Association between cardiac conduction defect and nonalcoholic fatty liver disease: a systematic review and meta-analysis.

BACKGROUND: Cardiovascular disease is a common comorbidity of patients with nonalcoholic fatty liver disease (NAFLD), particularly coronary artery disease and congestive heart failure. However, the relation between NAFLD and cardiac conduction defects has not been well studied. This systematic review and meta-analysis was conducted to identify all available studies on this association and summarize their results.
METHODS: A comprehensive literature review was conducted using MEDLINE and EMBASE databases through June 2020 to identify studies that compared the risk of a cardiac conduction defect among patients with NAFLD versus those without. Effect estimates from each study were extracted and combined using the random-effect, generic inverse variance method of DerSimonian and Laird.
RESULTS: Three cross-sectional studies with 3651 participants fulfilled the eligibility criteria and were included in this meta-analysis. The risk of a cardiac conduction defect was significantly higher among patients with NAFLD than in those without NAFLD, with a pooled odds ratio of 5.17 (95% confidence interval 1.34-20.01; I 2 96%).
CONCLUSION: A significantly greater risk of cardiac conduction defects among patients with NAFLD was observed in this meta-analysis. How this risk should be managed in clinical practice requires further investigation. Copyright: © Hellenic Society of Gastroenterology.

Introduction

Nonalcoholic fatty liver disease (NAFLD) is defined as the presence of more than 5% of hepatic fat accumulation without a history of excessive alcohol consumption or any secondary causes of liver disease [1]. A recent meta-analysis has estimated the pooled incidence of NAFLD to be approximately 50 per 1000 person-years in Asia and approximately 25 per 1000 person-years in Europe [2-4]. Several metabolic disorders, such as diabetes mellitus, obesity, dyslipidemia, hypertension and hyperuricemia, are more common among patients with NAFLD [5-9]. The relationship between NAFLD and these disorders is bidirectional and is linked by insulin resistance and chronic inflammation [10].

Cardiovascular disease is another common associated disorder among patients with NAFLD [11,12], particularly coronary artery disease and congestive heart failure, as has been observed by several epidemiologic studies [13-16]. Patients with NAFLD may also be at an increased risk of cardiac conduction defects, although the evidence for this particular type of cardiovascular disease is still relatively limited [17-19]. This systematic review and meta-analysis was conducted to identify all observational studies that investigated the risk of cardiac conduction defects among patients with NAFLD versus those without and to summarize their results together to better characterize this risk.

Materials and methods

Information sources and search strategy

We conducted a comprehensive literature review using the MEDLINE and EMBASE databases from inception to June 2020 and isolated all original studies relevant to the association between NAFLD and cardiac conduction defect. The comprehensive literature search was conducted by 3 investigators (KW, PP, and PU) individually, using search terms that included the following: “nonalcoholic fatty liver disease”, “nonalcoholic steatohepatitis”, “bundle branch block”, “heart block”, and “cardiac conduction defect”, as described in online supplementary data 1. No language limitation was applied. In addition, a manual search was performed for potentially valid studies using existing references cited in selected articles. This meta-analysis followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) statement provided as online supplementary data 2.

Selection criteria

Eligible studies had to be cross-sectional, case-control or cohort studies that investigated the relationship between NAFLD and cardiac conduction defects. They had to provide effect estimates—odds ratios (OR), relative risks, hazard ratios, or standardized incidence ratio—with 95% confidence intervals (CI). Inclusion was not limited by study size. In the case of overlap, we included the studies with the most comprehensive data.

Three investigators (KW, PP, and PU) individually reviewed selected articles for their eligibility. Disagreement was resolved by conference with all investigators to reach a mutual final decision. The quality of case-control and cohort studies was assessed using the Newcastle-Ottawa quality scale. The modified version of this scale as described by Herzog et al was used for cross-sectional studies [20].

Data abstraction

A specially designed table was used to collect the following data from each study: study title, name of first author, publication year, country or countries of origin of research, number of subjects, subjects’ demographic data, methods used to identify and verify NAFLD and cardiac conduction defect, 95%CI of adjusted effect estimates with covariates in multivariable analysis.

To ensure accuracy, 2 investigators (KW and PP) individually performed this data extraction process. The data table was finalized by the senior investigator (PU).

Statistical analysis

Data analysis was performed using the Review Manager 5.3 software from the Cochrane Collaboration (London, United Kingdom). The generic inverse variance method of DerSimonian and Laird was used to adjust point estimates from each study. This method weighted each study for the pooled analysis based on its variance [21]. A random-effect model was used, rather than a fixed-effect model, as we expected that the included studies would be somewhat different from each other with respect to the background populations and the methods used to diagnose NAFLD and cardiac conduction defect. Cochran’s Q test and I2 statistic were used to determine heterogeneity between the studies. Four intervals of I2 were used: less than or equal to 25% represents insignificant heterogeneity; 26-50% represents low heterogeneity; 51-75% represents moderate heterogeneity; and more than 75% represents high heterogeneity [22].

Results

Two hundred twenty-nine articles were initially identified using the described search strategy (95 articles from Medline and 134 articles from EMBASE). After the exclusion of 90 duplicated articles, 139 articles underwent title and abstract review. One hundred and twenty articles were excluded at this stage since they were case reports, case series, correspondence, review articles, in vitro studies, animal studies or interventional studies, leaving 19 articles for full-text review. Sixteen of these were excluded after the full-length review as they did not report the association of interest. Finally, 3 cross-sectional studies [17-19] with 3651 participants were included in the final analysis. A flowchart showing the literature search, review and selection process is shown in Fig. 1. The study characteristics and quality assessment using the modified Newcastle-Ottawa scale are presented in Table 1. It should be noted that the inter-observer reliability for the quality assessment was high, with a kappa statistics of 0.77.

