Keywords:
Performed at one institution, Not applicable, Prospective, Haemorrhage, Arteriosclerosis, Puncture, Arterial access, Angioplasty, Percutaneous, Fluoroscopy, Catheter arteriography, Vascular, Interventional vascular, Arteries / Aorta, Interventional Radiology
Authors:
T. Tsochatzis1, S. C. Spiliopoulos1, G. festas1, L. Reppas1, F. Christidi1, K. Palialexis1, E. Brountzos2; 1Athens/GR, 2Haidari/GR
DOI:
10.26044/ecr2020/C-12253
Results
Bleeding complications of peripheral EVT: The overall bleeding complication rate was 3.8% (20/530). Seventeen (3.2%) events were considered major and included active retroperitoneal bleeding in 4 (0.7%) patients and pseudoaneurysms in 13 (2.4%). The three minor complications were individual instances (0.2% each) of vessel perforation, large groin hematoma, and a large hematoma with arteriovenous fistula. One life-threatening retroperitoneal hemorrhage occurred during stent post-dilation in the external iliac artery; it was immediately resolved using a stent-graft. The other three retroperitoneal bleeding events were accessing site-related; in all, 16 (3.0%) major complications were related to the access site. The annualized incidences of overall bleeding and major bleeding complications were 1.9% and 1.6%, respectively. All bleeding events were successfully managed, and the bleeding-related mortality was 0%. Three bleeding events were identified during the procedure (before manual compression), and 15 were identified during the first 8 hours of hospitalization. The other 2 events occurred after discharge. One was a 2.3-cm pseudoaneurysm that clinically manifested with acute pain and access site swelling 6 days after the procedure. The other case involved active retroperitoneal bleeding 3 days after the procedure, detected following acute abdominal pain and hemoglobin decrease to 6.5 mg/dL from 10.1 mg/dL at baseline. No intracranial hemorrhage or gastrointestinal bleeding events were noted throughout the 30-day follow-up period. No bleeding event was noted in patients receiving DAPT with ticagrelor or those receiving warfarin and 100 mg/d ASA.
HAS-BLED Score: The main factors contributing to the HAS-BLED score are reported in Table 2. According to univariable analysis (Table 2), female sex(p<0.001) and age ≥75 years (p=0.01) were correlated with a significantly increased bleeding risk. Based on multivariable regression analysis, age ≥75 years was correlated with a significantly increased bleeding risk (HR 3.32,p<0.02), male sex (HR 0.193,p<0.001) and statin therapy(HR 0.245,p=0.01) were correlated with a significantly decreased bleeding risk. Hypertension, BMI, antegrade CFA access, and DAPT were not correlated with bleeding events. The diagnostic accuracy of the HAS-BLED score in detecting peripheral bleeding complications was not significant [area under the curve (AUC)=0.512,p=0.859; Figure 1A].
Newly Proposed PBS: The newly proposed PBS was formed from 6 covariates with a range of 0 to 8, with higher scores indicating an increased bleeding risk (Table 3). The mean PBS value in the cohort was 3.2±1.5; data per score value are reported in Table 5. According to the ROC analysis, the PBS demonstrated a sensitivity of 75.0 and a specificity of 78.6in detecting peripheral bleeding complications for a PBS value >4 (AUC=0.805,p<0.001; Figure 1B). The diagnostic performance of the PBS for different values is reported in Table 4.