Diabetes and Heart Failure in the Crosshairs: Where is the Target?

This article refers to ‘In-hospital and 1-year mortality associated with diabetes in patients with acute heart failure: results from the ESC-HFA Heart Failure Long-Term Registry’ by G. Targher et al., published in this issue on pages 54–65. In this issue of the Journal, Targher et al.1 present data from 6926 patients hospitalized for acute heart failure (HF) who were recruited in a multinational European registry snapshot acquired over 2 years ending in April 2013. Acute HF could be newly diagnosed or constituted worsening of previous HF; the goal was to evaluate the in-hospital and 1-year outcomes of this cohort focusing on the presence of previous or newly diagnosed diabetes. Strikingly, half of the cohort had diabetes and about one-fifth of the diabetics were newly diagnosed based on transient hyperglycaemia. Although we are not informed about the length of the index hospital admission, it is clear that diabetes—however diagnosed—was associated with a significantly higher hospital mortality [6.8 vs. 4.4%; adjusted-hazard ratio (HR) 1.774; 95% confidence interval 1.282–2.456, P < 0.001]. The current data are confirmatory of previous work derived from the Italian Network registry by some of the co-authors.2 Interestingly, although these mortality differences retain statistical significance, they become markedly attenuated at 1 year (HR 1.16) related to a striking catch-up phenomenon with a near fivefold increase in death among the non-diabetic population. The populations selected herein deserve some scrutiny regarding their representativeness and the validity of the risk adjustment employed. Patients with HF were recruited from clinics and cardiology wards and unlike many recent trials, those treated and discharged from emergency departments were not included. Somewhat surprisingly the large majority in both diabetic and non-diabetic groups had HF with reduced ejection fraction with a mean ejection fraction of 39%, likely reflecting the method of surveillance and recruitment at the sites involved. Unfortunately, given the powerful prognostic capacity of natriuretic peptides, they were not available in enough patients to be incorporated in the model: this would have been expected to provide additional insight, as noted by others.3 What can we learn that is applicable to current patient care? The striking association with an early mortality risk should give clinicians significant pause. Is this related to the risk of diabetes itself, or the comorbid factors often found in patients with diabetes or the severity of diabetes? On the last point, the mean glycated haemoglobin (HbA1c) of patients with diabetes was 7.4%, which is not ideal but far from poorly controlled on average, and 52.5% were already on insulin. As HbA1c was available from only one-third of patients in the diabetes cohort, it is possible that we are only seeing the tip of the iceberg, with surveillance and reporter bias playing a role. There is a potential for harvesting here too (i.e. the patients with diabetes presenting later in their disease process with HF, serving as a marker for more advanced disease). Could insulin or other diabetes therapy (the details of which are unavailable) be related to this increased mortality? Despite the often intense interest in tightly controlling in-hospital blood glucose levels, there is limited data supporting this practice. Indeed, randomized controlled trials have not shown an added advantage of monitoring and adjusting blood glucose in patients admitted with an acute illness despite the expense and invasive nature of this monitoring and therapy: paradoxically, it may be associated with harm in the sickest of patients.4 How then should we deal with a new or existing diagnosis of diabetes when a patient is admitted for HF in 2016? First and foremost, this is an excellent opportunity to review and optimize HF medications, as noted in Figure 1. This should take priority given the sizeable reduction in mortality that beta-blockers, mineralocorticoid receptor antagonists and angiotensin-converting enzyme inhibitors/angiotensin receptor blockers/angiotensin receptor neprilysin inhibitors offer. It should be noted that there appears to have been substantial room to have augmented this therapy in all patients in the current study. Second, scrutiny in hospital provides a key opportunity to enhance continuity of outpatient diabetic care including foot and eye care, which are often overlooked and is an important step in secondary prevention. Finally, a comprehensive review of the diabetes medical regimen for potential additions, subtractions, and dose adjustments is warranted. This includes reducing risk (e.g. discontinuing thiazolidinediones or sulphonylurea in appropriate patients), encouraging the use of first-line therapy (e.g. metformin) and considering the addition of emerging therapy (e.g. sitagliptin, empagliflozin). Uncertainty continues with other newer medications not showing benefit in patients with HF (e.g. liraglutide),5, 6 but which are efficacious in other states of diabetes. Acute HF associated with diabetes is likely comprised of multiple phenotypes that may have an impact on decisions for future care. There are at least three subtypes or aetiologies to consider. First, diabetes is associated with the development of macrovascular and microvascular coronary artery disease. Second, diabetes and hypertension, as is evident in the current study, commonly co-exist: this may promote left ventricular remodelling resulting in HF with preserved ejection fraction. Finally, the direct effects of diabetes on fatty acid metabolism, calcium handling, fibrosis, and mitochondrial energetics may promote HF with either reduced or preserved ejection fraction, independent of hypertension or coronary artery disease.7 It is this third subtype that is of the most interest because the prevalence of diabetes globally has been marked by a 30% rise in the global prevalence and disability-adjusted life years associated with diabetes over the past 20 years.8 Clearly, with the rising prevalence of diabetes, the associated complications are a major public health issue across the full spectrum of lower to higher income countries. Heart failure is one of the potential high cost outcomes of a higher prevalence of diabetes, with a reduction in quality of life, and heavy economic and health-system burden in emergency department and hospital use. Even with better hypertension control and coronary artery disease prevention and treatment, the need to attack the underlying substrate of diabetes as it pertains to the heart and vascular system is critical. Mechanistic investigation into the pathophysiology of the diabetes–HF axis will inform not only better application of current therapies but also guide the development of novel, more effective agents. A potential investigative agenda for this paradigm includes targeting not only the myocardium but also the kidney and vascular system by exploring inflammatory, injury and reparative processes. The results of such searching could well inform pharmacogenomically guided therapy. Finally, because acute HF carries a worse prognosis than acute coronary syndromes, it deserves priority billing and demands urgent attention.9 While the focus on diabetes in this context is welcome, we can expect that if our efforts succeed here, they may be applicable to a much broader population of HF patients with and without diabetes. Conflict of interest: none declared.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.