Article Text

Download PDFPDF

33 Absolute coronary blood flow and microvascular resistance modelling in patients with coronary artery disease
  1. Louise Aubiniere-Robb1,
  2. Paul Morris2,
  3. Julian Gunn2,
  4. Rebecca Gosling2,
  5. Rodney Hose3,
  6. Vignesh Rammohan3,
  7. Patricia Lawford3
  1. 1NHS Greater Glasgow and Clyde, Glasgow, UK
  2. 2Sheffield Teaching Hospitals
  3. 3University of Sheffield


Background Ischaemic heart disease causes reduced coronary blood flow (CBF). CBF is not routinely measured in the catheterisation laboratory. Cardiologists therefore rely upon surrogate indices of CBF like fractional flow reserve (FFR). Whilst valuable in guiding revascularisation decisions, FFR reports CBF as a fraction of a hypothetical and unknown value. It is therefore, semi-quantitative and cannot assess microvascular physiology. FFR-guided PCI is associated with incomplete symptom resolution in around 20% of patients. These limitations may be resolved by a new method (virtuQ™) which computes absolute coronary blood flow (aCBF) and coronary microvascular resistance (CMVR) from angiography and standard pressure wire measurements. The aims of our study were to establish the relationship between FFR and aCBF, and to investigate the contribution of CMVR to aCBF.

Methods virtuQ™ software was used to reconstruct the 3-D arterial anatomy of 229 vessels from 151 patients undergoing angiography and FFR assessment for coronary syndromes. aCBF and CMVR were computed by a numerical method based upon the reconstructed anatomy, pressure conditions and Navier-Stokes equations. The reduction in aCBF due to epicardial disease was also calculated. Percentage flow reduction (predicted by FFR) was compared to aCBF reduction in mL/min. A threshold for intervention for aCBF reduction was derived. Agreement between FFR and aCBF reduction was assessed by Cohen’s kappa (κ) statistic.

Results virtuQ™ computed all physiological parameters in 207/229 cases (90%). Physiological results are summarised in table 1. Calculated by regression intercept using an FFR-aCBF plot, the derived threshold for physiological significance for aCBF reduction was ≥23 mL/min (figure 1). Overall agreement between FFR and aCBF reduction was moderate (κ=0.70). Agreement between FFR and aCBF reduction was high in cases where FFR was >0.80 (90.0%) and perfect when FFR ≤0.70 (100%), but poor when FFR was 0.70-0.80 (68.2%). For cases in which FFR was ≤0.80 (n=109), 19.5% were discordant, associated with increased CMVR (1.24 vs 0.58 mmHg.min/mL, P<0.0001). A hybrid assessment strategy (FFR alone if >0.80 or ≤0.70, with aCBF used if FFR 0.70-0.80) increased agreement to near perfect (κ=0.90) and required aCBF to be measured in 32% of cases. CMVR was higher in females at baseline and hyperaemia (1.59 vs 1.21 and 0.86 vs 0.68 mmHg·mL·min-1, respectively, P<0.05).

Abstract 33 Table 1

Summary of physiological parameters computed in virtuQTM

Conclusions In the largest study of aCBF to date, 19.5% of cases identified as physiologically significant by FFR were identified as non-significant by aCBF reduction criteria. Pressure-flow discordance was associated with variability in CMVR and may explain why around 20% of patients experience persistent angina following FFR-guided PCI. Combined FFR + aCBF assessment may be valuable, particularly in the FFR grey zone. virtuQ™ may have a complementary role in selecting patients for PCI and help diagnose microvascular disease.

Conflict of Interest None

  • Flow
  • Resistance
  • Modelling

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.