Author + information
- Sohrab Fratz, MD, PhD⁎ ( and )
- Heiko Stern, MD, PhD
- ↵⁎Department of Pediatric Cardiology and Congenital Heart Disease, Deutsches Herzzentrum München (German Heart Center Munich), Lazarettstrasse 36, 80636 Munich, Germany
We want to compliment Clarke et al. (1) for their important work of thoroughly assessing the accuracy and reproducibility of right ventricular (RV) volume measurements by cardiac magnetic resonance in congenital heart disease. However, we think that 2 points warrant further comment.
First, Clarke et al. (1) considered a difference of 10 ml/m2 of different measurement methods of RV stroke volume to be clinically significant. Taking into account that the mean RV stroke volume in their study was around 60 ml/m2 with minimal values of around 20 ml/m2, a difference of 10 ml/m2 is fairly large. Translated into percentage values, this would amount to a 17% difference for the mean values and up to 50% for the smallest RV stroke volumes. Although clinical significance or relevance is a subjective measure, we think that 17% to 50% is too large. This problem could be addressed by comparing the percentage differences of the methods to each other.
Second, Clarke et al. (1) defined the first phase of each cine image as the end diastole. By doing this in patients with volume-loaded RV with wide QRS complexes, one will inadvertently underestimate the end-diastolic volume and stroke volume. This is because in patients with wide QRS complexes the computed trigger signal will be later than the upstroke of the QRS complex. Accordingly, the end diastole will not be found in the first phase of the cine image but in one of the last phases. Therefore, RV stroke volume is very likely to be significantly underestimated in this patient population as depicted in Figure 2 and Table 3 of Clarke et al. (1). Therefore, we suggest that the phase of the end diastole should be defined visually by the observer as the phase with the largest volume (2,3).
- American College of Cardiology Foundation