Authors: Rob Hirst, Andy Neill, Dave McCreary, Becky Maxwell, Chris Connolly / Codes: CAP37, PAP17, PAP6, PAP9, NeuP8, ResC7, SLO1, SLO5, TC2, TP7 / Published: 01/06/2023
Does etCO2 predict in-hospital mortality from triage?
Title of Paper
End-tidal carbon dioxide measured at emergency department triage outperforms standard triage vital signs in predicting in-hospital mortality and intensive care unit admission
Journal and Year
Academic Emergency Medicine
- ED Triage is all about trying to work out who do see first, and part of that it working out who is sickest.
- This is generally done quickly using a combination of history and vital signs.
- We all know that the traditional vital signs alone aren’t great predictors of mortality
- End tidal CO2 is a non-invasive measurement of metabolism, circulation and ventilation (all things that I think we can agree are important for life)
- Prospective, observational study with convenience sampling.
- Adults presenting to a tertiary ED through triage or EMS
- Exclusions: being resuscitated on arrival and non-urgent fast-track patients (both ends of the ED spectrum)
What The Did
- Standard vital signs recorded: HR, BP, RR, Temp, Sats.
- etCO2 measured in addition with nasal end-tidal catheter, recording the etCO2 on the 8th breath.
- If already measured by triage, treating team or EMS, was included in the analysis
- Primary: in-hospital mortality
- Secondary: admission to ICU
- Tertiary: correlation with other markers of metabolic acidosis like lactate, bicarb, anion gap.
Summary of Results
- 1091 patient analysed
- 1065 patients survived
- Only 26 had in-hospital mortality [not a great prevalence for calculating test characteristics] ****
- Survivors significantly younger (mean 55 vs 68 years)
What about the mean etCO2?
- Overall mean 34mmHg (4.53 kPa to translate from antipodean)
- Survivors mean 34mmHg
- Non-survivors mean 22mmHg (2.93kPa) [95%CI18-26mmHg; p<0.001]
AUC for predicting in hospital mortality
🤓 Area under the ROC curve – basically a good way to compare test performance between test. An AUC of 1.0 would be the perfect test, 100% sensitive and 100% specific.
|Vital Sign||AUC (in-hospital mortality)|
They adjusted for the other vital signs, and found etCO2 kept a strong odds ratio for predicting mortality. Didn’t, however, adjust for age.
What cutoff do they suggest be used if you were applying this in real life?
The authors’ suggest:
- Cutoff of etCO2 28mmHg (3.73 kPa): Sn 77%, Sp 79%, NPV 99%, PPV 8%, LR 3.7
- Cutoff of etCO2 32mmHg (4.27 kPa): Sn 81%, Sp 65%, NPV 99%, PPV 5%, LR 2.3
In this large, prospective study of an undifferentiated ED patient population, end-tidal carbon dioxide levels measured at triage were able to predict in-hospital mortality and intensive care unit admission. Furthermore, when compared to conventional vital signs at ED triage, end-tidal carbon dioxide provided the highest predictive measure for mortality and intensive care unit admission.
These results suggest that incorporating end-tidal carbon dioxide into initial patient assessments at triage may provide early and important information for ED clinicians and has implications for medical decision making, including the need for higher levels of care.
Clinical Bottom Line
Sure, it’s interesting and they’ve thrown out some nice numbers here, but I have one major issue: they haven’t adjusted for patient age anywhere that I can see in the paper. Yet, there are studies (as far back as 1998) that show that etCO2 decreases with age.
We know the non-survivor group was older, and we know the non-survivor group had a lower mean etCO2, so why haven’t they checked whether they had a lower etCO2 simply because they were older?
I’m not saying that there isn’t potential for this in the future, I’m just saying I think this means that this study doesn’t, on its own, tell us much.
It’s also worth noting that a portion of the nasal cannulas used in the study were provided for free by Covidien, who make them. They state that Covidien had no role in the study, but its obviously a potential source of bias / a conflict of interest.
Do we want our patients horizontal or vertical for lumbar puncture success?
Title of Paper
Is Lateral Decubitus or Upright Positioning Optimal for Lumbar Puncture Success in a Teaching Hospital?
Journal and Year
The Journal of Emergency Medicine. 2023.
- Lumbar punctures, the easiest hard procedure around.
- By this I mean, when they’re easy, they’re pretty easy but when they’re difficult they can just feel impossible.
- I was always taught that the upright position was best (unless you need opening pressures, which I am seldom interested in – though since our recording maybe Andy is convincing me otherwise…), and that does tend to be my go-to, particularly in patients with BMI on the higher end.
- Others swear by the lateral decubitus (which yes, Andy, allows for opening pressures), but can make the midline wander a little. It likely all depends on how you were taught – like so many of the things we do in medicine.
- So which is best? Is there a difference?
- Randomised, prospective trial
- Convenience sample of patients requiring lumbar puncture in a mixed adult and paediatric, academic emergency department.
- Included pregnant patients and neonates
- Excluded intubated, incarcerated or those unable to sit upright or recline to lateral decubitus position
- Lateral decubitus position vs upright position for lumbar puncture
- Powered to find 5% difference
- Primary: successful LP – defined as obtaining at least 1.5ml CSF
- Unsuccessful = no CSF, patient repositioned or change of operator necessary
- First pass success
- Number of insertions (through skin)
- Number of redirections
- Number of blood contaminated taps
Summary of Results
- 116 patients analysed
- 55 lateral decubitus
- 61 Upright
- Largely performed for ?meningitis (85%). Some for ?SAH (15%)
- Lateral decubitus: 85.5% successful [95%CI 73.8-92.4%]
- Upright: 80.3% successful [95%CI 68.7-88.4%]
- So possibly trending towards Lateral Decubitus in this patient group, though the CIs cross so not significant
- 20 failures, 65% of those in neonates.
Secondary: No differences in reinsertions, redirections or traumatic taps
- Didn’t adjust for BMI
- PGY2s had a better success rate than PGY3s
- ?More supervision
- ?sweet spot in confidence/competence curve
- ?PGY3’s get the harder LPs given to them
- Residents who have performed 31-50 LPs had 90% success rate
We conclude that lateral decubitus and upright position for emergency lumbar puncture yields similar success rates. This allows providers to position the patient as necessary for the clinical situation or for comfort. These results impact how lumbar puncture should be performed and taught.
Clinical Bottom Line
You need to be able to do both. It and doesn’t matter which one you are happiest with – the physicians in the study performed equally well in their preferred vs less comfortable technique.
I always use upright if the patient looks like they may be more difficult (large BMI) – assuming I don’t need an opening pressure.
But that’s not an option if you’re doing it on a tubed ?encephalitis.
Like with any procedure its important to train yourself to be versatile and practice both techniques so you have more to chose from when needed.