The Cardiac Decision Tool Fight is on. EDACS Vs ADAPT

download

Winter is upon us.
Corridors are full of patients waiting to be seen, waiting to be admitted, social care is creaking.
We have been told that the winter that breaks the NHS is coming.
Increasing frailty, increasing demand and expectation.
All of that.

But in amongst all of the chaos there’s a cohort of patients we could potentially improve the flow of through the ED, improve the risk stratification of, and facilitate early discharge planning.
And the great news is we see tons of it.
Suspected cardiac chest pain.

Your next patient is 58 years old and works as a teacher. He’s on no regular meds, and other than being a little ‘cuddlier’ than he’d like has no real medical problems. He had a sudden onset of epigastric pain which felt tight and ‘squeezing’, with radiation to both arms and jaw. He felt sweaty and his colleague commented that he ‘looked awful’. (make sure you have a look at this from Rick Body on cardiac risk assesment and read around how history exam affects likelihood of it being ACS) His pain last around 30 minutes (until the paramedics arrived and gave him some GTN and Aspirin). He feels fine now. A pre-hospital ECG was done which was normal.

download-1

This all happened around 90 minutes ago (yeah I know probably a bit optimistic) and he’s keen to know what you plan and if he can go, he can see that the department is full of elderly and critically unwell patients. Unfortunately, the centre where you work uses a 6 hour high sensitivty troponin T (HSTnT) sample on patients presenting with suspected cardiac chest pain with a repeat at 3 hours if the first is in a grey area (14-30mmol/L).
This guy could be here for the next 6 hours at least while you get his initial results back. Surely there’s an alternative?

clock-1060458_960_720

If you’ve been in a FOAMed community you cant have missed the late summer explosion in papers on this theme. There’s been TMACS, EDACS, ADAPT, Limit of Detection studies, and don’t forget the HEART score earlier in the year (which Ken Milne reviewed on his podcast here) it could all get a little overwhelming, and certainly a little bit acronym heavy!!

So the paper.

Title of Paper: Effectiveness of EDACS Versus ADAPT Accelerated Diagnostic Pathways for Chest Pain: A Pragmatic Randomized Controlled Trial Embedded Within Practice.

Published in Annals of Emergency Medicine 2016

Population: Adults presenting with chest pain and suspected ACS
Intervention: EDACS-ADP with HSTnI
Comparison: ADAPT-ADP with HSTnI
Outcome: Discharge at 6 hours and 30 day MACE

First up lets define a few things.

MACE: Major adverse cardiac event. In this case that’s:
Death
Revascularisation
Cardiac arrest
Ventricular arrhythmia
High degree AV block
Myocardial Infarction

The scores:

Both of the ADPs include the assessment of a normal ECG (i.e. non-ischaemic) and a High Sensitivity Troponin I measured at 0 and 2 hours with male and female cut offs at 34 and 16ng/L respectively.

ADAPT-ADP: A combination of normal ECG, normal troponin readings and a TIMI score of 0
EDACS-ADP: A score of <16 (scoring increasing points for increasing age, known CAD or multiple 'risk factors', some history features and being a dude - find a link to MDcalc here)

Anyone with ‘red flag’ symptoms of unstable symptoms, or abnormal vital signs were deemed as not ‘low risk’.

There were two primary outcome measures in this study (they wanted to see how effective the EDACS was at risk assessing and discharging folks with chest pain) – discharge at 6 hours and 30 day MACE. Remember that a study with two primary outcome measures is unlikely to have been powered for both so make sure you look for which one it is powered for.
A power calculation was done, the study having 90% power to detect a 13% difference in 6 hour discharge rate which is important when looking at any difference in the MACE rate. Have a listen to some real experts discussing this in more detail on the RCEM NewInEM podcast here the difficulty in powering for this outcome is that the numbers you’d need to detect something significant would be huge and unobtainable in the real world!

So what do we know about the patients and EDACS to date. The score is widely used in New Zealand and I understand in a number of places across Australia – I’d be really interested to know if anywhere in the UK is using it yet? If that’s you please get in touch and let me know how you’re getting on with it!!
The setting for this single centre study was Christchurch in New Zealand, their patients were anyone 18 years and upwards presenting with chest pain that the treating clinician warranted a cardiac work-up. They sadly didn’t include anyone who presented out of hours (2300-0800), which you’ll know fills me dread as I’m convinced our patients are different at night, I don’t know how or why, but anyone who’s done lots of nights will concur. I’m sure of it!

