OBJECTIVE To explore the differences in assessing obstetric anal sphincter injuries (OASI) between transperineal ultrasound (TPUS) and endoanal ultrasound (EAUS) and test relationships between ultrasound findings and anal incontinence (AI) symptoms. METHODS A group of 196 women with a history of vaginal delivery was recruited. OASI was detected in a set of 5 slices by EAUS and 8 slices by TPUS. OASI grading was performed on TPUS rules and EAUS rules. A "significant sphincter defect" was diagnosed by TPUS and EAUS using "2/3 rules." Symptoms of AI were determined using the St Mark's Incontinence Score (SMIS). Ultrasound findings were compared between the two methods and correlated with symptoms. RESULTS Of 196 women, 29 (14.8%) suffered from AI with a mean SMIS of 12.1 ± 4.5, and 70 (35.7%) women with a mean age of 57 years had suspected OASI on imaging. Twenty-one (10.7%) "significant defects" were diagnosed by TPUS and 24 (12.2%) by EAUS. OASI Grades on TPUS had good agreement with EAUS rules (k = 0.70, P < .001). Logistic regression analysis showed that OASI Grade on imaging and "significant sphincter defects" seen on both forms of imaging were associated with AI symptoms. The odds ratio was 46 and 38 for "significant defects" on TPUS and EAUS, and 14 and 7 for OASI 3b+ on TPUS and EAUS in predicting AI, respectively. CONCLUSIONS "Significant defects" diagnosed by EAUS or TPUS and OASI Grade 3b+ predict AI symptoms. The diagnostic performance of endoanal and exoanal ultrasound (EAUS and TPUS) appear to be very similar.