Vascular complications associated with intraaortic balloon pump supported percutaneous coronary intervention (PCI) and clinical outcomes from the British Cardiovascular Intervention Society National PCI Database.

Affiliation

Kinnaird T(1)(2), Anderson R(1), Gallagher S(1), Sharp ASP(1)(3), Farooq V(1), Ludman P(4), Copt S(5), Curzen N(6), Sirker A(7), Nolan J(8), Mamas M(2)(6).
Author information:
(1)Department of Cardiology, University Hospital of Wales, Cardiff, UK.
(2)Keele Cardiovascular Research Group, Institute of Applied Clinical Sciences, University of Keele, Stoke-on-Trent, UK.
(3)The University of Exeter, Exeter, UK.
(4)Department of Cardiology, Queen Elizabeth Hospital, Birmingham, UK.
(5)Division of Statistics, Biosensors SA, Morges, Switzerland.
(6)Department of Cardiology, University Hospital NHS Trust, Southampton, UK.
(7)Department of Cardiology, St. Bartholomews Hospital, London, UK.
(8)Royal Stoke Hospital, UHNM, Stoke-on-Trent, UK.

Abstract

INTRODUCTION: The impact of a vascular complication (VC) in the setting of intraaortic balloon pump (IABP) supported PCI on clinical outcomes is unclear. METHODS: Using data from the BCIS National PCI Database, multivariate logistic regression was used to identify independent predictors of a VC. Propensity scoring was used to quantify the association between a VC and outcomes. RESULTS: Between 2007 and 2014, 9,970 PCIs in England and Wales were supported by IABP (1.6% of total PCI), with 224 femoral VCs (2.3%). Annualized rates of a VC reduced as the use of radial access for PCI increased. The independent predictors of a VC included a procedural complication (odds ratio [OR] 2.9, p < .001), female sex (OR 2.3, p < .001), PCI for stable angina (OR 3.47, p = .028), and use of a glycoprotein inhibitor (OR 1.46 [1.1:2.5], p = .04), with a lower likelihood of a VC when radial access was used for PCI (OR 0.48, p = .008). A VC was associated with a higher likelihood of transfusion (OR 5.7 [3.5:9.2], p < .0001), acute kidney injury (OR 2.6 [1.2:6.1], p = .027), and periprocedural MI (OR 3.2 [1.5:6.7], p = .002) but not with adjusted mortality at discharge (OR 1.2 [0.8:1.7], p = .394) or 12-months (OR 1.1 [0.76:1.56], p = .639). In sensitivity analyses, there was a trend towards higher mortality in patients experiencing a VC who underwent PCI for stable angina (OR 4.1 [1.0:16.4], p value for interaction .069). Discussion and Conclusions Although in-hospital morbidity was observed to be adversely affected by occurrence of a VC during IABP-supported PCI, in-hospital and 1-year survival were similar between groups.

OUR Recent Articles