the risk of CVD. The development of new NAFLD was linked to a higher CVD risk compared to a persistent non-NAFLD. Conversely, reversal of NAFLD at second health screening reduced the risk of CVD compared to a persistent NAFLD status. Our stratified analysis showed that age, sex, obesity, medications, comorbidity status, and the presence of advanced fibrosis were associated with CVD risk in relation to newly developed NAFLD. The reduced CVD risk following the reversal of NAFLD was associated with age, sex, comorbidity status, and some medications. Interestingly, when recovered from NAFLD, the presence of obesity exhibited no significant difference in reducing CVD risk. Taken together, our study highlights a significant association between changes in NAFLD status and the risk of CVD.
Recent evidences highlights a close relationship between NAFLD and CVD. According to a meta-analysis, the relative risk for coronary artery disease is 2.26 in NAFLD patients (95% CI: 1.04–4.92).[18] Moreover, the risks for ischemic stroke (Odds ratio; 2.51; 95% CI 1.92–3.28) and hemorrhagic stroke(Odds ratio, 1.85; 95% CI, 1.05–3.27) are much higher among NAFLD patients.[19] These pieces of evidence suggest that NAFLD may act as a risk factor for CVD.
NAFLD is considered as hepatic manifestation of metabolic disorder.[4] NAFLD and metabolic disorders share several cardiometabolic risk factors that contribute to CVD.[20] For this reason, excluding the common risk factors shared by both diseases, it becomes challenging to discern whether there is an additional effect of NAFLD that independently increases the risk of CVD.[5] In this context, the findings of our study provide implications. According to the stratified analysis in our study, in the case of newly developed NAFLD, subgroups without traditional CVD risk factors, such as obesity, hypertension, diabetes, and dyslipidemia, exhibited a significantly higher risk of CVD than the subgroups with traditional CVD risk factors. Although obesity is known to be a major risk factor for CVD,[21] non-obese patients were found to exhibit a higher risk of CVD than obese patients when newly developing NAFLD. The results suggest that NAFLD has an additional effect on CVD risk beyond the traditional risk factors shared with CVD. Similarly, it was demonstrated that the reduction in CVD risk was consistent whether the participant was obese or not, as long as they recovered from NAFLD.
The mechanisms through which NAFLD increases CVD risk involve pathways that are concurrently associated with metabolic, cardiovascular, and hepatic functions. Several studies propose that NAFLD increases the risk of developing CVD through numerous pathophysiological mechanisms.[5] The underlying mechanisms linking NAFLD to CVD encompass various pathways, including atherogenic dyslipidemia,[22] endothelial dysfunction,[23] systemic/vascular inflammation,[24] oxidative stress,[25] systemic insulin resistance,[26] plaque formation[27] and instability,[28] and epigenetic alterations.[29] While NAFLD shares these common atherogenic pathophysiologic mechanisms, considering our findings that NAFLD independently increases the risk of CVD, it becomes essential to discern the specific mechanisms by which NAFLD contributes to this risk. Although the exact mechanism involved in the atherosclerotic changes of systemic blood vessels related to NAFLD has not been elucidated, there are several mechanisms thought to be related to it. First, in NALFD patients, impaired intrahepatic vascular function and hepatic fibrosis can lead to endothelial dysfunction and increased production of pro-thrombotic molecules and angiogenic factors that can affect the systemic vasculature.[30] Second, hepatokines play a key role in another mechanism linking NAFLD and CVD. Hepatokines are proteins that are secreted by hepatocytes to regulate systemic metabolic function.[31] Previous studies revealed that hepatokines, such as fetuin-A, fibroblast growth factor 21 (FGF-21), selenoprotein P, leukocyte cell derived chemotaxin 2 (LECT2), and angiopoietin-like protein, are closely related to CVD.[32] In this text, NAFLD, characterized by ectopic fat accumulation in the liver, is known to alter hepatokine secretion. In NAFLD patients, it is already known that serum levels of fetuin-A, FGF-21, angiopoietin-like protein and LECT2 are increased, while serum selenoprotein P level is decreased.[33–37] These altered hepatokines secretion patterns modify the systemic metabolic control function of the liver, altering metabolic risks, and consequently impacting the risk of CVD. To elaborate on the roles of hepatokines in detail, it is worth noting that fetuin-A exerts stimulatory effects on inflammatory responses in human umbilical vein endothelial cells, macrophage foam cell formation, and the proliferation and collagen production in human aortic smooth muscle cells, ultimately contributing to the development of atherosclerosis.[38] A high serum concentration of FGF-21 is closely associated with an increased risk of CVD.[33] However, when considering the potential preventive roles of FGF-21 in atherosclerosis, its increase might not be the foundation of atherosclerotic pathogenesis; rather, it could be increased as a compensatory response to atherosclerosis.[39] LECT2 is positively associated with hepatic inflammatory signaling, obesity, NAFLD, insulin resistance, and CVD risk.[40,41] Serum selenoprotein P level is inversely associated with CVD risk. Selenoprotein P reduces tissue oxidative stress by transporting selenium to vital tissues, binding heavy metals like Cd, As, and Hg, and protects cell membrane integrity by offering glutathione peroxidase activity.[42] All the mechanisms mentioned above are mediated by various CVD risk factors and NAFLD. Taking this into consideration, established NAFLD can be considered a risk factor for CVD, and recovery from NAFLD implies a departure from the major risk factors of CVD.
