Single-heartbeat 3D echocardiography (3DE) is more reliable than four-heartbeat 3DE for measuring left ventricular (LV) volume and ejection fraction in patients with atrial fibrillation, a study published Oct. 5 in Cardiovascular Ultrasound concluded. The authors found that single-heartbeat 3DE resulted in reduced inter- and intraobserver variation, required less time and reduced the obstruction caused by stitching artifacts.
Four-heartbeat 3DE is more commonly used to image LV volume and ejection fraction (EF) in patients with atrial fibrillation. Four-heartbeat 3DE involves taking ultrasound images of four cardiac cycles during end expiration breath-hold, whereas single-beat 3DE requires only one cardiac cycle to image left ventricle volume and function, requiring substantially less time than four-heartbeat 3DE.
Despite the higher spatial and temporal resolution of four-heartbeat 3DE, the authors sought to compare single-heartbeat with four-heartbeat 3DE because “the recently developed single-heartbeat method has some potential advantages. ... This modality will further advance assessment of LV by improving the speed of acquisition and reducing stitching artifacts. This is especially true for patients in atrial fibrillation,” wrote Kambiz Shahgaldi, PhD, of Karolinska University Hospital Huddinge in Stockholm, and colleagues.
Shahgaldi and colleagues used four-heartbeat and single-heartbeat 3DE imaging modalities on 70 patients, 50 with sinus rhythm (Group A) and 20 with atrial fibrillation (Group B). The researchers measured LV volume and EF, as well as inter- and intraobserver variability. Single-heartbeat imaging immediately followed the four-heartbeat 3DE and all images were analyzed by two readers, twice each at one-week intervals (to measure inter- and intraobserver variability). Each reader was blinded to the other’s measurements.
The authors found significant variations in measurement of end-diastolic volume (EDV), end-systolic volume (ESV) and EF for four-heartbeat images of Group B compared to single-heartbeat. Interobserver variability for EDV, ESV and EF measurements in Group B using four-heartbeat reached 10.4 percent, 15.2 percent and 17.9 percent, respectively. EDV, ESV and EF interobserver variations for single-heartbeat images were only 7.6 percent, 7.2 percent and 5.6 percent.
Intraobserver variability for EDV and EF measurements of Group B using four-heartbeat 3DE showed statistically significant variations as well, with mean values of 9 percent and 8.3 percent, respectively. These same intraobserver measurements using single-heartbeat imaging did not vary significantly from one another, with LV EDV and EF values of 4.5 percent and 4.8 percent, respectively.
ESV values for the same observer did not vary significantly in Group B, and no statistically significant intra- or interobserver differences in measurement were recorded for the sinus rhythm group (Group A).
“From a clinical point of view, perhaps the most important finding in this study is the lower variability of single-heartbeat in comparison to four-heartbeat when measuring LV volumes and EF in atrial fibrillation,” the authors highlighted. The authors also said that, with four-heartbeat necessitating imaging of four cardiac cycles, “EF measurements in atrial fibrillation patients are challenging in clinical practice due to the need of measuring an average of heart beats. …Thus, in atrial fibrillation patients particularly, four-heartbeat becomes impractical.”
The researchers recommended using a single algorithmic software program to measure LV volumes and EF in future studies in order to standardize measurements, as four-heartbeat volume and EF measurements were uniformly lower than single-heartbeat measurements in their study. “The only two possible explanations are either random findings or more probably systematic software error” the authors said, hypothesizing the latter.
Shahgaldi and colleagues ended by saying, “Single-beat 3D full volume acquisition seems to be superior to four-heartbeat ECG-gated acquisition in measuring LV volumes and EF in patients having atrial fibrillation. The variability is significantly lower both for EF and LV volumes.”