What’s my view?
Well how am I supposed to know?
Write a review
Well how objective can I be?
I like to wait to see how things turn out
If you apply some pressure
Mark Ronson ft Paul Smith (Maximo Park) – Apply Some Pressure
It was my first Friday night shift in the Emergency Department as a registrar. I was nervous; I had worked Monday to Thursday nights already, but the weekend promised something different.
Just after midnight, the red “standby” phone rang. It was a call I’d been anticipating and dreading in equal amounts; police officers allocated to the department had already told us about a call they’d overheard on their radio, requesting officers to attend reported gunfire – and now here it was. 19-year-old man, cardiac arrest, gunshot wounds to chest.
The case pre-dates the roll-out of structured trauma management across our region so there was no trauma team to contact but I did ask the nurse in charge to contact the ED consultant, who was oncall from home. We made preparations for the patient’s arrival as I tried desperately to remember something – anything – from the ATLS course I had attended four months previously.
But wait – the patient was in cardiac arrest. Ambulance control had told us the rhythm was PEA – but which algorithm should I use? ATLS? ALS? A mixture of both?
The patient was received into the resuscitation area with CPR in progress, time since arrest estimated at 20mins. He appeared very pale. He had been intubated with a size 8.0 ETT with apparent chest movement (difficult to assess whether there was equal air entry due to the noisy environment). His trachea was central and neck veins were flat. The monitor showed asystole and he had no palpable central pulses. There was a small entry wound on the upper left side of the chest with no visible exit wound. His pupils were fixed and unreactive.
So the guy is dead, right? But shouldn’t I at least try to address reversible causes of cardiac arrest? We tried in vain to obtain peripheral IV access; even external jugular access was unsuccessful. I knew that the most likely reason for his arrest was hypovolaemia from bleeding, but that if this was the case he was unlikely to survive. I also did not have the skills or knowledge to perform a thoracotomy. I could insert bilateral chest drains – but in ATLS scenarios, I remembered all too clearly, if you inserted a chest drain before getting IV access things would only get a lot more messy in every sense of the word. I (bravely, I thought) performed a pericardiocentesis (without ultrasound guidance – hadn’t been on that course either) on the basis that the initial rhythm had been PEA and relief of tamponade might have led to return of circulation. I was able to aspirate over 100ml of blood, at which point I concluded I was probably not in the pericardial sac.
There was still no organised electrical activity on the monitor. I considered stopping resuscitation at this point, but then the consultant arrived so we were able to proceed to thoracotomy via left lateral approach. There was a large volume of blood upon breaching the pleural space. There was a small hole in one of the major pulmonary vessels, which the consultant was able to quickly close; he then performed internal cardiac massage, but the preload was very small presumably due to the volume of blood lost into the thorax. After a total of 25mins resuscitation, we decided to discontinue CPR attempts.
This case is memorable for so many reasons, not least as the first time I had dealt with any sort of major trauma alone (at least initially) and also my first experience of managing penetrating trauma. I remember clearly trying to integrate the principles of ATLS and ACLS, finding each inadequate in isolation but incompletely compatible, leaving me unsure of what was indicated and what was inappropriate. I had been on both courses but had never properly considered that there might be a need to manage a patient with cardiac arrest secondary to specific trauma, and now as an ATLS instructor I still feel this is unclear in the current ATLS curriculum.
In my head, the patient was dead and unlikely to have a return of spontaneous circulation but I was unsure how far interventions should continue and what chance of success they offered. The patient was young – as, I understand, victims of gunshot wounds often are – and this undoubtedly added a heightened level of emotional investment.
There are many things I would do differently in this situation now, possessing a modest amount of ultrasound skill, an unbridled love for the EZ-IO and the luxury of a trauma team to summon to assist me. But what makes this case interesting is I am still not sure which algorithm I should be following, something which Karim Brohi’s recent EMCrit podcast presents an interesting perspective on.
Professor of Trauma Karim Brohi argues that in any traumatic cardiac arrest, secondary to blunt or penetrating trauma, cardiopulmonary resuscitation not beneficial. His rationale is that there is no role for chest compressions, and that they may cause harm, in haemorrhagic shock, tamponade or tension pneumothorax where preload is diminished, and that these patients benefit from fluids, bilateral thoracostomies and US to exclude tamponade, followed by ED thoracotomy if no cause is identified, with the presumption that opening the thorax will permit haemorrhage control from distal vessel (aortic) compression. For this case, it seemed likely that the cause of arrest was one of the above, and I certainly feel I could have made different decisions to better address the likely cause of arrest before proceeding to thoracotomy – but to omit chest compressions seems counter-intuitive. Do we need to move beyond the ALS rhetoric?
ED thoracotomy is a skill rarely practiced in any EDs and studies have recognised the role of practice in achieving and maintaining competence in practical skills. Traumatic cardiac arrest necessitating emergency department thoracotomy is widely accepted as carrying a poor prognosis, but positive outcomes do occur. In 1995, Fulton et al identified confusion about treating traumatic cardiac arrest, finding 6% survival from 245 patients. Their conclusion was that duration of arrest greater than 10mins was strongly associated with mortality. Efforts have been made since then to ascertain “best practice” for patients presenting in traumatic cardiac arrest. Should we perform chest compressions, or thoracotomy, or both, or neither?