This study possesses several strengths. Firstly, to the best of our knowledge, it is the first study to demonstrate a reduction in CVD risk through the reversal of NAFLD. Secondly, we conducted an analysis using a nationwide population-based database with repeated measurements to evaluate changes in NAFLD status. The identification of CVD events was based on national-level medical claims records. Furthermore, our study is based on a substantial population of approxiamtely 4.65 million adults, enhancing the reliability of our results. Lastly, the study accounted for various health behaviors and metabolic factors related to CVD, such as physical activity, smoking status, blood pressure, serum glucose, total cholesterol, comorbidities, and mediation history, thereby augmenting the reliability of our findings.
This study has several limitations. Firstly, changes in NAFLD status were defined using the FLI. While liver biopsy is considered the gold standard for diagnosing NAFLD, imaging tests, such as ultrasound, computed tomography, and liver elastography, can be used as alternatives. The diagnosis of NALFD status using the FLI may bdiffer from that obtained through liver biopsy or other imaging modalities. Since the NHIS database lacked information on liver biopsy or other imaging test results, further studies incorporating pathology records or imaging test results to investigate the association between changes in NAFLD status and the risk of CVD are needed. Second, in this study, the severity of NAFLD could not be evaluated due to the limitations of the aforementioned modality. Evaluating the severity of NAFLD independently, apart from its presence or absence, suggests the need for a follow-up study to evaluate the intrinsic impact of NAFLD on CVD risk. Lastly, the generalizability of our results is confined to Korean adults who underwent health screening through the NHIS. Additional evidence from other ethnic or multiethnic population-based cohorts is essential for a broader understanding.
5. Conclusion
Changes in NAFLD status are associated with the risk of CVD among Korean adults. Notably, recovery from NAFLD significantly reduced the risk of CVD independent of traditional cardiometabolic risk factors. Given the observed association between changes in NAFLD status and CVD risk in this study, improved monitoring of changes in NAFLD status and facilitating NAFLD recovery could hold significant clinical relevance as a means to lower future CVD risk.
Author contributions
Conceptualization: Yun Hwan Oh.
Data curation: Seogsong Jeong.
Formal analysis: Seogsong Jeong.
Funding acquisition: Yun Hwan Oh.
Investigation: Sun Jae Park.
Methodology: Joseph C Ahn.
Project administration: Joseph C Ahn, Sang Min Park.
Resources: Sang Min Park.
Software: Sang Min Park.
Supervision: Sang Min Park.
Validation: Sang Min Park.
Writing – original draft: Yun Hwan Oh.
Writing – review & editing: Sang Min Park.
Abbreviations:
- aHR
- adjusted hazard ratio
- CI
- confidence interval
- CVD
- cardiovascular disease
- FGF-21
- fibroblast growth factor 21
- ICD-10
- International Classification of Disease 10th Revision
- LECT2
- leukocyte cell derived chemotaxin 2
- NAFLD
- Nonalcoholic fatty liver disease
- NHIS
- National Health Insurance Service
YH and SJ contributed equally to this work.
This work was supported by a research grant from the Jeju National University Hospital in 2020 (2020-32).
The datasets generated during and/or analyzed during the current study are publicly available.
The authors have no funding and conflicts of interest to disclose.
How to cite this article: Oh YH, Jeong S, Park SJ, Ahn JC, Park SM. Reversal of nonalcoholic fatty liver disease reduces the risk of cardiovascular disease among Korean. Medicine 2023;102:44(e35804).
Contributor Information
Yun Hwan Oh, Email: swimayo@gmail.com.
Seogsong Jeong, Email: seogsongjeong@gmail.com.
Sun Jae Park, Email: swimayo@cauhs.or.kr.
Joseph C Ahn, Email: Ahn.Joseph@mayo.edu.
