Many thanks to Peter Sherren for spotting this paper published in Intensive Care medicine this month. Point of care ultrasound is one of my interests and only this week I met a patient whose life was certainly saved by it in the ED a few years ago (cracking diagnosis in the resus room of very unexpected cause of hypotension – patient would have died without diagnosis).
In the ED, the undifferentiated hypotensive patient presents a bit of a dilemma. Those of us who work in resus will know that the initial assessment of the really sick looking hypotensive, shocked patient is tough. Most, if not all of us will have made an initial diagnosis for the cause of the shock only to discover later (post imaging/invasive monitoring/time/post-mortem) that we got it wrong. We are therefore always on the look out for better ways of differentiating causes of shock so that we can intervene early and intervene accurately.
Ultrasound in the ED as a tool for the assessment of shock is not new. I first came across it in Cambridge round about 2007 with Paul Atkinson as my tutor. The ACES exam was published in the UK literature and it’s something we have adopted locally, but perhaps not as widely as we should. Similar methods are used around the world and EMCRIT has a nice review on the RUSH exam as a similar approach. Both of these approaches (and others) have their merits, but as a clinician I want to know if they work in practice, and by practice I mean in patients like the ones I see. I need studies that look at undifferentiated, shocked patients in the resus room.
So, the paper by Volpicelli et al looks at a group of patients that are like mine. 108 patients (non-trauma) with symptomatic undifferentiated hypotension. They looked to see if Ultrasound could identify the cause of hypotension in the ER by performing an ED exam and then following the patient up to see if they were right. This diagnostic cohort design is one of the simplest ways of looking at a diagnostic test.
[learn_more caption="So what was done?"] The patients were pretty similar to those that I would want to investigate.
- Age 18-95
- SBP < 100mmHg at presentation and on three other measurements.
- One or more of unresponsiveness, altered mental status, syncope, resp distress, asthenia, with fatigue/malaise
- severe chest/abdo pain
They then performed an ED USS that covered the following areas.
- Heart (kinesis and dilatation)
- Inferior Vena Cava (size and collapsibility)
- Lungs (PNX, pneumonia, failure)
- Abdomen (free fluid and AAA)
- Veins (any lower limb DVT)
So nothing too fancy there then – all things that I do in our ED. Once the scans were performed the operator decided on the likely cause of shock.This is a bit of a mixed bag and there is some cross over between categories, but the definitions for each of these diagnoses is given in the paper.
- Obstructive cardiac tamponade
- Obstructive pulmonary embolism
- Obstructive tension pneumothorax
Final diagnoses were decided through a structured notes review. [/learn_more]
[learn_more caption="What did they find?"] Firstly, shock assessment was quick. On average about 5 mins which is impressive (i.e. faster than me).
Secondly, in terms of agreement between initial diagnosis and final diagnosis the data looks to be fairly consistent.
So it looks as though there is pretty good agreement and in the analysis using kappa statistics looks pretty good.[/learn_more] [learn_more caption="So. Is it all good news then?"] Well, much as I love to believe papers that agree with what I think (and I think point of care USS is great) we do need to stop and think about how robust the findings are here.
1. The notes review team were blinded to the initial USS findings – but it is unclear if they were blinded to subsequent USS data. It does not look as though the clinicians were blinded though and this must surely have influenced other aspects of care that may have supported or refuted the initial impression from USS. This introduces the potential for circularity between the initial USS and the final diagnosis. This may lead to an increase in agreement.
2. The number of cases with some pretty important diagnoses is small. There were only three cardiac tamponades for instance. Significant differences in the performance of USS for some diagnoses may well be missed in a study of this size.
3. This sort of study can give us information on agreement, but what really matters is whether earlier diagnosis leads to better care and better outcomes for patients. We do not know that here, and it is one of the reasons why I advocate the use of randomised controlled trials for the final evaluation of diagnostic tests.
4. I would have been interested in the number of occurences where the USS changed the opinion of the treating team. I have had a number of cases where USS has completely changed my diagnosis of shock – leading to a significant change in patient care (hopefully for the better). This study did not compare initial clinical impression against USS outcome vs final diagnosis. In other words this paper does not tell me whether USS is performing better than that which we already know. [/learn_more] [learn_more caption="The bottom line"] I like point of care USS – a lot – I used it today in the resus room and I want the evidence to prove me right. I’m not sure that this paper on its own proves much, but as additional evidence that supports my feeling it is interesting. I might quote it in the future, but I’m still waiting for the definitive paper that tells me whether the addition of USS to shock assessment really makes a difference.[/learn_more] [learn_more caption="Read more here"]There’s loads of stuff out there and this is not a systematic review, but if you are interested then you might start with some of the following
- RCT on early vs delayed USS in resus: http://journals.lww.com/ccmjournal/Abstract/2004/08000/Randomized,_controlled_trial_of_immediate_versus.11.aspx
- Abdominal and Cardiac Evaluation with Sonography in Shock (ACES): an approach by emergency physicians for the use of ultrasound in patients with undifferentiated hypotension
- Nickson on USS from LITFL – read the comments!!!!!
- Check out the sonocave and the ultrasound podcast for the techniques tips and tricks that you need to develop as ED ultrasound gurus.[/learn_more]