Vascular access surveillance with blood flow monitoring:: a prospective study with 65 patients

Background: Periodic intra-access blood flow rate (Q(A)) monitoring is the preferred method for vascular access (VA) surveillance (NKF-K/DOQI, update 2000). Objectives: 1) To determine the ultrafiltration (UF) method accuracy for early detection of VA stenosis. 2) To evaluate the hemodynamic effect of elective VA intervention (angioplasty or surgery). 3) To define the impact of periodic QA monitoring using the UF method combined by elective VA intervention on VA thrombosis. Patients and methods: We prospectively monitored Q(A) during hemodialysis (HD) in 65 ESRD (mean age 64.9 +/- 11.4 years, 20% diabetes) patients over 1 year period. All patients undergoing HD in the Hospital de Mollet by arteriovenous fistula (89.2%) or graft 10.8%. Q(A) was measured at least every 4 months by the UF method using the Crit Line III Monitor. Fifty (77%) patients were included at the beginning of the study period and the remaining 15 (23%) were added later when they started HD. All patients with absolute Q(A) <700 ml/min or decreased >20% from baseline met criteria of positive evaluation (PE) and were referred for angiography (AG) plus subsequent preventive intervention (angioplasty or surgery) if VA stenosis greater than or equal to50%. We also studied 94 not Q(A) monitored patients since the beginning of the study period (mean age 64.6 +/- 13.7 years, 12.8% diabetes) that undergoing HD simultaneous in the Institut Nefrologic Granollers. Results: We performed 200 Q(A) measurements in 509 months of follow-up. The overall mean Q(A) was 1176.7 +/- 491.8 ml/min (range, 380.5-2904.0 ml/min). Three patients (4.6%) thrombosed VA. Nineteen (29.2%) patients had PE; none of them clotted VA. The AG was performed in 84.2% (16119) patients with PE and all of them (16116) showed VA stenosis greater than or equal to50%; 31.2% (5/16) were lost to follow-up (3 death, 2 transplantation); of the remaining explored patients (11/16), 72.7% (8/11) underwent intervention (3 angioplasty, 5 surgery). The mean Q(A) increased from 577.2 +/- 108.2 ml/min to 878.1 +/- 264.4 ml/min postintervention (p=0.005). The positive predictive value, negative predictive value, sensitivity and specificity of UF method for VA stenosis were 84.2%, 93.5%, 84.2% and 93.5%, respectively. VA thrombosis rate in our 50 beginners Q(A) monitored patients (mean age 64.5 +/- 11.4 years; 20% diabetes) was lower (2/50, 4%) compared to 94 not Q(A) monitored patients (16/94, 17%) (p=0. 024). Conclusions: 1) Q(A) monitoring using the UF method allows an early diagnosis of VA stenosis. 2) Serial Q(A) measurement by UF method can be used in assessing the functional response to corrective VA intervention. 3) Periodic VA surveillance by Q(A) measurements using the UF method combined with elective intervention results in reduced VA thrombosis.