Sodium–Glucose Cotransporter 2 Inhibitor Use and Risk of Liver-Related Events in Patients With Type 2 Diabetes: A Meta-analysis of Observational Cohort Studies Free

Metabolic dysfunction–associated steatotic liver disease (MASLD), formerly named nonalcoholic fatty liver disease, has become the predominant cause of chronic liver diseases worldwide (1). The global prevalence of MASLD is estimated to be ∼30%–40% among the general population (2), ∼60%–70% among people with type 2 diabetes mellitus (T2DM) (3), and ∼70–80% among those with obesity (4). Individuals with MASLD are at high risk of developing major adverse liver-related outcomes (MALOs), such as hepatic decompensation events, hepatocellular carcinoma (HCC), or liver-related mortality (5). Notably, in people with MASLD and liver fibrosis, the coexistence of T2DM further increases the long-term risk of MALOs (5,6).

The links between T2DM and risk of MASLD-related MALOs are complex (5,7). T2DM promotes liver disease progression through several mechanisms, such as insulin resistance, glucotoxicity, lipotoxicity, low-grade inflammation, and increased oxidative stress (5,7). MASLD poses a challenging therapeutic landscape where pharmacotherapies should address metabolic dysfunction and liver disease to reduce the risk of developing MALOs and extrahepatic clinical complications, such as cardiovascular disease, chronic kidney disease, or new-onset T2DM (8,9).

Sodium–glucose cotransporter 2 (SGLT2) inhibitors are a pivotal therapeutic option for managing T2DM (10). These inhibitors increase renal glycosuria and promote osmotic diuresis (10) by reducing renal ability to reabsorb glucose. This mechanism can be exploited to improve blood glucose levels, promote weight loss, and reduce blood pressure and plasma volume (10). Results from large cardiovascular and renal outcome trials have highlighted that SGLT2 inhibitors reduce risk of hospitalization due to heart failure and other adverse cardiovascular and renal outcomes (11,12). Recent meta-analyses have also shown that SGLT2 inhibitors improve liver fat content (assessed with magnetic resonance–based techniques) in individuals with MASLD, irrespective of T2DM status (13,14). These findings represent an attractive bonus for using SGLT2 inhibitors in people living with T2DM and MASLD. However, results from long-term randomized controlled trials examining the effect of SGLT2 inhibitor use on risk of MALOs are still lacking, and current evidence from real-world epidemiological studies supporting an association between SGLT2 inhibitor use and long-term risk of MALOs in people with T2DM is conflicting. Understanding the long-term effect of SGLT2 inhibitors on risk of MALOs is clinically very important, given the increasing use of these antihyperglycemic agents in populations at high risk for adverse liver-related events.

Therefore, the aim of this meta-analysis of observational active-comparator, new-user cohort studies is to quantify the long-term risk of developing MALOs for patients with T2DM who initiated SGLT2 inhibitors compared with risk for new users of other glucose-lowering agents.

We systematically searched PubMed, Web of Science, and Scopus from database inception to 1 January 2025 to identify observational cohort studies with examination of risk of developing long-term MALOs for patients with T2DM who initiated SGLT2 inhibitors compared with risk for those who initiated other antihyperglycemic agents. The search free-text terms can be found in Supplementary Material. Searches were restricted to human studies, and no language restriction was imposed. We also reviewed references from relevant original articles and review articles to identify additional eligible studies not covered by the original database searches. We performed this systematic review and meta-analysis of observational studies following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

The inclusion criteria of the meta-analysis were as follows: 1) observational active-comparator, new-user cohort studies including adult individuals with T2DM who initiated SGLT2 inhibitors (dapagliflozin, empagliflozin, canagliflozin, ipragliflozin, or ertugliflozin) with or without a coexisting diagnosis of MASLD or cirrhosis; 2) cohort studies with data reported on incidence rates of MALOs (as defined below) and/or liver-related mortality; 3) cohort studies that included a control arm with an active glucose-lowering drug for comparison; and 4) cohort studies with reporting of outcome comparisons between patient groups after appropriate adjustment for multiple potential confounding factors (mainly through propensity score matching, inverse probability treatment weighting, or other validated statistical methods for confounding control). The following comprised the exclusion criteria of the meta-analysis: 1) congress abstracts, case reports, case series, cross-sectional studies, case-control studies, reviews, commentaries, or editorials; and 2) phase 2 randomized controlled trials evaluating the effect of SGLT2 inhibitor use on (surrogate) hepatic endpoints for MALOs.

