DOI: 10.1161/circ.148.suppl_1.14742 ISSN: 0009-7322

Abstract 14742: Clinical Predictors of Bleeding Complications Following Percutaneous Ventricular Assist Device Insertion: Insights From the J-PVAD (Japanese Registry for Percutaneous Ventricular Assist Device) Registry

Madoka Sano, Toshiaki Toyota, Ryosuke Murai, Taiji Okada, Yasuhiro Sasaki, Tomohiko Taniguchi, Junichi Ooka, Kitae Kim, Atsushi Kobori, Natsuhiko Ehara, Makoto Kinoshita, Yutaka Furukawa,
  • Physiology (medical)
  • Cardiology and Cardiovascular Medicine

Introduction: Bleeding complications are significant concerns in clinical use of percutaneous ventricular assist devices (PVAD). This study aimed to identify predictors of bleedings in PVAD use.

Methods: We analyzed 3059 cases from the J-PVAD registry, a comprehensive national registry of PVAD use in Japan, excluding 53 cases without follow-up data. Patients were stratified based on whether they had acute coronary syndrome (ACS) on the onset or not, and incidence of bleeding was evaluated within each stratum, considering the clinically selected variables, including other mechanical circulatory support (MCS) use. Risk factors were explored through multivariate analyses. The primary outcome measure was any bleeding, while major bleeding was defined as a composite of cerebral bleeding, severe bleeding (defined as life-threatening, hospitalization-requiring, disability-causing, or congenital anomaly-inducing bleeding events), and blood transfusions.

Results: The median follow-up period was 32 days. The cohort included 1459 patients with ACS and 1575 without ACS. The incidence of any bleeding did not differ significantly between the ACS and Non-ACS stratum (30-day incidence: 25.4% and 27.6%, P = 0.57, respectively). We observed the same trends for major bleeding. Most any bleeding or major bleeding occurred on the day of insertion of PVAD or within a few days. The groups with MCS had significantly more bleeding complications than those without MCS in each stratum (Figure). In a multivariate analysis using the Cox proportional hazards model, adding MCS on PVAD was the most potent predictor of any bleeding in both strata, with the effects of other variables being modest by comparison.

Conclusions: Bleeding complications post-PVAD insertion are often severe and occur shortly after the procedure. Although the incidence of bleeding complications did not differ significantly according to the onset type, the MCS use strongly predicted bleeding complications.

More from our Archive