Acute scrotum in children is generally a symptom. Several pathologies of the testis, scrotum or groin may cause it, including inflammatory as well as ischaemic processes. Testicular torsion with ischaemia of the testis is an emergency requiring prompt surgical intervention. Even experienced paediatric surgeons and urologists may have difficulties differentiating an ischaemic from an inflammatory aetiology solely on the basis of clinical signs and symptoms. In cases of aetiologic uncertainty, operative exploration has to be performed to rule out ischaemia, which led in the past to a high rate of surgery. More recently, colour Doppler ultrasonography (CDUS) has become an imaging modality for the examination of the acute scrotum with the purpose of detecting ischaemia, thus reducing the need for explorative surgery. In this study we examined the reliability of CDUS, comparing the sonographical diagnoses with both the intraoperative findings and the clinical course. We present here 132 consecutive cases of acute scrotum in children, who were admitted from 01/98 through 10/02 to our hospital. All patients underwent CDUS. Patients with ultrasonographic signs of ischaemia of the testis (group A) and patients with demonstrated perfusion and suspicious or severe clinical symptoms (group B) were operated immediately. Patients with demonstrated perfusion without suspicious or severe symptoms (group C) were treated conservatively. Thirty-eight patients (28.8%) were treated operatively. Eleven cases of testicular torsion in group A (12 pat.) could be confirmed intraoperatively, in one case there was no testicular torsion but a severe inflammation of the testis with torsion of the appendix testis. In group B (26 pat.) operative exploration revealed one case of testicular torsion. In group B two cases of sonographically suspected torsions of the appendix testis also emerged as epididymitis intraoperatively. In the other cases postoperative diagnosis was in agreement with the preoperative CDUS. The clinical course of 94 conservatively treated patients (71.2%) (group C) confirmed in all cases a non-ischaemic aetiology. There was a total of 12 cases (9.1%) where acute scrotum was caused by testicular torsions. This relatively low rate underlines the necessity to differentiate between ischaemic and other aetiologies, thus avoiding operation in many patients. 71.2% of our patients could be treated conservatively. However, it is essential to exclude an ischaemic cause before opting for conservative therapy. Even though CDUS is a very valuable diagnostic tool in this sense, this exclusion cannot be ensured in every case and is highly dependent on the expertise and technique of the investigator. Therefore in cases with any doubts an operative exploration has to be performed to avoid the most dreaded complication in the treatment of acute scrotum.