Two investigators (A.M. and G.T.) independently reviewed the titles and abstracts of all studies initially identified according to the abovementioned inclusion criteria. For each study meeting the requirements of the first-round inclusion criteria, a full-text independent review was performed by both investigators. Eventual disagreements between investigators about the inclusion of eligible studies were resolved by a third independent investigator (R.M.).

For each eligible study, we extracted data on publication year, study design, sample size, study country, participants' characteristics, comparator drugs in the control arm, outcomes of interest, follow-up duration, and list of covariates used for statistical adjustments (after propensity score matching). In the case of multiple publications for the same database, we included the most up-to-date or comprehensive information.

Each eligible study was assessed for study quality with the Risk Of Bias In Non-randomized studies - of Interventions (ROBINS-I) tool (15) by two independent authors (A.M. and G.T.). Any disparities in scoring were reviewed, and consensus was obtained following discussion. The ROBINS-I tool is used to evaluate bias across several domains, including bias due to confounding, selection bias, classification of interventions, deviations from intended interventions, missing data, outcome measurement, and selection of the reported results (15). Based on the combined evaluation of these seven domains, studies were categorized as follows: low, moderate, serious, or critical risk of bias.

The primary outcome of the meta-analysis was a composite of MALOs, including hepatic decompensation events (gastric/esophageal variceal bleeding, hepatic encephalopathy, and ascites-related complications), HCC development, liver transplantation, or liver-related mortality. Secondary outcomes included the individual components of the primary composite outcome along with incident cirrhosis (from one study). MALOs were defined according to ICD codes in all included studies. The population of SGLT2 inhibitor users was included in the analysis for each comparison when SGLT2 inhibitors were compared with different active comparator drugs in the same cohort study. The hazard ratios (HRs) and 95% CIs of MALOs for each study that we used in the analysis were fully adjusted for several potential confounders after propensity score matching analyses (including demographics, anthropometric parameters, comorbidities, use of antihyperglycemic and other commonly prescribed medications, and laboratory results). Common and random-effects models based on log HRs and their SEs were used for pooled estimates. The inverse variance method was used for pooling.

A visual inspection of the forest plots was used to examine the possibility of statistical heterogeneity (16). The statistical heterogeneity among studies was assessed with the χ² test and the I² statistic, which estimates the percentage of variability across studies due to heterogeneity rather than chance alone (17). The proportion of heterogeneity accounted for by between-study variability was assessed with the I² statistic and adjudicated to be significant where I² index was >50% (17).

To examine possible sources of heterogeneity between the included studies and to test the robustness of the observed associations, we conducted subgroup analyses by type of comparator drug in the control arm, study country, baseline liver disease (MASLD) status, percentage of patients with compensated cirrhosis, and follow-up length. We also tested for the possibly excessive influence of individual studies using a meta-analysis influence test that eliminated each included study one at a time. Finally, we performed univariable meta-regression analyses to test the impact of age, sex, and percentage of patients with compensated cirrhosis at baseline on the effect size for the association between SGLT2 inhibitor use and risk of MALOs. A funnel plot and the Egger regression test (18) were used to evaluate asymmetry for potential publication bias for meta-analysis.

All statistical tests were two-sided, and a P value <0.05 was considered significant. We used R software (version 4.2.2, 2022) for all statistical analyses with the following packages: meta (version 8.0-1) and metafor (version 4.6-0).

This systematic review and meta-analysis protocol was registered in advance on OSF Registries (https://doi.org/10.17605/OSF.IO/4YTQB). This study involves human participants but was not approved by an ethics committee; approval from an ethics committee is unnecessary as this is a meta-analysis of published observational studies for which informed consent from participants was already obtained along with ethics approval by the local ethics committees.

All data generated or analyzed during this study are included in the published article (and its online supplementary files).

A PRISMA flow diagram summarizing the search and selection processes of the meta-analysis can be found in Supplementary Fig. 1. After examining the titles and abstracts and excluding duplicates, we identified 10 observational cohort studies from PubMed, Scopus, and Web of Science databases to assess their eligibility. Subsequently, we excluded two studies (19,20) because of unsatisfactory inclusion criteria, as detailed in Supplementary Table 1.

