Object Most CSF-diverting shunt systems have an access port that can be percutaneously tapped. Tapping the shunt can yield valuable information as to its function and whether an infection is present. The fear of causing a shunt infection by tapping may limit the physician's willingness to do so. The authors of this study investigate the risk of infecting a shunt secondary to percutaneous tapping. Methods Following institutional review board approval, CSF specimens obtained from tapping an indwelling CSF-diverting shunt during the 2011 and 2012 calendar years were identified and matched with clinical information. A culture-positive CSF sample was defined as an infection. If results were equivocal, such as a broth-only–positive culture, a repeat CSF specimen was examined. The CSF was obtained by tapping the shunt access port with a 25-gauge butterfly needle after prepping the unshaven skin with chlorhexidine. Results During the study period, 266 children underwent 542 shunt taps. With 541 taps, no clinical evidence of a subsequent shunt infection was found. One child's CSF went from sterile to infected 11 days later; however, this patient had redness along the shunt tract at the time of the initial sterile tap. Conclusions The risk of infection from tapping a shunt is remote if the procedure is done correctly.

IntroductionThe traditional methods of communication between nurses and physicians through paging or phone calls have been known to cause patient care interruptions, increase stress and workload, and cause burnout. Our study assessed the impact of using an electronic health record–based messaging system (Epic Secure Chat) for nonemergent communication between nurses and physicians in the emergency department (ED).MethodThis study was performed at a large urban academic ED. Surveys were distributed through e-mail using Google forms. Preimplementation and postimplementation surveys were performed from May 2019 (presurvey) to August 2019 (postsurvey).ResultThe number of nonurgent phone calls from registered nurse staff decreased after intervention (P < 0.001). The frequency of nonurgent calls disrupting workflow decreased after the intervention (P = 0.029). The number of calls that were appropriately alerted to increased but was not significant (P = 0.120), whereas the degree of burnout from nonurgent calls remained relatively the same (P = 0.841).ConclusionsThe findings provide insight into the implementation of an EMR messaging system in an ED setting. These findings suggest the messaging system can decrease unnecessary nonurgent calls from registered nurses and medical doctors, allowing for more efficient workflow in the ED. These findings suggest further research especially regarding burnout implemented on a larger scale and yield redesign suggestions in the ED.