Background: The diagnostic accuracy of ultrasonography for differentiating between benign and malignant adnexal masses is proportional to the expertise of the operator. However, we do not know whether improved diagnostic accuracy will affect the management of these tumours. We assessed the effect of the quality of gynaecological ultrasonography on the management of patients with suspected ovarian cancer in a randomised controlled trial.
Methods: 165 patients who were referred to the regional gynaecological cancer centre at Guy's and St Thomas' NHS Foundation Trust (London, UK), between June 7, 2004, and April 23, 2006, with suspected adnexal tumours met the inclusion criteria. Of these, 150 patients were randomly assigned to either level II (routine) ultrasonography (n=73) or to level III (expert) ultrasonography (n=77). The primary endpoint was the number of major surgical staging procedures (including a laparotomy and at least an oophorectomy and omental biopsy) in each group of the study. Secondary endpoints were: total number of surgical procedures; number of minimally invasive procedures (eg, operative laparoscopy or ultrasonography-guided cyst aspiration); number of additional diagnostic tests (eg, CT or laparoscopy); number of follow-up scans; diagnostic accuracy of level II and level III ultrasonography; and duration of hospital stay. All analyses were by intention to treat. This study is registered on the Current Controlled Trials website http://www.controlled-trials.com/mrct/trial/230201/ISRCTN02631195.
Findings: More major surgical staging procedures for suspected ovarian cancer were done in the level II group than in the level III group of the study (27 of 73 [37%] vs 17 of 77 [22%], respectively; difference between groups 15% [95% CI 0-29]; RR 1.68 [1.00-2.81]; p=0.049). The total number of surgical procedures was similar between the two groups: 35 of 73 (48%) in the level II group and 33 of 77 (43%) in the level III group (RR 1.12 [0.79-1.59]; p=0.53). The median duration of hospital stay for patients who were operated on was 6 days (range 3-13) in the level II group and 5 days (range 1-9) in the level III group (p=0.01). A likely histological diagnosis was provided to clinicians after ultrasonography for 76 of 77 (99%) patients in the level III group compared with only 38 of 73 (52%) patients in the level II group. 18 of 150 (12%) patients recruited were eventually diagnosed with ovarian malignancy. The sensitivity and specificity of ultrasonography was 2 of 5 (40%; [95% CI 6.5-84.6]) and 10 of 10 (100%; [34-100]), respectively, in the level II group and 7 of 8 (88%; [47-98]) and 27 of 28 (96%; [82-99]), respectively, in the level III group.
Interpretation: Improved quality of ultrasonography has a measurable effect on the management of patients with suspected ovarian cancer in a tertiary gynaecology cancer centre, and results in a significant decrease in the number of major staging procedures and a shorter inpatient hospital stay.