Analysis of Recent Papers in Hypertension Jan Basile, MD, Senior Editor

Cardiovascular (CV) trials in subjects with hypertension and coronary artery disease (CAD) have not been able to prove an advantage of any one specific antihypertensive agent and have not tested specific blood pressure (BP) levels and CV outcome. Angiotensin-converting enzyme (ACE) inhibitor therapy is currently recommended for individuals with both symptomatic and asymptomatic CAD, and calcium channel blocker (CCB) therapy is recommended to decrease symptoms of angina in the same patient population. In an effort to determine the effect of different antihypertensive drug treatments on CV events in patients with stable CAD and relatively normal BP, the Comparison of Amlodipine vs. Enalapril to Limit Occurrences of Thrombosis (CAMELOT) trial was conducted. This multicenter, double-blind, placebocontrolled, randomized clinical trial supported by the makers of amlodipine, was conducted at 100 study sites in the United States, Canada, and Europe. Study participants, 89% of whom were white, included men (75%) and women, aged 30–79 years (mean age, 58), with established CAD and a diastolic BP <100 mm Hg with or without treatment. To qualify, they required either coronary angiography for evaluation of chest pain or percutaneous coronary intervention. CAD was defined on angiography as ≥20% stenosis in one or more native coronary arteries. Patients with left main coronary artery obstruction >50%, left ventricular ejection fraction <40%, or symptoms of moderate to severe heart failure were excluded. A substudy, performed simultaneously in 274 patients, tracked atherosclerosis progression using intravascular ultrasound (IVUS). IVUS patients required a luminal narrowing of not more than 50% throughout a study vessel that had not been subject to angioplasty and was at least 30 mm in length. At baseline, 18% were diabetic, 28% were smokers, 38% had a previous myocardial infarction (MI), and 60% had hypertension. Approximately one third of the subjects had either one-, two-, or three-vessel CAD, respectively; 30% had undergone percutaneous intervention, and 7% had previous coronary artery bypass graft surgery. In terms of overall risk reduction, patients were aggressively treated, with 95% receiving aspirin, 83% on a statin (mean low-density lipoprotein cholesterol of 102 mg/dL), three fourths on β-blocker therapy, and one third on a diuretic. Less than 10% of subjects were on ACE inhibitor, angiotensin receptor blocker, or CCB therapy at entry, which, over a washout period, were discontinued before study initiation. All subjects participated in a 2-week placebo run-in period. A total of 1997 patients demonstrating 80% compliance during the runin period were then randomized to one of three treatment groups: amlodipine 5 mg once a day, enalapril 10 mg once a day, or matching placebo using a double-dummy design. After 2 weeks of therapy, if tolerated, subjects had the dose of study medication doubled. Crossover rates were low, with 7.4% of amlodipine patients receiving an ACE inhibitor and 1.7% receiving an angiotensin receptor blocker, while 6.1% of enalapril patients received a CCB. More patients in the placebo group received a CCB, ACE inhibitor, or angiotensin receptor blocker, although the exact percentages were not detailed. The study's prespecified composite primary end point was time to first occurrence of an adverse cardiovascular event defined as either CV death, nonfatal MI, resuscitated cardiac arrest, coronary revascularization (percutaneous or surgical), hospitalization for angina pectoris or heart failure, fatal or nonfatal stroke or transient ischemic attack, and any new diagnosis of peripheral vascular disease. The primary outcome was the incidence of adverse cardiovascular events in patients treated with amlodipine compared with placebo. Secondary outcomes included the incidence of adverse events for enalapril treatment compared with placebo and comparison of the amlodipine treatment group compared with the enalapril group. Additional prespecified secondary end points included all-cause mortality and the incidence of revascularization in vessels that had undergone previous stent placement. The end point of the IVUS substudy was the nominal change in percent atheroma volume in all sections of the targeted coronary artery. During the 5-year study period, 1856 subjects completed the protocol (655 placebo, 673 enalapril, and 663 amlodipine). The IVUS substudy was completed by 274 participants (95 placebo, 88 enalapril, and 91 amlodipine). Average sitting BP at baseline was 129/78 mm Hg. Mean BP increased 0.7/0.6 mm Hg in the placebo group and was significantly reduced equally in the enalapril