Кардиоваскулярная терапия и профилактика (Apr 2020)

Non-invasive fractional flow reserve: a comparison of one-dimensional and three-dimensional mathematical modeling effectiveness

  • D. G. Gognieva,
  • E. S. Pershina,
  • Yu. O. Mitina,
  • T. M. Gamilov,
  • R. A. Pryamonosov,
  • N. A. Gogiberidze,
  • A. N. Rozhkov,
  • Yu. V. Vasilevsky,
  • S. S. Simakov,
  • F. Liang,
  • V. E. Sinitsyn,
  • V. B. Betelin,
  • D. Yu. Schekochikhin,
  • A. L. Syrkin,
  • F. Yu. Kopylov

DOI
https://doi.org/10.15829/1728-8800-2020-2303
Journal volume & issue
Vol. 19, no. 2

Abstract

Read online

Aim. Comparative analysis of the diagnostic effectiveness of onedimensional (1-D) and three-dimensional (3-D) non-invasive methods for coronary fractional flow reserve (FFR) assessment based on the coronary computed tomography angiography (CCTA).Material and methods. We carried out a retrospective analysis of CCTA data for 13 patients (men — 9, mean age — 61,07±9,73). In the original research, coronary FFR of those patients was evaluated using the original 3-D HeartFlow® Analysis followed by a standard invasive FFR assessment. We estimated coronary FFR using the 1-D algorithm of the Laboratory of Mathematical Modeling (Sechenov University) and compared the diagnostic effectiveness of these methods.Results. In per-vessel analysis, the sensitivity and specificity of the 3-D approach were 90,91% (95% confidence interval (CI) 62,26-99,53) and 20% (95% CI 0,01026-62,46, p>0,9999), respectively; in per-patient analysis — 90% (95% CI 59,58-99,49) and 0% (95% CI 0-56,15, p>0,9999), respectively; area under the ROC curve was 93,75% (95% CI 80,26-100), p=2,0431e-10. For the 1-D approach, the same parameters in per-patient analysis were 88,89 % (95% CI 56,50-99,43) and 25% (95% CI 0,01282-69,94, p>0,9999), respectively; in per-vessel analysis — 100% (95% CI 72,25-100) and 33,33% (95% CI 0,05923-70, p=0,1250), respectively; area under the ROC curve was 84,54% (95% CI 63,93-100), p=0,001. Spearman’s rank correlation coefficient between the 3-D and 1-D techniques was 0,7326 (95% CI 0,35810,9041), p=0,0017.Conclusion. Although we have obtained lower values of area under the ROC curve, the sensitivity and specificity of experimental approach, as well as the correlation coefficient between models were rather high. However, further studies with higher statistical power are required.

Keywords