The value of computer analysis in predicting the long-term outcome of deep vein thrombosis

Background The use of standardised computerised ultrasound images is an objective and quantitative method of determining the echogenicity of thrombus. This method had been applied to study the natural history of 100 acute thrombi over a period of one year to. determine if early changes in echogenicity could indicate whether the thrombus would lyse, partially recanalise or remain occlusive. Methods. A consecutive series of 100 above knee deep vein thromboses (DVT's) were analysed over a period of one year. The presence of a DVT was initially diagnosed by duplex scanning and the patients underwent follow-up scans at one week, one month, six months and at one year. A grey scale image of the thrombus was transferred to a computer at each examination and its grey scale median (GSM) was measured. The mean GSM's were calculated for each examination and compared. At one year the patients were divided into groups according to their final outcome (i.e. lysis, recanalisation or occlusion) and the mean GSM values from each group were compared. Results. There were 100 proximal DVT's from 89 patients. At one year 14% of the patients had died and 23% were lost to follow-up. The mean GSM values increased over the one year period from 25.87+/-18.33 to a final value of 64+/-25.52 at one year. A total of 21 thrombi had fully resolved but there was no significant difference in their GSM values before resolution when compared to the other patients. Twenty-four patients had partially recanalised thrombi and 18 remained totally occluded. There was no significant difference in mean GSM values between these two groups until after six months when the permanently occluded venous segments had higher GSM values than those which partially recanalised. Conclusions. Measurement of GSM is an objective method of determining the degree of organisation of a thrombus and describes the subjective changes of individual thrombi. However, the organisation of a thrombus is a dynamic pro cess and mean GSM values did not reflect these changes. Early changes in GSM could not predict the final outcome of the thrombus i.e. lysis, recanalisation or occlusion.