Outline
Prevalence of Coronary Artery Disease in Heart Failure, 254
Prognostic Significance of Coronary Artery Disease in Heart Failure, 254
Pathophysiology of Acute Heart Failure in Patients With Coronary Artery Disease, 255
Pathophysiology of Chronic Heart Failure in Patients With Coronary Artery Disease and Reduced Ejection Fraction, 256
Left Ventricular Remodeling, 257
Myocardial Ischemia, 257
Hibernation/Stunning, 257
Endothelial Dysfunction, 260
Endothelial Vasodilators, 260
Endothelial Vasoconstrictors, 260
Clinical Manifestations, 260
Coronary Artery Disease and Diastolic Heart Failure, 261
Diabetes, Heart Failure, and Coronary Artery Disease, 262
Therapeutic Options, 262
Immediate Management of the Hospitalized Patient, 262
Long-Term Therapies for the Heart Failure Patient With Coronary Artery Disease, 263
Conclusions, 268
Despite significant progress in the prevention and treatment of cardiovascular disease over the past 30 years, national statistics indicate that the incidence and prevalence of heart failure (HF) continue to rise. This has occurred during a time period in which death rates from coronary artery disease (CAD) and stroke have declined. HF and CAD are both age-related conditions (the prevalence of HF is 1% between the ages of 50 and 59 years, but 10% above the age of 75 years). The increased survival after myocardial infarction (MI) and advances in medical and device therapies (e.g., β-blockers and implantable cardioverter-defibrillators [ICDs]) for the prevention of sudden cardiac death (SCD) have increased the pool of patients with both CAD and HF.
Prevalence of Coronary Artery Disease in Heart Failure
CAD has emerged as a dominant causal factor in HF ( see also Chapter 18 ). Survivors of acute MI, even when not complicated by HF, have a relatively high incidence of subsequent HF hospitalization. This is due not only to the initial ventricular insult caused by the MI but also the progressive nature of CAD ( Fig. 19.1 ). The Framingham Heart Study suggests that the factors contributing to HF are changing, as evidenced by a decrease in valvular heart disease and left ventricular (LV) hypertrophy but an increase in MI as a risk factor from 1950 to 1998. In this analysis, the odds of a prior MI as a cause of HF increased by 26% per decade in men and 48% per decade in women. In contrast, hypertension as a cause of HF decreased by 13% per decade in men and 25% in women, and valvular heart disease as a cause of HF decreased by 24% per decade in men and 17% in women.
In the Studies of Left Ventricular Dysfunction (SOLVD) registry, which enrolled 6273 patients, CAD was determined as the underlying cause of chronic HF in approximately 70% of patients, whereas hypertension was invoked as the primary cause in only 7% of cases. Of note, there was a history of hypertension in 43% of patients. There were striking racial differences observed in this registry. HF was considered due to CAD in 73% of white patients but only 36% of African-American patients ( see also Chapter 40 ).
Pooling data from 26 multicenter trials of chronic HF since 1986, with greater than 43,000 patients, revealed that 62% carried a diagnosis of CAD ( Table 19.1 ). This number may actually underestimate the true prevalence of CAD in this population because in clinical practice and in most studies there is no systemic assessment of coronary artery anatomy. In addition, most of these trials excluded patients with a recent MI, angina, or objective evidence of active ischemia. In a study of 136 patients (<75 years old) hospitalized with de novo HF, a review of the clinical, angiographic, and myocardial perfusion imaging data was used to determine that CAD was the primary cause in greater than 50% of cases. In this study alone, two-thirds of all patients who underwent angiography had obstructive CAD (defined as >50% luminal stenosis), although CAD was not considered the primary causal factor in all cases. In a recent analysis of a large US acute HF registry, myocardial ischemia was found to be a leading precipitating factor for hospitalization.
