Abstract and Introduction
Doxazosin remains a commonly used antihypertensive medication, although its use has been tainted by recent findings from the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). ALLHAT was a large, simple trial, designed in a fashion to closely mimic clinical practice as it occurs in high-risk hypertensive patients aged 55 years or older. Its goals were to determine whether the incidence of the primary outcome -- a composite of fatal coronary heart disease and nonfatal myocardial infarction -- differed between treatment with a diuretic (chlorthalidone) (12.5-25.0 mg/day) and treatment with each of three other types of antihypertensive drugs -- a calcium-channel blocker (amlodipine), an angiotensin-converting enzyme inhibitor (lisinopril), and a peripheral a-adrenergic blocker (doxazosin) (2-8 mg/day). Doxazosin was recently withdrawn from this trial after an interim analysis showed the secondary end point of combined cardiovascular disease to be 25% greater in patients on doxazosin than in those assigned to treatment with chlorthalidone. This finding was largely driven by congestive heart failure. The practicing clinician should not abandon doxazosin completely because of the ALLHAT findings, although these findings are indisputably important. These results represent an interim analysis and their application to clinical practice needs to occur carefully. A valued member of our therapeutic armamentarium need not be laid entirely to rest; rather, doxazosin should now be viewed as a secondary or tertiary antihypertensive therapy pending a more complete review of the ALLHAT data.
Doxazosin mesylate (Cardura) is one of several peripheral selective a1-adrenergic antagonists currently marketed in the United States for the treatment of hypertension.[1,2] The other currently marketed peripheral a-adrenergic antagonists, with proven utility in the management of hypertension, include prazosin (Minipress)[3,4] and terazosin (Hytrin).[5] Recently, peripheral a-adrenergic antagonists have emerged as effective treatment of the symptoms associated with benign prostatic hypertrophy.[6] The observation that the prostate contains a plentiful supply of a1-receptors has proven the basis for their therapeutic use in prostatic disease. As might be expected; this aspect of a-adrenergic antagonism has favorably impacted their use, particularly in the older hypertensive male.[7]
Doxazosin is structurally related to prazosin, the first peripheral a1-adrenergic antagonist, and not unexpectedly, when initially released it was considered as the first of a new generation of a1-adrenergic antagonists.[2,8] Doxazosin's mechanism of action is similar to that of other a1-adrenergic antagonists.[8] Its difference from other a1-adrenergic antagonists resides in its pharmacokinetics wherein once-daily dosing can occur irrespective of the time of day of dosing.[9] In addition, based on both its mechanism of action and timing of administration, doxazosin may reverse nocturnal nondipping blood pressure (BP) patterns[10] as well as blunt the morning surge in BP[11] that characterizes the early morning period of increased vulnerability to cardiovascular and cerebrovascular events.[12]
During the past decade, doxazosin has proved useful in the treatment of hypertension. It is particularly useful as add-on therapy in the resistant hypertensive.[13] Use of a-adrenergic antagonists as monotherapy in the treatment of hypertension has been limited by their tendency to increase plasma volume, a phenomenon that may be more evident at higher doses.[14] Despite a lengthy track record in the treatment of hypertension and many clinical studies that suggest a theoretical benefit in cardiovascular risk, no outcome data is available with a1-adrenergic antagonists to formally support their being used as first-step therapy in the general hypertensive population.[15] Thus, the findings from the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) were eagerly awaited because they would finally provide one large repository of information, which addressed the risk-benefit ratio of doxazosin in comparison to the accepted gold standard -- a thiazide-type diuretic.[16]
ALLHAT studied high-risk hypertensive patients aged 55 years or older. It was a large, simple trial, designed in a fashion to closely mimic clinical practice as it occurs in high-risk patients. Its goals were to determine whether the incidence of the primary outcome -- a composite of fatal coronary heart disease and nonfatal myocardial infarction differed between treatment with a diuretic (chlorthalidone) (12.5-25.0 mg/day) and treatment with each of three other types of antihypertensive drugs -- a calcium channel blocker (amlodipine), an angiotensin-converting enzyme inhibitor (lisinopril), and an a-adrenergic blocker (doxazosin) (2-8 mg/day). In this study, doxazosin titration occurred on a monthly basis. Secondary outcomes included all-cause mortality, stroke, and all major cardiovascular disease events. If patients did not meet the BP goal with the maximum tolerated dose of the initial medication, an open-label step 2 agent (atenolol, 25-100 mg/day; reserpine, 0.05-0.2 mg/day; or clonidine, 0.1-0.3 mg twice daily) or an open-label step 3 agent (hydralazine, 25-100 mg twice daily) could be added. In addition, a number of ALLHAT participants (n=10,337) participated in a randomized, open-label trial designed to determine whether lowering low-density lipoprotein cholesterol with the HMG-CoA-reductase inhibitor pravastatin reduced all-cause mortality compared to a control group receiving usual care. Patient enrollment in ALLHAT began in February, 1994 and active follow-up concluded in early 2002, with final study results to be reported in December 2002.[16]
