NPC Archive Item: Review of ALLHAT concludes that a thiazide is still a good first choice antihypertensive drug

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.
MeReC Rapid Review NPC Logo

27th May 2009

Thiazide diuretics are unsurpassed as initial therapy for reduction of cardiovascular (including stroke) and renal risk, a review of ALLHAT has concluded. ALLHAT, still the largest hypertension study conducted, compared a thiazide diuretic-based regimen with regimens based on an alpha-blocker, an ACE-inhibitor or a calcium channel blocker. The diuretic-based regime was superior at preventing heart failure, and new-onset diabetes associated with thiazides does not increase the risk of cardiovascular outcomes.

Health professionals should continue to follow the NICE guidance on hypertension. This recommends a thiazide diuretic or calcium channel blocker as first line treatment for most people with hypertension aged 55 years or older and black patients of any age. If there are no other compelling reasons to favour one over the other, good prescribing practice would suggest that a thiazide diuretic is preferable, as these are generally well tolerated and significantly less costly than calcium channel blockers.

What is the background to this?
The ALLHAT study was published in 2002. It recruited 42,418 patients aged 55 years or older with hypertension and at least one other cardiovascular (CV) risk factor such as type 2 diabetes, smoking, etc. Participants were randomised double blind to chlortalidone, lisinopril, amlodipine or doxazosin. These were titrated up to try to achieve a target blood pressure of less than 140/90 mmHg. Additional drug treatment (e.g. atenolol) could be added open-label. Follow-up was for a mean of 4.9 years. There were no statistically significant differences between the groups in relation to the primary outcome (fatal coronary heart disease [CHD] or non-fatal myocardial infarction [MI]). However, there were some advantages for the chlortalidone group with regard to secondary endpoints, notably heart failure. The doxazosin arm was stopped early after an interim analysis (median 3.3 years) found an excess risk of stroke and combined CV disease, particularly heart failure, with doxazosin compared with chlortalidone.

Since ALLHAT was published, other trials in hypertension have been published, notably the ASCOT study and also the ACCOMPLISH study, and the Blood Pressure Lowering Treatment Trialists’ Collaboration (BPLTTC) have published a series of meta-analyses. The ALLHAT authors have looked again at the ALLHAT data in the light of these other data.

What does this study claim?
The authors addressed several specific questions. Firstly, were differences in secondary outcomes seen in ALLHAT due to between group differences in blood pressure (BP) lowering? The relative risk (RR) of heart failure in the amlodipine, doxazosin and lisinopril groups was higher than in the chlortalidone group (RRs and 95% confidence intervals [CI] 1.38 [1.25 to 1.52]; 1.80 [1.61 to 2.02] and 1.20 [1.09 to 1.34] respectively), and the RR of stroke was also greater in the doxazosin and lisinopril groups than in the chlortalidone group (1.26, 95%CI 1.10 to 1.46 and 1.15, 95%CI 1.02 to 1.30 respectively) but similar in the amlodipine group (0.93, 95%CI 0.82 to 1.06). The ALLHAT authors’ analysis which adjusted for BP differences concluded that the differences in outcomes were not explained by BP differences. The BPLTTC analysis agreed that the difference in heart failure risk was not accounted for by BP differences. A specific study of the patients in ALLHAT admitted to hospital with heart failure seems to confirm the value of diuretics in this respect.

Secondly, they asked are the results applicable to other thiazide diuretics? Based on meta-analyses, they conclude that they are, as long as equivalent doses are used. They note that ASCOT and ACCOMPLISH used lower doses of thiazide diuretic.

Thirdly, they asked what are the implications of diuretic-associated diabetes? Fasting blood glucose levels increased in all groups in ALLHAT, but more so in the thiazide diuretic group and, after 4 years, the proportion of participants with levels consistent with diabetes was 11.6% in the chlortalidone group compared to 9.8% in the amlodipine group and 7.8% in the lisinopril group. However, development of diabetes, or a 0.56mmol/L increase in fasting blood glucose, was not associated with a statistically significant increase in risk of any CV outcome, all-cause mortality or development of end-stage renal disease. There were also no benefits from any of the other agents, compared to chlortalidone, in patients with metabolic syndrome.

So what?
This review of the largest trial of antihypertensive treatment to date confirms that diuretics are still a good first choice in most patients with hypertension. They are effective, well tolerated, low cost and generally acceptable to patients. This supports NICE guidance on hypertension. NICE guidance on type 2 diabetes recommends an ACE-inhibitor as first line treatment. Many people with diabetes will need more than one antihypertensive drug to achieve a desired blood pressure, and diuretics are a good choice for most people. A recently published meta-analysis of hypertension studies also concluded that all the main classes of blood pressure lowering drugs have a similar effect in reducing CHD events for a given reduction in blood pressure. That meta-analysis is discussed further on Matt Robinson’s Prescribing advice for GPs website.

More information on the management of hypertension can be found on the hypertension section of NPC

Please comment on this blog in the NPC discussion rooms, or using our feedback form.

Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you.