Postural hypotension in elderly patients given carvedilol

BMJ 1994; 309 doi: (Published 24 September 1994) Cite this as: BMJ 1994;309:775
  1. H Krum,
  2. E L Conway,
  3. J H Broadbear,
  4. L G Howes,
  5. W J Louis
  1. Department of Clinical Pharmacology and Therapeutics, Austin Hospital, Heidelberg, Victoria 3084, Australia
  1. Correspondence to: Dr Louis.
  • Accepted 28 July 1994

First dose postural hypotension is commonly observed with antihypertensive drugs,1,2 but its incidence may be overstated in clinical trials because of protocol requirements, which may include frequent venesection, fasting, acute withdrawal from caffeine, and frequent changes in posture.1 The ß blocker vasodilator drug carvedilol is associated with orthostatic hypotension, particularly in elderly people.*RF 234* We assessed the incidence of first dose hypotension in the absence of possible confounding factors linked to protocol.

Methods and results

The study group comprised 16 elderly patients (nine men; mean (SD) age 70 (6) mean weight 73 (17) kg) with uncomplicated mild or moderate hypertension whose supine diastolic blood pressures were 90-110 mm Hg after one week's washout with placebo after antihypertensive treatment. Previous studies showed that a sample of 10-18 subjects would produce an 80% chance of detecting a difference in response of 10 mm Hg between treatments at the 95% confidence level. The protocol was approved by the Austin Hospital Ethics Committee, and consent obtained from each patient.

The study was a randomised double blind crossover design with four study days with single doses - carvedilol 12.5 mg, carvedilol 25 mg, pindolol 15 mg and placebo - each separated by a washout of >=7 days. The patients were randomly allocated to eight different sequences of treatment.

After sitting for 10 minutes the patients received their allocated treatment and a standard breakfast They then sat for an hour, walked about during the second hour, and were supine during the third hour. Blood pressure and heart rate were recorded with patients sitting (supine during the third hour) then after one minute standing, at 0, 30, 60, 125, and 180 minutes after they had received the treatment.

The main outcome measures were δ supine blood pressure (blood pressure at time0-blood pressure at timet) and δ standing blood pressure (standing blood pressure at timet-supine blood pressure at timet), where timet was the time that the measurement was taken and time0 the value at baseline. These variables were analysed with three way analysis of variance, with patient, treatment, and period of study as factors, and treatment differences were estimated.5

Although all patients receiving placebo and pindolol could stand at all time points, two patients receiving carvedilol 25 mg could not stand at 125 minutes and two others receiving carvedilol 25 mg could not stand at either 60 or 125 minutes. Of these four patients, one could not stand at 125 minutes and another could not stand at either 60 or 125 minutes when they received carvedilol 12.5 mg. The table summarises differences between treatments in δ supine blood pressure and δ standing blood pressure two hours after each drug was given: carvedilol produced a dose dependent fall in supine blood pressure when compared with the placebo. Carvedilol 25 mg and pindolol produced similar reductions in supine blood pressure with minimal treatment differences. Dose dependent postural falls in standing blood pressure occurred with carvedilol. This orthostatic response was significantly greater with carvedilol 25 mg than with pindolol despite similar reductions in supine blood pressure.

Estimated mean differences* (95% confidence interval) in supine blood pressure and standing blood pressure between treatments two hours after they had been given to 16 elderly patients

View this table:


The incidence of first dose postural hypotension has been reported in pharmacokinetic studies as being as high as 40% with carvedilol.2,4 To avoid potential aggravating factors we did not take blood samples, the patients received their drugs with food, and each patient began the study seated.

Despite these precautions some patients could not stand after a single dose of carvedilol, and all patients had significantly greater falls in systolic and diastolic blood pressure in the standing position with carvedilol 25 mg than with the placebo, carvedilol 12.5 mg, and pindolol. Symptoms were obvious 60 and 125 minutes after carvedilol had been given, corresponding to the times of maximum hypotensive response to carvedilol (at 1-2 h).3 No symptoms of postural effects occurred, however, with pindolol.

Recommendations on dose for managing hypertension suggest starting with carvedilol 12.5 mg daily, with subsequent increases to achieve control of the blood pressure. On the basis of our results, however, a lower starting dose may be needed in elderly hypertensive patients to avoid first dose postural symptoms. Similar caution is probably necessary in other groups of patients likely to be senstive to combined (alpha) ß blockade - namely, those taking diuretic drugs or in heart failure.

This study was supported by Boehringer Mannheim GMBH, Germany.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract