|Skip left side navigation and go to content||
Expert Panel on a Hypertension Treatment Trial Initiative Meeting Summary
January 16, 2007
Complete Meeting Report (PDF file, 13 pages, 102 K)
A special ad hoc panel of scientific leaders in clinical hypertension research and clinical trials in cardiovascular diseases (CVD) was convened in Bethesda, Maryland, on January 16, 2007. The Institute asked the expert panel to:
The charge to the panel was to recommend a single hypertension trial that the NHLBI should initiate based on the following four evaluation criteria:
The following agenda of possible trial designs and presenters was developed prior to the meeting:
There was a strong consensus among the panel that the most important trial question is still whether treating to a systolic blood pressure lower than the currently recommended goal will reduce CVD mortality and morbidity. The design would involve 7,500 non-diabetic patients age 55 or older with systolic blood pressure (SBP) of 130-180 mmHg at baseline, and at least one other CVD risk factor, randomized to goals of <140 mmHg (“standard”) versus <120 mmHg (“lower”) SBP goals, and followed for a mean of 5 years. The CVD risk factors would include, but not be limited to, clinical coronary artery disease, peripheral arterial disease, and stage 3 chronic kidney disease. Alternative or complementary trial objectives were discussed and all had merit. The major alternative focused primarily on isolated systolic hypertension (ISH) below 160 mmHg (Stage 1 ISH). One of the attractions of this trial was the relative paucity of trial evidence on the effectiveness of treating ISH below 160 mmHg. Neither the proposed lower-goal trial nor the ongoing NHLBI Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial will answer completely the ISH question. One difficulty in conducting an ISH trial is that many patients with ISH are already on antihypertensive therapy, and withdrawal of treatment to confirm that they have ISH would be difficult and would likely make recruitment challenging. The lower-goal trial was given the highest priority.
Several complementary objectives were considered as possibly being incorporated into a factorial design or added as a third arm to the lower-goal trial. One factorial question discussed was what is the best combination of drugs to use in treating hypertension to either the current standard goal or a new lower goal. This factorial design was one of the recommendations of the 2003 workshop. While this remains an important question, there was an overall conclusion that trials completed since 2003 and those currently underway may provide additional useful information about the best combination of drugs for treating hypertension (the “second drug” question). The group concurred with the view that a factorial design with several different required drug combinations and with a lower-goal objective might make answering the primary goals question more difficult. The protocol-specified need in such a factorial trial to use specific drug combinations might reduce the likelihood of attaining an adequate difference in mean blood pressure between the lower-goal and standard-goal groups.
Another complementary objective discussed was a test of the efficacy of isosorbide dinitrate/hydralazine (IDH) as antihypertensive therapy. The rationale for the IDH approach was the additional benefit observed when IDH was added to guideline therapy for heart failure and the possibility that IDH might work by a different pathway (increased nitric oxide or reduced oxidative stress) than most other hypertensive agents. During the discussion it was noted that there is little to no experience with IDH therapy in hypertension trials, and its use may reduce the likelihood of attaining an adequate difference between the lower-goal and standard-goal groups. This issue may be addressed by raising the lower limit of the SBP inclusion criteria by 5-10 mmHg. There was some enthusiasm for exploring whether a lower-goal trial could incorporate drugs that work by less well-understood mechanisms, e.g. reduction of oxidative stress, with the caveat that only agents that had been used in hypertension trials and that would not jeopardize achieving the blood pressure goals would be considered for inclusion.
Another complementary objective discussed was the opportunity to test an intervention using omega-3 fatty acids - eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) - to reduce CVD event rates. Fish oil supplements contain EPA and DHA. The NHLBI and the NIH Office of Dietary Supplements Working Group Report (2004) on Future Clinical Research Directions on Omega-3 Fatty Acids and Cardiovascular Disease concluded that a definitive CVD events trial is needed. The recommended trial included both primary and secondary prevention populations. A trial of an omega-3 fatty acids (O3FA) intervention could be added as a factorial question to the lower-goal trial because the lower-goal trial would include both primary and secondary prevention populations, and the O3FA intervention would not have much effect on achieving the blood pressure goals. There was support for identifying for consideration other major questions similar to the proposed O3FA intervention that are ripe for a CVD events trial and could be factored into a lower-goal trial, as long as such a question would not greatly increase the cost of the trial and the intervention would not impact blood pressure.
