Abstract: FR-PO765
Predicting Vascular Access Thrombosis Using Vasc-Alert™ Based Scoring Algorithm
Session Information
- Hemodialysis: Vascular Access - II
November 03, 2017 | Location: Hall H, Morial Convention Center
Abstract Time: 10:00 AM - 10:00 AM
Category: Dialysis
- 603 Hemodialysis: Vascular Access
Authors
- Sands, Jeffrey J., Renal Systems Consulting LLC, Orlando, Florida, United States
- Astor, Brad C., University of Wisconsin, Madison, Wisconsin, United States
- Hirschman, Kim, Vasc-Alert LLC, Evanston, Illinois, United States
- Kennedy, John B., Vasc-Alert LLC, Evanston, Illinois, United States
- Besarab, Anatole, Stanford University, Palo Alto, California, United States
Background
We evaluated a Vasc-Alert™ based multifactorial scoring algorithm to estimate the risk of developing vascular access (VA) thrombosis within the subsequent 60-days. Vasc-Alert™ VA surveillance utilizes data from each dialysis treatment (Tx) to calculate intra-access venous (VAPR) and arterial (AAPR) pressure ratios to identify those VA at high risk for thrombosis needing further evaluation ± elective intervention. Alerts are generated when the VAPR>0.55 or the AAPR>0.65 (AVF) and >0.60 (AVG) on 3 consecutive Tx.
Methods
We identified 985 patients (263 AVF; 722 AVG), including 304 (81 AVF, 223 AVG) who experienced VA thrombosis (TH) and 681 (182 AVF, 499 AVG) without TH (N-TH), from 86 HD facilities with electronic download of Tx and VA intervention data with up to 120 days Tx data/patient. Records were divided into 15 day intervals (total intervals = 7655; 2049 AVF and 5606 AVG) and assessed to determine outcome (TH vs N-TH) within the subsequent 60 days. Bivariate repeated-measures logistic regression identified 5 factors significantly associated (p<0.05) with TH (mean AAPR, mean VAPR, VAPR slope, #VAPR alerts, #Tx with delivered/prescribed blood flow rate <90%). Risk scores were then assigned for specific ranges of each factor based upon their predictive value for TH (0-1 for Mean AAPR, VAPR slope, Tx Blood Flow Rate (BFR)<90% prescribed; 0-3 for mean VAPR) and summed for each interval resulting in an assigned cumulative risk score from 0-7.
Results
A total of 15.3% intervals were associated with thrombosis in the subsequent 60 days. The cumulative incidence of thrombosis was greater with higher cumulative score (see table below). Scores ≤1 were associated with a relatively low incidence (9.5% AVF, 13.0% AVG) and scores ≥3 with a high incidence (AVF: 25.4%, AVG: 23.2%) of thrombosis.
Conclusion
Risk scores based upon a Vasc-Alert™ scoring algorithm successfully identified VA with low or high probability of developing thrombosis within the next 60 days. Because these scores are Tx record based, they may be easily automated to help guide VA patient care through a population management model.
Cumulative Score
| Cumulative Score | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | Total | |
| AVF | Total Intervals | 571 | 561 | 311 | 250 | 194 | 106 | 50 | 6 | 2049 |
| % Thrombosis | 7.4 | 11.6 | 16.7 | 20.8 | 25.8 | 25.5 | 44.0 | 50.0 | 15.3 | |
| AVG | Total Intervals | 940 | 2663 | 1374 | 473 | 121 | 29 | 6 | 5606 | |
| % Thrombosis | 7.9 | 14.8 | 17.6 | 20.9 | 27.3 | 37.9 | 50.0 | 15.3 | ||
Funding
- Commercial Support – Vasc-Alert LLC