Image © Freepik
Matt is a Professor of Emergency Medicine at Edinburgh University and the Royal College of Emergency Medicine, and a Consultant and NRS Fellow at the Royal Infirmary of Edinburgh. He is Research Director of EMERGE, the Emergency Medicine Research Group Edinburgh, a multidisciplinary clinical research group delivering frontline applied health research which is the largest recruiter of EM research participants in the UK.
Matt’s research work focuses broadly on Emergency Medicine research in challenging areas of acute care such as syncope, palpitations, and now aortic dissection, as well as the innovative use of novel technology. He has published over 120 papers and personally been awarded over £8.4 million in research grants. When not a work, Matt is a below average runner, a passable golfer, and an undistinguished cellist in The Really Terrible Orchestra.
Mr S was a middle-aged gentleman who presented to our Emergency Department (ED) about six years ago when I was working in our Rapid Access Triage assessment area. This is essentially a triage area, common in many UK EDs, where a senior doctor is available to help identify potentially serious conditions early and rapidly fast track anyone who is unwell, or who could redirected to be better managed by another specialty or service elsewhere.
Mr S presented with a cold white left arm which had developed two hours previously. There were no other presenting symptoms and whilst I did consider acute aortic dissection (AAD), it looked very much like a straightforward case of an acute embolic event causing acute limb ischemia, a surgical emergency. Mr S went through to our ED, and I handed him over to an ED consultant colleague who also saw him and agreed that there wasn't anything else to suggest AAD. They referred to the vascular surgery team, and Mr S went to the operating theatre later that afternoon where he had a brachial artery thrombus removed. Mr S went back to the ward but tragically at 3:00 am suffered a cardiac arrest from which he was unable to be resuscitated. Mr S underwent a post-mortem examination which showed a type A AAD. It later transpired that Mr S, had presented two weeks previously to another department with chest pain, had undergone a troponin test which was normal, and he was discharged home with a diagnosis of non-specific chest pain.
Now I'm certainly not alone in having misdiagnosed a case of AAD. Data from the EDs of our two Edinburgh hospitals, showed that between 2011 and 2020, there were 26 patients whose diagnosis of AAD was delayed or missed. But it's not only us, 1 in 3 patients with AAD are misdiagnosed1,2, 1 in 4 patients are not diagnosed until over 24 hours after presentation to the ED3, and AAD is a common cause of fatality-related negligence claims4. The tragic thing is we know that prognosis is best when patients are treated early with mortality increasing 2% per hour of diagnosis delay5. AAD is a treatable condition with an 80% survival when diagnosed and treated on time.
Currently 4,000 patients a year in the UK suffer AAD. About half of these patients die before they reach hospital, but around 50% arrive at hospital6. AAD can affect people of any age, but it is more common as you get older. However, it is so important to remember that this condition also affects young people, with 25% of patients aged under 50 and half aged under 60. Although in the older AAD patient, atherosclerosis is the predominant underlying cause, in the younger patient connective tissue disease predominates. So, take home point, never use age as a reason to not consider AAD.
Now you might say, well AAD is rare, I'm just not going to see a case. Whilst it may not be as common as some other conditions such as acute myocardial infarction, AAD is not rare. In NHS Lothian between our two EDs, we see between about 15 patients a year with AAD, the commonest type being a type A dissection. Most of our ED consultants will see at least one case of AAD a year equating to around 30-40 during a career. Another statistic to put the condition into perspective is that 2,500 patients die within a month of being diagnosed with AAD, a number that's more than the number of people who die in the UK from road traffic accidents (1,800 per year), or from pulmonary embolism (PE) (2300 per year)6. This large cause of mortality is put into perspective when you think of the huge infrastructure we have for managing trauma, with developed pre-hospital systems and established major Trauma Centres.
So, what does the future hold. The Oxford Vascular study6 tells us that UK cases of AAD are set to rise over the next 10 years to over 5,500 and if the trajectory remains the same, almost 3,500 people will lose their lives every year by 2050, mainly due to the population aging and AAD being more predominant in the 40-to-70-year age group.
So, let’s now move onto pathophysiology, and firstly we're going to discuss the anatomy of the aorta. The aorta starts at the aortic root where the aortic valve is sited at the outflow track of the left ventricle and from where the coronary arteries originate. The aorta then continues as the ascending thoracic aorta up to the aortic arch where three main vessels originate, the combined right subclavian and right common carotid, followed by the left common carotid and finally the left subclavian artery. This vessel is especially important when we talk about the type of AAD, whether it is a type A or a type B, as this influences the patient’s management once diagnosed. An AAD that involves the aortic root or the ascending aorta up to the left subclavian artery, is classified as a type A AAD and is managed surgically with cardiothoracic surgeons replacing the arch of the aorta. Any AAD originating after this is classified as a type B AAD and is managed medically with blood pressure lowering to allow the AAD to become chronic and to settle, normally under cardiology teams in a Coronary Care Unit (CCU) environment.
