Heart failure (HF) is a leading cause of morbidity and mortality worldwide. In developed countries its prevalence is estimated at 1–2% and its incidence is rising.1 Currently, the main causes of HF are hypertension and coronary artery disease. As the most impactful risk factor for the development and progression of atherosclerosis, dyslipidemia,2 seems to have a role in HF pathophysiology in some patients.

HMG-CoA reductase inhibitors reduce cholesterol levels and subsequently diminish the risk of major vascular events and associated morbidity and mortality,3 as well the risk HF development.4,5 Two major randomized trials evaluated the impact of rosuvastatin in chronic HF patients in NYHA class II–IV and failed to provide an evident beneficial effect. Interestingly, only patients with HF due to ischemic heart disease (IHD) were included in the CORONA trial and IHD represented 40% of the GISSI-HF trial population, revealing that these results may be independent of HF etiology.6,7 Based on current recommendations, routine administration of statins in HF patients is not recommended, unless there is another indication for their use.1,7

Recent trials IMPROVE-IT, FOURIER, ODYSSEY and metanalyses including over 200000 patients reported a significant reduction in cardiovascular (CV) events with low-density lipoprotein cholesterol (LDL-C) levels below the previous established targets.8–10 Based on these results tighter LDL-C targets were recommended in the 2019 European Society of Cardiology (ESC) guidelines on Dyslipidemia and Cardiovascular Prevention.11,12 Importantly, in these trials only a minority of patients had co-existing HF (5% in IMPROVE-IT and 15% in ODISSEY).8,10

Previous studies reported that low cholesterol levels were independently associated with adverse outcomes and mortality in patients with chronic HF.13–15 A similar association was also observed in patients admitted for acutely decompensated HF, in whom total cholesterol was inversely related to, length of stay, symptoms at discharge, in-hospital mortality16 and also post-discharge mortality.17

Considering the mentioned correlation between low cholesterol levels and chronic HF prognosis it is unclear whether currently recommended CV risk-based LDL-C targets for general population should be applied to patients with established HF.

The aim of this study was to evaluate the association between LDL-C levels and prognosis in patients admitted for acutely decompensated HF and to establish the lowest LDL-C value associated with worse prognosis in this specific group of patients.

The present study evaluates the impact of on-admission LDL-C in a population of patients admitted for acute HF. In fact, a higher LDL-C level was significantly associated with better prognosis, as an increase in 1 mg/dL in LDL-C level was associated with a 4-fold increase in the likelihood of survival 1 year after discharge.

Determination of LDL-C association with prognosis in patients with high energetic demand conditions, as HF, is becoming more important. This importance arises from the fact that the most recent ESC guidelines on Dyslipidemia and Preventive Cardiology suggest even lower LDL-C targets than the previous ones.12 However, the establishment of these new targets was based on studies where HF patients may have been under-represented.10

Considering this relationship between LDL-C and mortality risk, we established an LDL-C cut-off value associated with worse outcomes. An increased risk of mortality was observed in patients with LDL-C below 88 mg/dL in the overall population. Even though this cut-off is lower than the suggested targets for moderate and low CV risk patients (100 mg/dL and 115 mg/dL, respectively), it is significantly higher than the proposed targets for high and very-high risk patients (55 mg/dL and 70 mg/dL, respectively). Moreover, as almost two thirds of the studied population presented at least high CV risk, theoretically benefiting from a stricter LDL-C control, a specific lowest LDL-C value associated with outcomes was deemed necessary for this subset of patients. The estimated value was 84 mg/dL, still significantly above the suggested LDL-C targets based on individual CV risk. It is important to stress that in this study the relationship between LDL-C and mortality was independent of HF etiology and also of individual CV risk.

We acknowledge that permitting more liberal LDL-C control in HF patients may be associated with a higher rate of thrombotic events, as higher LDL-C levels were associated with more thrombotic events in the present study. However, this trade-off may be beneficial as the net clinical benefit endpoint and mortality, principally, was positively influenced by an LDL-C level equal or above 88 mg/dL.