| Trial | Year | N | CAD |
|---|---|---|---|
| VHEFT-1 | 1986 | 642 | 282 |
| CONSENSUS | 1987 | 253 | 146 |
| Milrinone | 1989 | 230 | 115 |
| PROMISE | 1991 | 1088 | 590 |
| SOLVD-T | 1991 | 2569 | 1828 |
| VHEFT-2 | 1991 | 804 | 427 |
| SOVLD-P | 1992 | 4228 | 3518 |
| RADIANCE | 1993 | 178 | 107 |
| Vesnarinone | 1993 | 477 | 249 |
| STAT-CHF | 1995 | 674 | 481 |
| Carvedilol | 1996 | 1094 | 521 |
| PRAISE | 1996 | 1153 | 732 |
| DIG | 1997 | 6800 | 4793 |
| VEST | 1998 | 3833 | 2236 |
| RALES | 1999 | 1663 | 907 |
| DIAMOND | 1999 | 1518 | 1017 |
| Nesiritide | 2000 | 127 | 58 |
| COPERNICUS | 2001 | 2289 | 1534 |
| BEST | 2001 | 2708 | 1587 |
| Val-HeFT | 2001 | 5010 | 2880 |
| MIRACLE | 2002 | 453 | 108 |
| COMPANION | 2004 | 1520 | 842 |
| SCD-HeFT | 2005 | 2521 | 1310 |
| CARE-HF | 2005 | 813 | 309 |
| RethinQ | 2007 | 172 | 90 |
| Dronedarone | 2008 | 627 | 407 |
| Total | 43,444 | 27,074 |
Prognostic Significance of Coronary Artery Disease in Heart Failure
The presence of CAD in patients with HF has been shown to be independently associated with a worsened long-term outcome in numerous studies. Atherosclerosis is an important contributing cause of death in HF patients through a variety of mechanisms, including SCD, progressive ventricular failure, MI, renal failure, and stroke. In patients with HF, the long-term prognosis is directly related to the angiographic extent and severity of CAD. This has been demonstrated both in HF patients with LV systolic dysfunction and in those with preserved systolic function.
Recent data suggest that the mechanism of SCD may differ between ischemic and nonischemic HF, with acute coronary events representing the major cause of SCD in patients with CAD. In the Assessment of Treatment with Lisinopril and Survival (ATLAS) study, 54% of patients with chronic HF and CAD who died suddenly had autopsy evidence of acute MI. In another autopsy study of 180 patients with known ischemic cardiomyopathy, acute MI was responsible for 57% of the deaths. This study revealed that before autopsy data were available, many deaths as a result of acute MI in patients with HF were misclassified as caused by progressive HF or arrhythmias. In another study of patients with HF and left ventricular systolic dysfunction (LVSD) 25% of repeat hospitalizations were attributed to acute coronary syndrome (ACS). However, approximately 10% of patients with a history of HF who were subsequently hospitalized for ACS were originally classified as having a nonischemic cause. These findings further emphasize the importance of accurately assessing for the presence of CAD in patients with HF.
Pathophysiology of Acute Heart Failure in Patients With Coronary Artery Disease
Underlying Coronary Artery Disease
Patients hospitalized with acute HF differ from patients with chronic ambulatory HF with respect to prognosis and early management ( see also Chapter 36 ). Patients with CAD who develop acute HF do so with either an ACS or a non-ACS presentation. Although the majority of such patients do not have ACS, there is considerable overlap in these two presentations with respect to clinical characteristics ( Table 19.2 ) and potential therapies ( Table 19.3 ). However, the approach to the patient with ACS has become more standardized in clinical practice guidelines compared with the acute HF patient with a non-ACS presentation. Myocardial injury is common in both, but in ACS patients it is usually the principal cause of HF, whereas in non-ACS patients myocardial injury may be the result of worsening HF. In these latter patients, cardiac troponin levels are frequently elevated in patients with acute HF, representing myocardial injury ( see also Chapter 33 ). In the era of high-sensitivity troponin assays, the values of troponin in AHF patients often surpass the acute MI threshold (i.e., the 99th upper reference limit [URL]) and may demonstrate a typical rise and fall. Such events have been classified as “acute myocardial injury” rather than type II myocardial infraction in the new fourth universal definition of MI, although this is primarily a semantic distinction. Regardless, these acute elevations of troponin in HF are a marker of worse outcomes.
| AHFS and CAD | ACS Complicated by HF | |
|---|---|---|
| Dyspnea | Common | Common |
| Chest discomfort | Uncommon | Common |
| Prior HF | Common | Uncommon |
| BNP/N-terminal pro-BNP | Elevated | Elevated |
| Troponin | Normal or elevated a | Usually elevated |
| Left ventricular systolic function | Normal or depressed | Normal or depressed |
| Diagnostic testing for CAD b (ischemia/viability/angiography) | Uncommon | Standard (per guidelines) |
| Myocardial revascularization | Uncommon b | Standard (per guidelines) |
| Secondary prevention for CAD | Underused | Standard (per guidelines) |
| In-hospital mortality | Relatively low | Relatively high |
| Early after-discharge death or rehospitalization | High | High |
a Typically low-level elevation.
| AHFS and CAD | ACS Complicated by HF | |
|---|---|---|
| Immediate Therapies | ||
| Nitrates | Yes | Yes |
| Antiplatelet agents | Yes | Yes |
| Anticoagulation | No | Yes |
| Inotropes | Avoid if possible | Avoid if possible |
| Statins | Yes | Yes |
| Renin-Angiotensin System Modulation | ||
| ACE-I or ARB | Yes | Yes |
| Aldosterone blockade (if LVSD) | Yes | Yes |
| β-blockers | Yes | Yes |
| Early angiography/revascularization | Yes a | Yes a |
a If jeopardized myocardium present (ischemia or viability).
