Research Article

Caffeine restriction: effect on mild hypertension.

BMJ 1991; 303 doi: (Published 16 November 1991) Cite this as: BMJ 1991;303:1235
  1. T M MacDonald,
  2. K Sharpe,
  3. G Fowler,
  4. D Lyons,
  5. S Freestone,
  6. H G Lovell,
  7. J Webster,
  8. J C Petrie
  1. Department of Medicine and Therapeutics, Aberdeen Royal Infirmary.


    OBJECTIVE--To determine the effects on blood pressure of modifying dietary caffeine intake in patients with mild and borderline hypertension by monitoring ambulatory and clinic blood pressure. DESIGN--Four way, randomised, crossover trial of four consecutive two week dietary regimens: normal diet, caffeine free diet alone, caffeine free diet with decaffeinated instant coffee, caffeine free diet with caffeinated instant coffee (instant coffee phases conducted double blind). SETTING--Hospital hypertension clinic, Scotland. PATIENTS--52 patients (23 men; aged 26-67 years) with untreated borderline or mild hypertension (diastolic blood pressure 90-105 mm Hg) who normally drank a minimum of three cups of coffee daily. MAIN OUTCOME MEASURES--Mean ambulatory blood pressure over 24 hours; mean morning, daytime, and night time ambulatory blood pressure; sitting clinic blood pressure at 1700; plasma caffeine concentration at 1700 on the last day of each regimen. RESULTS--Mean 24 hour ambulatory blood pressure was not different between regimens. There was no difference in blood pressure variability between regimens. During the caffeine free diet alone morning ambulatory diastolic blood pressure was higher (2.8 mm Hg) than during the caffeine free diet with caffeinated coffee. Mean sitting clinic systolic blood pressure was higher at 1700 (4.7 mm Hg) with a caffeine free diet than with the caffeine free diet with caffeinated coffee (p less than 0.05). Dietary compliance as assessed by plasma caffeine concentration was excellent. There was no significant correlation between plasma caffeine concentration and blood pressure. CONCLUSIONS--Drinking caffeinated instant coffee over a two week period does not adversely influence blood pressure in patients with borderline or mild hypertension; abstinence is of no benefit.