group (4.9/2.4 mm Hg) and amlodipine group (4.8/2.5 mm Hg), both compared with placebo. The primary end point occurred in 23.1% of subjects in the placebo (n=151) group, who had been maintained on previous therapy, 16.6% in the amlodipine (plus other therapy) group (n=110), and 20.3% of patients in the enalapril (plus other therapy) group (n=136). Compared with the placebo group, there was a significant reduction in CV events in the amlodipine group (hazard ratio [HR], 0.69; 95% confidence interval [CI], 0.54–0.88, p=0.003). The most frequent component of the primary end point, coronary revascularization and hospitalization for angina, were both significantly reduced with amlodipine compared with placebo. As compared with placebo, enalapril was associated with a nonsignificant reduction in CV events (HR, 0.85 [95% CI, 0.67–1.07, p=0.16). When compared with enalapril, amlodipine was associated with a nonsignificant reduction in CV events (HR, 0.81; 95% CI, 0.63–1.04, p=0.10); however, there was a statistically significant decrease on amlodipine in the incidence of hospitalization for angina and a trend toward less coronary revascularization in patients who had undergone intervention at baseline. Additional post hoc analysis revealed there were no significant differences in nonfatal MI, nonfatal stoke, or all-cause mortality between the two medication groups. The IVUS substudy showed no significant differences. Mean percent atheroma volume changes were 0.5% for amlodipine, 0.8% for enalapril, and 1.3% for placebo. Paired analysis, comparing changes from baseline in each of the treated groups, showed progression on placebo, a nonsignificant trend for progression on enalapril, and a trend toward absence of progression for amlodipine. In the prespecified subgroup comparison of participants with systolic BP greater than the mean, the amlodipine group showed significantly slower progression compared with placebo with no treatment effects noted in the subgroup with baseline systolic BP below the mean. Discontinuation for treatment related adverse events over the 2 years of treatment were few and similar in all three groups, with a trend toward increased peripheral edema with amlodipine and cough and hypotension with enalapril. In patients with stable CAD and well controlled BP, the addition of the CCB amlodipine reduces the risk of composite CV events compared with placebo. A smaller and nonsignificant effect was observed by the addition of the ACE inhibitor enalapril when compared with placebo. The difference in results between the enalapril and amlodipine groups was mostly due to differences in the incidence of hospitalization for angina and revascularization. In the IVUS substudy, for patients with systolic BP greater than the mean, amlodipine treatment significantly slowed atherosclerosis progression but there were no overall significant changes between groups.—Nissen SE, Tuzcu EM, Libby P, et al. Effect of antihypertensive agents on cardiovascular events in patients with coronary disease and normal blood pressure: the CAMELOT study: a randomized controlled trial. JAMA. 2004;292:2217–2225. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) recommends a BP goal of <140/90 mm Hg for patients with uncomplicated hypertension and <130/80 mm Hg for those with diabetes or chronic kidney disease. Observational trials have suggested the optimal BP level for subjects with CAD may be substantially lower than the current recommendation of <140/90 mm Hg, but since there have been no outcome-based trials evaluating the effect of antihypertensive therapy to a specific BP level in these individuals, no specific recommendation has been issued by the JNC committee. Studies evaluating antihypertensive agents in patients with CAD have been conducted but have not exclusively enrolled hypertensive subjects and have not targeted specific BP goals. Both the Heart Outcomes Prevention Evaluation (HOPE) and the European Trial on Reduction of Cardiac Events with Perindopril in Stable Coronary Artery Disease (EUROPA) demonstrated a decrease in CV events attributed to the beneficial effects of ACE inhibitor therapy compared with placebo. In contrast, the Prevention of Events with Angiotensin-Converting Enzyme Inhibition (PEACE) study found in patients with stable CAD and well controlled BP (mean 133/78 mm Hg) who were receiving therapies for comprehensive risk reduction (including antiplatelet, β blocker, and lipid-lowering therapy) that additional ACE inhibitor therapy with perindopril did not improve outcome more than placebo. The results of these three trials suggested that it is the degree of risk of the patient and the aggressiveness of management of other risk factors in individuals