Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Hypertension and Its Relation to Heart Failure With a Preserved Ejection Fraction
Introduction
Hypertension (HTN) is the single most important and widespread cardiovascular diagnosis in the developed world and, as such, an enormous public health problem. With the recent change in guidelines reducing the systolic blood pressure (BP)–lowering target from 140 to 130 mmHg based on the landmark Systolic Blood Pressure Intervention Trial (SPRINT), it is now estimated that 46% of Americans should be considered to have treatable HTN. HTN is both a major risk factor for other forms of cardiovascular disease (e.g., coronary artery disease [CAD] and stroke) and a direct cause and substrate of cardiovascular disease (e.g., concentric left ventricular [LV] hypertrophy [LVH] and remodeling and heart failure [HF]). Despite recognition of the importance of HTN, there remains uncertainty about what constitutes the appropriate threshold for treatment. The SPRINT study results showing that targeting a systolic BP of 120 mmHg rather than 140 mmHg improved long-term outcomes begs the question, Should attempts be made to lower BP even further than 120 mmHg? In light of the SPRINT study, it is interesting to note that as recently as the 1960s there was some disagreement as to whether HTN should even be treated.
In view of the scope of the topic and the theme of this book, this chapter will focus on HTN and its relationship to concentric remodeling (CR), LV diastolic filling properties, and heart failure with a preserved LV ejection fraction (HFpEF). Of necessity, there are some weaknesses to this approach. First and most importantly, we largely disregard patients with a history of HTN without CR. Many of them undoubtedly have abnormal diastolic function, and a number clearly have HFpEF, but there are virtually no studies of ventricular or myocardial function or outcomes that have examined them as a distinct group. Second, many if not most patients with HTN and CR have comorbidities that contribute independently to CR and/or diastolic dysfunction. Thus we are usually not observing the effects of HTN in isolation. Third, many BP-lowering drugs have other properties (e.g., anti-inflammatory, antifibrotic) that contribute to their effects, so that HTN treatment typically involves more than simply lowering BP. Last, in considering HFpEF it is important to remember that not all patients have HTN, even after excluding specific diagnoses such as hypertrophic cardiomyopathy and amyloid heart disease. While small in number these patients have not been analyzed separately. The following case study serves as an introduction to the problem of HTN as it relates to these three main themes.
Case Study
A 58-year-old man is referred by his internist for evaluation of progressive dyspnea on exertion. He is sedentary, has longstanding HTN and obesity, and for many years has found that he becomes dyspneic going up more than one flight of stairs or carrying heavy bags of groceries. Over the last 6 to 8 months this complaint has been progressively worsening, such that he now becomes dyspneic walking on level ground at anything more than a normal pace for him. He is self-employed as a landscape architect and has had to cut down on his work because of his dyspnea. He also has obstructive sleep apnea (OSA), for which he is being treated with continuous positive airway pressure (CPAP). He sleeps on two pillows but denies paroxysmal nocturnal dyspnea or peripheral edema. He denies chest discomfort, palpitations, dizziness, or syncope. His HTN has been somewhat difficult to control over the years, and despite various drugs his systolic BP routinely ranges from 150 to 160 mmHg. He has been told that he has “borderline-elevated blood sugar.” He is a nonsmoker and does not consume significant amounts of alcohol.
His past medical history is significant only for gastroesophageal reflux disease manifest as occasional nonexertional epigastric burning relieved by a proton pump inhibitor, and osteoarthritis mainly affecting his knees.
His current medications include aspirin 81 mg daily, chlorthalidone 25 mg daily, metoprolol tartrate 50 mg twice daily, losartan 50 mg daily, furosemide 20 mg daily, atorvastatin 20 mg daily, and pantoprazole 40 mg daily.
On physical examination he appears in no acute distress and has central obesity. His BP is 155/95 mmHg (after sitting for 5 minutes), heart rate 75 bpm, and respiratory rate 14/min. He is 68 in. (150 cm) tall and weighs 235 lb (107 kg, BMI = 36). His chest is clear to auscultation, and his estimated jugular venous pressure is 7 cm. His carotid arterial pulses are normal. His cardiac point of maximal impulse cannot be palpated; S 1 and S 2 are normal, a fourth heart sound is present at the cardiac apex, and there are no murmurs. His abdomen is obese and nontender, and his extremities reveal 1+ bilateral pedal edema.
