Nazarian, S., et al. A Prospective Evaluation of a Protocol for Magnetic Resonance Imaging of Patients with Implanted Cardiac Devices. Annals of Internal Medicine; 2011;155:415-424.
Due to comorbidities, up to 75% of patients with implanted cardiac devices develop an indication for MRI, but MRI is avoided in most patients due to safety concerns. This study was a non-randomized, two-center trial that evaluated the actual effects of MRI on 438 prospectively enrolled patients with pacermakers (54%) and defibrillators (46%) from 02/2003 to 04/2010. Exclusion criteria included devices implanted prior to 1998, new devices (<6wks old: prone to dislodgement), those with abandoned or epicardial leads (prone to heating), and patients who were pacemaker dependent with and ICD (lack the asynchronous mode used for study patients in MRI). Vitals and cardiac monitors were observed by a nurse with backup from an electrophysiologist. Outcome measures were “reset” events, patient symptoms, and device variables (lead impedance, sensing, capture threshold, battery voltage). Scans included brain, heart, spine, abdomen/pelvis and extremities. Immediate and 6-month evaluations were undertaken. Long-term follow up was limited by 15% death (all deemed unrelated to pacer malfunction) and 25% loss to follow up. In 3 patients, the device reset to a backup mode without any immediate or long-term effects. There were both immediate and long-term changes in device variables; however, none required revision or reprogramming at long term follow up. The study concluded that MRI can be done safely in patients with selected cardiac devices given appropriate programming, though EP monitoring during MRI is essential given potential reset, programming changes or device variable changes. Essentially, know that MRI is an option and likely to be safe in most devices when set correctly, but that one shouldn’t feel confident ordering studies without EP backup immediately at hand.

- Niedermeyer

Puskarich, M et al.  Association between timing of antibiotic administration and mortality from septic shock in patients treated with a quantitative resuscitation protocol.  Crit Care Med. 2011 Sept; 39(9): 2066-71.
Guidelines recommend initiating broad-spectrum antibiotics within the first hour of recognizing severe sepsis and septic shock based on expert consensus and one large retrospective study, however optimal timing of antibiotic administration and its impact on outcome remain unclear. 
The aim of this study was to evaluate if the timing of antibiotic administration in relation to both triage time and time of shock recognition was associated with in hospital mortality.  It was conducted as a preplanned analysis of a recently completed prospective, parallel-group, nonblinded, randomized clinical trial of early sepsis.  From January 2007-January 2009 in 3 US EDs, 291 patients were enrolled using the following inclusion criteria: Age >17 yrs, had confirmed or suspected infection, two or more SIRS criteria, and hypoperfusion evidence by hypotension after fluid challenge or blood lactate >4.  Patients were excluded if they received antibiotics prior to arrival to the ED. 
The primary outcome was inhospital mortality.  As part of the protocol, all patients received broad-spectrum antibiotics according to hospital guidelines.  The study compared outcome of patients who received initial antibiotics in regards to:  1) time from triage (comparing hourly increments up to a max of 6 hrs) and 1) time of shock recognition defined as SBP <90 after 20cc/kg challenge or Lactate≥4 (comparing antibiotics given before or after shock recognition and hourly increments after recognition, up to 3 hours).  Results showed no increase in mortality with delays in antibiotic administration up to 6 hours after triage.  It did show an increase in mortality with antibiotic administration after shock recognition, when compared to antibiotic administration before shock recognition (23.8% and 11.8% respectively; odds ratio 2.35; 95%CI, 1.12-4.53).  However, among patients who received antibiotics after shock recognition, mortality did not change with hourly delays of antibiotic administration. 
Limitations of this study include that the setting of the study- in hospitals with experience with early quantitative resuscitation protocols and the results may not be generalizabe to hospitals without protocols, the majority of patients received antibiotics within 3 hrs, making it difficult to draw conclusions regarding longer time points, relatively small size, and the inability to identify the exact time of onset of shock.  Also to note, the only outcome was inhospital mortality, not length of stay, etc. 
In summary, inhospital mortality did not increase with delay in antibiotic administration from time of triage up to 6 hours, but did show an increase in mortality if given after shock recognition compared to before.  Guidelines recommend administering antibiotics within 1 hour of recognizing severe sepsis and septic shock; this article argues that once it is recognized, mortality does not increase with delay in antibiotics up to 3 hours after recognition, however if possible antibiotics should be administered before shock is recognized to reduce mortality.  This, however, will not change my practice, as I will continue to administer antibiotics as promptly as possible upon recognizing any severe infection, sepsis or shock. 

- Shutz

Holmes, et al. “Do Children with blunt Head Trauma and Normal Cranial Computed Tomography Scan Results Require Hospitalization for Neurologic Observation?” Annals of Emergency Medicine. 2011;58:315-322.
This was a prospective, multicenter observational cohort study looking at outcomes of children who had a normal head CT after sustaining head trauma. Their question, as suggested by the title, was whether these children could be safely discharged home, or if they required hospitalization and monitoring for further complications. Methods: This study utilized the PECARN network, and involved 25 EDs across the US from June 2004 through September 2006.  Children included in the study were those under 18 who sustained blunt head trauma (isolated or multisystem), and were undergoing cranial CT. Patients were excluded if their initial GCS was <14, if they had any traumatic findings on initial head CT, or if there was a history of coagulopathy or VP shunt. Traumatic findings included extra-axial blood, SAH, IVH, IPH/contusion, edema, DAI, pneumocephalus, or skull fracture.  Outcome measures were (1) traumatic findings on subsequent CT or MRI, and (2) neurosurgical intervention. Results: A total of 13,543 met the inclusion/exclusion criteria and were enrolled in the study.  Researchers were able to obtain successful telephone or mail follow-up with 79% of patients that were discharged home.
NPV for neurosurgical intervention with negative initial head CT and GCS 15 was 100% (95% CI 99.97 to 100%). NPV for neurosurgical intervention with negative initial head CT and GCS 14 was also 100% (99.6% to 100%).  Limitations: Not all patients enrolled received follow-up, so it is possible that some of those lost to follow-up had repeat imaging and/or neurosurgical intervention.  Not all patients received repeat imaging, so more patients may have developed traumatic findings (though it is unlikely that these would lead to neurosurgical intervention.  Researchers did not address reason for admission, so it is possible that these patients were admitted for reasons other than head trauma (other trauma, social reasons, etc.). Conclusions: This study shows that few children with initial normal head CT undergo further imaging, and that very few will have traumatic findings on subsequent imaging.  More importantly, this study demonstrates that children with minor head trauma and normal head CT, are at low risk for further neurosurgical intervention, and that hospitalization for serial neurologic exams is unnecessary. 

- Radke