CEBM Home Page Index to this Site How to use the site Teaching Resources and Activities The CAT Bank About the Book <I>How to Practice and Teach EBM</i> EBM Glossary The EBM ToolBox

HYPERTENSION, SYSTOLIC, ELDERLY - DIURETICS AND BETA BLOCKERS DECREASE STROKES

Clinical Bottom Line:
Treatment of isolated systolic hypertension with diuretics and/or beta blockers decrease the risk of stroke in the elderly.

Citation: SHEP co-operative research group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension. final results of the systolic hypertension n the elderly program. (SHEP). JAMA 1991;265:3255-64.
Three-part Clinical Question: In a patient with hypertension, will therapy with diuretics decrease the risk of stroke and death?
Search Terms: "stroke" and "hypertension" in Best Evidence

The Study:
Double-blinded concealed randomised controlled trial with intention-to-treat.
patients = 60 years with isolated systolic hypertension
Control group (N = 2371; 2371 analysed): placebos
Experimental group (N = 2365; 2365 analysed): 4 consecutive steps - chlorthalidone12.5 mg/d; increase to chlorthalidone 25 mg/d; addition of atenolol 25mg/d or reserpine 0.05 mg/d; increase to atenolol 50mg/d or reserpine 0.1 mg/d. goal of therapy was SBP<160 mm Hg and a reduction in SBP of at least 20 mmHg. Any patient with sustained SBP>220 mmHg or DBP> 90 mmHg was given active treatment

The Evidence:

Outcome

Time to Outcome

CER

EER

RRR

ARR

NNT

stroke

5 years

0.067

0.044

34%

0.023

43

95% Confidence Intervals:

15% to 54%

0.107 to 0.036

28 to 100

deaths

5 years

0.090

0.102

-13%

-0.012

-83

95% Confidence Intervals:

-32% to 5%

-0.029 to 0.005

NNT = 209 to INF NNH = 35 to INF;

Comments:
  1. decreased risk of nonfatal MI or coronary heart disease death with treatment RR 0.73 (0.57 to 0.94)
  2. only enrolled 1% of the patients that were screened - included the very well elderly an therefore each clinician must assess each patient's individual baseline risk of and individualise the NNT using this estimate
  3. at 5 years, 90% of the study group and 44% of controls were on active treatment and therefore the study likely underestimates the benefits of therapy since an intention to treat analysis was done
  4. ?was there some unblinding because of the need for some patients(more often the ones receiving diuretics) to receive potassium supplements

Appraised by: Straus 1998;
Expiry date: 1999


Click here to comment on this CAT		 Go Back to Search