The main characteristics of the eight cohort studies (21–28) included in the meta-analysis are summarized in Table 1. Overall, we included a total of 626,104 patients with T2DM (46.5% women; mean age 58 years), 228,298 of whom initiated SGLT2 inhibitors (SGLT2 inhibitor new users) and 397,806 who did not use SGLT2 inhibitors but initiated other glucose-lowering medications; among the latter group, 198,977 initiated dipeptidyl peptidase 4 (DPP-4) inhibitors, 80,500 initiated glucagon-like peptide 1 receptor agonists (GLP-1RA), 22,515 initiated metformin, and 95,814 initiated pioglitazone. During a median follow-up of 2.7 years (interquartile range 1.9–5.0 years), 12,239 patients developed incident MALOs (4,444 liver-related events in the SGLT2 inhibitor group and 7,795 in the non–SGLT2 inhibitor group).

All included studies had a retrospective observational active-comparator, new-user cohort design (21–28). Two studies included multicenter cohorts, and two were conducted in Asia (Japan and South Korea), two in the U.S., and two in the U.K. Four studies included GLP-1RA as the active comparator in the control arm, one metformin, one pioglitazone, and three DPP-4 inhibitors. Two studies included patients with T2DM and compensated cirrhosis, five studies included patients with T2DM and MASLD, and one study included patients with T2DM (with no data on MASLD status). Regarding the outcomes of interest, in three cohort studies only a combined outcome of major liver-related events was used, whereas detailed information on hepatic decompensation events was provided in five studies, on incident HCC in five studies, on liver transplantation in three studies, and on liver-related deaths in three studies. Notably, based on the ROBINS-I tool, all eligible cohort studies were judged to be at moderate risk of bias (especially for the domain of examining possible deviations from intended interventions), indicating that the certainty of evidence was relatively low.

Figure 1 shows a forest plot and pooled estimates for the risk of developing MALOs for patients with T2DM who initiated SGLT2 inhibitors (SGLT2 inhibitor new users) compared with risk for new users of other glucose-lowering agents, such as GLP-1RA, DPP-4 inhibitors, metformin, or pioglitazone. Overall, SGLT2 inhibitor use was associated with a significantly lower risk of developing MALOs in comparison with other glucose-lowering agents (random-effects HR 0.83, 95% CI 0.72–0.95; I² = 83.1%). With stratification of the results by the type of active comparator drug, SGLT2 inhibitor use was associated with a significantly lower risk of MALOs in comparison with DPP-4 inhibitors (random-effects HR 0.80, 0.68–0.93; I² = 13.3%), metformin (random-effects HR 0.58, 0.41–0.81; I² = 0%), and pioglitazone (random-effects HR 0.76, 0.72–0.81; I² = 52.2%) but not GLP-1RA (1.07, 0.98–1.18; I² = 39.7%; common-effects HR 1.08, 95% CI 1.01–1.14).

A sensitivity analysis with the one-study-removed (leave-one-out) approach to test the influence of each included study on the overall effect size showed that the exclusion of each study at a time from the pooled analysis did not have any effect on the significant association between SGLT2 inhibitor use and risk of MALOs (Supplementary Fig. 2). Stratifying the studies by country, we observed that SGLT2 inhibitor use was associated with a lower risk of MALOs in studies conducted in the U.K. and Asian countries but not in the case of multicenter cohorts or studies performed in the U.S. (Supplementary Fig. 3). With stratification of the studies by follow-up length, SGLT2 inhibitor use was associated with a lower risk of MALOs in studies with follow-up length <2.7 years but not in those with follow-up length ≥2.7 years (Supplementary Fig. 4). Although the results of this subgroup analysis should be interpreted with caution (mainly due to the wide 95% CIs across the included studies), this finding might be partly due to the different active comparator drugs in the control arm and the different study outcomes across studies. Conversely, the significant inverse association between SGLT2 inhibitor use and risk of MALOs was consistent after stratification of the studies by baseline liver disease (MASLD) status (Supplementary Fig. 5) or percentage of patients with compensated cirrhosis (Supplementary Fig. 6). Univariable meta-regression analyses showed no significant impact of female sex (β-coefficient 1.92, SE 1.14; P = 0.09), age (β-coefficient −0.03, SE 0.04; P = 0.43), or percentage of patients with compensated cirrhosis (β-coefficient 0.02, SE 0.18; P = 0.93) on the effect size for the association between SGLT2 inhibitor use and the risk of MALOs.