with stable CAD and normal left ventricular function that determines the additive benefit of ACE inhibitor therapy. INVEST (International Verapamil Slow Release/ Trandolapril Study) involved 22,576 patients with hypertension and CAD. Subjects were randomized to either a CCB-based strategy, which included sustained-release verapamil plus the ACE inhibitor trandolapril and the diuretic hydrochlorothiazide, or a β blocker/diuretic-based strategy, which included atenolol, hydrochlorothiazide, and trandolapril. There was no difference in BP control (>70% of patients with BP <140/90 mm Hg) or CV outcome between the verapamil-based compared with the atenolol-based treatment strategy, although the CCB-based strategy was more effective in reducing the frequency of angina. The main message from this study was that different antihypertensive agents that provide similar and effective BP reduction produce comparable protection against cardiovascular events. Like INVEST, CAMELOT involved subjects, all of whom had underlying CAD. It compared the ACE inhibitor enalapril, the dihydropyridine CCB amlodipine, and placebo, but only 60% of the subjects entered the study with a diagnosis of hypertension (100% of INVEST patients had hypertension). Accordingly, CAMELOT is a trial that involves antihypertensive agents that may have beneficial cardiovascular properties including BP reduction but CAMELOT is not a hypertension trial. As the trial did not randomize subjects to a particular BP goal but observed the reduction in BP that occurred with each randomized treatment, no specific BP target for benefit can be determined from CAMELOT. Finally, the trial enrolled participants based on diastolic BP <100 mm Hg, thereby limiting the ability to target the most appropriate systolic BP goal in those with CAD. There was an equivalent but significant 5/3 mm Hg reduction in BP in both the amlodipine and enalapril subjects at the end of the 2-year treatment. There was a significant reduction in CV events in the amlodipine group with BP reduced to 124/75 mm Hg, compared with placebo and a nonsignificant reduction in the enalapril group compared with placebo. It should be noted that the main end point of the composite end points accounting for this difference between amlodipine and placebo was for coronary revascularization rates and hospitalization for angina. There was no difference in the rate of “hard” CV end points such as nonfatal stroke, nonfatal MI, or all-cause mortality between the two active drugs. As in INVEST, which evaluated the CCB verapamil, the antianginal properties of amlodipine may have accounted for the majority of this benefit. Differences in actual as opposed to observed BP reduction may have influenced event rates as well. BP measurements were taken at baseline, 1 month, 3 months, and every 3 months thereafter for the total 2-year duration of the trial. The reported causal BP reductions with amlodipine and enalapril were reported to be similar, although details are not given regarding the timing and methodology of BP measurement. While 24-hour ambulatory measurements were not taken, other clinical trials, including HOPE, suggest that small casual BP readings as measured in CAMELOT may not adequately reflect the large BP differences that actually occur when 24-hour ambulatory blood pressure monitoring (ABPM) is used. This may be true given the fact that enalapril has a relatively short half-life (approximately 11 hours) compared with amlodipine (approximately 50 hours) and both were given once a day. It is possible that once-daily enalapril does not provide adequate blockade of the renin-angiotensin-aldosterone system over a 24-hour period in subjects with CAD; previous heart failure trials in those with systolic dysfunction showed a clinical benefit with enalapril when used twice daily. The current study suggests there is no particular risk to lowering BP to 124/75 mm Hg in patients with CAD. It does not, however, define the optimal BP level necessary to improve outcome. In addition, it is unclear whether the background therapy or doses and frequency of medication given in the study may at least partially account for the effects seen. Well designed randomized clinical trials that evaluate specific BP levels using specific antihypertensive agents are necessary before we can evaluate the benefits of a given achieved BP level or a particular therapeutic strategy. Defining the benefits of a particular IVUS strategy only in people above a certain BP in subgroup analysis, even if pre-specified, does not prove cause and effect. The results of CAMELOT, the Antihypertensive and Lipid lowering Treatment to Prevent Heart Attack Trial (ALLHAT), INVEST, and the Controlled Onset Verapamil