His laboratory data include a normal complete blood count (CBC), serum blood urea nitrogen (BUN) 26 mg/dL, creatinine 1.4 mg/dL, estimated glomerular filtration rate (GFR) 55 mL/min/1.73 m , and normal electrolytes with serum K+ 4.6 mEq/L. His fasting blood glucose is 105 mg/dl, and hemoglobin A1c is 6.3%. His total cholesterol is 210 mg/dL, low-density lipoprotein (LDL) 82 mg/dL, high-density lipoprotein (HDL) 38 mg/dL, and triglycerides 180 mg/dL. His N-terminal pro-brain natriuretic peptide (NT-proBNP) is 1050 pg/mL (normal <300).
His electrocardiogram (ECG) reveals normal sinus rhythm, borderline left axis deviation, and minor nonspecific ST-T abnormalities. A chest radiograph shows normal lung fields and pulmonary vascularity and a slightly enlarged cardiac silhouette.
An echocardiographic Doppler study reveals LV ejection fraction (EF) of 0.60 to 0.65, LV end-diastolic volume index 55 mL/m 2 (normal range 35–75), LV mass index 138 g/m 2 (normal range 49–115), left atrial (LA) volume index 39 mL/m 2 (normal <34), E/A ratio 0.9, and average E/e′ 13 (normal <10). An echocardiographic image illustrating this patient’s concentric LVH is shown in Fig. 20.1 .
His 6-minute walk test result is 340 m.
Following this initial evaluation and a discussion of diagnostic and treatment possibilities with the patient, spironolactone 25 mg daily is added to his regimen, and the dose of losartan is increased to 100 mg daily. In addition, a combined exercise-pharmacologic stress test with nuclear imaging is scheduled. Myocardial perfusion is subsequently reported as normal.
The patient returns in 2 weeks. He feels that his dyspnea may be mildly improved. He has not had any medication side effects. Repeat blood work obtained 4 days previous does not reveal any significant change in his renal function or electrolytes. His physical examination is unchanged except that his BP is now 128/80 mmHg. No further medication changes are made. A home exercise conditioning and weight-loss program is recommended to the patient as his insurance plan will not reimburse him for a formal cardiac rehabilitation program.
On return in 2 months, he has succeeded in losing 10 lb and he is taking a 2-mile daily walk, which takes about 1 hour to complete. He feels that his dyspnea is definitely improved, although still a significant problem. His BP is now 124/80 mmHg. On a repeat 6-minute walk test he walks 390 m.
This patient fulfils published diagnostic criteria for HFpEF, as delineated in Chapter 7 . Thus he has dyspnea on exertion in the absence of other medical conditions that can cause dyspnea (with the exception of obesity), concentric LVH with evidence of diastolic dysfunction, and a modestly elevated NT-proBNP. The results of the initial 6-minute walk test confirm that he has poor exercise tolerance. Although not routinely assessed as of yet, most such patients demonstrate abnormal elevations of LV filling pressures during upright bicycle exercise, further confirming the diagnosis of HF.
This patient has the most common comorbidity associated with HFpEF: HTN, which has not been optimally controlled. Like most patients with HFpEF he also has other comorbidities, including obesity, OSA, chronic kidney disease (CKD), and apparent insulin resistance. HTN, especially if poorly controlled, is of course an important stimulus for concentric LVH. However, insulin resistance with associated hyperinsulinemia and most likely OSA are also contributory. Thus the substrates for HFpEF and LVH overlap.
HTN and concentric LVH are associated with myocardial abnormalities that cause diastolic dysfunction, specifically slowed actomyosin relaxation kinetics, altered Na-Ca handling, changes in titin phosphorylation and possibly isoforms, and increased collagen content and crosslinking, which are discussed in more detail later in the chapter. It is not known whether a history of HTN alone without CR or LVH is sufficient to cause any or all of these diverse myocardial abnormalities.
With respect to the management of HFpEF, there is an indication for stress testing in this patient as the incidence of coexistent CAD is high, he has multiple CAD risk factors, and his dyspnea on exertion could represent an angina equivalent.
