Abstract
Coagglutination was compared with counterimmunoelectrophoresis (CIE) for sensitivity and specificity in the detection of pneumococcal antigens in sputum. Initial sputum samples from patients with pneumococcal pneumonia (less than 12 h of antibiotic therapy) were positive for antigens in 37 of 44 cases (84%) by either test. There was a decline in the number of positive results with sputum samples obtained during continuing antibiotic therapy, but the decline was greater with CIE (only 29% of samples were positive at 3 days of therapy) than with coagglutination (61% of samples were positive at 3 days of therapy) (P less than 0.05). Sputum from 3 of 11 patients (27%) and from 2 of 11 patients (18%) with nonpneumococcal pneumonia was positive for pneumococcal antigens by CIE and coagglutination, respectively, indicating a similar degree of non-specificity. Coagglutination produced the same results as CIE with sputum from patients with chronic bronchitis but without pneumonia; 9 of 23 of these patients were positive. Coagglutination was simpler to perform than CIE and required only a fraction (about 1/30) of the antiserum required for CIE. These advantages, plus the greater sensitivity of coagglutination with sputum samples obtained during antibiotic therapy, suggest that coagglutination is preferable to CIE.