Figure 2A shows a forest plot and pooled estimates for risk of hepatic decompensation events for patients with T2DM who initiated SGLT2 inhibitors compared with risk for new users of other glucose-lowering agents. SGLT2 inhibitor use was not associated with a significantly lower risk of hepatic decompensation events in comparison with other glucose-lowering agents (random-effects HR 0.89, 95% CI 0.73–1.08; I² = 86%).

Figure 2B shows a forest plot and pooled estimates for risk of liver transplantation for patients with T2DM who initiated SGLT2 inhibitors compared with risk for new users of other glucose-lowering agents. SGLT2 inhibitor use was not associated with a lower risk of liver transplantation in comparison with other glucose-lowering agents (random-effects HR 0.82, 95% CI 0.42–1.60; I² = 65.2%).

Figure 3A shows a forest plot and pooled estimates for risk of new-onset HCC for patients with T2DM who initiated SGLT2 inhibitors compared with risk for new users of other glucose-lowering agents. SGLT2 inhibitor use was associated with a marginally lower risk of developing HCC in comparison with other glucose-lowering agents, although without statistical significance (random-effects HR 0.74, 95% CI 0.54–1.03; I² = 35.4%).

Figure 3B shows a forest plot and pooled estimates for risk of liver-related deaths (mainly attributable to complications of cirrhosis [i.e., liver failure] or HCC) for patients with T2DM who initiated SGLT2 inhibitors compared with risk for new users of other glucose-lowering agents. SGLT2 inhibitor use was associated with a significantly lower risk of liver-related death in comparison with other glucose-lowering agents (random-effects HR 0.64, 95% CI 0.50–0.82; I² = 0%).

As shown in Supplementary Fig. 7, the funnel plot for risk of MALOs was relatively symmetric on the visual inspection. The Egger regression test showed no statistically significant asymmetry of the funnel plot (P = 0.373), thus suggesting that publication bias was unlikely.

This is the first meta-analysis of real-world observational cohort studies examining the association between SGLT2 inhibitor use and the long-term risk of developing MALOs for individuals with T2DM, regardless of MASLD status. The main and novel findings of our meta-analysis, which incorporated eight observational active-comparator, new-user cohort studies from different countries with aggregate data on ∼626,000 middle-aged and older patients with T2DM who initiated SGLT2 inhibitors (n = 228,298) compared with new users of other glucose-lowering agents (n = 397,806), are as follows: 1) SGLT2 inhibitor use was associated with significantly lower risk of developing MALOs (combined end point) over a median follow-up period of 2.7 years (pooled random-effects HR 0.83, 95% CI 0.72–0.95) in comparison with other glucose-lowering agents; 2) this significant risk reduction in long-term MALOs was evident in comparing use of SGLT2 inhibitors with DPP-4 inhibitors, metformin, or pioglitazone but not GLP-1RA (for which similar or only marginally better hepatic effectiveness was seen in comparison with SGLT2 inhibitors); and 3) regarding the individual liver-related outcomes of interest, SGLT2 inhibitor use was significantly associated with reduced risk of liver-related mortality (36% reduction) and marginally associated with lower risk of incident HCC (26% reduction), liver transplantation (18% reduction), and hepatic decompensation events (11% reduction).

Reports from many randomized clinical trials of SGLT2 inhibitors were of a significant reduction in hemoglobin A_1c (ranging from ∼0.5% to 1.0%) after 1 year of treatment, regardless of baseline therapy (12). However, SGLT2 inhibitors provide clinical benefits beyond glycemic control, exerting significant benefits for risk of developing major adverse cardiovascular and renal outcomes for patients with T2DM and those with heart failure, regardless of T2DM status (11,12). Some meta-analyses showed that SGLT2 inhibitors may also improve liver fat content (measured with magnetic resonance imaging–based techniques) in individuals with MASLD (13,14). These findings support the use of SGLT2 inhibitors in people with T2DM and MASLD. However, there are no placebo-controlled randomized trials with specific examination of the effect of SGLT2 inhibitors on liver histological end points in people with MASLD. In a randomized, 48-week, open-label, active-controlled trial (involving 40 participants with biopsy-confirmed MASLD who were randomly assigned to receive once daily 20.0 mg tofogliflozin or 0.5 mg glimepiride), the investigators found that tofogliflozin and, to a lesser extent, glimepiride led to metabolic and liver histological improvements in patients with T2DM and MASLD (29). To date, real-world epidemiological evidence on the potential hepatoprotective effect of SGLT2 inhibitor use on long-term MALOs is still conflicting. We attempted to address this knowledge gap in our meta-analysis, which shows for the first time that SGLT2 inhibitor use, in comparison with use of other glucose-lowering medications (except for GLP-1RA), is significantly associated with a lower risk of developing long-term MALOs for people with T2DM, regardless of MASLD status.

