Quality of Life and Antihypertensive Drug Therapy

Joel Handler, MD

In This Article

QOL Studies

A review of the literature demonstrates that there are many poor QOL studies: studies where dropouts due to adverse effects were not counted, studies where only a dozen patients were included, studies where scale validity was in question, and studies promoted by pharmaceutical companies as marketing tools. QOL measures in chronic disease have been applied to Alzheimer's disease, Parkinson's disease, heart failure (HF), human immunodeficiency virus, acute myocardial infarction (MI), asthma, cancer, and hypertension. Patients are interested in prolonging life, but there has been more and more emphasis on health-related QOL.

One of the good scales with regard to health-related QOL measurement of hypertension is the Subjective Symptoms Assessment Profile (SSAP).[2] This was developed in 1990 with data from place-bo-controlled trials. Thirty-nine questions evaluating several different domains were noted. When this scale was applied to patients who were unaware of their BP, including patients with hypertension and individuals who were normotensive, there were significant differences in the performance of the three groups. Included in the scale were symptoms of emotional distress, specifically insomnia, fatigue, lethargy, cardiac symptoms, dizziness, and self-assessment of sexual performance and interest, labeled "sex life." This was a scale that had a high correlation with side-effect profiles and performance with other scales that are believed to be reliable. Thus, the Subjective Symptoms Assessment Profile was believed to have good validity and reliability; there was internal consistency over time and with repeated administration. Looking at the scale, there were three different areas where hypertensive patients appeared to perform less well that normotensives (Figure 1). Emotional distress, cardiac symptoms, and "sex life" were different in hypertensive patients not knowledgeable of their BP and not on antihypertensive therapy.

Results from the Subjective Symptoms Assessment Profile questionnaire showing 95% confidence intervals for differences between "borderline" hypertensives (diastolic blood pressure 85-95 mm Hg) and hypertensives. Significant differences were demonstrated for emotional distress, cardiac symptoms, and sex life. Reprinted with permission from J Clin Res Pharmacoepidemiol. 1990;4:205-217.[2]

Another study looked at normotensive, treated, and untreated patients with hypertension who were referred to a London hypertension clinic ( Table 1 ).[3] Hypertensive patients performed less well in these scales with regard to unsteadiness, nocturia, waking headache, and blurred vision, on or off therapy, when compared with normotensives. Using the Nottingham Scale, which is another good QOL scale, there also appeared to be significant differences between normotensive and hypertensive individuals (Figure 2).[4] There was poor performance with the hypertensive patients who did not know their BP with regard to emotions and sleep disturbance.

Nottingham Health Profile summary of scores for quality of life domains out of a maximum of 100 for three study groups. *Statistical significance compared to normotensives. Reprinted with permission from J Hypertens. 1989;7:885-890.[4]

The appraisal scales just discussed represent hypothesis-generating material. These data suggest that patients not knowledgeable about their BP and hypertensive status perform less well than normotensive patients; QOL was impaired. The question then becomes, "Do trials using drugs with acceptable tolerance profiles improve the QOL of patients with hypertension?" A few studies have examined this relationship.

The Systolic Hypertension in the Elderly Program (SHEP) study[5] evaluated QOL responses at baseline and at the study end by treatment group (Figure 3). Patients on active therapy compared with placebo did at least as well and perhaps better in some QOL measures. With aging (these patients were 60 years of age and older), there was some decline in ability to perform more strenuous leisure activities, as one would expect, but with drug treatment for hypertension, measures of mood and cognition were stable, and some of the nearly 5000 patients described improvements in QOL on drug therapy.

Systolic Hypertension in the Elderly Program (SHEP) results revealing quality of life responses to thiazide-based drug treatment and placebo at baseline and end of study using three questions. Reprinted with permission from Arch Intern Med. 1994;154:2154-2160.[5]

In the Treatment of Mild Hypertension Study (TOMHS),[6] there were five active drug arms in addition to placebo. This was an excellent QOL study. Seven different domains were examined -- gen-eral health, energy or fatigue, mental health, general functioning, satisfaction with physical ability, social functioning, and social contacts. Patients ranked the individual QOL domains in terms of importance to them; this weighting of the scales based on ranking led to a "global statistic." Patients on drug treatment compared with placebo performed significantly better in the global statistic ( p =0.007). Both the placebo and medication groups were also on an active weight reduction and exercise program. Those in the placebo group who lost weight and maintained an exercise program reported an improved QOL compared with participants in the placebo arm who were unable to achieve weight reduction or exercise maintenance. Nonetheless, when the active lifestyle placebo patients were compared with the drug treatment arm, the drug treatment patients felt significantly better.

In the TOMHS blinded trial, more than 900 patients in five different treatment arms were randomized against placebo.[6] Each of the five drug treatment arms was compared with placebo with regard to QOL performance. A calcium channel blocker (CCB) (amlodipine), a diuretic (chlorthalidone), an angiotensin-converting enzyme (ACE) inhibitor (enalapril), a cardioselective ß blocker (acebutolol), and a peripheral α adrenergic blocker (doxazosin) were evaluated. There were only two drugs that significantly outperformed placebo in terms of QOL measures: the diuretic and the ß blocker. Significant QOL improvement occurred throughout the study, with surveys at 3 months, 12 months, and 48 months. These data suggest that hypertension may not be a truly asymptomatic disease and that drug treatment, particularly with well tolerated agents, can actually improve QOL.

In the Hypertension Optimal Treatment (HOT) study,[7] there were several different diastolic pressure targets. Patients were divided into target groups of diastolic BP >90, 86-90, 81-85, and <80 mm Hg. As the BP was lowered, there was a proportional improvement in QOL, as measured by the Psychosocial General Well Being Index scale (Figure 4). This was particularly true for measures having to do with cardiac symptoms and headache.

The Hypertension Optimal Treatment (HOT) study results. PP=per protocol; ITT=intention to treat; DBP=diastolic blood pressure (mm Hg). Reprinted with permission from Blood Press. 1997;6:357-364.[7]

We conclude from these observations that hypertension may not always be an asymptomatic disease. Scales show that patients with hypertension may not feel quite as well as patients with normotension. In several drug treatment trials, patients felt better when their BP was lowered, despite being treated with medication. Yet it is not a cut and dried, simple concept. In fact, when you try to delineate symptoms that are specific for hypertension, it is very difficult to do so.

The National Health and Hypertension Nutrition Evaluation Survey[8] divided more than 6000 patients into different systolic BP ranges: <140, 140-159, and >160 mm Hg. Specific symptoms were evaluated -- headache, epistaxis, tinnitus, dizziness, and fainting. There was no difference in these symptoms related to BP ( Table 2 ).

An interesting study in a London clinic evaluated how patients felt when they were certain that their BP was elevated. Fifty different "predictors" generally failed to accurately predict BP; however, measures that were more commonly reported by predictors, such as headaches, dizziness, lightheadedness, feeling hot all over, and blurred vision were more typical of underlying anxiety-related symptomatology.[9]

In some patients with an underlying anxiety or depressive disorder, it will be difficult to keep them on medication because of numerous perceived symptoms that they attribute to medication and that may instead be secondary to an anxiety state. In such individuals, medication compliance is best achieved by making a diagnosis and treating the underlying anxiety or depression problem, rather that changing antihypertensive drugs.[10]