Several observational studies and some meta-analyses have reported that NAFLD, as detected by indirect biomarkers of steatosis, ultrasonography or liver biopsy, is associated with an increased risk of prevalent and incident chronic kidney disease (CKD) in patients with and without T2DM, independent of established cardio-metabolic risk factors, diabetes-related variables and other potential confounders [28,30]. In a recent 2022 meta-analysis of 13 longitudinal studies for a total of 1,222,032 patients (~28% with NAFLD as detected by biomarkers, International Classification of Diseases [ICD] codes, imaging techniques or biopsy) and 33,840 new cases of incident CKD stage (defined as CKD stage ≥3 and/or overt proteinuria) over a median follow-up of nearly 10 years, our research group reported that NAFLD was associated with a 43% increased risk of incident CKD (random-effects hazard ratio 1.43, 95% confidence interval 1.33 to 1.54; I² = 60.7%), independent of age, sex, obesity, hypertension, T2DM and other CKD risk factors [16]. In a 2018 meta-analysis, the same research group documented that such association was slightly higher when the analysis was restricted to cohort studies involving exclusively patients with diabetes mellitus (random-effects hazard ratio 1.56, 95% confidence interval 1.07–2.05; I² = 0%) [31]. Interestingly, accumulating observational studies using vibration controlled transient elastography (VCTE), as non-invasive method to evaluate the degree of liver fibrosis, also reported an independent association between liver stiffness and renal dysfunction. In this regard, for instance, in a 2022 systematic review and meta-analysis of seven cross-sectional studies for a total of 7736 individuals with NAFLD, Ciarduillo et al. showed that liver fibrosis (as assessed by VCTE) was associated with an increased risk of prevalent CKD (defined as eGFR < 60 mL/min/1.73 m² and urinary albumin to creatinine ratio ≥30 mg/g) (random-effects odds ratio 2.49, 95% confidence interval 1.89–3.29; I² = 46.5%), as well as with an increased risk of prevalent albuminuria (random-effects odds ratio 1.98, 95% confidence interval 1.29–3.05; I² = 46.5%) [32]. However, it should be noted that, at present, only few observational studies on this topic have used liver biopsy for the diagnosis of NAFLD, which is the reference standard for diagnosing and staging NAFLD [1,2]. Conversely, most available studies on this topic have used liver ultrasonography, which is to date the recommended first-line imaging method for detecting NAFLD in clinical practice [1,2], able to accurately detect mild-to-moderate hepatic steatosis, as assessed by liver biopsy [7].

Notably, the presence of NAFLD may be even associated with CKD progression [33]. In a cohort study of nearly 1500 CKD patients who underwent periodic health check-ups, Jang et al. showed that age- and sex-adjusted decline in eGFR values was higher in patients with NAFLD (as detected by ultrasonography) when compared with those without NAFLD [34]. In that study, interestingly, the decline in estimated eGFR related to NAFLD was even higher in patients with higher NAFLD fibrosis score (which is an indirect marker of advanced liver fibrosis), in those with proteinuria and/or low eGFR values at baseline and in those who were active smokers or had hypertension at baseline [34]. Although additional studies are needed, preliminary evidence also indicates that the improvement in liver histology in NAFLD patients is associated with improved kidney function [33,35].

Observational studies involving patients with and without T2DM have reported that the presence of the G allele of rs738409 in the PNPLA3 gene is associated with lower eGFR values and/or higher prevalence of CKD, even after adjustment for the presence of NAFLD and other cardio-renal risk factors [28,30,33,36,37,38,39]. In a cross-sectional study including 157 Italian patients with T2DM, who underwent liver ultrasonography and kidney function assessment, our research group reported that the presence of the G allele of rs738409 in the PNPLA3 gene was associated with an increased risk of CKD (defined as <60 mL/min/1.73 m² and/or abnormal albuminuria), independent of liver disease severity, cardiorenal risk factors and other potential confounders [37]. Interestingly and notably, in that study, the authors also found that PNPLA3 mRNA expression was greatest in the liver and renal cortex, thereby suggesting that the PNPLA3 rs738409 variant might contribute, at least in part, to the impaired kidney function in these patients [37]. These findings have also been confirmed in some cohorts of children and adolescents [40,41,42].