These data from ALLHAT would have allowed doxazosin to have been directly compared, for the first time, to the gold standard for hypertensive therapy from an evidence-based medicine perspective -- that is, a thiazide-type diuretic, which in ALLHAT was chlorthalidone.[17,18] There was considerable rationale for the inclusion of a peripheral a-adrenergic antagonist, such as doxazosin, as one of the treatment limbs of ALLHAT. Doxazosin, as well as other a-adrenergic antagonists, had previously been demonstrated to reduce BP in a manner comparable to that seen with hydrochlorothiazide.[1,19] In addition, these drugs had been well documented to favorably modify the insulin resistance and the hyperlipidemia features of the dysmetabolic syndrome,[20,21,22,23] as well as to have beneficial effects on fibrinolysis,[24] red blood cell deformability,[25] endothelial function,[26,27] as well as to inhibit the proliferation and migration of human vascular smooth muscle cells.[28] A number of these findings emerged even as ALLHAT was in progress and simply supported prior suppositions. Thus, it was anticipated that doxazosin would confer additional cardiovascular benefit beyond its ability to reduce BP as a consequence of its favorably effecting metabolic risk factors known to increase the risk of coronary artery disease.[29]
Thus, there was considerable surprise and maybe more so, disappointment, when the National Heart, Lung, and Blood Institute (NHLBI) announced that doxazosin was being withdrawn from the trial after an interim analysis of ALLHAT showed a 25% greater rate of a secondary end point, combined cardiovascular disease, in patients on doxazosin than in those on chlorthalidone, largely driven by the occurrence of congestive heart failure (CHF).[30] There was no difference in the primary end point, fatal coronary heart disease, or nonfatal myocardial infarction, at the time of the interim analysis; moreover, calculations indicated that the difference in CHF rate was unlikely to change in a meaningful enough way by the end of the trial to ethically justify continuation of this study limb. Following independent data reviews on January 6 and 21, 2000, the director of the NHLBI accepted a recommendation to discontinue the doxazosin treatment arm in the BP component of the trial. It was determined that participants assigned to the doxazosin group should be informed of their BP treatment assignment and that the major clinical findings regarding this treatment and its comparison agent, chlorthalidone, should be reported as soon as possible. The ALLHAT Data and Safety Monitoring Board specifically stressed the importance of continuing the remainder of the BP and lipid-lowering trial.
The decision to discontinue the doxazosin arm of the antihypertensive trial was made after thorough deliberation and was reached only after several factors had been considered (Table Table 1 and Table 2 ). An important, and as of yet unresolved, aspect of ALLHAT is the precision of the criteria employed for the definition of CHF, particularly since CHF episodes were not centrally adjudicated. This is of particular concern since peripheral edema was one of the several criteria employed to reach the diagnosis of CHF. Peripheral edema is a not too uncommon finding with direct vasodilators and in most cases of treatment with these drugs does not reflect developed CHF.
The mechanism behind these findings remains enigmatic, particularly in light of the fact that doxazosin had previously been shown to be an effective therapy in the treatment of CHF.[31] It became important to verify these findings since ALLHAT was a large, simple trial and did not have routine central adjudication of major end points though a 10% sample of all coronary events and stroke were reviewed centrally. Heart failure episodes were not originally reviewed centrally, but a sample was selected when these treatment differences began to emerge and reviewed blindly by the Trial Endpoints Committee based on the hospital discharge data of these events. This review showed that the investigators were for the most part using the preset criteria for CHF in their diagnosis; alternatively, there was no documentation of how the diagnosis of heart failure was reached in nonhospitalized cases.[32] These reviews were quality control reviews only and were not used to change the diagnosis if there was disagreement with the diagnosis assigned by the treating physician. Moreover, in evaluating the ejection fraction values in discharge summaries the coordinating center for ALLHAT found that two thirds of these patients had a value below 40%.
The authors of the original publication speculated on several possible mechanisms for this inordinately high rate of CHF in the study population. For example, a blockers increase plasma volume, may increase plasma norepinephrine levels, and when left ventricular (LV) hypertrophy exists, they do not regress LV mass as well as diuretics. In addition, there were BP differences between the two treatment limbs favoring the chlorthalidone treatment group, although information on the mean daily dose of doxazosin was not given in the original report. Whether these physiologic responses to a blocker therapy contributed to the increased rate of CHF cannot be determined from ALLHAT.
