Three months before this study, susceptibility for malignant hyperthermia (MH) had been tested in 15 pigs. In all pigs, MH was triggered by administration of 1% halothane. Brain electrical activity was examined during therapy of MH with and without administration of dantrolene. From the EEG, power densities in selected frequencies and the median frequency of the power spectrum were calculated. Therapy was started when severe respiratory changes were observed (PaCO2 > 10 kPa, mixed venous oxygen tension (PvO2) < 4 kPa). At this time, heart rate exceeded 150 beat min-1, mean arterial pressure (MAP) was less than 60 mm Hg and median frequency was less than 2 Hz. EEG was isoelectric (n = 6) or showed slow polymorphic delta-activity. For therapy, administration of all anaesthetics was terminated, 100% oxygen was delivered and ventilation was increased four-fold. Acidosis was treated by administration of sodium bicarbonate 2-4 mmol litre-1 kg-1. Animals were allocated randomly to one of two groups: group I (control, n = 7) received no dantrolene; group II (n = 8) received dantrolene 2.5 mg kg-1 i.v. All variables were measured over a period of 60 min after therapy: EEG, HR and MAP were recorded continuously and blood-gas tensions, arterial potassium and glucose concentrations and pH were measured every 150 s. In group I (no dantrolene) minor, transient improvements in EEG activity were noted, but all animals died within 15-25 min after the start of therapy. In dantrolene-treated animals, EEG total power and median frequency increased within 5 min.(ABSTRACT TRUNCATED AT 250 WORDS)