The findings of this meta-analysis have some important clinical implications. From a clinical perspective, the hepatic effectiveness of SGLT2 inhibitors on long-term MALOs is clinically important, given the increasing burden of MASLD worldwide (1), the high prevalence of advanced forms of MASLD (e.g., metabolic dysfunction–associated steatohepatitis and cirrhosis) in patients with T2DM (3), and the adverse effect of liver-related complications on overall survival rates in patients with T2DM and MASLD (5,6). In this meta-analysis, we also observed that the hepatoprotective effect of SGLT2 inhibitors was primarily driven by a 36% risk reduction in liver-related deaths. However, this finding should be interpreted with caution because in most pooled analyses of individual liver-related outcomes, the included cohort studies were limited and did not permit a detailed analysis regarding the type of active comparator drug in the control arm. In this meta-analysis, we also observed similar or tendentially lower hepatic effectiveness of SGLT2 inhibitors in comparison with GLP-1RA in reducing long-term risk of MALOs. GLP-1RA (especially semaglutide 2.4 mg/week) have shown promising effects on liver histology outcomes in phase 2 and phase 3 placebo-controlled randomized controlled trials in individuals with biopsy-confirmed metabolic dysfunction–associated steatohepatitis and liver fibrosis (5,30) (although it should be noted that the maximum dose of semaglutide usually given for treatment of T2DM is 1.0 mg weekly). Increasing evidence suggests that GLP-1RA may have stronger metabolic benefits than SGLT2 inhibitors (12). GLP-1RA can also provide additional hepatic benefits through greater weight loss and effects on hepatic stellate, endothelial cells, and immune cells (30,31). Moreover, the follow-up duration of most eligible cohort studies was relatively short, preventing adequate observation of some hepatic-related events. Follow-up duration could be particularly relevant for HCC development, which occurs over many years. Finally, it should also be noted that most studies were conducted in general populations of patients with T2DM rather than with a focus on high-risk patient groups with cirrhosis. In this regard, except for the two cohort studies by Simon et al. (21) and Huynh et al. (22) that included only patients with T2DM and cirrhosis, the proportion of patients with compensated cirrhosis at baseline varied from 0% to 4.2% in the studies included in the meta-analysis, which would have potentially reduced our ability to detect meaningful differences in the incidence rates of specific (individual) liver-related outcomes. Hence, additional cohort studies are needed to further corroborate these findings.

A detailed discussion of the putative underlying mechanisms underpinning the association between SGLT2 inhibitor use and risk of long-term MALOs in people with T2DM is beyond the scope of this meta-analysis. Nonetheless, these mechanisms are complex and not yet fully understood. SGLT2 inhibitors may reduce risk of liver disease progression by promoting weight loss (32) and improving systemic insulin resistance (11,13). Additionally, SGLT2 inhibitors promote sodium and glucose excretion by inhibiting SGLT2 in renal proximal tubules, inducing a diuretic response (∼7%–10% reduction in plasma volume after a 3-month treatment (11,13,27,33). This diuretic effect reduces fluid imbalance, which is relevant in patients with cirrhosis (11,13). Furthermore, SGLT2 inhibitors lower blood pressure (∼4 mmHg systolic and ∼2 mmHg diastolic blood pressure reduction) (11), positively affecting the liver, heart, and kidneys. The combination of negative energy balance through glycosuria (∼200–250 kcal/day in urine [34]) and substrate switching toward lipids as a source of energy expenditure may also contribute to hepatoprotection of SGLT2 inhibitors (11,13). Finally, experimental data indicate that SGLT2 inhibitors reduce low-grade inflammation and oxidative stress (11,13) while increasing pancreatic glucagon secretion (11), thereby improving adipocyte dysfunction and hepatic lipotoxicity (13,27). Recently, Chung et al. (24) performed Mendelian randomization analyses for two large European cohorts (the UK Biobank and FinnGen databases) to investigate potential causal links between SGLT2 inhibition and risk of liver-related complications. These investigators found that the genetically predicted effect of SGLT2 inhibition was associated with a nearly 30% lower risk of developing cirrhosis.

