RESUMO
BACKGROUND: We evaluated a dipstick test for rapid detection of Shigella sonnei on bacterial colonies, directly on stools and from rectal swabs because in actual field situations, most pathologic specimens for diagnosis correspond to stool samples or rectal swabs. METHODOLOGY/PRINCIPAL FINDINGS: The test is based on the detection of S. sonnei lipopolysaccharide (LPS) O-side chains using phase I-specific monoclonal antibodies coupled to gold particles, and displayed on a one-step immunochromatographic dipstick. A concentration as low as 5 ng/ml of LPS was detected in distilled water and in reconstituted stools in 6 minutes. This is the optimal time for lecture to avoid errors of interpretation. In distilled water and in reconstituted stools, an unequivocal positive reaction was obtained with 4 x 10(6) CFU/ml of S. sonnei. The specificity was 100% when tested with a battery of Shigella and different unrelated strains. When tested on 342 rectal swabs in Chile, specificity (281/295) was 95.3% (95% CI: 92.9% - 97.7%) and sensitivity (47/47) was 100%. Stool cultures and the immunochromatographic test showed concordant results in 95.5 % of cases (328/342) in comparative studies. Positive and negative predictive values were 77% (95% CI: 65% - 86.5%) and 100% respectively. When tested on 219 stools in Chile, Vietnam, India and France, specificity (190/198) was 96% (95% CI 92%-98%) and sensitivity (21/21) was 100%. Stool cultures and the immunochromatographic test showed concordant results in 96.3 % of cases (211/219) in comparative studies. Positive and negative predictive values were 72.4% (95% CI 56.1%-88.6%) and 100 %, respectively. CONCLUSION: This one-step dipstick test performed well for diagnosis of S. sonnei both on stools and on rectal swabs. These data confirm a preliminary study done in Chile.