Figure 1

Literature review process

Table 1

Main characteristics of the studies included in this meta-analysis

Association between NAFLD and cardiac conduction defect

The pooled analysis found a significantly greater risk of a cardiac conduction defect among patients with NAFLD, with a pooled OR of 5.17 (95%CI 1.34-20.01), as demonstrated in Fig. 2. The heterogeneity between the studies was high, with an I2 of 96%.

Figure 2

Forest plot of all studies

NAFLD, nonalcoholic fatty liver disease; CI, confidence interval; df, degrees of freedom

Evaluation for publication bias

Evaluation for publication bias was not performed as the number of included studies was too small.

Discussion

This study is the first systematic review and meta-analysis that comprehensively combined the data from all available studies on the association between NAFLD and cardiac conduction defect. The pooled analysis found an approximately 5-fold greater risk of cardiac conduction defects among patients with NAFLD compared to individuals without. The reasons behind this increased risk are not well understood but there are some possible explanations.

The first explanation is related to the increased level of inflammatory cytokines among patients with NAFLD [23,24]. Deposition of adipose tissue in the liver induces low-grade systemic inflammation through the release of proinflammatory cytokines such as interleukin (IL)-1, IL-6, tumor necrosis factor, and transforming growth factor (TGF)-β. These cytokines are arrhythmogenic, as demonstrated by several studies [25,26]. For instance, in a rodent study, IL-1β was shown to prolong the repolarization phase and interfere with potassium and calcium fluxes, causing disturbance of cardiac conduction [27]. Another in vitro study found that an increased level of TGF-β can induce fibrotic changes in cardiomyocytes, causing arrhythmias and conduction defects [28].

Second, fatty accumulation in the liver could be a marker of fatty deposition in other organ tissues, including myocardium and pericardium [29]. This ectopic fatty tissue acts as an electrical insulator, which may cause heart block and arrhythmia if it infiltrates the conduction system [18,30,31]. Moreover, this fatty infiltration can subsequently transform into fibrous tissue, which would further increase the severity of a conduction defect [32].

It is also possible that NAFLD does not directly cause conduction defects. The observed association could be a consequence of the increased risk of coronary artery disease [33] and heart failure [34] among patients with NAFLD, as both conditions can cause structural changes of the heart [35,36]. The current systematic review and meta-analysis could not exclude the role of these potential confounders, as the primary studies did not comprehensively adjust their effect estimates for them.

Even though we conducted a comprehensive literature search and retrieved high quality studies, as suggested by the high Newcastle-Ottawa scores, the study had some limitations and our results should be interpreted with caution. First, the statistical heterogeneity was high in this meta-analysis. We believe that the differences in the study population, the definition of cardiac conduction abnormality, and the variables used for adjustment in the multivariate analysis across the studies were the main source of the between-study variation. For instance, the study by Iscen et al [17] found only right bundle-branch block, a finding without any clinical significance in young adults. On the other hand, the other 2 studies included patients with cardiac conduction defects that are severe and need special cardiological management (i.e., first-degree block, Mobitz type 1 block, left bundle-branch block, bifascicular block and nonspecific intraventricular block in the study by Mangi et al [18], and first-degree atrioventricular block, second-degree block, third-degree block, left bundle-branch block, right bundle-branch block, left anterior hemiblock and left posterior hemiblock in the study by Mantovani et al [19]). This would also make it difficult to determine the clinical implications of the observed association. Second, the temporal relationship between NAFLD and cardiac conduction defect could not be clearly established, given the cross-sectional nature of the included studies. It is still possible that the association between NAFLD and cardiac conduction is confounded by comorbidities, such as diabetes mellitus and other cardiovascular diseases, given that one of the 3 included studies did not adjust its effect estimate for any potential confounders. Further prospective cohort studies with comprehensive adjustment for potential confounders are still required. Third, we could not perform a publication bias analysis because of the limited number of studies included. Thus, publication bias in favor of studies with positive results may have been present. Fourth, none of the included studies provided data on subgroups of patients with nonalcoholic steatohepatitis. Further studies are still needed to investigate the risk in this specific subgroup.

In summary, the current study demonstrated a significantly greater risk of cardiac conduction defects among patients with NAFLD compared with individuals without. Physicians who provide care to patients with NAFLD should be aware of this association, since cardiovascular mortality is their most common cause of death. Nonetheless, further studies are still required to investigate whether the association is causal and to determine the role of screening and preventative measures in clinical practice.

What is already known:

• Cardiovascular disease is the most common cause of death in patients with non-alcoholic fatty liver disease (NAFLD)

• Previous studies showed that patients with NAFLD may have a higher risk of cardiac conduction defectss

• However, the results from those studies varied considerably

What the new findings are:

• This meta-analysis of 3 cross-sectional studies with 3651 participants showed that the risk of cardiac conduction defects among patients with NAFLD was 5 times higher than in those without NAFLD

• However, subgroup analysis for each type of cardiac conduction defect could not be performed because the data from the primary studies were limited

Click here for additional data file.