References
- [1].LaBrecque DR, Abbas Z, Anania F, et al. World Gastroenterology Organisation global guidelines: nonalcoholic fatty liver disease and nonalcoholic steatohepatitis. J Clin Gastroenterol. 2014;48:467–73. [DOI] [PubMed] [Google Scholar]
- [2].Chalasani N, Younossi Z, Lavine JE, et al. The diagnosis and management of nonalcoholic fatty liver disease: practice guidance from the American Association for the Study of Liver Diseases. Hepatology. 2018;67:328–57. [DOI] [PubMed] [Google Scholar]
- [3].Singh SP, Singh A, Misra D, et al. Risk factors associated with non-alcoholic fatty liver disease in Indians: a case–control study. J Clin Exp Hepatol. 2015;5:295–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Caballería L, Pera G, Auladell MA, et al. Prevalence 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. [DOI] [PubMed] [Google Scholar]
- [5].Stahl EP, Dhindsa DS, Lee SK, et al. Nonalcoholic fatty liver disease and the heart: JACC State-of-the-Art Review. J Am Coll Cardiol. 2019;73:948–63. [DOI] [PubMed] [Google Scholar]
- [6].Targher G, Byrne CD, Lonardo A, et al. Non-alcoholic fatty liver disease and risk of incident cardiovascular disease: a meta-analysis. J Hepatol. 2016;65:589–600. [DOI] [PubMed] [Google Scholar]
- [7].Hamaguchi M, Kojima T, Takeda N, et al. Nonalcoholic fatty liver disease is a novel predictor of cardiovascular disease. World J Gastroenterol. 2007;13:1579–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Henson JB, Simon TG, Kaplan A, et al. Advanced fibrosis is associated with incident cardiovascular disease in patients with non-alcoholic fatty liver disease. Aliment Pharmacol Ther. 2020;51:728–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Kleiner DE, Brunt EM, Wilson LA, et al. Association of histologic disease activity with progression of nonalcoholic fatty liver disease. JAMA Netw Open. 2019;2:e1912565–e1912565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Cheol Seong S, Kim Y-Y, Khang Y-H, et al. Data resource profile: the national health information database of the National Health Insurance Service in South Korea. Int J Epidemiol. 2017;46:799–800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Son JS, Choi S, Kim K, et al. Association of blood pressure classification in Korean young adults according to the 2017 American College of Cardiology/American Heart Association Guidelines with subsequent cardiovascular disease events. JAMA. 2018;320:1783–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Lee J, Lee JS, Park S-H, et al. Cohort profile: the national health insurance service–national sample cohort (NHIS-NSC), South Korea. Int J Epidemiol. 2017;46:e15–e15. [DOI] [PubMed] [Google Scholar]
- [13].Bedogni G, Bellentani S, Miglioli L, et al. The Fatty Liver Index: a simple and accurate predictor of hepatic steatosis in the general population. BMC Gastroenterol. 2006;6:1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Liver EAftSoT, Diabetes EAftSo. EASL-EASD-EASO clinical practice guidelines for the management of non-alcoholic fatty liver disease. Obes Facts. 2016;9:65–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Lee Y-H, Bang H, Park YM, et al. Non–laboratory-based self-assessment screening score for non-alcoholic fatty liver disease: development, validation and comparison with other scores. PLoS One. 2014;9:e107584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics-2016 update: a report from the American Heart Association. Circulation. 2016;133:e38–360. [DOI] [PubMed] [Google Scholar]
- [17].Harrison SA, Oliver D, Arnold HL, et al. Development and validation of a simple NAFLD clinical scoring system for identifying patients without advanced disease. Gut. 2008;57:1441–7. [DOI] [PubMed] [Google Scholar]
- [18].Mahfood Haddad T, Hamdeh S, Anantha Narayanan M, et al. Nonalcoholic fatty liver disease and the risk of clinical cardiovascular events: a meta-analysis. Circulation. 2015;132(suppl_3):A14339–A14339. [DOI] [PubMed] [Google Scholar]
- [19].Hu J, Xu Y, He Z, et al. Increased risk of cerebrovascular accident related to non-alcoholic fatty liver disease: a meta-analysis. Oncotarget. 2018;9:2752–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Adams LA, Anstee QM, Tilg H, et al. Non-alcoholic fatty liver disease and its relationship with cardiovascular disease and other extrahepatic diseases. Gut. 2017;66:1138–53. [DOI] [PubMed] [Google Scholar]
