Skip to main content
Department of Emergency Medicine

Department of Emergency Medicine

2013 - August R3 Journal Review

Holmes JF et al. Identifying Children at Very Low Risk of Clinically Important Blunt Abdominal Injuries. Ann Emerg Med. 2013 August 62(2).107-16.
This was a study with the goal of deriving a prediction rule to identify children with blunt torso trauma who are at very low risk for intra-abdominal injury requiring acute intervention who can safely not undergo CT scans. The study was a prospective, observational cohort study of children with blunt torso trauma in the Pediatric Emergency Care Applied Research Network (PECARN). The study was conducted at 20 participating emergency departments of children who sustained blunt torso (thorax and abdomen) trauma from May 2007 to January 2010. Inclusion criteria: decreased level of consciousness in association with blunt torso trauma, paralysis, multiple nonadjacent long bone fractures, MVC >40mph, ejection or rollover, auto vs ped/bike >5mph, falls >20ft high, crush injury to torso, physical assault involving the abdomen, physician concern for abdominal trauma resulting in any of the following screening tests: abdominal CT, FAST, labs, CXR or pelvic XR. Patients were excluded if they had injury >24 hours before presentation, penetrating trauma, preexisting neurologic disorders impeding reliable exam, known pregnancy or transfer from another hospital with CT or DPL results. Physicians recorded history and physical exam findings before CT (if needed). All CT results were from the site’s board certified radiologist. Medical records were reviewed of all hospitalized patients. A telephone interview was conducted at least seven days after the original ED visit to identify patients who later received a diagnosis of intra-abdominal injury. For those who were not able to be contacted, a mail survey was sent and if needed the medical records, quality improvement records and local morgue records were reviewed to identify patients who were later diagnosed with IAI or died.

The outcome of interest was IAI (intra-abdominal injury) requiring acute intervention (defined at IAI causing death, laparotomy, angiographic embolization, blood transfusion or needing IVF for 2 or more nights with pancreatic or GI injuries). IAI was defined as radiologically or surgically discovered injury to the spleen, liver, urinary tract, GI tract, pancreas, gallbladder, adrenals, intra-abdominal vasculature or traumatic fascial defect.

The data underwent statistical analysis using binary recursive partitioning to determine a clinical decision rule. Of the 14,882 patients eligible, 12,044 patients were enrolled. Median age was 11.1 years old. 50% of the children were admitted and 50% discharged home from the ED. Telephone follow-up was successful for 74% of patients discharged from the ED and the remainder had follow-up as described above. Abdominal CT scans were obtained for 46% of patients. 761 patients (6.3%) were diagnosed with IAI with the highest number of injuries to the spleen (39%) and liver (37%). The primary outcome of IAI which underwent acute intervention was identified in 203 patients. Sixteen of the 6,053 patients (0.3%) initially discharged from the ED were later found to have IAI with 2 requiring acute intervention with one of them who had a CT scan which was read as normal initially. The prediction rule showed that the children most at risk for IAI consisted of the following 7 variables in descending order of importance: evidence of abdominal wall trauma or seat belt sign, GCS <14, abdominal tenderness, evidence of thoracic wall trauma, complaints of abdominal pain, decreased breath sounds, and vomiting. Of the six patients who were considered very low risk (no rule variables present) and had IAI, five of the six had lab abnormalities. Therefore those children without any history or physical exam findings in the prediction rule are at very low risk of 0.1% for IAI needing acute intervention. The rule had a sensitivity of 92.5%, specificity of 44.1%, a PPV 10%, and a NPV 98.9% to identify children with IAI. This study was limited by the lack of use of labs or FAST but this rule is based entirely on history and physical which makes it very generalizable to clinicians who might not have labs or ultrasound at their disposal. Of note, the rule is not meant to scan every child with one variable present but to aid in not scanning those children without any variables present as these children have less than a 1% chance of having an IAI requiring intervention.

- Laws


Westfall M et al. Mechanical Versus Manual Chest Compressions in Out-of-Hospital Cardiac Arrest: A Meta-Analysis.  Crit Care Med  July 2013.41(7):1782-9
I chose this study because we’ve all seen on the ride-a-long with EMS – lack of resources and people on hand to help.  They need IV access, BVM, continuous chest compressions someone to drive to the nearest hospital at the very least!  There is a reason ACLS has changed from ABC’s to the CAB’s…

This study was a meta-analysis, based on 12 studies based on mechanical vs manual cardiopulmonary resuscitation.  Eight of the studies compared “load-distributing band” CPR vs manual while the remaining four compared “piston-driven” CPR vs manual.  All cases met the requirement of being initiated in the pre-hospital setting.  Of note, these studies were composed of a mix of model types including either randomized, historical or case-control studies.  Overall, 6,538 subjects were included amongst the 12 studies.  Primary endpoint in all studies was ability to achieve ROSC.