In acute HF, a high LV diastolic pressure can result in subendocardial ischemia (even in the absence of epicardial coronary disease). Experimental evidence suggests that troponin release is correlated with both LV loading conditions and microvascular dysfunction. Excessive neurohormonal activation can exacerbate ischemia via increased cardiac contractility and reduced coronary perfusion because of endothelial dysfunction. In addition, patients with acute HF and CAD often have hibernating or stunned myocardium. Together, all of these factors may result in myocardial injury.
Low systemic blood pressure combined with elevated LV diastolic pressure reduces coronary perfusion, and in this setting, the autoregulation between coronary artery perfusion pressure and coronary vasoactive tone may be lost or impaired in patients with obstructive epicardial CAD. This may contribute to myocardial injury (as reflected by cardiac enzyme elevation) and worse outcomes. This may help to explain why patients with acute HF and underlying CAD have a worse outcome than those without CAD and have improved outcomes if they have a history of myocardial revascularization.
Acute Coronary Syndromes
Approximately 10% to 20% of patients with ACS have concomitant acute HF, and approximately 10% of ACS patients develop HF in-hospital. In the EuroHeart Survey II on HF, 42% of all de novo HF cases were due to ACS. Patients with ACS and ST-segment elevation typically have a high degree of myocardial injury. ACS patients with HF but without ST-segment elevation also have significant cardiac enzyme elevation but a smaller degree of injury. The short-term risk of adverse outcomes in ACS patients with HF is directly proportional to the level of troponin elevation. Most of these patients do not have a history of HF or LVSD.
Patients with ACS complicated by HF have markedly increased short- and long-term mortality rates compared with ACS patients without HF. ACS patients who develop HF after the initial presentation have even higher mortality rates. The prognosis of ACS complicated by HF is directly related to the Killip class. Compared with Killip class I patients, patients with an ACS in Killip class II or III HF are four times more likely to die in-hospital. The risk goes up to 10-fold for patients with cardiogenic shock (Killip class IV). Among ACS patients who recover from transient HF, the majority develop recurrent HF.
Pathophysiology of Chronic Heart Failure in Patients With Coronary Artery Disease and Reduced Ejection Fraction
HF in the setting of CAD is a heterogeneous condition with several possible factors contributing to clinical manifestations of HF and LVSD and/or diastolic dysfunction. First and foremost, the sequelae of MI, with loss of functioning myocytes, development of myocardial fibrosis, and subsequent LV remodeling, result in chamber dilation and neurohormonal activation that lead to progressive deterioration of the remaining viable myocardium. This is a well-recognized clinical process that can be ameliorated after acute MI by the use of angiotensin-converting enzyme (ACE) inhibitor therapy, beat-blocking agents, mineralocorticoid receptors, and myocardial revascularization. Second, the majority of patients surviving MI have significant atherosclerotic disease in coronary arteries other than the infarct-related artery. Thus superimposed on the LV with irreversibly damaged myocardium, there is often a considerable degree of jeopardized myocardium served by a stenotic coronary artery either within the infarct zone or remote from the infarcted tissue. This may result in myocardial ischemia/hibernation, contributing to LV dysfunction and the risk of recurrent MI producing further deterioration in LV function or SCD. Finally, endothelial dysfunction, a characteristic feature of atherosclerotic CAD, may also contribute importantly and independently to the progression of LV dysfunction ( Fig. 19.2 ).
Left Ventricular Remodeling (see also Chapter 12 )
LV remodeling is the process by which the LV’s size, shape, and function are altered in response to acute injury and/or hemodynamic overload. As reviewed in detail in Chapter 12 , LV remodeling occurs secondary to mechanical, neurohormonal, and genetic factors ( Fig. 19.3 ). The severe loss of myocardial cells after acute MI results in an abrupt increase in loading conditions that induces a unique pattern of remodeling involving the infarct zone, the infarct border zone, and the remote noninfarcted myocardium. Myocyte necrosis initiates a process of reparative changes, which consist of dilation, hypertrophy, and the formation of a collagen scar. Other factors may influence this process, including the location and transmurality of the infarct, the extent of myocardial stunning beyond the initial infarction, infarct-related artery patency, and local trophic factors. Postinfarction remodeling has been arbitrarily divided into an early phase (within 72 hours) and a late phase (beyond 72 hours). In patients with transmural MI, the early phase involves expansion of the infarct, with thinning and bulging that may result in ventricular rupture, aneurysm, mitral insufficiency, and ventricular tachyarrhythmias. Late remodeling involves the LV globally and is associated morphologically with dilation, hypertrophy, and myocyte hypertrophy (see Fig. 19.3 ). Importantly, LV remodeling creates a de novo mechanical burden for the heart and can integrated processes contribute independently to the progression of HF.
