Detox or Dehospital?
Flower K, Past A, Sussman J, et al. Validation of triage criteria for deciding which apparently inebriated persons require emergency department care. Emerg Med J. 2011;28(7):579–584.
In major metropolitan areas, EMS is often called to assess publicly intoxicated patients who have an “altered mental status” and may or may not require emergency department (ED) care. Transport of the patients whose only condition is inebriation taxes the EMS system and unnecessarily overcrowds EDs.
Using a checklist of criteria, the crews in this study attempted to predict which inebriated patients actually needed an ED visit. I particularly like the fact that the researchers also compared the paramedics’ “gut feelings,” with the official criteria.
San Francisco paramedics completed the checklist on 99 inebriated patients during a one-month period. The retrospective records review yielded mixed results.
The majority (~80%) of the apparently inebriated patients didn’t need care in the ED. Unfortunately, trying to predict which patients needed the ED in advance proved to be difficult. The key in this prediction tool is that it needs to be “sensitive” enough to identify who needs the ED while not casting too wide of a net that it catches a bunch of the patients who really didn’t need the visit. A good analogy would be fishing with a net: If you cast the net too wide, you’ll catch a lot of fish, but you may not be very accurate (“specific”) in capturing the exact type of fish you wanted.
The paramedics’ opinions alone were accurate (specific) 80% of the time for those who didn’t require ED care. But they weren’t very sensitive (39%), meaning the paramedics would have left 61% of patients who did need care. The criteria checklist was better (72% sensitivity) at catching patients who needed ED care, but only had 43% specificity for those who didn’t.
In either case, the end result is that patients who needed care would have been missed. The authors point out that in hindsight, it would have been better to modify two criteria: age greater than 55 and heart rate greater than 83 (94% sensitivity with 61% specificity). This still would have meant transporting a lot of people who didn’t need care, but it wouldn’t have left behind the ones who did need it.
Readers beware: The criterion in the research study wasn’t validated, and further analysis is needed to discover a better prospective triage tool.
CPR before Defibrillation?
Stiell IG, Nichol G, Leroux BG, et al. Early vs. later rhythm analysis in patients with out-of-hospital cardiac arrest. ROC Investigators; N Engl J Med. 2011;365(9):787–797.
This is one of two major reports published in the September issue of the New England Journal of Medicine from a long list of accomplished researchers in the Resuscitation Outcomes Consortium (ROC). This carefully constructed, large, prospective, randomized and multi-institutional trial has all the hallmarks of gold-standard research.
The study randomized 9,933 patients to receive either “early” ECG rhythm analysis (only 30–60 seconds of CPR before possible defibrillation), or “later” analysis after three minutes of CPR and shocks. They then watched to see which of the two groups was more likely to survive to hospital discharge with favorable neurological outcomes. Their conclusion: no difference. The groups had the same outcomes (only 5.9% survived).
Some interesting fine print exists in this study: The research showed that patients who had bystander CPR and were in ventricular fibrillation (v fib) or ventricular tachycardia (v tach) should be defibrillated as soon as possible because more of them survived in the “early” group.
Their secondary data analysis also seems to show improved outcomes if patients with pulseless electrical activity receive “later” rhythm analysis (meaning more CPR up front). Similarly, patients with an active impedance threshold device (ITD) benefited from early, high-quality compressions and later analysis. This brings us to the second study in the same issue.
Aufderheide TP, Nichol G, Rea TD, et al. A trial of an impedance threshold device in out-of-hospital cardiac arrest. N Engl J Med. 2011;365(9):798-806.
Dr. Aufderheide is the principal author of this report on the ITD’s effectiveness during cardiac arrest. Using the same patient population as the study above, the ROC analyzed whether patients with an active ITD were more likely to have good outcomes than those treated with a sham ITD, that is, a visually identical but actually non-working ITD. Again, they conclude there was no difference.
Reading between the lines ends up being rather important in both of these studies. The authors describe the difficulty and number of variables involved in cardiac arrest research and some major limitations due to the realities of implementing high-quality resuscitation standards in the uncontrolled prehospital environment across 10 different EMS systems.
Reporting of the exact intervention times and delays in the application of ITDs, CPR or rhythm analysis also clouds the results of a study based on time-sensitive devices and interventions. In some cases, crews didn’t remove the ITD on return of pulse and failed to obtain adequate mask seals. According to the paper, the ITD was placed, on average, 11–14 minutes after the patient arrested. Given this delay, it’s not surprising the results were neutral. A complex study design involving many different EMS systems can lead to protocol non-compliance, crippling a well-intended study.
Despite all these limitations, the authors point out that when high-quality compressions were delivered 60–71% of the time, with the active ITD, a significant improvement in survival to discharge was observed. JEMS
Schmidbauer W, Ahlers O, Spies C, et al. Early prehospital use of non-invasive ventilation improves acute respiratory failure in acute exacerbation of chronic obstructive pulmonary disease. Emer Med J. 2011;28(7):626-627.
The American College of Emergency Physicians (ACEP) held its annual conference in October. Keep an eye out for the full report on the following abstracts:
Schulz K. Paramedics do not forego treatment of infants. Ann Emerg Med. October 2011;58(Research Forum Abstracts_358):S298.
These researchers showed that EMS rarely responds to infants less than 1 year old (0.3% of all cases), and of those, only 10% were given a medication. The authors wondered whether paramedics could have done more for these cases.
The retrospective review of the hospital records exonerated the paramedics. The authors conclude that neither education nor discomfort of the paramedics appears to play a role because “it’s rare that infants are given medication or procedures in the ED that could have been given out of hospital,” the study says. The most common EMS miss was albuterol and IV administration.
Bier S, Hermstad E, Trollman C, et al. Army flight medic performance of advanced emergency medical technician procedures: Indicated versus performed. Ann Emerg Med. October 2011;58(Research Forum Abstracts_#359):S299.
These authors wondered whether Army helicopter medics with an EMT-P certification were performing a greater percentage of indicated skills than helicopters staffed with EMT-Bs.
They reviewed 406 cases, of which 22% were considered “critical care” transports. EMT-Bs failed to perform indicated procedures 35% of the time vs. paramedic staff, who failed 3% of the time. The authors suggest that the Army consider staffing air medical transport units with standards similar to civilian crews.
This article originally appeared in December 2011 JEMS as “Detox or Dehospital? Study measures providers’ predictions about transporting inebriated patients.”
Visit www.pcrfpodcast.org for audio commentary.