These results are to be considered in the light of previously published data. Rauchhaus et al., observed a significant benefit of higher total cholesterol in chronic HF survival.20 Additionally, in a study with over 1000 patients with advanced HF, a higher total cholesterol level was associated with a two-fold increase in survival likelihood.14 A long-term study with a 20-year follow-up reported unfavourable prognosis in HF patients with lower LDL-C levels, irrespective of their of the HF etiology.21 Also, in a post hoc analysis of the EVEREST study, total cholesterol and triglycerides were inversely correlated with hospitalization and mortality rates during follow-up.22 On the other hand, a cohort with 422 patients with DCM suggested that although lower cholesterol levels were associated with HF severity, a significant impact on mortality was not shown.23

Besides the direct impact on mortality, in the studied population, having an LDL-C level below the recommended target according to individual CV risk, was also associated with characteristics that are traditionally considered markers of worse prognosis or disease severity in HF patients such as increased age, lower hemoglobin or higher NT-proBNP. Also, a likelihood of higher rate of admissions was observed in patients with lower LDL-C levels. Interestingly, statin treatment did not impact LDL-C levels or the rate of patients that reached their LDL-C target according to individual CV risk. Although it is well established that by reducing ischemic events, statins lessen cardiac damage and subsequently HF development,24 similarly to previous studies,6,7 in this cohort, statins had no impact on mortality nor on other clinical outcomes. Considering all these findings, one may speculate that in HF, LDL-C levels are mostly influenced by HF related factors and not by lipid lowering therapy and, consequently, LDL-C might not be seen as a therapeutic target and instead be considered another marker of adverse prognosis in HF.24 Considering this, other CV risk markers and new metabolic targets should be sought in HF patients.

Although the nature of this relationship remains undefined, similar associations have been described in other chronic diseases such as chronic kidney disease, chronic pulmonary obstructive disease and cancer,25 as well as with advanced age.26 Different theories have been proposed. HF promotes increased metabolic demand and so one may consider LDL-C a surrogate for malnutrition or even cachexia. Although an association between low cholesterol, low albumin and worsened nutritional status had previously been reported,27 the relationship between low cholesterol levels and cachexia is inconsistent.20 Likewise, in the present cohort no relation between LDL-C level and BMI was observed and there were no differences in BMI between patients that had LDL-C below their individual recommended target and patients that did not. However, non-survivors presented a significantly lower BMI, with a median BMI of 25.0 (23.0–27.6) kg/m2, supporting the existence of obesity paradox in HF. Altogether, these data suggest that cachexia may have not played a major pathophysiological role in the impact of LDL-C in HF prognosis. Moreover, in chronic HF a consistent upregulation of inflammatory markers (C-reactive protein, TNF-α and many cytokines) is detected and associated with HF severity and poor outcomes.28,29 Considering the relationship between LDL-C and classic HF risk markers and prognosis presented in this study, one may speculate that this pro-inflammatory status may promote catabolism which, in its initial phases way before cachexia ensues, can be identified by low LDL-C levels. Another possible explanation is that congestion may promote a reduction in LDL-C both through hepatic congestion that leads to impaired lipogenesis30 and through gut edema causing cholesterol malabsorption. The fact that in the present cohort patients with LDL-C below their recommended target had higher levels of NT-proBNP supports both hypotheses.

However, these results should be interpreted in light of certain limitations. Firstly, these results derive from a single center cohort and must be externally validated, especially the LDL-C cut-off value associated with outcomes. The authors propose to act accordingly. On the other hand, this study evaluated the association with prognosis of a single determination of LDL-C and, consequently, conclusions regarding the effect of actively reducing LDL-C cannot be draw. The follow-up was limited to one-year which possibly promoted an underestimation of the association with prognosis of higher levels of LDL-C on the thrombotic events rate. Finally, most causes of death were not report and therefore not available for analysis, which prevented the establishment of any correlation between LDL-C levels and the causes of death.

In conclusion, this study examined the association with prognosis of LDL-C in patients admitted for acutely decompensated HF. The findings shed light on the relationship between LDL-C levels and mortality risk in this specific population. These results demonstrated that higher LDL-C levels were associated with improved survival outcomes, with each 1 mg/dL increase in LDL-C showing a four-fold increase in the likelihood of survival one year after discharge. Importantly, the study identified the lowest LDL-C value, below which there was a markedly increased risk of mortality. For the overall population, this cut-off was determined to be 88 mg/dL, while for high and very-high CV risk patients, it was 84 mg/dL. These cut-off values are significantly higher than the recommended LDL-C targets based on individual cardiovascular risk. These findings challenge the current guidelines and raise questions about the application of standard LDL-C targets to patients with established HF. Nevertheless, these results should be externally validated, and further research is warranted to better understand the role of LDL-C in HF prognosis as well as to refine guidelines for lipid management in this specific patient population.

The authors have no conflicts of interest to declare.