Outcomes Following VASCADE® Closure Device Versus Manual Compression in Peripheral Vascular Interventions: A Propensity Matched Study
Authors: Yulanka Castro-Dominguez, MD; Ahmad Arham, MD; Zain Ahmed, MD; Aaron Brice, MD; Yasin Hussain, MD; Steffne Kunnirickal, MD; Keith Love, MD; Carlos Mena-Hurtado, MD; Kim Smolderen, MD; Sameer Nagpal, MD.
Vascular Outcomes Research Program, Yale School of Medicine
Category: Peripheral Artery Disease
Word count: 399 (limit is 400 words)
Background: Patients with peripheral arterial disease may be at higher risk of complications from use of vascular closure device (VCDs) after interventional procedures. The aims of this study were to describe the patient characteristics by use of VASCADE® (Cardiva Medical, Inc; Santa Clara, California) closure device vs manual compression (MC) for hemostasis in patients undergoing peripheral vascular interventions (PVI) from a large single center database to facilitate real-world comparative effectiveness research
Methods: We reviewed all patients undergoing PVI from a single-center electronic medical record database from January 2014 to September 2020. This institutional database was then linked with the Vascular Quality Initiative registry for the corresponding procedures using indirect patient identifiers. Access-site hemostasis method (MC vs. VASCADE) was derived from the linked dataset. Baseline demographic and procedural characteristics were compared before and after propensity score matching. Student t test and Chi-square tests were used for continuous and categorical variables, respectively. A 1:1 propensity matching algorithm was applied using 19 demographics and medical history variables to balance characteristics between both groups. A precision of 0.01% for finding the nearest propensity score matching was applied. Success of matching was assessed via visual inspection of the graphical distribution of scores and repeating the descriptive analyses to examine differences in demographic and procedural characteristics.
Results: After linking the datasets, a total of 1493 unique procedures were included. Of these, 714 (47.8%) received MC and 779 (52.2%) received a closure device, 404 (51.9%) of which were VASCADE. Before matching, patients receiving VASCADE had a higher BMI (28.8 vs 27.8; p = 0.03), were more likely to have diabetes (63.1% vs. 56.7%; p = 0.04), and more than 1 access site (11.5% vs 6.9%; p = 0.03). They were also less likely to have had a prior inflow endovascular intervention (12.4% vs 20.1%; p<0.01), prior lower extremity bypass (2.0% vs 6.9%; p<0.01), and common femoral endarterectomy (4.3% vs 13.6%; p<0.01. Propensity score matching yielded 381 well balanced pairs with no significant differences, except prior history of common femoral endarterectomy, prior leg bypass and more than one access site (Table 1).
Conclusion: VASCADE closure device is commonly used for access hemostasis in patients undergoing PVI. There are differences in comorbidities and prior vascular interventions among patients receiving VASCADE vs MC. Propensity score matching to account for baseline differences in between treatment groups is feasible and will allow for comparative effectiveness research on VCDs in this higher risk cohort.
