Abstract

OBJECTIVE: (1) To determine the accuracy of accident and emergency (A&E) doctors' diagnosis of radio-opaque ureteric calculi on plain abdominal radiographs; (2) to study the predictive value of haematuria with a history suggestive of ureteric colic. DESIGN: A prospective study of all patients seen in a three month period with a provisional diagnosis of ureteric colic. Intravenous urography (IVU) was used as the gold standard for diagnosis of ureteric calculi. SETTING: The accident and emergency department and medical unit of a large teaching hospital. SUBJECTS: 60 patients who were admitted with an initial diagnosis of ureteric colic, 51 subsequently undergoing intravenous urography. RESULTS: A&E doctors achieved a calculated sensitivity of 29% (95% confidence intervals 13% to 49%) and a specificity of 73% (52% to 90%) for identification of renal calculi on plain abdominal radiograph, compared with figures of 68% (48% to 84%) and 96% (78% to 100%) respectively for consultant radiologists. The difference between these results was highly significant (P = 0.0011). No patient with a definitive diagnosis of ureteric colic had a negative result for haematuria on urinary dipstick analysis. CONCLUSIONS: A&E doctors are poor at identifying radio-opaque ureteric calculi on plain abdominal radiographs. If haematuria is absent on urinalysis then ureteric colic is an unlikely diagnosis.