The most edition of the Advanced Trauma Life Support Course (2008) advocates resuscitative thoracotomy in pulseless patients with penetrating trauma, and asserts that “closed heart massage for cardiac arrest or PEA is ineffective in patients with hypovolaemia”. This section of the manual, which goes on to state that for patients with blunt injuries arriving pulseless thoracotomy is “not indicated”, is somewhat controversial. The guidance stems from a statement by the National Association of EMS Physicians and the American College of Surgeons Committee on Trauma, published in the Journal of the American College of Surgeons in 2003. The paper includes a position statement regarding patients who suffer cardiac arrest out-of-hospital as a result of trauma; they recommend that for penetrating trauma:
“Victims of penetrating trauma found apneic and pulseless by EMS, based on their patient assessment, should be rapidly assessed for the presence of other signs of life, such as pupillary reflexes, spontaneous movement, or organized ECG activity. If any of these signs are present, the patient should have resuscitation performed and be transported to the nearest emergency department or trauma center. If these signs of life are absent, resuscitation efforts may be withheld.”
Given that our patient was in PEA arrest on paramedic arrival, it seems that transport and resuscitation would have been deemed appropriate in this case. The Journal of the American College of Surgeons’ 2003 review goes on to suggest that patients with penetrating trauma have a greater chance of survival than in other patient groups. But the blunt versus penetrating trauma debate is contentious; Lockey et al performed a 10-year retrospective analysis of the outcomes of traumatic cardiac arrest in the HEMS database, finding 7.5% survival to hospital discharge. Of interest, their study identified 13 survivors to discharge who would not have been resuscitated at all had the American guidelines been followed. Willis et al found just 4 survivors from their retrospective cohort of 195, but 2 were from penetrating trauma and 2 from blunt trauma, leading them to advocate for consideration of the role of CPR and further resuscitation in any traumatic arrest.
Survival advantage in penetrating versus blunt trauma seems to be supported by the literature although studies are small and retrospective. There are many concerned with Emergency Department Thoracotomy (EDT). A review in 2000 suggested that overall survival of EDT was 7.4% with an impressive 92.4% of those surviving being neurologically intact. When subgroups were analysed, survival rose to 8.8% in the penetrating trauma group. Moriwaki et al looked exclusively at 477 victims of blunt trauma in Japan, giving aggressive treatment and finding only 3% survived to discharge. 86% of their patients underwent EDT and most died within 24h of arrival at hospital. Of interest, the patients who survived included those presenting in asystole.
Moore et al also found survivors in the group presenting with asystole, but all had cardiac tamponade at thoracotomy. In their groups patients with stab wounds to the left ventricle were most likely to have positive outcomes, followed by those with gunshot wounds to the lung.
So it seems that ATLS principles agree that the patient should have undergone EDT, but that the evidence suggests that the odds of positive outcome were strongly stacked against him. Cera et al found 9% survival to discharge from a cohort of 195 out-of-hospital traumatic cardiac arrest patients, of whom 14/15 had reactive pupils at the time of arrival to ED, a prognostic factor which also weighs against my patient.
Studies have supported the value of ultrasound in traumatic arrest, with 100% of the otherwise small proportion of patients surviving to hospital discharge having organised contractile activity visible in one study and sensitivity of ultrasound cardiac motion for predicting survival to hospital admission in patients with penetrating trauma 100% in another paper.
So the jury is out on management of cardiac arrest in trauma, and I still don’t know what to think. But I can’t get away from the fact that Professor Brohi’s approach seems sensible. What would I do, then, if a similar patient presented to my ED tonight?
I think that, having obtained IO access, I would either perform a thoracostomy or to insert two large bore chest drains. Either approach, in this case, would have immediately identified massive left haemothorax as the cause of his cardiac arrest. Given the length of time pre-hospital without an output, it might then have been appropriate to proceed to EDT or to stop the resuscitation. In reality, the patient was unlikely to have a positive outcome, but it is possible to forsee a situation where a ventricular or vessel repair could be undertaken quickly enough to facilitate effective volume replacement and possibly successful resuscitation. One of the challenges to me, as an Emergency Medicine trainee, is to find ways and opportunities to prepare myself to perform EDT when it is needed. Thankfully, there is a wealth of online resources and videos which utilise “safe” teaching methods; as a rarely-performed procedure which has the potential to significantly improve outcomes, there is a duty to be prepared to perform EDT in the Emergency Department when it is indicated. Would I perform the thoracotomy? Probably not yet, but I would be prepared to assist – especially after spending some time trying to learn from Andy Neill’s excellent collection of learning resources.
It seems that traumatic cardiac arrests, therefore, should be approached very differently from the standard ACLS/ALS cardiac arrest teaching. Traumatic cardiac arrest carries a very low survival rate, irrespective of the circumstances and care delivered. The prognostic evidence for any particular mechanism, course of action or presenting ECG finding is low quality due to small patient numbers and low proportion of survivors. Beyond the Airway-Breathing-Circulation approach, including IO access where venous access cannot be obtained quickly, consideration of the mechanism of injury may help adapt response and interventions to prevent death; it seems there may be little benefit from chest compressions when underlying pathology has not been addressed. Ultrasound has potential as an adjunct to resuscitation in ruling-in the presence of tamponade, pneumothorax or intra-abdominal free fluid. After ultrasound and insertion of chest drains, Emergency Department thoracotomy may be indicated in either blunt or penetrating trauma; the patient is likely to die even if EDT is performed but very unlikely to survive if it is not.
Future editions of the Advanced Trauma Life Support course would be improved if exploration of the management of traumatic cardiac arrest – and when to discontinue resuscitation – was included in greater depth.
And now, my esteemed colleagues – over to you. What would you do? Can anyone hold back from the urge to apply some pressure?
Before you go please don’t forget to…