The relative effect of each treatment to achieve ROSC was based on a random effects model.    Out of the 6,538 subjects in the studies collectively, 1,824 had ROSC events (palpable pulse and measurable BP for at least one minute). 

In summary: 
• General Mechanical CPR > manual CPR
o (odds ratio [OR], 1.53 [95% CI, 1.32, 1.78]; p < 0.001)
• load-distributing band CPR > manual CPR
o (odds ratio, 1.62 [95% CI, 1.36, 1.92], p < 0.001).
• piston-driven CPR ≈ manual CPR
o (OR, 1.25 [95% CI, 0.92, 1.68]; p = 0.151)
• ROSC rates for load-distributing band = 8.3%
• ROSC rates for piston-driven CPR = 5.2%

As a meta-analysis there, are inherent limitations – although tests were done to limit the heterogeneity of studies, there will always be differences in the details of studies that may include confounders.  Ideally, we’d be comparing apples to apples – instead, we may be comparing a Granny Smith apple to a Fuji apple.  One example is that some of the piston-driven studies were done in Europe, while many of the band studies were done in the US – may account for differences in protocol, response times, etc.  Another limitation is that obtaining ROSC has many other contributors besides just good CPR (time down, bystander initiated CPR, early defibrillation if indicated, reversible causes, etc).  Also, information about quality of manual CPR was not documented – however, this may reflect real life variances. 

If there is a light-weight, easy to use device out there, I feel a mechanical CPR device deserves a large randomized controlled trial – it will likely benefit the patient, pre-hospital medics and ED physician. 

- Murphy



Zahed R et al. A new and rapid method for epistaxis treatment using injectable form of tranexamic acid topically: a randomized controlled trial. Am J Emerg Med. July 2013. 31(9): 1389-92.
This was an prospective, unblinded, single-center, randomized controlled trial that compared the efficacy of topical tranexamic acid (TXA) to anterior nasal packing (ANP) in the treatment of anterior epistaxis. The study was performed in one Iranian ED by EM residents all trained in the application of either cotton pledgets soaked in injectable TXA(500mg per 5ml) or “usual” anterior nasal packing with cotton pledgets soaked in epinephrine (1:100000) + lidocaine (2%) and followed by cotton pledgets covered with tetracycline. The study had quite significant exclusions, including major trauma, posterior epistaxis, known history of bleeding disorder, international normalized ratio greater than 1.5, shock, and visible bleeding vessel. Of note, the study did include patients with recurrent treatment for anterior epistaxis.

In total, 216 patients were randomized to the two groups, with 107 allocated to the TXA group and 109 to the “usual” ANP group. The investigators assigned four primary outcomes to establish comparative efficacy, with the first primary result showing that within 10 minutes of treatment, bleedings were arrested significantly more in the TXA group compared to the ANP group (71% to 31.2%). In addition, a significantly larger percentage of TXA group patients were discharged within 2 hours compared to the ANP group (95.3% to 6.4%). The study also showed a significant difference in rebleeding rates at 24 hours and 1 week, with the TXA group being superior to the ANP group at both end points (4.7% to 12.8%, 2.8% to 11%). Finally, their last primary outcome showed significantly higher patient satisfaction scores in the TXA group compared to the ANP group.

There were no significant differences in minor complications between the two groups and there were no serious adverse events recorded during the study. Of note, the two groups were comparable with regard to basic characteristics without significant differences, except the TXA group had a significantly higher percentage of patients with a history of epistaxis (58.1% to 13.6%).

This study concludes that the injectable form of TXA, when applied topically, may be a superior treatment to their “usual” anterior nasal packing for idiopathic anterior epistaxis. The authors describe some of their study’s limitations, including their inability to be truly blinded due to the practical difference in the two nasal packing therapies, as well as their exclusion of posterior epistaxis. They also mention the fact that they did not stratify the severity of anterior epistaxis and acknowledge this shortcoming as a limitation. However, the significant, but subtle difference in their patient demographics, with the TXA group containing a much larger percentage of patients with a history of epistaxis(as mentioned above), suggests a possible difference in epistaxis severity and etiology between their two randomized groups.

In regards to practical application of this study to our patient population and our regular practice, I concur that topical TXA may be a promising treatment option for epistaxis. However, their “usual” anterior nasal packing may not be the standard of practice for every emergency physician, with topical alpha agonists (oxymetazoline), cocaine, silver nitrate, surgicel, and nasal tampons(rhino rocket) all evidence-based therapies. Additional studies comparing TXA to some of these therapies could further establish TXA as viable alternative or even as an adjunct. Finally, this study chose to exclude various higher risk attributes and etiologies of epistaxis, but the speed and simplicity of topical TXA suggests that TXA may be ideal as either the first-line treatment of as an adjunct for these exact situations, with further studies likely indicated.

- Colby