Patients who undergo mechanical ventilation for severe asthma are at risk of developing diffuse muscle weakness because of acute myopathy. The relative importance of corticosteroids and neuromuscular paralysis in causing the myopathy is controversial, and it is uncertain whether the chemical structure of the drug used to induce paralysis influences the risk of myopathy. Using a retrospective cohort study design, we evaluated 107 consecutive episodes of mechanical ventilation for severe asthma to assess (1) the incidence of clinically significant weakness in patients treated with corticosteroids alone versus corticosteroids with neuromuscular paralysis, (2) the influence of the duration of paralysis on the incidence of muscle weakness, and (3) the relative risk of weakness in patients paralyzed with the nonsteroidal drug atracurium versus an aminosteroid paralytic agent (pancuronium, vecuronium). The use of corticosteroids and a neuromuscular blocking agent was associated with a much higher incidence of muscle weakness as compared with the use of corticosteroids alone (20 of 69 versus O of 38, p < 0.001). The 20 weak patients were paralyzed significantly longer than the 49 patients who received a neuromuscular blocking agent without subsequent weakness (3.4 +/- 2.4 versus 0.6 +/- 0.7 d, p < 0.001). Eighteen of the 20 weak patients had been paralyzed for more than 24 h. The incidence of weakness was not reduced when paralysis was achieved with atracurium as opposed to an aminosteroid neuromuscular blocking agent. In conclusion, corticosteroid-treated patients with severe asthma who undergo prolonged neuromuscular paralysis are at significant risk for the development of muscle weakness, and the risk of weakness is not reduced by use of atracurium.