Taken together, these data strongly indicate that patients with NAFLD, especially those with severe forms, have an increased risk of developing CKD, independent of several cardio-renal risk factors and other confounders [28,33]. Interestingly, novel data also suggest that MAFLD criteria might identify patients with CKD better than NAFLD criteria [43]. However, seeing the observational nature of all studies available so far, it is essential to underline that a causal relationship between NAFLD and incident CKD cannot be proven yet [28,33].

Some observational studies [44,45,46], although not all [47,48], have documented an association between NAFLD and the risk of prevalent distal symmetric polyneuropathy in T2DM patients, independent of multiple cardio-metabolic risk factors and diabetes-related variables. In a cross-sectional study involving roughly 400 outpatients with T2DM attending five Italian diabetes centers, who underwent liver ultrasonography, vibration controlled transient elastography (by FibroScan^®) and evaluation of microvascular diabetic complications, Lombardi et al. documented that significant liver fibrosis (i.e., liver stiffness measurement [LSM] ≥ 7.0 and 6.2 kPa with M and XL probes, respectively) was independently associated with higher prevalence of microvascular diabetic complications (28% in patients with LSM < 7.0/6.2 kPa vs. 50% in patients with LSM ≥ 7.0/6.2 kPa, p < 0.001), including distal symmetric polyneuropathy (3% in patients with LSM < 7.0/6.2 kPa vs. 14% in patients with LSM ≥ 7.0/6.2 kPa, p < 0.05) [46]. Accumulating evidence also suggests the existence of an association between hepatic steatosis (as detected by imaging techniques) and cardiac autonomic dysfunction in patients with and without T2DM [49,50]. For instance, in a recent cross-sectional study including 173 individuals with T2DM and 183 age- and sex-matched nondiabetic controls from the Cooperative Health Research in South Tyrol (CHRIS) study, Targher et al. reported that individuals with T2DM and NAFLD (on ultrasonography) and individuals with NAFLD alone, but not those with T2DM alone, had an increased risk of cardiac sympathetic/parasympathetic imbalance (as assessed by low- to high-frequency power ratio and other heart rate variability measures obtained by a 20 min resting electrocardiogram), when compared with those without NAFLD and T2DM [50].

However, although this evidence is interesting, additional research is needed to corroborate these findings in larger populations and, more willingly, in longitudinal studies.

Some cross-sectional studies have investigated the relationship between NAFLD (as detected by imaging techniques) and the risk of prevalent diabetic retinopathy in patients with T2DM, reporting inconsistent results [44,51]. In this regard, a 2021 meta-analysis of nine cross-sectional studies for a total of 7170 patients with T2DM (57% with NAFLD on ultrasonography) reported no association between NAFLD and risk of prevalent diabetic retinopathy (random-effects odds ratio 0.94, 95% confidence interval 0.51–1.71; I² = 96%) [52]. In addition, in that meta-analysis, subgroup analyses suggested that in China, Korea and Iran, T2DM patients with NAFLD had a decreased risk of diabetic retinopathy when compared with those without NAFLD, whereas in Italy and India, T2DM patients with NAFLD had an increased risk [52]. As suggested by the authors of that meta-analysis, the aforementioned results should be interpreted with caution, because of the high heterogeneity observed and the differences in the results seen across various countries. Hence, additional research is needed to better explore this issue [52].