There was extensive discussion of the lower-goal trial with regard to which patient populations and what secondary outcomes should be included. One important scientific gap that the lower-goal trial should address is the lack of current scientific information on whether a lower SBP goal would reduce cardiovascular events and slow the progression of kidney disease patients with stage 3 chronic kidney disease and less than moderate proteinuria (< 300 mg albumin/d). (There is strong evidence that targeting SPB in patients with higher levels of proteinuria, equivalent to spot urine albumin greater than 300 mg/g, should be lower than the usual goal of <140 mmHg, and that such treatment should include an ACE inhibitor). Although extra recruitment efforts would be needed to achieve the proposed goal of including 3500 patients with stage 3 non-diabetic chronic kidney disease (CKD), the group felt that it was feasible and desirable to include a CKD subgroup with estimated glomerular filtration rate (eGFR) between 30 and 59 ml/min, with subgroup analyses planned in advance for the CKD subgroup as a whole and separately for those with spot urine with an albumin-to-creatinine ratio less than or greater than 30 mg/g. A factorial design in the CKD subgroup to examine the benefits of initial therapy using an ACE inhibitor versus a diuretic would be of interest, but would likely complicate the study design for the main question. The group concluded that there should be further discussion between NHLBI and NIDDK about the possibility of including an adequate number of CKD patients. There was support for incorporating into the lower-goal trial as study endpoints both kidney events and measures of decline in renal function. Moderate or high-risk cardiovascular patients should be a requirement for the study, but a balance should be found between ensuring that patients have an elevated cardiovascular risk and avoiding too restrictive eligibility criteria, which would limit recruitment or reduce generalizability of trial results. Patients with treated and untreated ISH would comprise a large component of the patients in the lower-goal trial, and it was recommended that a pre-specified analysis of this group be conducted.
For the lower-goal trial, a number of different drug regimens were discussed, including the therapeutic strategy used in the ACCORD trial, an IDH strategy, and a strategy for using angiotensin converting enzyme inhibitors in all patients with stage 3 CKD. No specific drug regimen was recommended for use in the trial, although there was support for the ACCORD approach because it is consistent with JNC 7 recommendations and because this approach allows the withdrawal of medications in the standard goal group if the SBP becomes too low. Because of the epidemiological evidence and the magnitude of the problem of cognitive decline and dementia in the elderly population, there was substantial support for including quality measures of cognitive function. The hypothesis that the lower SBP goal would reduce the rate of cognitive impairment and new onset of structural white matter abnormalities could also be evaluated in the study.
In conclusion, there was a clear consensus that a hypertension trial comparing a lower SBP goal with a standard goal in patients without diabetes was the top priority. The lower-goal trial should include a substantial subgroup of CKD patients and appropriate kidney disease outcome measures. The panel also recommended that the eligibility criteria be broad, including patients with isolated systolic hypertension. Consideration should also be given to evaluating the effect of nitric oxide enhancing therapy if there is a feasible intervention that would not threaten the goal of achieving a substantial SBP delta between the two randomized groups. The panel also encouraged inclusion of measures of cognitive impairment and, if feasible, structural abnormalities of the central nervous system white matter. If the addition of an O3FA intervention as a factorial in a hypertension goals trial would not increase the cost of the trial substantially, then it deserves consideration because the O3FA intervention should not have a large effect on blood pressure, and the O3FA question has substantial public health and scientific importance.
1The National Heart, Lung, and Blood Institute Working Group on Future Directions in Hypertension Treatment Trials. Major Clinical Trials of Hypertension What Should be Done Next? Hypertension. 2005;46:1-6.