Author + information
- Kosaku Goto, MD, PhD,
- Gary S. Mintz, MD,
- Claire Litherland, MS,
- Alexandra J. Lansky, MD,
- Giora Weisz, MD,
- John A. McPherson, MD,
- Bernard De Bruyne, MD, PhD,
- Patrick W. Serruys, MD, PhD,
- Gregg W. Stone, MD and
- Akiko Maehara, MD∗ ()
- ↵∗Cardiovascular Research Foundation, 111 East 59th Street, 12th Floor, New York, New York 10022
Using multilesion, multivessel, multi-imaging modality data from PROSPECT (Providing Regional Observations to Study Predictors of Events in the Coronary Tree) (1), we examined differences between angiographic and intravascular ultrasound (IVUS) lumen dimension measurements and factors that affected the discrepancy. Baseline grayscale and IVUS virtual histology studies in 3,017 nonculprit lesions (658 patients) were coregistered to the angiographic roadmap by an individual blinded to the IVUS and quantitative coronary angiography (QCA) analyses; IVUS and QCA measurements were performed at separate core laboratories (Cardiovascular Research Foundation, New York, New York) (2). The primary comparison was IVUS versus QCA in-lesion minimal lumen diameter (MLD) using a Passing-Bablok approach with differences between IVUS- and QCA-measured MLD assessed using Kendall’s concordance coefficients and Bland-Altman plots. A model with a generalized estimating equation approach was used to compensate for potential cluster effects of multiple lesions in the same patient. IVUS pullback lengths were left anterior descending coronary artery (LAD) 72.9 ± 26.6 mm, left circumflex coronary artery (LCX) 61.2 ± 25.1 mm, and right coronary artery (RCA) 82.9 ± 30.6 mm; distances from the ostia were grouped in tertiles to identify proximal, mid, and distal segments. Statistical analyses, except Passing-Bablok regressions, were conducted using SAS version 9.2 (SAS Institute, Cary, North Carolina); Passing-Bablok analyses were conducted using R version 3.2.1 (R Foundation for Statistical Computing, Vienna, Austria) using the ggplot2 package version 1.0.0 and mcr package version 1.2.1.
There was a strong concordance between IVUS MLD and QCA MLD (Kendall’s W = 0.89, p < 0.001) (Figure 1A); however, QCA underestimated MLDs in small vessels and overestimated MLDs in large vessels compared with IVUS. The “intersection MLD” was 3.8 mm, with smaller vessels having an MLD <3.8 mm and larger vessels an MLD >3.8 mm. The Bland-Altman plot of IVUS minus QCA MLD (Figure 1B) indicated that QCA underestimated MLD compared with IVUS: mean difference 0.21 ± 0.48 mm and median difference 0.23 mm (interquartile range [IQR]: −0.05 to 0.52]. In lesions with IVUS MLD >QCA MLD, the median difference was 0.39 mm. In lesions with IVUS MLD <QCA MLD, the median difference was −0.26 mm. The overall discrepancy was 16.5 ± 13.9% or a median of 13.1% (IQR: 6.2% to 22.5%).
There was a strong concordance between IVUS versus QCA MLD in the LAD, LCX, and RCA, separately: LAD, QCA MLD 2.55 ± 0.80 mm versus IVUS MLD 2.71 ± 0.60 mm, Kendall’s W = 0.88, p < 0.001; LCX, QCA MLD 2.50 ± 0.73 mm versus IVUS MLD 2.72 ± 0.60 mm, Kendall’s W = 0.89, p < 0.001; and RCA, QCA MLD 2.73 ± 0.70 mm versus IVUS MLD 2.99 ± 0.61 mm, Kendall’s W = 0.88, p < 0.001. There was also a strong concordance in each proximal, middle, and distal subsegment (p < 0.001; data not shown); however, the mean discrepancy between IVUS MLD and QCA MLD in the proximal LAD was smaller than the mid or distal LAD (median 0.07 mm [IQR: 0.74 mm], 0.24 mm [IQR: 0.56 mm], 0.30 mm [IQR: 0.52 mm], respectively) or the proximal, mid, or distal LCX (median 0.21 mm [IQR: 0.49 mm], 0.22 mm [IQR: 0.56 mm], 0.21 mm [IQR: 0.54 mm], respectively) or RCA (median 0.22 mm [IQR: 0.61 mm], 0.30 mm [IQR: 0.53 mm], 0.31 mm [IQR: 0.55 mm], respectively) subsegments. Conversely, scatter was greatest in the proximal LAD subsegment (range [min–max MLD difference]): LAD, proximal 3.89 mm (−2.61 to 1.29 mm), mid 3.07 mm (−1.38 to 1.69 mm), distal 3.24 mm (−1.82 to 1.42 mm); LCX, proximal 3.16 mm (−1.40 to 1.75 mm), mid 3.43 mm (−1.14 to 2.29 mm), distal 2.75 mm (−1.41 to 1.34 mm); and RCA, proximal 3.41 mm (−1.47 to 1.94 mm), mid 2.87 mm (−1.28 to 1.59 mm), distal 3.26 mm (−1.60 to 1.66 mm).
An IVUS MLD minus QCA MLD difference >0.23 mm (median) and an IVUS >QCA MLD were associated with longer lesions (p = 0.001, p = 0.0016, respectively). An IVUS MLD minus QCA MLD difference ≤0.23 mm (median) and IVUS ≤QCA MLD were associated with larger plaque burden (p = 0.0006, p < 0.0001, respectively) and a larger amount of dense calcium (p < 0.0001, p = 0.0004, respectively), but also larger reference vessels (p < 0.0001, p < 0.0001, respectively) with less severe angiographic diameter stenosis (p < 0.0001, p < 0.0001, respectively). Plaque rupture and necrotic core were predictors of only IVUS MLD minus QCA MLD >0.23 mm and ≤0.23 mm (data not shown).
Only nonculprit lesions in acute coronary syndromes patients were analyzed, and culprit lesions were not included. The size of the IVUS catheter may have contributed to some of the difference between QCA and IVUS.
In conclusion, in small coronary arteries, QCA underestimated lumen diameters compared with IVUS, whereas in large coronary arteries, QCA overestimated lumen diameters compared with IVUS. Among proximal, mid, and distal subsegments of the 3 arteries, the discrepancy between IVUS and QCA MLD was least in the proximal LAD; however, the scatter was greatest.
Please note: Dr. Mintz is a consultant for Volcano Corporation, Boston Scientific, InfraReDx, and ACIST; has received fellowship/grant support from Volcano Corporation, Boston Scientific, and InfraReDx; and has received honoraria from Boston Scientific, InfraReDx, and ACIST. Dr. Weisz is on the advisory board of Angioslide, AstraZeneca, Calore Medical, Corindus, Medivizor, and Medtronic. Dr. De Bruyne receives grant support and consulting fees to his institution from St. Jude Medical. Dr. Maehara has received grant support from Boston Scientific; is a consultant for Boston Scientific and ACIST; and has received speaker's fees from St. Jude Medical. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Sotirios Tsimikas, MD, served as the Guest Editor for this article. (PROSPECT: An Imaging Study in Patients With Unstable Atherosclerotic Lesions; NCT00180466)
- American College of Cardiology Foundation