To determine the clinical predictors of Neisseria gonorrhoeae infection in children examined for sexual abuse.
Retrospective review of a prospective management plan.
A 240-bed children's hospital with 36,000 emergency department visits per year.
In 1988, a Pediatric Emergency Medicine Department protocol was introduced for the examination of children who present with complaints suggestive of sexual abuse.
From January 1990 through December 1991, the records of all children less than 12 years of age examined for suspected sexual abuse were reviewed. Vaginal/urethral, oral, and rectal cultures for N gonorrhoeae were performed in 316 children. Seven children (2.2%) had a total of 12 positive cultures: seven vaginal/urethral, four rectal, and one oral. Evidence of vaginal/urethral discharge on physical examination was the best predictor of N gonorrhoeae infection (sensitivity, 100%; specificity, 88%; positive predictive value, 16%; and negative predictive value, 100%). Historical and physical evidence of discharge was significantly associated with N gonorrhoeae infection (P < .0006 and P < .000001, respectively).
Children less than 12 years of age examined for sexual abuse who did not have evidence on physical examination of vaginal or urethral discharge were found to have a 100% probability of having negative vaginal/urethral, oral, and rectal N gonorrhoeae cultures. These findings do not support the practice of obtaining cultures for N gonorrhoeae routinely in all children who present for evaluation of possible sexual abuse.