A large body of evidence now supports the existence of a strong and independent association between NAFLD and sub-clinical myocardial remodelling and dysfunction, heart valve diseases (i.e., aortic-valve sclerosis and mitral annulus calcification) and cardiac arrhythmias (mainly atrial fibrillation) in patients with and without T2DM (Table 1) [13,22,23,53,54,55,56]. For instance, in a cross-sectional study involving 222 outpatients with T2DM (~70% with NAFLD on ultrasonography), our research group showed that NAFLD was associated with increased risk of left ventricular diastolic dysfunction (as evaluated by trans-thoracic echocardiography), independent of established CVD risk factors, diabetes-related covariates and other confounders [57]. Some recent observational studies using biopsy or vibration-controlled transient elastography (by FibroScan^®) also observed a graded relationship between functional and structural myocardial abnormalities and NAFLD severity in patients with and without T2DM [22]. A 2019 meta-analysis of 16 observational studies further confirmed that NAFLD (as detected by imaging techniques or liver biopsy) was independently associated with many functional and structural myocardial abnormalities, including higher left ventricle mass, higher left ventricular end diastolic diameter, higher left atrium diameter and the ratio between left atrial volume and body surface area, higher posterior wall and septum thickness, lower E/A wave ratio, higher E/E′ ratio, longer deceleration time and longer relaxation time [58]. Interestingly, recent observational studies also indicated that NAFLD (as detected by ultrasonography) was associated with a reduction in global longitudinal strain, which is a relatively novel echocardiographic parameter strongly associated with adverse cardiovascular outcomes [59,60,61].

Relating to heart valve calcifications, some cross-sectional studies have shown an independent association between NAFLD and risk of aortic valve sclerosis (AVS) and mitral annulus calcification (MAC) in patients with and without T2DM [22,62]. For instance, in a study involving nearly 250 consecutive outpatients with T2DM (~70% with NAFLD on ultrasonography), our research group documented that NAFLD was strongly associated with cardiac calcifications in both the aortic and mitral valves, even after adjustment for established CVD risk factors, diabetes-related covariates and other confounders [62]. These findings may be clinically relevant, as functional and structural myocardial abnormalities and AVS/MAC are strongly associated with all-cause and cardiovascular mortality in patients with and without T2DM [63].

Relating to cardiac arrhythmias, several observational studies and some meta-analyses [64,65,66,67] have documented that NAFLD (as detected by imaging techniques) is associated with prevalent and incident permanent atrial fibrillation (AF) in patients with and without T2DM (Table 1) [22]. Notably, AF is, at present, the most frequent cardiac arrhythmia observed day by day in clinical practice and, importantly, it is strongly linked to adverse cardiovascular outcomes [22]. In a recent meta-analysis of five observational studies for a total of roughly 240,000 adult individuals with and without T2DM, our research group documented that NAFLD (as detected by imaging techniques) was associated with higher prevalence and incidence of AF [66]. Interestingly, in a recent retrospective longitudinal study including 267 patients (33% with NAFLD as detected by ultrasonography and 17% with T2DM at baseline) undergoing AF ablation, Donnellan et al. reported that NAFLD was associated with increased arrhythmia recurrence rates following AF ablation, during a mean follow-up of nearly 2.5 years [68]. Other observational studies and meta-analyses, also enrolling T2DM patients, have reported that NAFLD (as detected by ultrasonography) was associated with an increased risk of prolonged QTc, ventricular arrhythmias or conduction defects, independent of established cardiovascular risk factors, diabetes-related covariates and other confounders [22,67,69,70,71,72]. Interestingly, in a 2021 meta-analysis of 19 observational studies, Gong et al. confirmed that NAFLD (as detected by indirect markers of steatosis or imaging techniques) was independently associated with higher risks of prolonged QT interval (random-effects odds ratio 2.86, 95% confidence interval 1.64–4.99), premature atrial/ventricular contraction (random-effects odds ratio 2.53, 95% confidence interval 1.70–3.78) and heart block (random-effects odds ratio 2.65, 95% confidence interval 1.88–3.72) [67]. These data are clinically relevant, because NAFLD-related cardiac arrhythmias complications might contribute to explaining, at least in part, the increased risk of fatal and non-fatal CVD events observed in NAFLD patients.