Each of these possible mechanisms requires some comment. Alpha blockers, if given in high enough doses, will limit sodium excretion in a clinically relevant fashion, whereas diuretics act in an opposite fashion.[33] ALLHAT could not determine whether the doxazosin treatment group was excessively volume expanded and thereby prone to the development of CHF. The maximum doxazosin dose was 8 mg in ALLHAT, a dose not typically associated with excessive fluid retention and/or edema.[34] Therefore, alterations in salt-and-water handling would seem an unlikely explanation for these findings. Alpha blockers have been suggested to stimulate both the renin-angiotensin and sympathetic nervous system axes, but this occurs in a compound and situation-specific fashion,[34,35] although, long-term treatment with a blockers appears not to alter activity in the renin-angiotensin axis. The minimal effect of these compounds on the renin-angiotensin axis has proved useful in dynamic testing of this axis when BP control is necessary.
Sympathetic activation may be observed with a blockers.[36] This phenomenon is probably dependent on the time course of a blockade. For example, prazosin administration, which is followed by a fairly rapid onset of action, clearly activates the sympathetic nervous system.[37] Doxazosin has not been studied in a similar fashion. Although not studied in a formal manner, its slower onset of action would suggest it to be less prone to activation of the sympathetic nervous system. To this end, long-term studies with doxazosin have not shown any significant alteration in pulse rate. Most studies with a blockers have not directly measured sympathetic activity and instead employed pulse rate as a surrogate for sympathetic activation.[34,38] In addition, the pressor action of arginine vasopressin is maximized after a1-adrenergic blockade[39] and it has been observed that under certain conditions arginine vasopressin can become an important vasopressor factor in patients with hypertension[40] and/or CHF.[41] Experimental evidence suggests that, in the presence of functional baroreflexes, the small elevation in arginine vasopressin following a1-adrenergic blockade[39] produces an increase in systemic resistance with a strong reflex suppression in cardiac output.[42] How exactly and even whether the potential for these neurohumoral changes relates to the findings in ALLHAT cannot be determined at this time.
Also, LV hypertrophy can be viewed as a precursor condition to CHF, being preceded in most cases by hypertension. Regression of LV mass with antihypertensive therapy is commonly sought and is believed, though not proven, to slow the progression to systolic CHF. Alpha blockers are known to regress LV mass.[43,44] When studied in comparison to other antihypertensive medication classes, a blockers have not been as effective in reducing LV mass. [43,44] For example, in the Treatment of Mild Hypertension Study (TOMHS), LV mass was reduced in the chlorthalidone treatment arm compared with placebo, while the change in the doxazosin treatment arm could not be distinguished from placebo.[43] Finally, a1-adrenergic blockade blocks heat shock protein expression in the myocardium in response to norepinephrine stimulation, thus leaving the heart vulnerable to injury. This process is one wherein heat shock proteins act as cellular chaperones to proteins, lipids, and nucleic acids by reducing oxidant stress, preventing apoptosis and suppressing proinflammatory cytokines.[45] This concept remains one of several mechanistic hypotheses that cannot be retrospectively tested in the cohort of ALLHAT subjects having developed CHF while on doxazosin.
Does the manner in which doxazosin influences volume status, neurohumoral activation, BP control, and regression of LV mass provide a logical explanation for the findings of ALLHAT? If each of these is viewed as separate variables then there is no one factor that can adequately explain the observed findings. Alternatively, if these physiologic potentials are viewed as interdependent variables then the answer probably lies therein. What remains to be more carefully elucidated is the exact interplay of these variables, which may, in turn, importantly relate to the elusive issue of patient susceptibility to the development of heart failure. To date, the patient subset analysis of ALLHAT has not identified a specific patient susceptibility to these events since the observed findings were evenly distributed across all patient groups.
When the Kaplan-Meier plots for CHF are examined, it is apparent that the divergence in CHF development appeared very early on, within the first few months of therapy, and that the greatest separation in event rate had occurred by the end of the first year. This raises the question of whether discontinuation of prior therapy might have played a role. Ninety percent of patients were on antihypertensive therapy prior to entry into the trial; these were stopped, and study medication substituted with no interim washout period. Although no information on prior therapy is given, it is reasonable to presume that a number of patients would have been on either a diuretic or an angiotensin-converting enzyme inhibitor. Such therapy, although being prescribed for hypertension, may actually have been treating early-stage CHF. Cessation of such therapy, followed by substitution therapy with doxazosin, may have then allowed heart failure symptoms to become manifest. Conversely, in the chlorthalidone group, such symptoms may have remained latent under the influence of diuretic therapy. The utility of diuretic therapy in preventing heart failure is now well accepted and it is possible in this trial that doxazosin was not a drug that caused heart failure but rather one that simply did not treat it.
The fact that CHF is a serious condition, not uncommonly fatal within a year or two of onset, also raises the question of why there was no difference in the death rate between the two treatment groups despite a very clear difference in the rate of heart failure development. It should be appreciated that longer follow-up may reveal differences in death rates between treatment limbs, which were not apparent with this interval analysis of the data set. Also, although the authors state that the result for hospitalized CHF was similar to that for all CHF, they do not state what percentage of the CHF was hospitalized -- and presumably more severe.
