Objective:To evaluate the effect of real time feedback provided by smartphone application on cardiopulmonary resuscitation (CPR) performance. Methods: Participants were randomised in two groups based on whether chest compression with or without the assistance of the smartphone application. Both groups performed hands-only CPR on a mannequin for 4 minutes. Data on CPR performance of both groups was compared. To assess the reliability the feedback value, we compared the CPR data from Skillmeter and data from smartphone. A questionnaire survey to participants about the usefulness of the application was also evaluated. Results: Twenty-one subjects were recruited for the study. We found no significant difference in mean chest compression rate (103.35.0/min vs. 107.11.7/min; p=0.133) and depth between the two groups (47.3 [39.3, 56.2] mm vs. 45.8 [40.3, 49.9] mm; p=0.085). The proportion of adequate compression depth over the total compression was significantly higher in the group using the smartphone (38.1% vs. 22.2%; p=0.034). The CPR data displayed on smartphone application in mannequin's chest was not different from Skillmeter software. The majority of the participants considered the application easy to use, but holding the smartphone during CPR hampered compression. Conclusions: Real-time audio-visual feedback on CPR depth and rate using a smartphone application can help to maintain the adequate chest compression depth in prolonged CPR. A better method to hold the smartphone may maximise the feedback effect on CPR quality. (Hong Kong j.emerg.med. 2014;21:153-160) 目的：探討智能手機應用程式實時反饋對 CPR 表現的影響。方法：受試者根據是否使用智能手機應用程 式幫助胸部按壓隨機分為兩組。兩組在人體模型進行心肺復甦術 4 分鐘。比較兩組的 CPR 表現數據。為 了評估反饋的可靠性，我們比較了 Skillmeter 和智能手機的 CPR 數據。以問卷調查，了解參與者對應用 程式實用性的評價。結果：二十一受試者納入研究。我們發現兩組之間在平均胸部按壓率（ 103

We report a case of hydrothorax after receiving intravenous (IV) contrast material by power injector through right subclavian central venous catheter (CVC) line. A 38-year-old woman presented to the local emergency department with hypotension after a pedestrian accident. After resuscitation, CVC was inserted into her right subclavian vein and fluid was administered well before computed tomography (CT) enhancement. Contrast-enhanced CT scan showed a large amount of extravasation of contrast material and fluid collection in the thoracic and pericardial cavities which was not shown in non-enhanced CT scan. During operation, vascular perforation was found in right subclavian vein. This case highlights that emergency physicians must keep in mind the possibility of vessel injury after CVC insertion and contrast material is preferably injected via peripheral IV access.