Understanding cardiohepatic interactions in patients with heart failure with mildly reduced or preserved ejection fraction

This article refers to ‘Liver tests, cardiovascular outcomes and effects of empagliflozin in patients with heart failure and preserved ejection fraction: The EMPEROR-Preserved trial’ by M. Böhm et al., published in this issue on pages 1375–1383.

Heart failure (HF) is a global epidemic affecting more than 64 million people worldwide.¹ Acute and chronic liver injury, as evidenced by elevations in liver function tests (LFTs) are a common manifestation of acute and chronic HF and may reflect systemic hypoperfusion or chronic passive, venous congestion, respectively.² Several recently published reports have demonstrated the potential utility of LFTs in HF prognostication.^3-11 Specifically, reduced levels of albumin and elevated levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), bilirubin, and alkaline phosphatase (ALP) have traditionally been associated with increased mortality and HF hospitalizations.^3-11 However, these prior studies examined the role of LFTs in those with acute HF irrespective of left ventricular ejection fraction (LVEF) or ambulatory HF with reduced ejection fraction (HFrEF). There are sparse data examining the relationship between LFTs and chronic HF with mildly reduced (HFmrEF) or preserved ejection fraction (HFpEF). In this issue of the Journal, Böhm et al.¹² describe the interaction between empagliflozin, LFTs, and HF outcomes in participants with ambulatory HFmrEF/HFpEF in the EMPEROR-Preserved trial.

EMPEROR-Preserved enrolled 5988 patients with HF and an ejection fraction >40% and randomized them in a double-blind, 1:1 ratio to receive empagliflozin 10 mg daily or matching placebo, both in addition to usual guideline-directed medical therapy (GDMT) for HF. Patients with known clinically significant acute or chronic liver disease (i.e. defined by serum levels of AST/ALT or ALP more than three times the upper limit of normal at screening) were excluded from the study population. Direct and indirect liver tests including AST, ALT, bilirubin, ALP, and albumin were serially measured at follow-up visits of 4, 12, 32, 52, 76, 100, 124, 148, and 172 weeks. For this analysis, LFTs were divided into subgroups according to tertiles and normal or abnormal values according to established reference ranges. The authors examined the association with LFT elevations and the primary endpoint of a composite of time-to-first adjudicated hospitalization for HF or cardiovascular death. The authors then evaluated the effects of empagliflozin versus placebo on LFT changes and the primary endpoint across tertiles and for normal or abnormal values.

At baseline, a minority of patients had abnormally elevated values for both the placebo and empagliflozin group – AST (6.1% and 6.4%), ALT (4.9% and 4.9%), ALP (13.0% and 13.4%), bilirubin (4.9% and 5.6%), and albumin (0.2% and 0.2%). Elevated bilirubin in the third tertile (>0.6 mg/dl) or abnormal range (>1.2 mg/dl for females and males), and elevated ALP in the third tertile (>89 U/L) or abnormal range (>104 U/L for females, >129 U/L for males) were associated with higher event rates for the primary outcome. Specifically, there is an almost twofold increase in risk in the primary outcome with abnormally elevated bilirubin and ALP. Low albumin (<4.2 g/dl) was also associated with an almost twofold increase in risk in the primary outcome. No significant associations were detected for AST. Elevated ALT in the third tertile (>20 U/L) was associated with a 28% decrease in risk in the primary outcome and better prognosis. Empagliflozin did not significantly affect LFTs over the study period. Additionally, there was no heterogeneity of treatment effect with empagliflozin for all LFTs across tertiles and for normal versus abnormal values.

There are several novel insights from this post-hoc analysis that merit further mention. This study is the first to specifically examine the association between LFTs and clinical outcomes in ambulatory HFmrEF/HFpEF. Previously published reports based on landmark clinical trials have generally focused on the acute care settings or ambulatory HFrEF.^3-11 This analysis extends previously published data to ambulatory HFmrEF/HFpEF patients and identified a similar association between elevated bilirubin, elevated ALP, and low albumin and greater risk of adverse cardiovascular outcomes. Notably, this study did not observe a significant association between AST and the composite of hospitalization for HF and cardiovascular death. Prior data on this association have been conflicting,^3-11 but this remains a somewhat unexpected finding. Finally, this study is the first to report that elevated ALT is associated with lower risk of adverse cardiovascular outcomes. Because HFpEF is an age-related disease with a high comorbidity burden, it is possible ALT behaves more like a marker of nutrition and synthetic function rather than perfusion in the HFpEF population.¹¹

Notably, empagliflozin did not have a discernible impact on LFTs during follow-up. Previous studies have demonstrated the salutary effects of empagliflozin in the heart and kidneys.^{13, 14} Empagliflozin's cardioprotective and nephroprotective effects are likely multifactorial and may involve mitigating deleterious cardiorenal oxidative stress, apoptosis, and fibrosis pathways, as well as promoting cellular regeneration.^{13, 14} It is interesting these investigators did not find an improvement in LFTs with empagliflozin suggesting that the salutary effect on end-organ damage may not apply to the liver, which is congruent with findings from prior studies like DAPA-HF.¹⁰ However, it is entirely possible that sodium–glucose cotransporter 2 inhibitors may preserve or improve liver function in ways not adequately assessed by the surrogate laboratory assessments reported here; whether assessment of liver function and histology using more advanced (and potentially invasive) methods would yield different results may be an area of further study, particularly given the emerging evidence regarding hepato-cardio-renal interplay.

Interestingly, the treatment effect of empagliflozin was not modified by baseline liver parameters. However, it is important to note that the study population did not include clinically significant liver disease at baseline. The authors excluded patients based on abnormal LFTs (see definition above), which may not completely capture patients with underlying liver disease. This highlights the lack of evidence on the safety and efficacy of empagliflozin in liver disease. Additionally, it is possible that empagliflozin has a greater treatment effect in certain types of liver disease, such as non-alcoholic fatty liver disease which shares similar comorbidities with HFpEF. Prior studies have demonstrated empagliflozin's benefits in liver steatosis and liver fibrosis,¹⁵ and there are ongoing clinical trials studying this interaction.

In conclusion, Böhm et al.¹² enhance the current understanding of the prognostic implications of LFTs in HF and provide new insights into the associations between elevated LFTs and ambulatory HFmrEF/HFpEF. In particular, the interaction between elevated ALT and improved outcomes is highly intriguing and suggests nutritional markers may have added prognostic value in the setting of HFmrEF/HFpEF. As the population prevalence of HFmrEF/HFpEF continues to grow, especially given the established comorbid overlaps between the HF state, fatty liver disease, and progression of renal dysfunction, a deeper understanding of these complex interactions is paramount and may lead to unique insights into the management of these complex patients. In this comprehensive post-hoc analysis, Böhm and colleagues empower us to think more deeply about these interactions.

Conflict of interest: none declared.