Findings coming from autopsies and serum of SARS patients suggest an important immune-inflammatory implication in the evolution to respiratory distress. Conditions such as HIV infection or treatment with immunosuppressors (in cancer or autoimmune diseases) are not among the bad prognosis factors for development of distress. To date, there have been no reported case fatalities in children, probably due to their more immature immune system. Our conclusions follow: (1) The milder form of SARS in children and the apparent protective factor that immunosupression represent rules out a significant viral cytopathic effect (they would be the most affected). (2) The evidence for immune implication in distress strongly supports immunomodulators for therapy: phosphodiesterase inhibitors (due to their down-modulating activity on proinflammatory cytokines); inhaled corticoids (aimed at producing a local immunomodulation); teophylline or nedocromil sodium (which prevents inflammatory cell recruitment into the airway wall). (3) An early immunomodulatory therapy, based on the levels of proinflammatory cytokines and clinical parameters to evaluate the respiratory function such as arterial oxygen saturation, could prevent the occurrence of distress. (4) Vaccine design should consider the immune origin of distress. (5) Physicians should be aware of mildly symptomatic patients (children, immuno-compromised hosts) to avoid transmission to immunocompetent adults.