Yanni Evan, Tsung James W, Hu Kevin, Tay Ee Tein
Department of Emergency Medicine, NYU Langone Medical Center, New York, NY.
Department of Emergency Medicine, Mount Sinai Health System, New York, NY.
Ann Emerg Med. 2023 Nov;82(5):566-572. doi: 10.1016/j.annemergmed.2023.04.003. Epub 2023 Jun 2.
This study aimed to determine the level of agreement among pediatric emergency medicine (PEM) physicians in whether various point-of-care ultrasound (POCUS) video clips represent cardiac standstill in children and to highlight the factors that may be associated with the lack of agreement.
A single, online, cross-sectional, convenience sample survey was administered to PEM attendings and fellows with variable ultrasound experience. PEM attendings with an experience of 25 cardiac POCUS scans or more were the primary subgroup based on ultrasound proficiency set by the American College of Emergency Physicians. The survey contained 11 unique, 6-second video clips of cardiac POCUS performed during pulseless arrest in pediatric patients and asked the respondent if the video clip represented a cardiac standstill. The level of interobserver agreement was determined using the Krippendorff's α (Kα) coefficient across the subgroups.
A total of 263 PEM attendings and fellows completed the survey (9.9% response rate). Of the 263 total responses, 110 responses were from the primary subgroup of experienced PEM attendings with at least 25 previously seen cardiac POCUS scans. Across all video clips, PEM attendings with 25 scans or more had an acceptable agreement (Kα=0.740; 95% CI 0.735 to 0.745). The agreement was the highest for video clips wherein the wall motion corresponded to the valve motion. However, the agreement fell to unacceptable levels (Kα=0.304; 95% CI 0.287 to 0.321) across video clips wherein the wall motion occurred without the valve motion.
There is an overall acceptable interobserver agreement when interpreting cardiac standstill among PEM attendings with an experience of at least 25 previously reported cardiac POCUS scans. However, factors that may influence the lack of agreement include discordances between the wall and valve motion, suboptimal views, and the lack of a formal reference standard. More specific consensus reference standards of pediatric cardiac standstill may help to improve interobserver agreement moving forward and should include more specific details regarding the wall and valve motion.
本研究旨在确定儿科急诊医学（PEM）医生对于各种床旁超声（POCUS）视频片段是否代表儿童心脏停搏的一致程度，并突出可能与缺乏一致性相关的因素。
对具有不同超声经验的PEM主治医师和住院医师进行了一项单一的在线横断面便利样本调查。根据美国急诊医师学院设定的超声熟练程度，有25次或更多心脏POCUS扫描经验的PEM主治医师是主要亚组。该调查包含11个独特的、时长6秒的儿科患者心脏停搏期间进行的心脏POCUS视频片段，并询问受访者该视频片段是否代表心脏停搏。使用Krippendorff's α（Kα）系数确定各亚组间的观察者间一致性水平。
共有263名PEM主治医师和住院医师完成了调查（回复率为9.9%）。在263份总回复中，110份回复来自有至少25次既往心脏POCUS扫描经验的经验丰富的PEM主治医师这一主要亚组。在所有视频片段中，有25次或更多扫描经验的PEM主治医师有可接受的一致性（Kα=0.740；95%CI 0.735至0.745）。壁运动与瓣膜运动相对应的视频片段的一致性最高。然而，在壁运动发生而瓣膜运动未发生的视频片段中，一致性降至不可接受的水平（Kα=0.304；95%CI 0.287至0.321）。
对于有至少25次既往报告的心脏POCUS扫描经验的PEM主治医师，在解读心脏停搏时观察者间总体有可接受的一致性。然而，可能影响缺乏一致性的因素包括壁运动和瓣膜运动之间的不一致、欠佳的视野以及缺乏正式的参考标准。更具体的儿科心脏停搏共识参考标准可能有助于提高未来观察者间的一致性，并且应包括关于壁运动和瓣膜运动的更具体细节。