Myocardial Ischemia
Under basal conditions, episodes of reversible myocardial ischemia caused by a severe coronary artery stenosis superimposed on the LV with depressed systolic function may produce transient worsening of LV function. This exacerbates dyspnea on exertion and fatigue. In many patients, these HF symptoms, stimulated by exercise, represent an anginal equivalent that may occur in the absence of chest pain.
Transient LV dysfunction can aggravate symptoms during stress or spontaneous ischemia in patients with CAD and HF. Ischemia can also produce a rapid and massive increase in the concentration of all three endogenous catecholamines (norepinephrine, epinephrine, dopamine) in the myocardial interstitium, which is mediated by inhibition of neuronal reuptake mechanisms. High myocardial catecholamine concentration may have a deleterious effect on cardiac myocytes.
Ischemia may also lead to myocyte apoptosis, which may result in progression of LV dysfunction without a clear ischemic event. This situation also indicates that ischemia from a chronic stenosis can produce substantial myocyte loss in the absence of significant necrosis or fibrosis. Ischemia may also cause an increase in endothelin production that may have a negative effect on LV function. Aggressive medical and surgical interventions designed to ameliorate ischemia appear to have a substantial impact on limiting apoptosis.
Hibernation/Stunning
Episodes of transient myocardial ischemia may cause prolonged LVSD that persists after the ischemic insult itself has resolved. This process is termed stunning, which is similar to more severe and protracted myocardial stunning that results from coronary occlusion and reperfusion ( Fig. 19.4A ). Recurrent episodes of myocardial ischemia that produce repetitive myocardial stunning may contribute to overall LV dysfunction and HF symptoms.
Another important mechanism for systolic dysfunction with additive effects on LV performance is myocardial hibernation. Once considered a process in which myocardial contraction is downregulated in response to chronic reduction in myocardial blood supply, the current evidence supports the hypothesis that persistent contractile dysfunction in patients with chronic CAD represents a process of programmed disassembly of contractile elements following repeated episodes of reversible ischemia (see Fig. 19.4B ). Thus rather than a “protective” mechanism, hibernation represents a disadvantageous process that, left uncorrected, may lead to apoptosis and myocyte loss, replacement fibrosis, graded and reciprocal changes in alpha- and beat-adrenergic receptor density, progressive LVSD, and the risk of ventricular arrhythmias ( Fig. 19.5 ). This process may affect a substantial number of HF patients. Among patients with HF, CAD, and LVSD, approximately 50% have evidence of viable but dysfunctional myocardium.
Diagnosis
Hibernating myocardium should be suspected in all patients with CAD and chronic LV dysfunction of any degree, regional and global. Up to 50% of patients with CAD and chronic LV dysfunction have significant areas of dysfunctional but viable myocardium. Hibernating myocardium can be determined with the use of imaging techniques that detect myocardial contractile reserve, preserved metabolic activity, or cell membrane integrity within the region of dysfunctional myocardium. Intact perfusion, cell membrane integrity, and intact mitochondria can be evaluated with single-photon emission tomography using thallium 201–and/or technetium 99m–labeled traces. Preserved glucose metabolism can be assessed by positron emission tomography using F18-fluorodeoxyglucose. Contractile reserve can be unmasked by infusion of low-dose dobutamine during echocardiography. The use of these techniques has been associated with improved survival in patients with chronic HF and significant viability who underwent myocardial revascularization.
Cardiac magnetic resonance imaging is also an established technique to assess myocardial viability and the potential for recovery of LV function. Resting cine MRI can be used to assess LV end-diastolic wall thickness. An end-diastolic wall thickness less than 5 to 6 mm is a marker of transmural MI and virtually excludes the presence of viable myocardium. In dysfunctional myocardium with preserved end-diastolic wall thickness (6 mm), detection of contractile reserve during low-dose dobutamine infusion confirms the presence of viable myocardium. Gadolinium-based contrast agents have been used to detect nonviable myocardium because these agents accumulate selectively in areas of scar tissue. It should be noted that this technique is extremely sensitive in detecting scar tissue (with very high spatial resolution), but the absence of scar tissue does not permit discrimination between normal tissue and hibernating or stunned myocardium.
