Auditing the outcome from vascular surgery with regard to graft occlusion is made difficult by variations in the type of surgery performed and the case mix. These difficulties are compounded when attempting to compare units. In the present study we have attempted to develop a scoring system to predict the risk of graft occlusion, and thus compensate for these variables. Prospectively collected data from 214 consecutive patients undergoing vascular reconstructive surgery (233 arterial grafts) were analysed. Graft occlusion occurred in 82 patients (35.2%). Using a multivariate linear regression analysis of these data a five-factor, five-grade scoring system has been devised (GORA: Graft Occlusive Risk Assessment). Logistic regression analysis of the observed risk of occlusion with this derived score produced the following relationship between the odds ratio of occlusive risk and GORA score: (logeR/1 - R = (0.229 x score) - 4.165). The score was then validated in a different group of 186 patients (196 arterial grafts). In both groups the score was found to predict accurately the risk of graft occlusion (P < 0.001). There was no significant difference in the receiver operating characteristic curves between the estimation and validation groups.

Full text is available as a scanned copy of the original print version. Get a printable copy (PDF file) of the complete article (644K), or click on a page image below to browse page by page. Links to PubMed are also available for Selected References.