[Assessment of micturition cystourethrography. Intra- and inter-observer variation]

The reliability of any method of investigation depends upon the accuracy and reproducibility of the results of the investigation. The accuracy of voiding cystoureterography (VCU) which is greatly dependent on the radiographic assessment cannot be assessed because no standard answers exist. The reliability of the method may be assessed in the form of intra- and inter-observer variations. VCU investigations from 24 women with incontinence were assessed by two independent radiologists. Fifteen of the radiological examinations had previously been described by one of the radiologists so that intra-observer variation could be assessed by these. Inter-observer variation was 70% (95% confidence limits 51-89%), calculated from the diagnoses anterior and posterior suspension defects, combined defects and normal conditions. The corresponding intra-observer variation was 53% (95% confidence limits 27-78%). The radiographic criteria for subdivision of suspension defects into anterior and posterior defects are, theoretically, very simple but appear to be difficult to attain in practice. The indications for employing a form of examination where assessment of the result of examination is so obscure should be very weighty.