I am on vacation this month and I am trying hard to make it a real vacation. So I am not reading any medical literature, even if I have a minute while wait in line at the Colosseum (yes, that has happened before.) Instead of my usual articles of the month, covering the most recent papers I have been reading, I am going to summarize a few classic emergency medicine papers. Most people probably know all of these already, but it is good to review the evidence behind our practice occasionally. Enjoy…
ARDSnet: The rise of low tidal volumes
Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome Network. N Engl J Med. 2000;342:(18)1301-8. PMID: 10793162
This is an RCT of 861 mechanically ventilated patients with ALI or ARDS, designed as a 2×2 trial (half of which examined ketoconazole, but that arm of the trial was stopped due to lack of efficacy.) They randomized patients to the now famous ARDSnet protocol of low tidal volumes to limit plateau pressures or a traditional ventilation strategy. The ARDSnet protocol resulted in a decrease in mortality (31.0% versus 39.8%, p=0.007).
Bottom line: Follow the protocol for your intubated patients. (Copy available here)
GUSTO II: Cath versus lytics
Andersen HR, Nielsen TT, Rasmussen K, et al. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med. 2003;349:(8)733-42. PMID: 12930925 [free full text]
This is a substudy of GUSTO II. It is a prospective multicenter RCT that assigned 1138 patients presenting within 12 hours of their STEMI to either primary angioplasty or thrombolytic (t-PA). For their primary outcome, a composite of death, non-fatal reinfarction, and non-fatal stroke at 30 days, angioplasty had better outcomes (8.0% versus 13.7% p<0.001). This effect was entirely from non-fatal re-infarction, as stroke and death were unchanged – a problem with composite outcomes. Interestingly, and something that we don’t tend to talk about a lot, or at least I was never taught, there was no difference in that composite outcome at 6 months (14.1 vs 16.1% statistically insignificant.)
Bottom line: Angioplasty provides some early benefit over fibrinolytics, but we may be over-emphasizing its benefit. For many centers and specific patients, lytics may still be the best option. (See, I am not just totally against t-PA. I am just for evidence.)
Analgesics for abdominal pain
Mahadevan M, Graff L. Prospective randomized study of analgesic use for ED patients with right lower quadrant abdominal pain. Am J Emerg Med. 2000;18:(7)753-6. PMID: 11103723
I only know the medical world after this study was published, but many people probably still remember the days when surgeons wouldn’t let us treat patients’ pain because it would ruin the abdominal exam. This is a randomized, double blind trial of 68 adult patients suspected of appendicitis, given either tramadol or placebo. Of course, pain was lower in the group that received pain medication (although not by a lot). Not only was the analgesic group examinable, but actually had more specific exams for appendicitis.
Bottom line: If patients are in pain, doctors treat it. I am not sure what surgeons do.
NEXUS: A pain in the neck?
Hoffman JR, Mower WR, Wolfson AB, Todd KH, Zucker MI. Validity of a set of clinical criteria to rule out injury to the cervical spine in patients with blunt trauma. National Emergency X-Radiography Utilization Study Group. N Engl J Med. 2000;343:(2)94-9. PMID: 10891516 [free full text]
Jerry Hoffman. Nexus. This is classic emergency medicine. We should all know the criteria:
- No midline cervical tenderness
- No focal neurological deficit
- Normal alertness
- No intoxication
- No painful, distracting injury
This was a prospective, multi-centre observational study that included 34,069 patients who had imaging of the cervical spine after blunt trauma and found 818 cervical spine injuries. The decision instrument was 99% sensitive (95%CI 98-99.6%) with a negative predictive value of 99.8% (95%CI 98.0-99.6%). Of course, you do have to accept the specificity of 12.9%. Only 1 of the 8 patients missed had a clinically significant injury that required a surgical intervention.
Bottom line: You can remove c-collars quickly and safely in many patients. If you are EMS, you can probably even prevent them from going on in the first place.
Cage match: NEXUS versus the Canadian C-spine rule
Stiell IG, Clement CM, McKnight RD, et al. The Canadian C-spine rule versus the NEXUS low-risk criteria in patients with trauma. N Engl J Med. 2003;349:(26)2510-8. PMID: 14695411 [free full text]
This is a prospective cohort of 8283 alert trauma patients comparing NEXUS and Canadian c-spine rule (CCR). There were 169 (2%) clinically important c-spine injuries. Unfortunately, in 10% of patients physicians did not properly apply the CCR – they did not assess range of motion as defined. Of course, if a decision instrument is easily misinterpreted (even with the Hawthorne effect of a study) that will affect its utility in practice. How you interpret this study depends entirely on what you do with those patients. If you exclude them, the CCR looks great (sensitivity of 99.4% and specificity of 45.1%). However, if you include them, the sensitivity drops to 95.3% and specificity is 50.7%. This compares with NEXUS with a sensitivity of 90.7% and a specificity of 36.8%. Obviously, neither test performed quite as well as we would hope in this cohort.
Bottom line: It is important to know the specifics of clinical decision instruments, including inclusion and exclusion criteria. I still use a combination of both these tools in clinical practice.
Dexamethasone for croup
Bjornson CL, Klassen TP, Williamson J, et al. A randomized trial of a single dose of oral dexamethasone for mild croup. N Engl J Med. 2004;351:(13)1306-13. PMID: 15385657 [free full text]
This is a multi-centre, double-blind, RCT that included 720 children with mild croup who were randomized to either dexamethasone 0.6mg/kg to a max dose of 20mg or placebo. The children receiving dexamethasone had less “return to medical care” – 7.3% versus 15.3%, p<0.001. The dexamethasone group also had slightly lower croup scores and slept about 1 hour a day more than the placebo group.
