ABSTRACT
BACKGROUND: Most episodes of acute diarrhea are self-limiting. Most patients require neither a stool culture nor antibiotic therapy. This study was designed to investigate the predictors of a stool-culture positivity and indicators for antibiotic therapy in acute infectious diarrheal patients. METHODS: Medical records of 139 adult cases were retrospectively reviewed during the 12 months between January 1999 and December 1999. We used the Chisquare-test to analyze the data for statistical significance. RESULTS: Fecal leukocytes were examined in 104 cases; 30 tested positive (28.8%). Eighty-five stool cultures were taken, and 24 yielded enteric pathogens, such as Salmonella, Shigella, V. Cholera and V. Parahaemolyticus. Cultures from patients treated between August and October, with fever above 37.6degrees C, or with symptoms of abdominal pain had higher yields(44.2% vs 11.9%, p=0.001, correlation coefficient=0.359; 36.8% vs 10.7%, p=0.012, correlation coefficient=0.273;, 38.1% vs 18.6%, p=0.046, correlation coefficient=0.216 ; respectively) and when combined with fecal leukocytes had a sensitivity of 100% and a specificity of 90.1-95.1%. CONCLUSION: The primary variables (season, fever, and abdominal pain) were excellent predictors of stool culture positivity and indicators for antibiotic therapy, especially when combined with fecal leukocytes