Our meta-analysis has important limitations strictly inherent to the studies included. First, the observational design of the cohort studies does not permit drawing any conclusions about a definitive causal relationship between SGLT2 inhibitor use and long-term risk of MALOs. Moreover, the retrospective nature of these studies may also introduce selection bias despite appropriate confounding adjustment(s). Second, using aggregate data limited our ability to perform detailed subgroup analyses based on individual patient characteristics or to adjust for all potential confounding factors. Third, although we used a random-effects model, interpreting some meta-analysis results might require caution, given the observed medium-high heterogeneity. Based on our subgroup analyses, the observed between-study heterogeneity could reflect differences in study populations, study country, type of active comparator drug in the control arm, follow-up duration, and prevalence of underlying chronic liver diseases. In this regard, it is important to note that two (21,22) of the eight eligible studies included all individuals with T2DM and compensated cirrhosis. In addition, the study by Chung et al. (24) had the highest baseline prevalence of alcoholic liver disease (∼25%), which is a risk factor for overall mortality and MALOs in patients with and without T2DM (35). Finally, the lack of data on patients with compensated advanced chronic liver disease at baseline and the lack of detailed information on duration of diabetes (available only for a few cohorts) and types of SGLT2 inhibitors and dosages used, as well as the lack of data on individuals without T2DM, may further limit the interpretation and generalizability of these results.

Notwithstanding these limitations, this meta-analysis also has important strengths. It represents the first synthesis of real-world evidence studies on association between SGLT2 inhibitor use and risk of developing clinically relevant MALOs, such as hepatic decompensation events, HCC development, and liver-related mortality. The large number of participants included in the meta-analysis (although there is a relatively limited number of studies available thus far) provides sufficient statistical power to quantify the magnitude of the association between SGLT2 inhibitor use and risk of long-term MALOs in people with T2DM. Furthermore, the included studies had active-comparator, new-user cohort designs (that help mitigate biases arising from inclusion of prevalent users), with use of robust statistical methods for confounding adjustment mainly based on propensity score techniques, thus enhancing the reliability of the findings. Finally, visual inspection of the funnel plot did not reveal any significant asymmetry, suggesting that risk of publication bias was low.

In conclusion, this meta-analysis of real-world observational active-comparator, new-user cohort studies provides evidence that SGLT2 inhibitor use is significantly associated with a lower risk of developing long-term MALOs for people with T2DM. In waiting for results from phase 3 randomized controlled trials designed to assess the benefits of SGLT2 inhibitors on long-term liver-related events, these findings suggest a beneficial effect of SGLT2 inhibitors in preventing clinically relevant MALOs beyond their recognized cardiovascular-kidney-metabolic benefits. The hepatic effectiveness of SGLT2 inhibitors was greater than that of DPP-4 inhibitors, metformin, or pioglitazone but not GLP-1RA (for which similar or only marginally better hepatic effectiveness was seen in comparison with SGLT2 inhibitors). Future well-designed cohort studies with an emulated target trial design and longer follow-up durations are needed to better characterize the impact of SGLT2 inhibitor use on long-term MALOs.

Funding. C.D.B. is supported in part by the National Institute for Health Research (NIHR) Southampton Biomedical Research Centre (NIHR203319). S.P. is supported in part by grants from Italy Ministry of Education, Universities and Research (MIUR) under the National Recovery and Resilience Plan (Piano Nazionale di Ripresa e Resilienza [PNRR]) M4C2I1.3 Heal Italia project PE00000019 CUP B73C22001250006, the Italian PNRR-MAD-2022-12375656 project, Research Projects of National Relevance (Progetti di Rilevante Interesse Nazionale [PRIN]) 2022 project 2022L273C9, and the project RF-2021-12372399. G.T. is supported in part by grants from the University School of Medicine of Verona, Verona, Italy.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. A.M. and G.T. were involved in the conception of the study and the analysis and interpretation of the results and wrote the first draft of the manuscript. A.M., R.M., M.G.L., V.F., and G.T. were involved in the conduct of the study and searched the published articles. G.P., S.P., N.S., H.T., and C.D.B. were involved in the interpretation of the results and contributed to the discussion. All authors edited, reviewed, and approved the final version of the manuscript.

Handling Editors. The journal editors responsible for overseeing the review of the manuscript were Elizabeth Selvin and Csaba P. Kovesdy.