Bottom line: A NNT of 14 to prevent further visits is your primary benefit in mild croup.
Dexamethasone for croup: But what dose?
Geelhoed GC, Macdonald WB. Oral dexamethasone in the treatment of croup: 0.15 mg/kg versus 0.3 mg/kg versus 0.6 mg/kg. Pediatr Pulmonol. 1995;20:(6)362-8. PMID: 8649915
This is an RCT of admitted pediatric patients with croup comparing dexamethasone at doses of 0.15mg/kg, 0.3mg/kg, and 0.6mg/kg. There was no difference in length of hospital stay, use of epinephrine, croup scores, or representations for medical care.
Bottom line: Dexamethasone at 0.15mg/kg is probably just as good as the 0.6mg/kg we have all been taught.
Rehydration – isn’t that what the GI tract was designed for?
Fonseca BK, Holdgate A, Craig JC. Enteral vs intravenous rehydration therapy for children with gastroenteritis: a meta-analysis of randomized controlled trials. Arch Pediatr Adolesc Med. 2004;158:(5)483-90. PMID: pubmed
This is a meta-analysis of 16 RCTs involving 1545 children comparing enteral to intravenous rehydration in the treatment of gastroenteritis. (Unfortunately, I have been told by medical-legal types that I am never allowed to make the diagnosis of “gastroenteritis”, so I am not sure who I will apply this study to.) Oral rehydration has significantly fewer adverse events including death and seizure (relative risk 0.36 95%CI 0.14-0.89) and significantly reduced hospital stay (mean decrease of 21 hours). There was no difference in the treatment effect or weight gain. The failure rate for enteral therapy was 4%.
Bottom line: You should almost never place an IV in a pediatric gastroenteritis patient.
Steroids for meningitis
de Gans J, van de Beek D; European Dexamethasone in Adulthood Bacterial
Meningitis Study Investigators. Dexamethasone in adults with bacterial
meningitis. N Engl J Med. 2002 Nov 14;347(20):1549-56. PMID: 12432041 [free full text]
This is a multi-centre, prospective RCT of 301 adult patients suspected of having meningitis and having either cloudy CSF, bacteria on CSF gram stain, or a CSF white count >1000. Patients were randomized to either placebo or dexamethasone 10mg IV q6h for 4 days, with the first dose give 20 minutes before or concurrently with antibiotics (initial antibiotics treatment was with amoxicillin alone). 7% of the steroid group died as compared to 15% of placebo (p=0.04; relative risk 0.48 95%CI 0.24-0.96). There was no difference in hearing loss or focal neurologic abnormalities. Note that steroids and antibiotics were given only after waiting for the CSF results.
Bottom line: Steroids decreased mortality, but did not affect neurologic outcomes
However, although this study is considered a classic, it is at odds with the bulk of the literature.
Brouwer MC, McIntyre P, Prasad K, van de Beek D. Corticosteroids for acute bacterial meningitis. The Cochrane database of systematic reviews. 6:CD004405. 2013. PMID: 23733364
This review covers 25 studies involving 4121 participants. Steroids did NOT provide a statistically significant mortality advantage (RR 0.90, 95%CI 0.80-1.01). However, steroids did results in less hearing loss (RR 0.74 95%CI 0.63-0.87).
Bottom line: Unfortunately steroids will probably not save any lives. Given the potential delay to antibiotics if steroids are used as they were in the de Gans study, it is unclear how important the hearing changes are. The steroids for meningitis question is not definitively answered, but any benefits are likely to be small.
Sepsis: early goal directed therapy
Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345:(19)1368-77. PMID: 11794169 [free full text]
This paper is now infamous and certainly created its share of controversy. It was a randomized trial of 263 patients with severe sepsis who were randomized to a specific treatment protocol or standard care. Rivers was able to show a significant mortality benefit, 30.5% versus 46.5% (p=0.009). However, we now know that the specifics of his protocol were mostly irrelevant, you just need to care for your sepsis patients.
Bottom line: Dr. Rivers pushed sepsis care forward around the world, but there is no reason to be using this protocol anymore.
Restrictive transfusion policy
Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med. 1999;340:(6)409-17. PMID: 9971864 [free full text]
This is a multi-centre RCT based in Canada that included 838 adult ICU patients with anemia Hb≤ 90 (excluding chronic anemia and patients with active blood loss.) They were randomized to either a restrictive transfusion strategy (transfuse with a Hb <70; target 70-90) or a liberal strategy (transfuse with a Hb < 10; target 100-120). There was not a statistical significance in 30 day mortality (18.7% in restrictive versus 23.3% in liberal). The liberal group had higher in-hospital mortality and cardiac events (secondary outcomes.)
Bottom line: This was the first of many studies showing we give too much blood.
OPALS: What is the value of ACLS?
Stiell IG, Wells GA, Field B. Advanced cardiac life support in out-of-hospital cardiac arrest. The New England journal of medicine. 351(7):647-56. 2004. PMID: 15306666 [free full text]
This is a prospective multicenter before and after trial that compared outcomes with basic life support paramedic crews (who had defibrillators) to advanced crews with full ACLS training including medications. 5638 adult patients with out of hospital cardiac arrest were included. The advanced life support paramedics resulted in more ROSC (12.9% vs 18%) and more admissions to hospital (10.9% vs 14.6%), but without any change in survival to hospital discharge (5.0 vs 5.1%).
Bottom line: This is one of the many studies that indicate ACLS and particularly the use of medications in cardiac arrest don’t work, but might actually be harmful.
Cheesy joke of the month
Why didn’t skeleton cross the road?
He has no guts