Current issues of ACP Journal Club are published in Annals of Internal Medicine


Decreased caffeine consumption did not reduce blood pressure in mild hypertension

ACP J Club. 1992 Mar-April;116:38. doi:10.7326/ACPJC-1992-116-2-038

Source Citation

MacDonald TM, Sharpe K, Fowler G, et al. Caffeine restriction: effect on mild hypertension. BMJ. 1991 Nov 16;303:1235-8.



To determine whether dietary caffeine restriction is a useful nonpharmacologic treatment for patients with mild hypertension.


Randomized, 4-way crossover trial of 4 consecutive dietary caffeine regimens, each lasting for 2 weeks, and double-blinded for freeze-dried caffeinated and decaffeinated coffee.


Hospital hypertension clinic in Scotland.


Patients referred to the clinic, who had an untreated sitting diastolic blood pressure (BP) of 90 to 105 mm Hg, were recruited if they consumed ≥ 3 cups of coffee daily and had no other illnesses. 50 of 52 patients (23 men and 27 women, age range 26 to 63 y) completed the 8-week trial.


The dietary regimens were 2-week periods of the participant's usual diet, a caffeine-free diet, a caffeine-free diet with ≥ 3 cups/d of decaffeinated instant coffee, and a caffeine-free diet with ≥ 3 cups/d caffeinated instant coffee.

Main outcome measures

24-hour ambulatory BP measurements taken every 15 min during the day and every hour at night on day 13 of each dietary regimen; sitting BP measurements taken at the clinic on day 14. Dietary compliance was monitored by plasma caffeine content, which was compared with the weight of coffee consumed during the coffee phases of the trial.

Main results

Mean ambulatory systolic and diastolic BPs did not differ among regimens at the end of the study: Mean BP for the usual-diet regimen was 134/86 mm Hg compared with 134/87 mm Hg for the caffeine-free diet alone, and 134/86 mm Hg and 133/86 mm Hg for the decaffeinated and caffeinated coffee regimens, respectively. Diastolic BP was slightly lower between 0600 h and 1000 h for the caffeinated coffee regimen than for the caffeine-free diet (86.4 mm Hg [95% CI 84.6 to 88.1 mm Hg] vs 89.2 mm Hg [CI 87.4 to 90.9 mm Hg], P < 0.01). Clinic systolic BP was also lower for the caffeinated coffee regimen than for the caffeine-free diet (139.7 mm Hg [CI 135.6 to 143.8 mm Hg] vs 144.5 mm Hg [CI 140.0 to 149.0 mm Hg], P < 0.05). Plasma caffeine concentration indicated excellent dietary compliance and was not correlated with BP measurements.


Short-term restriction of dietary caffeine did not reduce blood pressure in patients with mild hypertension.

Source of funding: Wellcome Trust.

Address for article reprint: Dr. T.M. MacDonald, Department of Clinical Pharmacology, Ninewells Hospital and Medical School, Dundee DD1 9SY, Scotland, United Kingdom.


Nonpharmacologic interventions are often recommended for the initial treatment of mild hypertension. These recommendations sometimes include withdrawal of all caffeine.

Most previous studies supporting caffeine elimination did not study elimination directly but showed that adding caffeine can produce pressor responses by vasoconstriction and cardiac stimulation. There may be a modest additive effect when caffeine intake accompanies emotional stress, mental work, or physical exercise. Infrequent users have larger BP elevations (4 to 5 mm Hg) in response to caffeine consumption than habitual users, but stabilization occurs within 2 to 3 days of continued use.

Two randomized, crossover trials are pertinent to this study. The first found no difference in BP between drinking compared with abstaining from filter-brewed, caffeinated coffee among 21 normotensive, habitual users over 2 months (1). The other observed 1- to 2-mm Hg declines in BP among 45 normotensive habitual users drinking decaffeinated instead of caffeinated coffee over 6 weeks (2). This study by MacDonald and colleagues is helpful because they selected a uniform group of untreated, mildly hypertensive persons who were habitual coffee drinkers. Other methodologic strengths include both clinic and 24-hour ambulatory BP measurements, plasma caffeine concentrations to document compliance, and randomization of the order of the interventions for each patient. Although there were small (2 to 3 mm) but statistically significant increases in 2 of the morning BP measurements in the group on the caffeine-free diet alone, there was no overall effect of caffeine restriction on BP.

The available information suggests that discontinuing coffee or its caffeine will not improve BP levels among habitual drinkers seeking to lower mild or borderline hypertension.

Peter Rudd, MD
Stanford University Medical CenterStanford, California, USA


1. Rosmarin PC, Applegate WB, Somes GW. Coffee consumption and blood pressure: a randomized, crossover clinical trial. J Gen Intern Med. 1990;5:211-3.

2. van Dusseldorp M, Smits P, Thien T, Katan MB. Effect of decaffeinated versus regular coffee on blood pressure. A 12-week, double-blind trial. Hypertension. 1989;14:563-9.