The Society for Academic Emergency Medicine’s Annual Meeting was held June 1-5, 2011 in Boston, Massachusetts. The Research Forum featured 655 Abstracts, and 24 Innovations in Emergency Medicine education presentations. We selected the 10 we thought had the most potential to impact the practice of emergency physicians.
The Impact Rating Scale
* Interesting, practice changing for some
** Moderate impact, practice changing for many
*** High impact, practice changing for most
**** A must read, practice changing for all
***** Landmark project, practice changing for all
++ Independently selected by both authors as a “Top Abstract”
****
Association Between the Timing of Antibiotic Administration and Outcome in Patients With Septic Shock++
Puskarich et al [Abstract #1]
What were they looking for?
To determine association between time to initial antibiotics and mortality in patients treated with ED early sepsis protocol.
Methods
Multi-center, 300 patient trial. Patients with severe sepsis were included. Primary outcome was in-hospital mortality.
Results
Delay in antibiotics had no effect on mortality up to 6 hours after triage. However, delay in antibiotic administration until after shock onset as compared to before was associated with an increased likelihood of death (OR 2.4, 1.1-4.5).
Why this could change your practice
Give patients with suspected sepsis antibiotics prior to the onset of severe sepsis. This study shows the great importance of promptly recognizing and then treating early sepsis.
***
Waiting With an Emergency: Short-term Mortality and Hospital Admission Following Departure From Crowded Emergency Departments
Guttmann et al [Abstract #41]
What were they looking for?
To determine among non-admitted ED patients whether presenting during either crowded shifts or LWBS had a higher risk of death or hospitalization within 7 days of ED departure.
Methods
Observational, using databases to identify non-admitted ED patients from 2003-2008 in Ontario, Canada.
Results
Among high-acuity patients, the rate of 7-day death or hospitalization increased from 2.3% to 3.1% for patients seen during shifts with a mean ED LOS 6hrs (among low-risk patients the risk increased from 0.7% to 1.3%). LWBS was not associated with higher risk.
Why this could change your practice
The risks of ED crowding are significant and place patients who are discharged home at significantly higher risk of death or hospitalization. Surprisingly, this finding extends to low-risk discharged patients as well, underscoring the need for hospital administrators to work to reduce ED LOS as a patient safety measure.
***
Emergency Department Utilization Among Recently Insured and Uninsured Adults
Ginde et al [Abstract #160]
What were they looking for?
To compare ED usage of recently-insured and uninsured to stable insured and uninsured adults.
Methods
157,630 adult patients analyzed from the National Health Interview Study. Primary outcome was ≥1 ED visit during the past 12 months.
Results
29.5% of the recently-insured compared to 20.2% of the stable insured had at least 1 ED visit. 23.6% of the recently-uninsured compared to 17.5% of the stable uninsured had ≥ 1 ED visit. Recent change in insurance status was associated with higher ED utilization for both the recently-insured (OR 1.34) and the recently-uninsured (OR 1.29).
Why this could change your practice
Health policy changes – through altering the number of patients with insurance – could affect the number of patients coming to the ED to seek care. Increases in ED volume may be commonplace if this study’s predictions pan out.
*****
Identifying Children at Very Low Risk of Intra-abdominal Injuries (IAI) Undergoing Acute Intervention
Holmes et al [Abstract #416]
What were they looking for?
To develop a clinical prediction rule that identifies children with blunt abdominal trauma (BAT) who are at very low risk for injuries that require intervention.
Methods
12,044 pediatric patients, mean age of 9.8 yo, prospectively enrolled from 20 EDs. Acute intervention defined as therapeutic laparotomy, angiographic embolization, blood transfusion or IV fluid administration for ≥ 2 days in those with pancreatic or duodenal injuries.
Results
Rule identified 197/203 (97%, 95% CI 95-99%) of those patients with IAI who required acute intervention. The rule had a negative predictive value of 5,028/5,034 (99.9%, 95% CI 99.8-100%)).
Why this could change your practice
The clinical prediction rule derived was:
1. Abdominal pain
2. History of vomiting
3. Evidence of abdominal wall trauma
4. GCS < 14
5. Abdominal tenderness
6. Evidence of thoracic wall trauma
7. Decreased breath sounds
If these parameters were absent, the pediatric patient with BAT was at very low risk for intra-abdominal injury requiring acute intervention.
**
iPad Use at the Bedside Can Decrease Time Spent at a Computer
Horng et al [Abstract #256]
What were they looking for?
To determine whether physician iPads would decrease the time spent at a computer workstation.
Methods
Prospective cohort. Primary outcome measure was the time spent using the ED workstation computer.
Results
Clinician use of an iPad when working clinically was associated with a 39 minute decrease in time spent per shift using the computer workstation. Physicians who used the iPad logged into the computer workstation 5.1 fewer times per shift.
Why this could change your practice
The authors note the potential benefit of using an iPad would be more time spent at the bedside with the patient. This, as well as efficiency metrics, will of course need to be examined in future studies but for now iPads certainly hold some promise to get EPs out from our workstations.
**
Ibuprofen Prevents Altitude Illness: A Prospective, Double-blind, Randomized Controlled Trial
Lipman et al [Abstract #28]
What were they looking for?
To determine efficacy of treatment with ibuprofen vs placebo for acute mountain sickness (AMS).
Methods
Double-blind, randomized, placebo-controlled trial comparing 600mg of Ibuprofen taken in 4 consecutive doses starting 6 hours prior to ascent, three times daily vs placebo. Eighty-six (44 I and 42 P) low-altitude dwelling, non-acclimatized, healthy volunteers were enrolled at 4000 feet, drove to 11,700 feet, then hiked and spent the night at 12,500 feet.
