THE RELIABILITY OF X-RAY MEASUREMENTS TAKEN USING THE PACS (PATIENT ARCHIVING AND COMMUNICATION SYSTEM) OF ACUTE DISTAL RADIUS FRACTURES.

Abstract

Sixteen observers measured eight anatomic parameters on digitalised images of six acute distal radial fractures using the Patient Archiving Communication System (PACS) software and repeated the measurements two weeks later. Inter and intra observer variability for each parameter was calculated using intraclass correlation coefficients (ICC) and tolerance limits (TL). Highest inter-observer agreement was demonstrated in dorsal tilt (ICC 0.858; TL ± 14.2°) with poor agreement on the size of the gap and step. When compared with the results of a similar study published 10 years ago looking at observer variability in x-ray measurement of healed distal radial fractures, the reliability of computerized measurements is not significantly different to those achieved by manual techniques (dorsal tilt inter-observer TL on PACS ± 16° compared with TL ± 15° using ruler and protractor). These results suggest the current guidelines in the literature for acceptable radiological reduction limits based on < 10° change in palmar tilt, < 2mm radial shortening, < 5° change in radial angle and < 1–2mm articular step for acute distal radius fractures cannot be reliably measured