Clinical Implications
The presence of viable but dysfunctional myocardium can be used to predict a favorable response to myocardial revascularization and pharmacologic therapy. The restoration of blood flow with revascularization or treatment with agents that improve endothelial function and blood flow, such as statins and beat-blockers, may improve contractility in a hibernating area. In contrast, agents such as dobutamine and milrinone, especially at high doses, may precipitate myocardial necrosis and are associated with worse long-term outcomes in patients with CAD and HF. Hibernating myocardium is associated with global alterations in LV volume and shape, not just impairment of underperfused segments. This explains why myocardial revascularization of hibernating territories can promote reverse remodeling globally.
However, the clinical importance of routine viability testing remains controversial based on the results of the viability testing in the Surgical Treatment for Ischemic Heart Failure (STICH) trial, described in more detail later.
Endothelial Dysfunction
Available data suggest that the coronary endothelium plays an important role not only in the control of blood flow and vascular patency but also in the physiologic modulation of myocardial structure and function. Thus endothelial dysfunction, an inherent component of the pathophysiology of atherosclerotic CAD, may directly affect ventricular function.
Endothelial Vasodilators
The endothelial release of nitric oxide relaxes vascular smooth muscle cells in association with activation of guanylyl cyclase and increased levels of cyclic glucose monophosphate. NO is the most potent endogenous vasodilator and is responsible for the maintenance of vasovascular tone. NO also inhibits smooth muscle cell proliferation and migration, leukocyte adhesion, and platelet aggregation.
Endothelial Vasoconstrictors
The major endothelin-derived vasoconstrictive substances include angiotensin II and endothelin. Angiotensin II is a potent vasoconstrictor that also exerts a variety of effects on vascular structure and function. Studies of angiotensin II indicate the involvement of the renin-angiotensin system in many aspects of vascular homeostasis. Angiotensin II increases the production of plasminogen activator inhibitor type 1, the primary endogenous inhibitor of tissue plasminogen activator, and promotes vascular growth in addition to stimulating the production of other growth factors. Angiotensin II also enhances platelet aggregation, sensitizes the platelets to the effects of direct platelet agonists, and stimulates the production of endothelin. Endothelin is the most potent endogenous vasoconstrictor yet identified and promotes proliferation of smooth muscle cells and secretion of extracellular matrix, which contribute to the formation of atherosclerotic plaque.
Disordered endothelial function in patients with CAD stimulates vasoconstriction, smooth muscle migration and proliferation, increased lipid deposition in the vessel wall, and possibly coronary thrombosis. This promotes myocardial ischemia, which may further contribute directly or indirectly to the progression of LV dysfunction. The release of endothelin is also increased in failing myocardium. Angiotensin II contributes to the release of endothelin and the excessive degradation of NO. Taken together, these observations make a case for an interplay between the failing myocardium and the coronary endothelium that potentiates the progression of both CAD and LV dysfunction.
Properties of the normal endothelium serve to relax vascular tone and inhibit smooth muscle growth, platelet aggregation, and leukocyte adhesion. Many drugs that reduce mortality and reinfarction in patients with CAD have the potential to improve endothelial function, including lipid-lowering agents, ACE inhibitors, nonselective β-blockers, and aspirin. For example, marked reduction of serum cholesterol is associated with a rapid recovery of endothelial function, improvement of myocardial perfusion, and reduction of myocardial ischemia. An improvement in tissue perfusion is an important goal in patients with HF in terms of both the peripheral and the coronary circulation.
In summary, endothelial dysfunction may further reduce blood flow, promote progression of coronary atherosclerosis, and have direct negative effects on the myocardial cells and the interstitium.
Clinical Manifestations
Reinfarction
Patients with HF and CAD are at increased risk for reinfarction. In clinical trials, the rate of infarction or reinfarction is relatively low, with a fatal MI rate of 3%. However, in one study, more than half of patients with HF and CAD who died suddenly had autopsy evidence of an acute ischemic event (e.g., coronary clot, recent infarct), suggesting that the number of patients with plaque rupture is not accounted for in clinical trials.
