We measured quantitative immunoglobulins (IgG, IgA, IgM, and IgG subclasses) in 101 unselected asthmatic patients. We identified hypogammaglobulinemia in 12 patients primarily involving IgG (dose-related) without a strong prediction for any IgG subclass. IgA and IgM were also suppressed but to a lesser extent. This prevalence of hypogammaglobulinemia (.12 +/- standard error of .03) is significantly greater than that seen in the normal population (approximately .025 +/- .017, P = .01). Hypogammaglobulinemia was strongly associated with use of systemic corticosteroids (P = .0001). A cumulative steroid dose of greater than or equal to 5 mg/day for at least 2 years was found in 10/12 patients with hypogammaglobulinemia compared with 37/89 patients without hypogammaglobulinemia (P = .024). No significant increase in the number of infectious episodes was seen in the hypogammaglobulinemic patients. To assess the significance of hypogammaglobulinemia in asthmatics, we assessed responses to tetanus and pneumococcal vaccine in three groups of asthmatics: (1) those with total IgG less than 400 mg/dL who had been on chronic oral steroids, (2) those with total IgG between 855 and 1199 mg/dL who were currently receiving oral steroids, and (3) those with total IgG between 855 and 1199 mg/dL who were not receiving oral steroids. All patients responded normally to tetanus vaccine, but three of eight patients in the hypogammaglobulinemic group showed impaired responses to pneumococcal vaccine. Patients with impaired pneumococcal responses were not clearly distinguishable on the basis of sinus disease or pneumonia. We conclude that although many patients with severe, steroid-dependent asthma experience repeated episodes of bronchitis or exacerbations of sinusitis, these problems are rarely associated with an impairment in specific antibody production. IgG subclass deficiencies are not common in this patient population. A very small subgroup of patients manifest a more severe hypogammaglobulinemia (IgG less than 400 mg/dL) or an inordinate frequency of infectious episodes. Given that bronchitis or sinusitis can be attributed to factors other than hypogammaglobulinemia in these patients, an assessment of specific antibody production in response to pneumococcal vaccination is warranted. A small but significant percentage of such patients will demonstrate impaired responses. These patients should be considered at increased risk for bacterial infections and should, therefore, be monitored closely for infectious episodes.