You are viewing an older, archived article. There may be more up to date articles on this subject, try a new search

High Blood Pressure

Bottom Lines
Rationales References
Issue 6 Contents


If ß-blockers are not indicated which antihypertensive is first choice?

With the publication of the Antihypertensive and Lipid Lowering to prevent Heart Attack Trial (ALLHAT) study11 the issue of which blood pressure lowering medicine to use appeared to be resolved, except a ß-blocker was not included in the study. The comparator medicines were a thiazide (chlorthalidone), an ACE inhibitor (lisinopril) and a calcium channel blocker (amlodipine). Doxazosin was originally in the comparator group but this arm was discontinued early due to a significantly higher rate of heart failure compared to chlorthalidone.

There was no significant difference between the three medicines for cardiovascular endpoints – stroke, angina, coronary revascularisation, peripheral arterial disease end stage renal disease or all cause mortality, or cancer and gastrointestinal bleeds. However there did appear to be less protection against heart failure for the calcium channel blocker and ACE inhibitor compared to the thiazide. The ACE inhibitor also appeared to be less protective for stroke than the thiazide (Table 1).

Table 1: ALLHAT Study. Risk of progressing to future medical condition11
  Thiazide CCB NNT Thiazide ACEI NNT
Heart failure 7.7% 10.2% 40 7.7% 8.7% 100
Combined CVD       30.9% 33.3% 42
Stroke       5.6% 6.3% 142
NNT (number needed to treat) = Number of people treated with a thiazide instead of the comparator that would prevent one person having the event over six years.

Adverse effects

There were significantly more metabolic changes in the thiazide group but overall these did not translate into more cardiovascular events or all cause mortality. All groups had a reduction in total cholesterol at four years (0.49 mmol/L, 0.54 mmol/L and 0.53 mmol/L for the thiazide, calcium channel blocker and ACE inhibitor respectively).

The change in average serum potassium concentration at four years reduced by 0.1 mmol/L for the thiazide group, and increased by 0.1 mmol/L for the calcium channel blocker and ACE inhibitor groups. More people had a serum potassium concentration < 3.5 mmol/L in the thiazide group (8.9% compared to 1.9% in the calcium channel blocker group and 0.8% in the ACE inhibitor group).

The thiazide increased the average blood glucose 0.16 mmol/L compared to 0.03 mmol/L for the calcium channel blocker group (not significant) and a reduction of 0.08 mmol/L in the ACE inhibitor group (p=0.001).

The messages from the ALLHAT Study were:

  • All people should be initiated on a thiazide as the first-line therapy for hypertension. For people requiring combination therapy, a thiazide should be included in the combination.
  • Multiple blood pressure lowering medicines are usually required.
  • ACE inhibitors do not offer a unique advantage in people with uncomplicated diabetes (no microalbuminuria).
  • Significantly higher rate of heart failure in the calcium channel blocker group compared to thiazide group (10.2% vs. 7.7%; NNT = 40).
  • Higher rate of combined cardiovascular disease in ACE inhibitor group compared with a thiazide (33.3% vs. 30.9%; NNT = 42).
  • The potential adverse effects of the thiazide did not outweigh the benefits.

Unlike the ALLHAT study, the ASCOT-BPLA study in 200512 did include a ß-blocker. It used the composite end point of combined fatal coronary heart disease plus non-fatal myocardial infarction. It showed there was no significant difference between the ACE inhibitor (perindopril) added to the calcium channel blocker (amlodipine) compared to the thiazide (bendroflumethiazide) added to atenolol. However, all cause mortality was significantly less (NNT = 650 for one year) in the amlodipine plus perindopril group, as was the incidence of stroke and cardiovascular events and procedures.

Controversial aspects of the trial included:

  • The use of a ß-blocker first, then a thiazide when a thiazide would normally be first-line, particularly as 63% of the study group was older than 60 years.
  • The choice of atenolol as the ß-blocker raised doubts about whether the results could be extrapolated to all ß-blockers.
  • The difference in blood pressure between the two groups (2.7 mmHg) could account for the difference in event rates.

The issue of the blood pressure difference was investigated in a paper published in the same journal.13 Using multivariate analysis other factors accounted for approximately 50% and 40% of the difference between the groups for cardiovascular events and stroke respectively. HDL-cholesterol was the biggest contributor to cardiovascular events and blood pressure was the biggest contributor to stroke. The differences between the groups became no longer significant after adjustment for these differences.

Other factors to consider in the ASCOT-BPLA study were that only 19% of the people were on aspirin and 10% to 11% were on lipid lowering medicines. This compares to 36% on aspirin in the ALLHAT study. Additionally 17 to 18% of women were on HRT in the ALLHAT Study. This is contrary to the New Zealand guidelines where we expect people with a cardiovascular risk of more than 15% to be on aspirin and a statin as well as a blood pressure lowering medicine. Use of these adjunctive medicines may well reduce the importance of the choice of blood pressure lowering medicine.

Number 33,357
Mean 4.9 yrs follow up
Mean 5.5 yrs follow up
Population Older than 54 years with hypertension plus at least one other CHD risk factor 40 - 79 years with hypertension plus at least three other CV risk factors
Medicines Chlorthalidione, lisinopril, amlodipine Amlodipine up to 10 mg, then add perindopril versus atenolol up to 100 mg, then add bendroflumethiazide
Endpoints Combined fatal CHD or non-fatal MI Combined fatal CHD or non-fatal MI
BP achieved
  • Target 140/90 mmHg
  • Achieved goal for 68% thiazide, 66% calcium channel blocker, 61% ACE inhibitor
  • ACE inhibitor systolic blood pressure 2 mmHg greater than with thiazide
  • Target 140/90 mmHg
  • Achieved goal for 53% of participants
  • Amlodipine arm achieve mean of 2.7 mmHg less than atenolol arm
  • NSD - primary endpoint
  • NSD - all cause mortality
  • Heart failure significantly increased with ACE inhibitor, calcium channel blocker
  • CVA and combined cardiovascular disease significantly increased with ACE inhibitor
  • Less increase in blood glucose in the ACE inhibitor group
  • NSD – primary endpoint
  • All cause mortality reduced in amlodipine group (NNT ~ 650 for 1 year)
  • Total cardiovascular events and procedures, and CVA were significantly reduced in the amlodipine group
  • Risk of diabetes was significantly reduced in amlodipine group
NSD = No significant difference
Page  1 | 2 | 3 | 4 | 5 | 6 | 7 Page 4