As discussed here and elsewhere in this book (see Chapter 35 ) no treatment has been clearly shown to favorably affect long-term outcomes in HFpEF. Therefore current guideline recommendations are general measures and include loop diuretics for volume overload, optimal BP control, treatment of comorbidities, and weight loss and exercise. The patient in the case study has no overt evidence of volume overload other than the elevated NT-proBNP. Thus there is no clear indication for a loop diuretic for decongestion, but in light of his CKD and relatively resistant HTN, furosemide is appropriate as a component of BP control. Since his BP is poorly controlled, increasing his losartan dose is also entirely appropriate. Aldosterone blockers have unquestioned benefit in HF with reduced EF (HFrEF), but the situation is less clear in HFpEF. As discussed in more detail later in the chapter, the TOPCAT trial of spironolactone in HFpEF did not meet its primary end points, but there was a signal of possible benefit. Thus, while there is no definite proof of efficacy, there may be a role for spironolactone in patients who fit the TOPCAT entry criteria. Like many patients with HFpEF, he is also receiving a beta blocker, which is typically prescribed for HTN and is one of the drug classes recommended for BP control as part of guideline-directed care of HFpEF. However, as with other treatments there is once again no proof that beta blockers provide long-term benefit. Moreover, we and others have argued that beta blockers are not preferred for the treatment of HTN in general and may actually be harmful in patients with a normal LV EF. Thus a case could be made for discontinuing metoprolol and substituting other BP-lowering drugs if necessary. Exercise conditioning programs (some including an explicit weight-loss component) have been shown in a number of relatively small studies to improve exercise capacity in HFpEF and of course are also beneficial in treating most of the common comorbidities, including HTN. While difficult to achieve, this is an important therapeutic recommendation for this patient.
Epidemiology of Hypertension in HFpEF
The prevalence of a history of HTN in patients with HFpEF has been somewhat variable. An early registry of hospitalized patients reported an incidence of 78%. In contrast, HTN was present in 96% of patients in the Mayo Clinic outpatient registry. In CHARM-Preserved HTN was present in ∼65%, in the PEP trial in 79%, and in SENIORS in 77%. These three early randomized trials had EF inclusion cutoffs below 35% to 40%. In more recent large clinical trials, each of which required a LV EF of 45% or more, the prevalence of HTN has been consistently much higher. Thus in I-PRESERVE it was 88%, in TOPCAT 91%, and in the ongoing PARAGON trial it is 96%. In view of the fact that a modern diagnosis of HFpEF requires a LV EF of 50% or more, it seems likely that the true incidence is more in line with the Mayo registry and the more recent clinical trials (i.e., >90%).
Pathophysiology of Diastolic Dysfunction in HyperTeNsion and HFpEF
Abnormalities of the relaxation and filling properties of the left ventricle are common in patients with asymptomatic HTN and constitute a core component of the pathophysiology in patients with HFpEF. Systemic HTN and other forms of pressure overload (PO) (e.g., aortic stenosis) cause an increase in LV mass/volume ratio, consisting of either an absolute increase in LV mass with a normal end-diastolic volume or an increase in mass/volume ratio without an absolute increase in mass. These two variants are referred to as concentric remodeling.
As discussed, the great majority of patients with HFpEF have a history of HTN. In large clinical trials, greater than 50% have had CR, although in the Mayo Clinic community registry only 42% had CR. A small number of HFpEF patients have eccentric remodeling. While HFpEF patients without CR appear to have abnormal LV diastolic properties, understanding how or even if HTN contributes to diastolic dysfunction and the development of HFpEF in the absence of its signature cardiac end-organ effect is very difficult, and there are no reports focusing on myocardial determinants of diastolic dysfunction in such patients. In addition, while animal models of PO are useful in providing mechanistic insights, they frequently do not simulate human disease very well. This may be a particular problem with rodents because their underlying cardiovascular physiology is so different. Lastly, as noted earlier, despite the high prevalence of HTN in HFpEF patients it is important to keep in mind that other comorbidities may contribute to the development of CR and/or diastolic dysfunction (e.g., hyperinsulinemia and other effects associated with type 2 diabetes mellitus [T2DM]/insulin resistance, obesity, OSA, CAD, and normal aging). It is impossible to know how much these comorbidities, as opposed to HTN per se, influence the development of CR. For these reasons we once again focus our discussion on diastolic function in patients with HTN and CR, understanding the difficulty of isolating the effects of HTN. When possible, we distinguish between patients with and without HFpEF and the effects of HTN versus comorbidities. In addition, we divide LV diastolic properties into the rate and completeness of active relaxation and passive, or fully relaxed, filling properties. A schematic representation of the time course of HTN and common comorbidities and their progression to concentric remodeling in relation to LV diastolic dysfunction and HFpEF is presented in Fig. 20.2 .
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here
