Ketamine with and without midazolam for emergency department sedation in adults. Sener S, Eken C, Schultz CH, Serinken M, Ozsarac M. Ann Emerg Med. 2011 Feb;57(2):109-114. This is a prospective, double-blind, placebo-controlled, that used a 2×2 factorial trial at a single center.  The study enrolled consecutive patients and after exclusion, 182 total ED patients in study 18-50 years old and ASA class I or II only. Study subjects had a mean age of 35, 70% were male and the sedation was usually for an orthopedic procedure. Patients with significant cardiovascular disease, central nervous system lesions or injuries, psychiatric disorders, pregnancy, ocular trauma or disease, thyroid disease, acute pulmonary infections, conditions requiring stimulation of the posterior oropharynx and those that had solid food in the previous 4 hours or clear liquids in the previous 2 hours. The incidence of recovery agitation in adults receiving ketamine (1.5mg/kg IV or 4mg/kg IM) with and without concurrent administration of IV midazolam (0.03mg/kg). In addition, they compared the incidence of adverse respiratory events, nausea and vomiting between groups receiving IV and IM ketamine. Respiratory adverse events was defined as pulse oximetry less than 90%, apnea for >20 seconds or laryngospasm. Recovery agitation was 8% versus 25% in the IV treatment groups receiving midazolam compared to IV ketamine without midazolam (total difference 17%; [95% CI] 6%-28%) resulting in a NNT of 6. The overall reduction of agitation was similar in the IM groups (9% with midazolam and 28% without midazolam) although sedation time in the IM group was greatly increased with midazolam (from 36 to 49 minutes. Overall the sedation times in the IV groups were similar with or without midazolam (24 minutes). Other adverse events (respiratory, nausea and vomiting) were similar between IV and IM Ketamine. Limitations include: the exclusion of patients 50 years and older and blinding may have been penetrable based on time of onset of action regarding which drug was or was not given (IV Ketamine faster onset than IM ketamine). Also, perceived agitation was binary coding, so minimal agitation may have been recorded as agitation when in fact it was minimal. This means estimation of agitation may be greater than actual clinical effect and thus reducing the true difference. In conclusion, the results of this could greatly expand the role of IV ketamine use for adults who require moderate sedation. Low incidence of respiratory compromise, easy dosing make this an ideal agent if it were not for the concerns about agitation and emergence. 

- Noel Hastings 

Silbergleit RS et al.  Intramuscular versus intravenous therapy for prehospital status epilepticus.  N Engl J Med 2012;366:591-600. This was a multicenter, double-blind, randomized, non-inferiority trial that compared the efficacy of pre-hospital intramuscular midazolam to that of intravenous lorazepam in the treatment of status epilepticus.  Also known as the Rapid Anticonvulsant Medication Prior to Arrival Trial (RAMPART), the study was conducted through 33 EMS agencies and 79 receiving hospitals across the US.  Inclusion criteria included (1) all adults and (2) children estimated to weigh 13kg or more, (3) having convulsive seizures for at least 5 minutes at the time of EMS arrival.  Exclusion criteria included seizures precipitated by major trauma, hypoglycemia, cardiac arrest, or heart rate less than 40; known allergy to midazolam or lorazepam; pregnancy; prisoner; being treated as part of another study; or opted out by wearing an alert tag marked “RAMPART declined.” All patients received, in double-blind fashion, an IM injection of either midazolam or a placebo, followed by IV injection of either lorazepam or placebo.  The dosage for <40kg was 5mg midazolam or 2mg lorazepam; for >40kg, it was 10mg midazolam or 4mg lorazepam.  The study kits also included a voice recorder; paramedics verbalized when they administered the IM treatment, when they obtained IV access, and when they administered the IV medication.  Obtaining intraosseous access was considered equivalent to IV access for this study.  There were 448 patients (IM midazolam) and 445 patients (IV lorazepam), included in “intention to treat” analysis; they also excluded some patients for a “per-protocol” analysis, due to incorrect dosage given, eligibility violations, or incorrect administration of study medication.  As the primary outcome, for those receiving IM midazolam, seizures were absent on ED arrival in 73%, compared to 63% for those receiving IV lorazepam (p<0.001 for noninferiority AND for superiority).  These results were similar for the intention-to-treat and per-protocol groups.  It was faster to administer the IM dosage, but it took longer to act; overall, the time period to cessation of seizures was similar.  There were more treatment failures due to inability to administer the IV lorazepam; 31 patients did not receive the medication because of failure to obtain access. 5 patients receiving IM midazolam did not receive therapy because of malfunctioning injectors.  As secondary outcomes, there were slightly lower rates of hospitalization and ICU admission among those receiving IM midazolam.  The length of hospital stay, length of ICU stay, frequency of intubation, and recurrent seizures were all similar and not statistically different between the two groups.  This study supports the use of IM midazolam as a more reliable, faster, and at least as (if not more) effective therapy for status epilepticus.  It does not require IV placement, and also does not require refrigeration like lorazepam does, thus is easier to stock. Though the intranasal and buccal routes are also possible, there is the possibility of the patient spitting/blowing the medication out with these routes.  The main limitation of this study is that those patients who did receive IV lorazepam, had slightly more delay than typical because all patients received IM injection first.  It might have been more effective if administered earlier.  Also, the dosages used in this study are higher than we typically use for our patients in status.  Nevertheless, this study seems to clearly support the use of IM midazolam, and would be directly applicable to our EMS system.  We should also consider it as an effective option for patients in the ED with status and poor IV access.

- Dan Naylor