BACKGROUND: Two-dimensional strain echocardiography (2DS) has been used to assess ventricular function in several disease states. In previous studies of 2DS, strain analysis was usually performed offline by experienced echocardiographers. The applicability of 2DS in busy clinical labs would be enhanced if 2DS could be reproducibly measured by sonographers at the time of the echo exam. In this study we compared the reproducibility of strain measurements between sonographers at the time of the echo exam with those performed offline by an experienced echocardiographer. METHODS: Apical left ventricular (LV) B-mode images were acquired in 98 consecutive patients being evaluated for aortic stenosis. 2DS analysis was performed at the time of the exam by a sonographer. The same images were analyzed offline by an experienced echocardiographer. Global longitudinal strain (GLS) results were analyzed for interobserver reproducibility. Additionally, the regional longitudinal strain (RLS) of 20 randomly selected patients was analyzed for intraobserver reproducibility. RESULTS: Acceptable data quality was available in 97.8% of the segments measured at the time of the exam and in 96.9% at the workstation. Interobserver reproducibility of the global peak strain was high (r = 0.855, P < 0.001). Additionally, applying cutoffs for separating normal from abnormal GLS revealed good agreement between sonographer and experienced echocardiographer [kappa analysis (κ= 0.739, P < 0.001)]. Overall RLS intraobserver reproducibility was high (raw mean adjusted r = 0.915). CONCLUSION: The GLS in aortic stenosis patients can be reliably measured at the bedside by a sonographer without additional benefit of offline analysis.
BACKGROUND: Two-dimensional strain echocardiography (2DS) has been used to assess ventricular function in several disease states. In previous studies of 2DS, strain analysis was usually performed offline by experienced echocardiographers. The applicability of 2DS in busy clinical labs would be enhanced if 2DS could be reproducibly measured by sonographers at the time of the echo exam. In this study we compared the reproducibility of strain measurements between sonographers at the time of the echo exam with those performed offline by an experienced echocardiographer. METHODS: Apical left ventricular (LV) B-mode images were acquired in 98 consecutive patients being evaluated for aortic stenosis. 2DS analysis was performed at the time of the exam by a sonographer. The same images were analyzed offline by an experienced echocardiographer. Global longitudinal strain (GLS) results were analyzed for interobserver reproducibility. Additionally, the regional longitudinal strain (RLS) of 20 randomly selected patients was analyzed for intraobserver reproducibility. RESULTS: Acceptable data quality was available in 97.8% of the segments measured at the time of the exam and in 96.9% at the workstation. Interobserver reproducibility of the global peak strain was high (r = 0.855, P < 0.001). Additionally, applying cutoffs for separating normal from abnormal GLS revealed good agreement between sonographer and experienced echocardiographer [kappa analysis (κ= 0.739, P < 0.001)]. Overall RLS intraobserver reproducibility was high (raw mean adjusted r = 0.915). CONCLUSION: The GLS in aortic stenosispatients can be reliably measured at the bedside by a sonographer without additional benefit of offline analysis.