FUNCTIONAL-EVALUATION OF VASCULAR ACCESSE S FOR HEMODIALYSIS

No definitive methods have been established so far to predict the fail ure of the vascular access. We have examined the functional behaviour of 45 vascular accesses, by a) doing kinetics of venous pressure (VP) and % recirculation (R) at different blood flows; b) selectively picki ng up those patients with an abnormal VP to perform a fistulogram. The results disclosed that: Over the whole of the data both VP and % R fi tted a linear distribution as a function of the blood flow (p < 0.001) . A linear relationship was found, at the analized flows, between VP a nd % R (r = 0.295, p < 0.001). However, the predictability value of th ese data was small, since the relationship between % R and VP was not linear in some individual patients. The mean VP rose with flow in the 3 types of accesses (radial and arm arteriovenous fistulae and Goretex ) but it was higher at all flows in the goretex (p < 0.01). However, t here were no differences in the slope of increase of % R between the 3 types of accesses. Two patterns were identified using the VP/Flow and % R/Flow curves: A) Normal VP with high % R. B) High VP with hig % R. Five out of the 8 patients with pattern A had an alteration on the ar terial side of the access. All of the 4 patients with pattern B had al terations on the venous side of the access. In conclusion: 1. The VP d ata are not a useful predictor of % R in individual cases. 2. The valu es of VP are higher in Goretex,accesses, compared with arteriovenous f istulae. However, there are no differences in the slope of increase of VP or % R. 3. In terms of predictability of the presence and type of vascular lesion, the pattern of enhanced % R in the absence of increas e of VP may suggest the presence of an alteration of the arterial inle t, whereas the simultaneous increase of VP and % R may suggest venous obstruction. 4. The data on VP and % R are meaningful only if the flow at which they were measured is indicated.