Table 1. Baseline Patient and Procedural Characteristics
| Prior to Propensity Matching | After Propensity Matching | |||||
| MC (n=714) | VASCADE (n=404) | p-value | MC (n=381) | VASCADE (n=381) | p-value | |
| Demographics | ||||||
| Age, yrs | 70.5 ± 11.0 | 70 ± 11.2 | 0.03 | 69.2 ± 10.6 | 69.3 ± 11.1 | 0.87 |
| White race | 567 (79.4) | 304 (75.4) | 0.24 | 302 (79.3) | 289 (75.9) | 0.59 |
| Hispanic ethnicity | 70 (9.8) | 32 (7.9) | 0.29 | 30 (7.9) | 32 (8.4) | 0.79 |
| Female | 280 (39.2) | 142 (35.1) | 0.18 | 139 (36.5) | 138 (36.2) | 0.94 |
| BMI, kg/m2 | 27.7 ± 6.0 | 28.8 ± 10.4 | 0.03 | 28.6 ± 6.4 | 28.3 ± 6.1 | 0.54 |
| Comorbidities | ||||||
| Hypertension | 680 (95.2) | 387 (96.0) | 0.54 | 364 (95.5) | 366 (96.1) | 0.86 |
| Diabetes | 405 (56.7) | 255 (63.1) | 0.04 | 241 (63.3) | 235 (61.7) | 0.65 |
| Chronic lung disease | 201 (28.2) | 87 (21.5) | 0.02 | 87 (22.8) | 84 (22) | 0.79 |
| ESRD on dialysis | 42 (5.9) | 27 (6.7) | 0.59 | 24 (6.3) | 23 (6) | 0.88 |
| Smoking history | 570 (79.9) | 314 (77.7) | 0.38 | 308 (80.8) | 299 (78.5) | 0.42 |
| Coronary artery disease | 415 (58.1) | 231 (57.3) | 0.79 | 233 (61.2) | 223 (58.5) | 0.46 |
| Prior CABG | 181 (25.4) | 106 (26.3) | 0.74 | 105 (27.6) | 104 (27.3) | 0.94 |
| Prior PCI | 232 (32.6) | 125 (30.9) | 0.57 | 132 (34.6) | 120 (31.5) | 0.40 |
| Congestive heart failure | 162 (22.7) | 88 (21.8) | 0.73 | 94 (24.7) | 83 (21.8) | 0.35 |
| CVA | 76 (17.5) | 49 (17.8) | 0.92 | 42 (18.9) | 45 (17.6) | 0.71 |
| PAD history | ||||||
| Prior PVI inflow treatment | 143 (20.1) | 50 (12.4) | <0.01 | 53 (13.9) | 50 (13.1) | 0.75 |
| Prior CFA endarterectomy | 59 (13.6) | 12 (4.3) | <0.01 | 28 (7.3) | 11 (4.3) | <0.01 |
| Prior CEA/CAS | 111 (15.5) | 50 (12.4) | 0.15 | 56 (14.7) | 49 (12.9) | 0.46 |
| Prior leg PVI | 379 (53.2) | 188 (46.5) | 0.03 | 188 (49.3) | 185 (48.6) | 0.82 |
| Prior leg bypass | 49 (6.9) | 8 (2.0) | <0.01 | 26 (6.8) | 8 (2.1) | <0.01 |
| Prior major amputation | 6 (0.8) | 6 (1.5) | 0.32 | 5 (1.3) | 5 (1.3) | 1 |
| Prior minor amputation | 65 (9.1) | 34 (8.4) | 0.7 | 35 (9.2) | 31 (8.1) | 0.61 |
| Procedural characteristics | ||||||
| Indication | ||||||
| Claudication | 396 (55.5) | 222 (55.0) | 0.87 | 212 (55.6) | 211 (55.4) | 0.94 |
| Rest pain | 66 (9.5) | 38 (9.9) | 0.85 | 26 (7) | 36 (9.9) | 0.15 |
| Tissue Loss | 100 (14.4) | 50 (13.0) | 0.61 | 60 (16.1) | 49 (13.4) | 0.31 |
| Acute ischemia | 8 (1.2) | 4 (1.0) | 0.86 | 2 (0.5) | 4 (1.1) | 0.40 |
| Urgency | <0.01 | 0.65 | ||||
| Elective | 680 (95.4) | 399 (98.8) | 379 (99.5) | 378 (99.4) | ||
| Urgent or Emergent | 33 (4.6) | 5 (1.2) | 2 (0.5) | 3 (0.8) | ||
| >1 access site | 52 (7.3) | 47 (11.6) | 0.02 | 26 (6.9) | 44 (11.5) | 0.03 |
| Ultrasound guided access | 25 (34.7) | 19 (31.1) | 0.66 | 15 (38.5) | 18 (31.6) | 0.49 |