Investigation of Cardiovascular Endpoints (CONVINCE) study prove the safety and effectiveness of both long acting dihydropyridine and nondihydropyridine CCB therapy in high-risk patients, especially those with CAD and hypertension. BP should be reduced below 140/90 mm Hg in subjects with CAD and further reduction to the levels achieved in CAMELOT appear to be without risk. A specific BP target or a compelling indication for a specific antihypertensive agent in hypertensive and CAD patients cannot be recommended based on the results of CAMELOT. While hypothesis generating, only future well designed clinical trials will change the landscape of JNC 8. Both computerized tomographic angiography (CTA) and magnetic resonance angiography (MRA) are gaining widespread acceptance as screening tests for renal artery stenosis (RAS). When compared with conventional angiography, the reported accuracy of these studies has generally been high, but to date the available studies have been relatively small and perhaps subject to bias. Using contemporary imaging protocols, the Renal Artery Diagnostic Imaging Study in Hypertension (RADISH) Study Group undertook a prospective comparison study at three large hospitals in the Netherlands to determine the accuracy of MRA and CTA compared with the reference standard, digital subtraction angiography (DSA). All patients with hypertension between the ages of 18 and 75 with a diastolic blood pressure >95 mm Hg were routinely screened at participating internal medicine outpatient clinics for clinical clues suggesting possible renovascular hypertension (RVH). Patients were eligible if they exhibited at least one clinical clue suggestive of RVH and had no contraindications to the diagnostic tests or possible intervention if RAS was confirmed. Included patients were scheduled to undergo all three diagnostic tests: MRA, CTA, and DSA. Each center was equipped with state-of-the art imaging equipment, imaging protocols, and experienced radiologists and technicians. Multiple radiologists who were blinded to any clinical information or other test results examined each image. For each patient, the observers recorded the number of renal arteries, the percentage of luminal narrowing, the nature of the stenosis (atherosclerosis or fibromuscular dysplasia [FMD]), the location of the stenosis, and the level of confidence in the diagnosis. The severity of stenosis on CTA and MRA was categorized on a 5-point scale (grade 1, 0%–19%; grade 2,20%–49%, grade 3, 50%–74%; grade 4, 75%–99%; and grade 5, total occlusion). The most severe stenosis for each patient was used for statistical analysis. Overall sensitivity, specificity, and positive and negative predictive values were calculated for each test. Four hundred two patients met criteria for inclusion in the study, and 356 underwent all three studies and were included in the analysis. The mean age of those patients included in the analysis was 52 years and the mean BP was 183±25/107±15 mm Hg. Review of the DSA images showed that 20% had some degree of RAS; 63% of those were judged to be atherosclerotic; and 36% FMD. Seven percent of all patients had grade 4 or 5 stenosis (>75% luminal obstruction) on DSA. As compared with DSA, the overall sensitivity was 64% (95% CI, 55%–73%), specificity was 92% (CI, 90%–95%), positive predictive value was 68% (CI, 59%–77%) for CTA, and negative predictive value was 91% (CI, 88%–94%). As compared with DSA, the overall sensitivity of MRA was 62% (CI, 54%–71%), specificity was 84% (CI, 81%–87%), positive predictive value was 49% (CI, 40%–58%), and negative predictive value was 90% (CI, 87%–93%). In subgroup analyses, both CTA and MRA had lower sensitivity but higher specificity for patients with FMD; the accuracy of both tests increased marginally in the subgroup of patients older than 60 years of age. Additionally, there was often a significant rate of disagreement between these experienced radiologists as to the presence and degree of RAS using both CTA and MRA. The authors conclude that neither CTA nor MRA is reproducible or sensitive enough to rule out RAS in patients with hypertension.