To assess the value of color Doppler ultrasound in distinguishing minimal pleural effusion from pleural thickening, a prospective analysis was done on the ultrasonographic findings in 51 patients. Real-time, gray-scale, and color Doppler chest ultrasound examinations were carried out by different sonographers who had no clinical information concerning the patients. The sonographer evaluated the images for internal echogenicity of the effusion, pleural lesions that change shape with respiration, and movable septa and echo-densities in pleural space in conventional gray-scale as well as color signal in color Doppler ultrasound. Of the 35 patients with true effusion, 33 had positive color signal (sensitivity 94.3%, 95% confidence intervals [CI] 89 to 98.6%); in 16 patients without effusion, none had color signal (specificity 100%, 95% CI 83 to 100%). Although real-time, gray-scale ultrasound is also sensitive for detecting minimal effusion (sensitivity 100%, 95% CI 92 to 100%), it is less specific (specificity 68.7%, 95% CI 46 to 91.5%). Five of 16 examinations showing fluid-like lesions were found to be deceptive. With relatively high sensitivity and specificity, this method proved to be a useful diagnostic aid to real-time, gray-scale ultrasound for diagnosis of minimal or loculated effusion.