Another area of concern is the single centre nature of the study, single centre studies can struggle to be externally valid, in this case the centre was far more ethnically diverse than our place. 76% were NZ European with the 24% spread across Asian, Maori, Pacific Islanders, other European and ‘Other’ – this combination means they probably don’t represent our patients.

Both groups had the same management – Arrival ECG, high sensitivity troponin I and a score assessment (modified TIMI or EDACS). If the ECG was deemed normal and the first HSTnI were ‘normal’ then they could progress to the 2 hour blood sample. If all of these were negative along with an EDACS of <16 or TIMI 0 then they were sent home with a view to an outpatient stress test. The EDACS score has been derived and validated well and externally validated in a Canadian population (We reviewed this on the RCEM podcast here). A score of 16+ makes you outside of low risk. One sure fire way to be higher risk is to be an older man. In fact if you’re a 61 year old male you aren’t low risk. I guess that’s one thing to think about if implementing this strategy into your place. Whats the demographic of where you work?

I was massively impressed with the ability to get an early outpatient stress test – in the study 85% had one within 3 days. That’s incredible and something we can only dream of here. Does this affect my reading or desire to implement this paper into practice? No, not as an isolated item but it certainly does make me wish I work in Christchurch!!

Laufband (Threadmill)

They found an absolute difference in the numbers of patients classed as low risk in favour of EDACS-ADP vs ADAPT-ADP (slightly fewer after reclassifying some as unstable) 41.6% Vs 30.5% with a difference of 11.1% (95% Confidence interval 2.8-19.4%) which is quite a broad confidence limit in my mind, although any decrease in admission rates would be welcomed, especially with (as they found) no difference in MACE rates (0 patients had 30 day MACE in either group).

Despite the difference in absolute number identified in the EDACS group as low risk there was no significant difference in discharge rate by 6 hours – this may be related to the fact the EDACS represented a change in practice (as the authors postulate in their discussion) or it may be that a significant number of the low risk EDACS patient had known coronary artery disease, and this may have influenced the clinicians discharge decision making. It’d be interesting to see the discharge timings now the EDACS is firmly ingrained in the culture of the department. I’d imagine a review of that process wont be too far away!!

In Summary

I really like the look of the EDACS score. This is the second paper I’ve reviewed in the past couple of months that has tested it with good safety and it looks to me like a more easily applied score than TIMI (not needing to know about degree of coronary artery stenosis, nor caring about spurious aspirin use) – I’m intrigued to see if the EDACS has been validated in a UK population or if anywhere is using it without UK validation.

I’d also be interested to know if it performs as well with a HSTnT as HSTnI – I’ll bow to the FOAMed world’s collective knowledge of variance in performance between the two types of High Sensitivity Trops – I really don’t know if one exists – please comment and let me know!!

I don’t think this paper means I’m going to adopt their protocol tomorrow, but I am going to explore the idea of implementing a change in my practice a little further.

Thanks for reading

You can access a one minute wonder for your department here.

Keep fighting the good fight.

Chris

References

Getting your chest pain evaluation right: #UMECS16

SGEM#151: Groove is in the HEART Pathway

November 2016 New in EM

October 2016 New in EM

2 Comments Add yours

  1. John Pickering says:

    Hello Chris,
    I just came across this post of yours… thanks for taking notice of our study and reporting on it so well.
    I can confirm that the EDACS pathway is in place at multiple hospitals throughout NZ now including some with hsTnT.
    You asked about performance differences between hsTnT and hsTnI – I do think there are differences which we have noted when we have tested the ESC guidelines head to head. Happy to send you references if you need them.
    You also wish you worked in Christchurch – you are most welcome!

    1. foamshed says:

      Hi John

      Thanks for touching base glad you enjoyed the post!

      Sorry for the delay – I’ve managed to clear the Junk in the inbox and found your comment! The site would be snowed under with dodgy auto-links if everything went straight on from the comment box!!

      Yes-would be interested to read your work on HSTnT Vs HSTnI – either an email or post the links on here if you’re happy to share widely!!

      I don’t need convincing about the benefits of NZ – sadly my other half and kids settled in school make a move more tricky/politically difficult! (I’ll keep badgering!!)

      Cheers

      Chris

Leave a Reply

Your email address will not be published. Required fields are marked *