Sudden Cardiac Death
The risk of SCD after MI has significantly declined in recent years. However, the occurrence of HF post-MI is associated with a markedly increased risk of SCD. In several clinical HF trials, SCD accounted for 20% to 60% of deaths, depending on the severity of HF. In the Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF), 64% of patients in New York Heart Association (NYHA) class II who subsequently died had SCD compared with 59% of patients in class III and 33% of patients in class IV. Several factors have been implicated in the high rate of SCD in patients with HF. These include subendocardial ischemia, ventricular hypertrophy, stretching of myocytes, a high sympathetic tone, abnormal baroreceptor responsiveness lowering the threshold for a malignant arrhythmia, potassium and magnesium depletion, and coronary artery emboli from atrial or LV thrombi. However, CAD probably contributes directly to SCD. Patients with CAD and systolic HF have dilated hearts, large regions of myocardial scar, and obstructive epicardial coronary stenosis. CAD and its major structural consequences (i.e., plaque rupture, thrombosis, and infarction) constitute the most common structural basis of SCD.
Uretsky and colleagues reported the relative importance of an acute coronary event as a trigger for SCD in patients with HF who were studied in the ATLAS trial, which included 3164 patients with moderate to severe systolic HF. There were 1383 deaths (43.7%) during the follow-up period of 3 to 5 years. An autopsy was performed in only 188 patients, and the postmortem data were available in only 171 patients (12.4% of the total patients who died). Patients who died in this study were older and had both more symptoms and a higher prevalence of CAD than the surviving patients. The patients who died and did not undergo autopsy were similar to those who died and were subjected to autopsy. Acute coronary findings were observed in 54% of the patients with significant CAD who died suddenly ( Fig. 19.6 ). The ATLAS study was the first to demonstrate that recent coronary events are frequently unrecognized in patients with moderate to advanced HF symptoms who die suddenly, especially in patients with CAD.
Other studies have documented a high frequency of plaque rupture or coronary thrombosis in patients with CAD who suffered SCD. However, it should be noted that these studies reported a much higher incidence of ruptured plaque, ranging from 57% to 81%, than the ATLAS autopsy study. However, the prevalence of clinical acute coronary findings in the same series ranged from 21% to 41%, which was similar to the ATLAS study. Because the autopsy findings reported in ATLAS were based on routine clinical examinations, it is unlikely that the examinations involved the degree of detail necessary to observe ruptured plaque and small thrombi. Therefore it is possible that the rate of acute coronary events may have been even higher than reported. This study underlines the importance of strategies to prevent and treat acute coronary events to successfully prevent SCD in patients with HF. For example, in the ATLAS trial, two-thirds of patients had CAD but only 40% of this group was taking aspirin.
Among patients with HF who receive an ICD for the primary prevention of SCD, those who receive shocks have a markedly increased short-term risk of death compared with those who do not receive shocks. This risk may be much higher in patients with CAD. A recent analysis of the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) revealed that patients with HF due to CAD who received an appropriate ICD shock had a threefold increased risk of mortality compared with patients with HF and a nonischemic cause who received an appropriate ICD shock. This suggests that among patients with HF, those with CAD can develop fundamental alterations in the underlying arrhythmic substrate that predisposes them to SCD.
Coronary Artery Disease and Diastolic Heart Failure (See Also Chapter 39 )
The vast majority of HF trials conducted over the past 30 years have studied patients with LVSD. However, HF with relatively preserved systolic function is present in approximately half of all patients hospitalized with HF. Among patients with HF and preserved systolic function, approximately 60% have documented CAD. Over the past two decades, the relative proportion of patients with HF and preserved systolic function has risen steadily relative to those with LVSD. Patients with HF and preserved systolic function tend to be older than those with HF and LVSD. Thus the relative rise in this category of HF is reflective of an aging population. This rise has also corresponded to increased rates of CAD, hypertension, diabetes, and atrial fibrillation in this population. Among patients hospitalized with HF, the early and long-term risk of death is similar for patients with preserved systolic function and LVSD. However, patients with HF and preserved systolic function are more likely to die from other cardiac comorbidities, including CAD, rather than progressive HF when compared with patients with HF and LVSD.
When systolic function is preserved, it is assumed that the majority of these patients have HF signs and symptoms on the basis of abnormal LV diastolic function. A variety of factors predispose to abnormalities in diastolic functional behavior of the LV and lead to elevating filling pressures, impaired forward output, or both, despite normal systolic function. Myocardial ischemia is one of the leading factors. Pulmonary congestion can be caused by reversible episodes of ischemia, which impair LV relaxation and increase LV filling pressure.
The prognosis in patients with HF and preserved systolic function in the presence of CAD may be directly related to the angiographic burden of CAD. Data from the Duke experience demonstrated that patients with HF and preserved systolic function have a worse 5-year survival if they have left main or three-vessel CAD versus those with one- to two-vessel CAD. Similarly, according to the Coronary Artery Surgery Study (CASS) registry, the 6-year survival rate of patients with normal ejection fraction and HF symptoms was 92% in patients with no CAD, 83% in patients with one- or two-vessel CAD, and 68% in patients with three-vessel disease.
