A double-blind study comparing placebo and haem arginate was conducted in 12 patients with acute intermittent porphyria. 2 days after admission in attack patients were randomised to receive intravenous haem arginate 3 mg/kg per 24 h for 4 days or placebo. 9 patients were readmitted with a further attack and were given the alternative treatment. Before randomisation the paired attacks were of similar severity with respect to urinary porphobilinogen (PBG) excretion and clinical manifestations. With haem arginate the median PBG excretion of the 9 patients with two attacks (normal range 0-16 mumol per 24 h) fell significantly from 332 mumol per 24 h (range 137-722) on admission to a median lowest level of 40 (range 22-105). On placebo, median PBG excretion was 382 (range 196-542) on admission, falling to 235 (range 128-427). Median duration of admission after the start of treatment was 11 days (range 2-28) for placebo and 8 days (3-26) for haem arginate. Median total analgesic requirement between the start of treatment and discharge was 8150 mg pethidine equivalents (range 0-17,650) with placebo versus 6425 (range 50-20,650) with haem arginate. Phlebitis occurred in 5 patients on haem arginate and in 2 on placebo. Haem arginate effectively reduces porphyrin precursor overproduction in the acute porphyric attack but this reduction is not accompanied by striking resolution of the clinical manifestations of the attack.