Over the last few decades, it has become increasingly evident that the leading cause of death in NAFLD patients is CVD [22,23,73,74,75]. In this regard, using data from the National Vital Statistics System multiple-cause mortality data (2007–2016), Paik et al. reported that CVD was the main cause of death among US patients with NAFLD, as detected by ICD codes [74]. In a meta-analysis of 45 observational studies for a total of approximately 8 million individuals followed up to 13 years, Younossi et al. also estimated that the pooled CVD-specific mortality rate among NAFLD patients with or without T2DM was nearly 5 per 1000 person-years [3]. Several longitudinal studies and some meta-analyses confirmed that patients with NAFLD (as detected by imaging techniques, ICD codes or liver biopsy) have an increased risk of developing fatal and non-fatal CVD events, even after adjustment for several traditional CVD risk factors, diabetes-related variables, specific medications and other potential confounders (Table 1) [22,23,54,55,56,76,77,78,79]. In a 2021 meta-analysis of 36 longitudinal studies for a total of 5,802,226 adults and 99,668 incident cases of fatal and non-fatal CVD events over a median follow-up of 6.5 years, our research group reported that NAFLD (as detected by imaging techniques, ICD codes or liver biopsy) was associated with a 45% increased risk of fatal or non-fatal CVD events, independent of age, sex, body mass index, waist circumference, presence of T2DM and other cardiovascular risk factors (random-effects hazard ratio 1.45, 95% confidence interval 1.31–1.61; I² = 86.2%) [14]. Such risk further increased in patients with severe forms of NAFLD, especially those with advanced fibrosis [14]. Another 2021 meta-analysis confirmed that NAFLD (as detected by imaging techniques, ICD codes or liver biopsy) was independently associated with increased risk of myocardial infarction (random-effects odds ratio 1.66, 95% confidence interval 1.39–1.99), ischemic stroke (random-effects odds ratio 1.41, 95% confidence interval 1.29–1.55) and heart failure (random-effects odds ratio 1.62, 95% confidence interval 1.43–1.84) [26]. In this regard, it is important to underline that the magnitude of cardiovascular risk is strongly related to the severity of NAFLD [25,80,81,82]. For instance, in a nationwide, matched cohort study of 10,568 Swedish individuals with biopsy-confirmed NAFLD (11% with T2DM at baseline) who were followed for a median period of 14 years, Simon et al. reported that, when compared to 49,925 adults of the general population (3% with established T2DM at baseline), mortality rates from CVD were significantly elevated in patients with simple steatosis (adjusted-hazard ratio 1.25, 95% confidence interval 1.16–1.35), and that these risks progressively increased in patients with NASH without fibrosis (adjusted-hazard ratio 1.66, 95% confidence interval 1.38–2.01), in those with non-cirrhotic fibrosis (adjusted-hazard ratio 1.40, 95% confidence interval 1.17–1.69) and also in those with cirrhosis (adjusted-hazard ratio 2.11, 95% confidence interval 1.63–2.73) [80]. Similar findings were also documented in cohorts involving NAFLD patients with T2DM [22,23,53,54,55,56].

To date, data regarding whether the improvement of NAFLD may reduce the incidence of cardiovascular complications are scarce. Although some retrospective studies enrolling Asian adults without pre-existing CVD have reported that the improvement or resolution of NAFLD (on ultrasonography) could be associated with a reduced risk of (carotid) atherosclerotic development in patients with and without T2DM [56,83], we believe that additional information on this issue is needed. In addition, it is important to underline that current evidence also indicates that histologic resolution of NASH could be associated with beneficial changes in risk factors for CVD [56,83], thus suggesting a potential favorable effect on cardiac complications.

Lastly, novel evidence also suggests that MAFLD criteria might identify patients with CVD better than NAFLD criteria [84].