There is a need for reappraisal on whether systolic function is truly normal at the time when HF symptoms are present in patients diagnosed with HF and “normal” systolic function. The majority of studies of this syndrome did not report the timing of the evaluation demonstrating normal systolic function relative to the episodes of HF itself. In other studies, the evaluation was performed days to weeks after the episode. Transient ischemia may cause regional systolic dysfunction, which in many patients was severe and extensive enough to cause a brief but profound reduction in global LV function. The pathophysiologic changes in regional and global systolic function form the basis for exercise radionuclide ventriculography and exercise echocardiography as diagnostic tests for myocardial ischemia due to CAD. It may be that many patients with apparently normal systolic function and HF caused by CAD do not have isolated diastolic dysfunction but instead have transient systolic and diastolic dysfunction at the time when myocardial ischemia induces HF symptoms.
Diabetes, Heart Failure, and Coronary Artery Disease (See Also Chapter 48 )
In terms of cardiovascular risk, a diagnosis of diabetes is comparable to a diagnosis of CAD. The prevalence of documented CAD in diabetic patients has been shown to be as high as 55%, compared with 2% to 4% for the general population. Diabetic patients with a history of MI have a markedly worse prognosis than individuals with only one of these conditions. A significant number of patients with HF have diabetes: 23% in the CONSENSUS (Cooperative North Scandinavian Enalapril Survival Study) trial, 25% in SOLVD (Studies of Left Ventricular Dysfunction), 20% in V-HeFT (Vasodilator Heart Failure Trial), 20% in ATLAS (Assessment of Treatment with Lisinopril and Survival), 27% in RESOLVD (Randomized Evaluation of Left Ventricular Dysfunction), 42% in the OPTIMIZE-HF (Organized Program To Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure) registry, and 44% in the ADHERE (Acute Decompensated Heart Failure National Registry). Diabetes is an independent risk factor for the development of HF. In the Framingham Heart Study, the relative risk for developing HF in diabetic patients was 3.8 for men and 5.5 for women, respectively, compared with nondiabetic patients. The risk of developing HF in diabetic patients has been directly related to glycemic control. In the United Kingdom Prospective Diabetic Study (UKPDS), for each 1% increase in glycosylated hemoglobin level, the risk of HF rose by 12%.
The presence of diabetes in patients with HF is associated with substantially higher mortality rates. Several studies suggest that the increased risk in diabetic patients with HF compared with nondiabetic patients with HF is limited to individuals with concomitant CAD. Diabetic patients with CAD also have worse outcomes following myocardial revascularization. Derangements associated with diabetes, including hyperglycemia, insulin resistance, dyslipidemia, inflammation, and thrombosis, contribute to the development of hypertension, endothelial cell dysfunction, accelerated atherogenesis, and coronary thrombosis. In addition, diabetic patients exhibit more complex and diffuse anatomic patterns of CAD, including more lipid-rich plaques and intracoronary thrombi but less compensatory vascular remodeling.
Diabetes directly contributes to HF in patients with LVSD, diastolic dysfunction, or both. The increased mortality in patients with HF in the presence of diabetes has been observed in patients with either LVSD or preserved systolic function. Ventricular dysfunction in patients with HF and diabetes has been termed diabetic cardiomyopathy and is the result of the complex interplay between the sympathetic nervous system and the renin-angiotensin-aldosterone system (RAAS), increased levels of circulation cytokines, alterations in heart rate variability, and increased oxidative stress. Chronic hyperglycemia leads to the glycation of collagen and elevated serum levels of advanced glycation end products, which results in increased myocardial stiffness. Pathologically, this cardiomyopathy is characterized by myocyte atrophy, interstitial fibrosis, increased periodic acid–Schiff (PAS)-positive material, intramyocardial microangiopathy, and depletion of myocardial catecholamines. In diabetic patients with HF and LVSD, myocardial fibrosis and the deposition of advanced glycation end products predominate, whereas in those with HF and preserved systolic function, increased cardiomyocyte resting tension is a more important mechanism. There is increased recognition that therapies for diabetes may interact with the risk of both incident HF and the risk of disease progression in patients with known HF ( see also Chapter 17, Chapter 48 ).
