To pan-scan or not pan-scan REACT2.

Its been a few weeks since the last post and there’s a few reasons for that.

Largely because its the summer and I’ve been outside with the kids a bit more than usual, trying to get my work-life balance tipped a bit more towards home, 6 months into being a consultant. It’s also hard to not have your mind focused by reading and listening to Rob Rogers talk about his own MI.

This blog and supporting podcast from the St Emlyns crew can help us all find a little perspective and help us think about who we are….

There was also this……


Which got in the way of work somewhat….

Date of JC: 17/8/16

Title of Paper: Immediate total body CT scanning versus conventional imaging and selective CT scanning in patients with severe trauma (REACT2): a randomised control trial.

Published in The Lancet 2016.

The next patient in resus is a 29 year old guy who’s crashed his car at speed, he was wearing a seatbelt, airbags were deployed. His passenger was unrestrained and killed on impact, the paramedics tell you it was pretty harrowing. A slick primary survey is completed, he has some abdominal pain and a seatbelt mark across his chest, with a resp rate of 24 and Saturation in air of 98, his heart race is 110 and BP 134/76 and GCS is 15. You wonder whether you should ‘pan-scan’ him head to pelvis or focus your investigations with a chest x-ray, FAST and pelvic x-ray to help you decide whether you should move on to a CT.


Population: Adults with potentially severe injuries
Intervention: Immediate whole body CT scanning
Comparison: Conventional imaging with selected CT scanning
Outcome: In-hospital mortality

Generalisability: This was a study run in the Netherlands and Switzerland who have a slightly different ED/Trauma set up to the UK. Teams were made up of a trauma surgeon, anaesthetist, radiologist and support staff (its not clear who they mean). My understanding of the pre hospital set up in continental Europe is that it’s slightly different to ours in the UK with more pre-hospital clinician time there. I don’t think this means we can discount this study as not applicable to our population but its certainly food for thought. They list inclusion criteria for reasons around Vital signs, mechanism of injury and clinician suspected significant injury in a bi to detect and study those who had severe or life threatening injury. This caused some discontent at the face-face JC, we felt that we often had a cohort of patients who had fallen from standing and had severe or life threatening injury without any initial physiological derangement – the elderly. Again I don’t think this is reason to ignore this paper but again its worth taking in context of ‘the changing face of trauma‘ paper which shows us that the average age of a trauma patient in your ED is 60…..20 years older than the patients studied here.


Design: This was a prospective RCT, it was a pragmatic study, which tried to test the pan-scan against probably what is ‘real world’ practice.
The consent was deferred at enrolment and sought as soon as possible after the resuscitation had taken place – this seems to be something that is becoming more and more commonplace in ED research with the currently running ECLIPSE trial and the C3PO trial in heart failure utilising this to good effect.
The randomisation was computer based which was good to see but it was impossible to blind the clinician or patient to the group they were allocated to.

A power calculation was done with an 80% power to detect a 5% mortality difference with an alpha of 5%. They recruited the required 539 patients to detect this difference if it was there (541 in each arm).

Intervention/Comparitor: The pan-scan approach was a head-pelvis CT which included a split bolus contrast study of the chest-pelvis – we’re not radiologists but we think this is different to a ‘Bastion’ style scan that is undertaken at our place, but hands up, we didn’t know the significance of this – if you are a radiologist please let us know in the comments at the bottom!! The comparitor arm was any combination of imaging that the team leader thought appropriate, including a pan-scan (in fact 43% of the patients had one of these), or a combination of x-rays, FAST and focused CT scanning – we felt we probably lay somewhere between the two groups – thinking we did more pan-scans than 43% but didn’t think that all patients meeting the inclusion criteria would get a pan-scan every time.


Primary outcome: In-hospital mortality: No difference
Secondary Outcomes: There were lots of these (make sure you read the paper to digest them all) but of note:

Radiation dose: reported as higher in the pan scan group but the IQRs crossed (20.6mSv (20.6-20.9) Vs 20.6mSv (9.9-22.1)) and sadly no 95% CI were reported.

Time: Time to end of imaging and time to diagnosis was quicker in the Pan-scan group (again no 95% CI) with 30 mins (24-40) vs 37 mins (28-52) and an 8 minute difference in time to definitive diagnosis (50 vs 58 mins). We had a few concerns with the timing of all this – in reality how quik is it in your place to get a set of primary survey x-rays done? and a FAST done and logged on to the IT system? with a head scan and neck scan? We also thought that 30 minutes for to get someone through a pan-scan when the scanner’s in the room was a little slow……

The only other food for thought we felt was the primary outcome measure itself, and whether its the primary outcome measure we wanted – mortality is a great thing to measure in a clinical trial. its pretty definitive, its definitely patient centred. so why are we grumpy yorkshirefolk? Well i think its this – if I choose to scan someones chest and not their belly, I wonder when I reflect back later or next week – did i miss something? and if I did whats the outcome going to be for the patient in terms of morbidity? I’d like to know if selective scanning in my practice is likely to lead to missed significant injuries (from a patient perspective) as I think those that are apparently ‘badly injured’ will end up in the pan-scan (as in this trial) anyway.

All in all this was a well conducted trial with a few issues raised around generalisability and the study outcomes – the room was not split however, we couldn’t find a reason not to do a Pan-Scan on the patients who were young, had a worrying mechanism of injury, abnormal physiology or some specific injury concerns – whilst there’s no mortality difference, we’ll scan them and diagnose them quicker and probably wont expose them to a significantly higher amount of radiation.

As always you can download the One minute Wonder here.

Please let me know what you think of the post and the site.

Keep fighting the good fight.



Life and Learning from Rob Rogers on the other side of that ECG. St.Emlyn’s

Be well and be a better critical care clinician. St.Emlyn’s

2 Comments Add yours

  1. Lisa says:

    The Bastion scan is a split bolus scan. You inject one bolus of contrast, which will be in the venous phase on the final scan, wait, then inject a higher speed bolus that will trigger the scan in the arterial phase. So you get both contrast phases on the same scan. Saves doing two separate scans. Halves the radiation dose but 50mls more contrast.

  2. foamshed says:

    Awesome-Thanks Lisa!!

    I’m obviously busy watching the monitor in the scan room so have never noticed how the contrast is given……

    Cheers for engaging!

Leave a Reply

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