Anatomical risk models for paravalvular leak and landing zone complications for balloon-expandable transcatheter aortic valve replacement

Catheter Cardiovasc Interv. 2017 Oct 1;90(4):690-700. doi: 10.1002/ccd.26987. Epub 2017 May 4.

Abstract

Background: Though several anatomical characteristics have been reported separately as risk factors for paravalvular leak (PVL) and landing zone (LZ) complications after transcatheter aortic valve replacement (TAVR), multivariate risk models are needed.

Methods: Patients that underwent balloon-expandable TAVR with multidetector cardiac computed tomography (MDCT) sizing were studied. MDCT images were analyzed and the association between anatomical factors and ≥mild PVL, ≥moderate PVL, and LZ complications (annular rupture, requirement of new permanent pacemaker, and coronary obstruction) was determined, and subsequently competing predictive models were developed and validated.

Results: A total of 316 consecutive TAVR patients were included. Median age was 82.0 years (74.0-87.0) and STS score was 8.3% (5.4-10.9). Factors associated with ≥mild PVL included TAVR with Sapien/Sapien XT vs. Sapien 3 (OR = 2.50, 95% CI = 1.24-5.07), LVOT nontubularity (OR = 1.02, 95% CI = 1.01-1.04), LZ calcification (OR = 1.01, 95% CI = 1.00-1.01), and low cover index (OR = 0.94, 95% CI = 0.91-0.96). Factors associated with LZ complications included LZ calcification (OR = 1.01, 95% CI 1.00-1.01), leaflet asymmetry (OR = 1.01, 95% CI 1.01-1.02), and cover index (OR = 1.09, 95% CI 1.03-1.14). Predictive models for ≥mild PVL (AUC = 0.71, 95% CI = 0.66-0.77), ≥moderate PVL (AUC = 0.75, 95% CI = 0.65-0.84), and LZ complications (AUC = 0.77, 95% CI = 0.67-0.87) were created using procedural details and anatomical data from the MDCT. Clinical variables were not included as they were poorly correlated with the occurrence of PVL and LZ complications. For each outcome, the area under the curve (AUC) of the multivariate model was superior to the model consisting only of individual factors.

Conclusions: A model using procedural/anatomical characteristics derived from MDCT predicts ≥mild PVL, ≥moderate PVL, and LZ complications post-TAVR. Incorporation of anatomical risks into clinical practice may help stratify patients before TAVR. © 2017 Wiley Periodicals, Inc.

Keywords: annular rupture; annular sizing; conduction disturbances; paravalvular leak; transcatheter aortic valve replacement.

MeSH terms

  • Aged
  • Aged, 80 and over
  • Aortic Valve / diagnostic imaging
  • Aortic Valve / physiopathology
  • Aortic Valve / surgery*
  • Aortic Valve Insufficiency / diagnostic imaging
  • Aortic Valve Insufficiency / etiology*
  • Aortic Valve Insufficiency / physiopathology
  • Balloon Valvuloplasty / adverse effects*
  • Clinical Decision-Making
  • Female
  • Heart Valve Prosthesis
  • Humans
  • Logistic Models
  • Male
  • Multidetector Computed Tomography
  • Multivariate Analysis
  • Odds Ratio
  • Retrospective Studies
  • Risk Assessment
  • Risk Factors
  • Transcatheter Aortic Valve Replacement / adverse effects*
  • Transcatheter Aortic Valve Replacement / instrumentation
  • Treatment Outcome