Immunoglobulin treatment for whooping cough (pertussis) is widely believed to be ineffective although there are sound reasons for regarding the condition as a toxin-induced disease. We wondered whether the lack of success with pertussis immunoglobulins might be attributable to inadequate dose, so we designed a randomised, double-blind, placebo-controlled trial of two immunoglobulin preparations. The study was conducted at three Swedish hospitals. We enrolled 73 children aged less than 36 months who were admitted with a clinical diagnosis of whooping cough. On admission they were assigned to one of three groups: (a) monocomponent pertussis toxoid vaccine; (b) two-component acellular vaccine also containing filamentous haemagglutinin; or (c) 20% albumin solution (placebo). The immunoglobulins had a high antitoxin content and had been raised with acellular pertussis vaccines. Diagnosis of pertussis was confirmed by laboratory tests and the follow-up was completed in 67 children. The main study group consisted of 47 children with less than or equal to 14 days of disease before therapy. Duration of whoops post-treatment was 8.7 days (95% Cl 4.8, 12.6) in the 33 children receiving immunoglobulin vs 20.6 (95% Cl 11.9, 29.3) in the 14 receiving placebo (p = 0.0041). Mean number of whoops during the first week of follow-up was also significantly reduced (p = 0.0196). We found that early treatment was important, since the effect on duration of whoops was most pronounced when disease duration before treatment was less than or equal to 7 days. There were no significant differences between the two immunoglobulin preparations. High-dose specific pertussis immunoglobulin with a high antitoxin concentration has a beneficial effect in the treatment of whooping cough.