Results
The 2 groups were well matched and the incidence of AMS was less in the Ibuprofen group (43% vs 69%, OR 2.94, NNT 3.9). The study investigators found no significance in prevention of headache severity or oxygen desaturation change.
Why this could change your practice
For all the weekend warriors heading off for overnight trips to the mountains, here is a nice trial assessing Ibuprofen use for the preventative treatment of AMS.. Ibuprofen is much more accessible than dexamethasone and acetazolamide. For longer trips to high altitude it is unclear if this benefit would persist with continued treatment.
***
A Triage Rule to Identify Patients in Need of an Immediate 12-Lead Electrocardiogram in
the Emergency Department to Diagnose ST-Elevation Myocardial Infarction (STEMI)
Glickman et al [Abstract #31]
What were they looking for?
To identify need for immediate triage EKG in ED for STEMI
Methods
An all-inclusive statewide ED population of 8.1 million patient visits in North Carolina was evaluated using a classification and regression tree analysis with patient age and chief complaints.
ECG triage rule:
-Age ≥ 31 with chest pain
-Age ≥ 50 with dyspnea, altered mental status, or upper extremity pain
-Age ≥ 62 with syncope or weakness
-Age ≥ 82 regardless of complaint
Results
3,575,178 patient visits were evaluated and 6,464 (0.18%) were diagnosed with STEMI. Almost 22% presented without chest pain and almost 47% presented without chest pain in the group > 80 years of age.
ECG triage rule for STEMI:
Sensitivity 91.7%
(95% CI 90.8–92.6%)
NPV 99.98%
(95% CI 99.98–99.98%) for STEMI
Why this could change your practice
When it gets busy, how can we tell who at triage needs an EKG and who needs it right away? Extended door to EKG times have been shown to delay time to reperfusion therapy and increase morbidity and mortality. Atypical presentations complicate the fact that chest pain alone cannot be the sole reason to get an EKG. Here is a potentially useful study to prioritize patients that need an immediate EKG for the evaluation of STEMI. The investigators also evaluated a more simplified version of the rule using age points of ≥ 30, ≥50, and ≥ 80. This was noted to have similar performance to the derived rule. Whether or not this changes reperfusion times needs to be determined, but intuitively it seems that it should.
****
Normalization of Vital Signs Does Not Reduce the Probability of Pulmonary Embolism (PE)++
Kline et al [Abstract #232]
What were they looking for?
To quantify and compare the change in HR, RR, and pulse oximetry (SaO2) in patients undergoing diagnostic testing for PE.
Methods
Prospective, non-interventional, single center study of ED patients with at least one symptom, one sign, and one risk factor for PE and for whom CT pulmonary angiography (CTPA) was performed.
Results
180 patients enrolled and 35 (19.4%) were PE+. At least two sets of vital signs were recorded in 174 (97%); the mean time between triage to the second (D1) and triage to the third (D2) VS sets were 176±127 and 263±144 min, respectively. No significant differences were found between PE+ vs. PE- for either the mean values for any VS, or the means of their relative changes during D1 and D2. At triage, 12/35 PE+ and 53/155 PE- patients had HR>99 and 8/12 (67%) PE+, and 33/52 (65%) PE – patients normalized their HR.
Why this could change your practice
In patients in whom you have a pretest concern for PE, normalization of the vital signs does not necessarily lower the patient’s risk of a PE. In this study, PE+ patients showed similar changes in VS over time compared with PE – patients, and similar rates of normalization.
***
Findings of Chronic Sinusitis (CS) on Brain CT
Are Not Associated With Acute Headaches
Kroll et al [Abstract #260]
What were they looking for?
To compare the prevalence of (CS) findings on CT in patients discharged with a diagnosis of non-traumatic headache (NT-HA) with those who were discharged with a diagnosis of head injury (HI).
Methods
Retrospective cross-sectional study of consecutive head CT scans obtained for patients who were discharged home with NT-HA or HI for findings consistent with CS (mucosal thickening), indeterminate (IND) (polyps, concha bullosa), or no findings of sinusitis. Patients with findings of air fluid levels were excluded.
Results
485 consecutive patients were enrolled:
225 NT-HA group and 260 HI group
In the NT-HA, 63 (28.0%) patients had positive CT scan findings (CS and IND), compared with 61 (23.5%) in the HI group, p=0.30.
Why this could change your practice
Headaches are frequently encountered problems in the ED. Findings of CS on CT scan are common. Clinicians should be cautious when attributing the cause of the headache to “sinusitis” when evidence of CS is present on CT.
***
Elderly Emergency Department Patients With Pain Are Less Likely to Receive Pain Medication: Results From a National Survey 1999–2008
Platts-Mills et al [Abstract #395]
What were they looking for?
To determine whether elderly patients presenting with pain were less likely than younger patients to receive any analgesic medication or an opioid.
Methods
Cross-sectional national survey data on ED visits over 10 years (1999–2008) collected by National Hospital Ambulatory Medical Care Survey.
Results
Patients age 75 or older with a pain-related visit were less likely to receive any analgesic (49%, 95% CI 48%-50%) than patients 18–44 (66%, 95% CI 65%-66%), 45–64 (63%, 95% CI 63%-64%), or 65–74 (55%, 95% CI 53%-56%). The same pattern of oligo-analgesia was seen for opioid medication. No analgesic medication was given to 48% of patients 75 or older with moderate pain vs. 32% of patients 18–44 and 30% of patients 75 or older with severe pain vs. 23% of patients 18–44.
Why this could change your practice
Don’t forget to medicate elderly patients with analgesics and opioids when they are in pain. This is a nice study showing that elderly adults who present to the ED with pain are less likely to receive pain medication than younger patients, even after controlling for pain severity, sex, and race.