Therapeutic Options
Recognition that progression of CAD may contribute importantly to progression of HF, in at least a subset of patients, shifts the focus from medical management designed solely to reduce neurohormonal activation and alleviate congestive symptoms to a strategy designed to use aggressive secondary prevention measures. Those efforts to slow the progression of CAD include attention to reducing the risk of acute coronary events by plaque stabilization, reducing ischemia, and enhancing endothelial function. It is noteworthy that the classes of drugs that have shown conclusively to improve survival in HF—ACE inhibitors, angiotensin receptor blockers (ARBs), β-blockers, and aldosterone antagonists ( see also Chapter 37 )—address those factors. The beneficial effects of these drugs may relate as much to their vascular protective effects as to their neurohormonal blocking effects. Patients hospitalized with HF are frequently undertreated for CAD. For example, ACS patients with acute HF are less likely to receive antiplatelet agents, β-blockers, ACE inhibitors, or statins than ACS patients without HF. In addition to pharmacologic therapy, myocardial revascularization, surgical therapy, and cardiac device therapy may play an important role in the treatment of patients with HF in the setting of CAD.
Immediate Management of the Hospitalized Patient
The immediate management of acute HF usually occurs in the emergency department ( see also Chapter 36 ). There is considerable overlap in the presentation and management of acute HF patients with CAD and ACS versus CAD and non-ACS (see Tables 19.2 and 19.3 ). In patients with underlying CAD who are not hypotensive, nitrates may provide a rapid reduction of myocardial ischemia and improve coronary perfusion. In patients with severe pulmonary edema, the combination of high-dose nitrates and low-dose diuretics (vs. low-dose nitrates and high-dose diuretics) led to significantly decreased rates of mechanical ventilation and MI. In a large acute HF registry, the use of intravenous nitroglycerin or nesiritide was associated with lower in-hospital mortality compared with treatment with dobutamine or milrinone. However, compared with intravenous nesiritide in acute HF patients, of whom greater than 60% had documented CAD, intravenous nitroglycerin has been associated with less deterioration of renal function and a trend toward less mortality at 30 days.
Inotropes may be particularly harmful when used in HF patients with CAD. Experimentally, the use of dobutamine in a model of HF with hibernating myocardium led to increased myocardial necrosis. Hospitalized HF patients with troponin elevation have significantly higher in-hospital mortality when inotropes are used. In the Outcomes of a Prospective Trial of Intravenous Milrinone for Exacerbations of Chronic Heart Failure (OPTIME-CHF) trial, use of the phosphodiesterase inhibitor milrinone in patients with CAD was associated with increased postdischarge mortality compared with a placebo. In general, a decrease in coronary perfusion as a result of a decrease in blood pressure and/or an increase in heart rate resulting from inotropes with vasodilator properties, or inotropes used in conjunction with vasodilators, may be particularly deleterious in HF patients with CAD. In a recent series of 112 inotrope-dependent patients with stage D HF patients not eligible for transplantation, there was no difference in the adjusted mortality rate observed between those treated with milrinone and those treated with dobutamine.
Long-Term Therapies for the Heart Failure Patient With Coronary Artery Disease
Renin-angiotensin-aldosterone system modulators
The RAAS regulates sodium balance, fluid volume, and blood pressure, which has a profound impact on HF and CAD ( see also Chapter 5 ). The use of ACE inhibitors or ARBs is strongly indicated in HF patients with LVSD and is also indicated for the secondary prevention of cardiovascular events in all patients with CAD.
Endothelial dysfunction plays a fundamental role in many forms of cardiovascular disease and is the final common pathway through which most cardiovascular risk factors contribute to inflammation and atherosclerosis. Angiotensin II is a powerful vasoconstrictor and also stimulates smooth muscle cells (hyperplasia), fibroblast proliferation, collagen deposition, inflammation, and thrombosis. All these maladaptations can be mitigated by the use of ACE inhibitors or ARBs. In the SOLVD and the SAVE (Survival and Ventricular Enlargement) trials, the ACE inhibitors enalapril and captopril not only reduced overall mortality in patients with CAD but also reduced the rate of nonfatal MI and unstable angina. In the SAVE trial, a 25% decrease in MI with captopril occurred despite the selection criteria, which excluded patients with residual ischemia who were considered at great risk of reinfarction. The reduction of acute ischemic events would not have been anticipated only on the basis of the hemodynamic or neurohormonal effects of ACE inhibitors. Moreover, in the SOLVD trial, the reduction of unstable angina and MI with enalapril was not evident until more than 6 months after randomization. This suggests that the beneficial effects of enalapril on ischemic events was not due to an immediate effect related to a primary or secondary reduction in LV afterload.
The addition of eplerenone, an aldosterone antagonist, to optimal medical therapy in ACS patients with acute HF and LVSD significantly reduced death and rehospitalization in the Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS) trial ( Fig. 19.7 ). The reduction in death corresponded with a decrease in SCD, which may be due to the inhibition of myocardial fibrosis.