- [21].Mathieu P, Lemieux I, Després JP. Obesity, inflammation, and cardiovascular risk. Clin Pharmacol Ther. 2010;87:407–16. [DOI] [PubMed] [Google Scholar]
- [22].DeFilippis AP, Blaha MJ, Martin SS, et al. Nonalcoholic fatty liver disease and serum lipoproteins: the Multi-Ethnic Study of Atherosclerosis. Atherosclerosis. 2013;227:429–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Villanova N, Moscatiello S, Ramilli S, et al. Endothelial dysfunction and cardiovascular risk profile in nonalcoholic fatty liver disease. Hepatology. 2005;42:473–80. [DOI] [PubMed] [Google Scholar]
- [24].Libby P, Everett BM. Novel antiatherosclerotic therapies. Arterioscler Thromb Vasc Biol. 2019;39:538–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Pacana T, Cazanave S, Verdianelli A, et al. Dysregulated hepatic methionine metabolism drives homocysteine elevation in diet-induced nonalcoholic fatty liver disease. PLoS One. 2015;10:e0136822. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Shulman G. Ectopic fat in insulin resistance. Dyslipidemia, and Cardiometabolic Disease [accessed on 24 October 2017]. [Google Scholar]
- [27].Coulon S, Francque S, Colle I, et al. Evaluation of inflammatory and angiogenic factors in patients with non-alcoholic fatty liver disease. Cytokine. 2012;59:442–9. [DOI] [PubMed] [Google Scholar]
- [28].Tripodi A, Fracanzani AL, Primignani M, et al. Procoagulant imbalance in patients with non-alcoholic fatty liver disease. J Hepatol. 2014;61:148–54. [DOI] [PubMed] [Google Scholar]
- [29].Dongiovanni P, Petta S, Maglio C, et al. Transmembrane 6 superfamily member 2 gene variant disentangles nonalcoholic steatohepatitis from cardiovascular disease. Hepatology. 2015;61:506–14. [DOI] [PubMed] [Google Scholar]
- [30].Francque S, Laleman W, Verbeke L, et al. Increased intrahepatic resistance in severe steatosis: endothelial dysfunction, vasoconstrictor overproduction and altered microvascular architecture. Lab Invest. 2012;92:1428–39. [DOI] [PubMed] [Google Scholar]
- [31].Stefan N, Häring H-U. The role of hepatokines in metabolism. Nat Rev Endocrinol. 2013;9:144–52. [DOI] [PubMed] [Google Scholar]
- [32].Yoo HJ, Choi KM. Hepatokines as a link between obesity and cardiovascular diseases. Diabetes Metab J. 2015;39:10–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [33].Zhang Y, Yan J, Yang N, et al. High-level serum fibroblast growth factor 21 concentration is closely associated with an increased risk of cardiovascular diseases: a systematic review and meta-analysis. Front Cardiovasc Med. 2021;8:705273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [34].Ke Y, Liu S, Zhang Z, et al. Circulating angiopoietin-like proteins in metabolic-associated fatty liver disease: a systematic review and meta-analysis. Lipids Health Dis. 2021;20:55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [35].Polyzos SA, Kountouras J, Mavrouli M, et al. Selenoprotein P in patients with nonalcoholic fatty liver disease. Exp Clin Endocrinol Diabetes. 2019;127:598–602. [DOI] [PubMed] [Google Scholar]
- [36].Liu S, Xiao J, Zhao Z, et al. Systematic review and meta-analysis of circulating fetuin-a levels in nonalcoholic fatty liver disease. J Clin Transl Hepatol. 2021;9:3–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [37].Yoo HJ, Hwang SY, Choi J-H, et al. Association of leukocyte cell-derived chemotaxin 2 (LECT2) with NAFLD, metabolic syndrome, and atherosclerosis. PLoS One. 2017;12:e0174717–e0174717. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [38].Naito C, Hashimoto M, Watanabe K, et al. Facilitatory effects of fetuin-a on atherosclerosis. Atherosclerosis. 2016;246:344–51. [DOI] [PubMed] [Google Scholar]
- [39].Cheng P, Zhang F, Yu L, et al. Physiological and pharmacological roles of FGF21 in cardiovascular diseases. J Diabetes Res. 2016;2016:1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [40].Jensen-Cody SO, Potthoff MJ. Hepatokines and metabolism: deciphering communication from the liver. Mol Metab. 2021;44:101138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [41].Wei M, Liu J, Pan H, et al. Plasma Leukocyte Cell-Derived Chemotaxin 2 (LECT2) as a risk factor of coronary artery disease: a cross-sectional study. Angiology. 2021;73:265–74. [DOI] [PubMed] [Google Scholar]
- [42].Saito Y, Sato N, Hirashima M, et al. Domain structure of bi-functional selenoprotein P. Biochem J. 2004;381:841–6. [DOI] [PMC free article] [PubMed] [Google Scholar]