—Vasbinder GBC, Nelemans PJ, Kessels AGH, et al. Accuracy of computed tomographic angiography and magnetic resonance angiography for diagnosing renal artery stenosis. Ann Intern Med. 2004;141:674–682. Multiple small, well-designed clinical trials have demonstrated that both MRA and CTA have sensitivity and specificity of over 90% for the detection of angiographically significant RAS. A recent meta-analysis concluded that MRA and CTA have significantly greater accuracy than other commonly used noninvasive imaging studies. Others have questioned these same studies as they have included highly selective patient populations and have often excluded patients who might have FMD. In addition, a significant publication bias has been suggested where centers with positive data are more likely to have their data accepted for publication than centers with less positive results. The RADISH investigators try to provide us with a more real-world assessment of the accuracy of these diagnostic tests. However, like all of the literature surrounding this vexing clinical problem, there are significant limitations to the present study. This study used very broad inclusion criteria, only 7% of subjects had evidence of RAS >75% that would be expected to result in clinically significant RVH. This pretest probability is likely lower than what is generally seen in the population undergoing screening for RVH in the United States. If the investigators had narrowed their study population to a group with a higher prevalence of disease, even if the reported sensitivities and specificities were unchanged, the positive predictive values could have been significantly higher. Furthermore, up to one third of subjects found to have RAS on DSA had FMD, which is higher than the usual percentage seen. Both MRA and CTA have lower accuracy in the setting of distal and branch stenoses which is often seen with FMD. When the present analysis was limited to patients with atherosclerosis or age greater than 60 years, both sensitivity and specificity did improve, although still not to the level reported in previous publications. Finally, since DSA does not allow for lesions to be examined from all directions and provides no functional information, it has been called into question as an appropriate reference standard for renal artery imaging. The most important question to ask of an imaging test is does it predict clinical improvement with intervention, not whether it provides the same anatomic detail as DSA. Despite these significant limitations, the inter-observer variability in determining the degree of stenosis with both MRA and CTA is troubling and calls into question the widespread utilization of these tests. How should one then use these data clinically? In making the diagnosis of RVH, the most important first step is clinical judgment. Limiting work-up to a subgroup of hypertensive patients with a high clinical index of suspicion (pretest probability) can have a dramatic effect on positive predictive values (systolic/diastolic bruit, resistant hypertension, new onset hypertension, etc.). Second, since FMD and atherosclerotic disease are very different disorders, and tend to present in different patient populations, they require different diagnostic strategies. In patients with a high pretest probability of RAS due to FMD, it may be reasonable to proceed directly to DSA since these patients are generally at low risk of complications and no noninvasive study offers comparable accuracy. For those patients with suspected FMD, but a more moderate clinical suspicion, it may be more reasonable to choose a “functional” test like angiotensin-converting enzyme inhibitoraugmented renography. Given the low sensitivity reported here and in other studies for detecting FMD by CTA or MRA, these tests should generally not be used in these patients. In patients with suspected atherosclerosis, DSA should once again be considered for those patients with the highest likelihood of clinically significant RAS; however, that enthusiasm should be tempered somewhat by remembering that the risk of complications with DSA is higher in the atherosclerotic population. In patients with suspected atherosclerotic RAS and a more moderate pretest probability of disease, it may be reasonable to consider using CTA and MRA (or duplex Doppler) as the next step, but with the knowledge that false negatives and false positives remain more likely. A completely normal study likely obviates the need for further work-up in most, but probably not all of these patients, and a positive study certainly does not come with any assurances that the lesion will be as prominent on DSA, or more importantly, that the patient would receive any benefit from intervention. In summary, this report nicely puts the utility of both CTA and MRA into context. It suggests that although subject to certain limitations, these two tests continue to have an important role in carefully selected patients. However, given the relatively low reproducibility, when using CTA or MRA, it may be important to get a consensus opinion from local radiologists and other specialists. With the rapid pace of technological advance, we can assume that the accuracy of these tests will continue to improve in the coming years.

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.