Is computed tomography of the body overused? An audit of 2068 attendances in a large acute hospital

Computed tomography (CT) delivers a substantial proportion of the medical radiation dose to the population. The National Radiological Protection Board (NRPB) recommends that "CT examinations should only be carried out after there has been proper clinical justification for the examination of each individual by an experienced radiologist". The aim of this study was to review the indications for body CT in a large acute hospital in order to identify groups of patients in whom CT might not be "clinically justified". We reviewed the age, sex and diagnosis of all patients undergoing CT of the abdomen and pelvis (including lumbar spine) in our department in 1992. All requests had been checked by a senior radiologist before booking. CT was performed in 1078 male patients and in 990 females. Patients examined for benign disease were younger (mean age 52 years) than patients with malignant disease (mean age 57 years). Of 915 examinations performed for benign disease, 293 (32%) were orthopaedic in nature and included 255 (87%) for back pain. CT was carried out for benign non-orthopaedic disease in 97 women under 50 years old, usually following other tests with inconclusive or normal results. 49 of these examinations were reported as normal. Of 191 patients who had more than one CT examination, 158 (83%) were undergoing follow-up for malignant disease. All 39 patients who underwent CT on three or more occasions had life-threatening conditions, either malignancy or complicated acute pancreatitis. We have been unable to find substantial numbers of patients in whom CT was performed inappropriately. Substitution of magnetic resonance imaging (MRI) for CT of the spine will help population dose reduction. The benign condition responsible for the largest radiation dose from CT is complicated acute pancreatitis, and it may be possible to substitute MRI for CT in these patients. Women under 50 years old with normal CT form an important group for clinical audit.