If we look at a cross-sectional slice through the aorta, we see that there are three main layers. The inner intima, the media, a reasonably thick layer compared to the others made up of more than 50 alternating layers of elastin and smooth muscle cells, and the outer adventitia layer. Essentially in AAD, you get a small tear in the intimal inner lining of the aorta which allows blood into the middle media area. Because the media is weaker than the other two walls, the blood, coming straight out of the heart and therefore under pressure, tracks up and down through the media separating the layers of the intima and the adventitia. The blood in the aortic media then pushes the dissection flap into the middle of the aorta, separating the true from the false lumen.
AAD is now commonly referred to as Acute Aortic Syndrome (AAS) which is made up of 4 conditions: Type A AAD, Type B AAD, intramural aortic haematoma and penetrating aortic ulcer. In intramural haematoma, blood leaks into the aortic media at low pressure, forming a thrombus that pushes the outer wall of the aorta outward, leaving a relatively normal appearing aortic lumen. A penetrating atherosclerotic ulcer allows blood to enter the aortic media, but atherosclerotic scarring of the aorta typically confines the blood collection, often resulting in a localised dissection or pseudoaneurysm.
AAS is a dynamic process, as the calibre of the true and the false lumens is very dependent on the pressure in both these lumens, and this pressure will determine whether the dissection flap moves
more towards the true lumen or towards the false lumen. When blood moves into the false lumen, a few things can happen. Pressure building up in the false lumen, can lead to rupture, re-entry tear, branch vessel occlusion or true lumen collapse. If the false lumen blood ruptures out of the aorta through the adventitia layer, the result is a bloody pericardial effusion, mediastinal haematoma or haemothorax, all normally fatal conditions. It is thought that 7% of out of hospital cardiac arrests are due to Type A aortic dissection7.
In a re-entry tear, the blood tracks back into the true aorta through a further tear in the intima, creating a double lumen channel meaning blood can go through either lumen and arrive back into the normal distal aorta. Branch vessel occlusion is essentially where the false lumen has blood within it which surrounds branches that come off the aorta. When there is a high enough pressure in the false lumen, it collapses the true lumen and restricts perfusion to the vessels coming off the aorta. This phenomenon explains why AAS can present with bizarre symptoms like stroke as branch vessel occlusion can temporarily occlude the left carotid artery for example, or you might find that there's a STEMI (ST elevation myocardial infarction) presentation because the left coronary artery is occluded. The patient may also present with limb ischemia, as in the case of the first patient we discussed, due to left subclavian artery occlusion. With AAS being a dynamic process, pressure changes between the true and the false lumens can lead to occlusion or reperfusion of different areas at different times. The pressure in the true and false lumen, and subsequent blood flow through these stabilises over a period of minutes, hours or at the most, a few days. Finally, in true lumen collapse, the pressure in the false lumen exceeds the pressure in the true lumen impeding distal perfusion in the true lumen resulting in distal organ ischemia.
At this point it is important to discuss the difference between AAS, and an abdominal aortic aneurysm (AAA) as they are very different pathologies, presenting in very different ways, with very different treatment. It is common for these two conditions to be confused in EDs with ED staff commonly discussing the process of doing bilateral blood pressures in both arms when they are considering an AAA. In AAS, where the patient has obliterated the true lumen to one arm (commonly the left), but not to the other (commonly the right), a different blood pressure in each upper limb may be recorded, or the clinician may not be able to feel a pulse on one side. An AAA is a slow growing dilatation of the aorta that happens over years. When the AAA gets to about five or six centimetres, the risk of it rupturing becomes significant, presenting with sudden onset abdominal or back pain with haemodynamic collapse secondary to blood leaking into the abdomen or tracking into the retroperitoneal space. AAA rupture in an area of aneurysmal dilatation, is very different to AAS. Patients who suffer AAS, whilst not having a normal aorta pathologically, do not commonly have an aortic aneurysm.
So now let's talk about the symptoms of AAS, which are consistent with the three main pathophysiological processes at play here. Firstly, dissection of the aortic wall is extremely painful. It's a sudden thing that happens in seconds, with the dissection peeling away the aortic media. The pain is sudden and intense but might settle once the dissecting process has stopped. It may be described as a ripping or tearing pain but not universally. Secondly, a contained rupture leading to pericardial effusion, mediastinal haematoma or haemothorax will cause intense physiological instability and perhaps breathlessness and/or hypotension. Finally, there are the end organ symptoms associated with malperfusion, such as STEMI and stroke mimics which may be transient, may recur and which may affect different organs.
So now we have discussed the pathophysiology of AAS, we are going to address why it is so difficult to diagnose. Chest pain is the commonest AAS presenting complaint (80%)8. Back (40%) and abdominal pain are not uncommon8, but there are two million chest, back or abdominal pain presentations to English EDs a year9, overwhelmingly due to causes other than AAS. 1 in 980 ED patients with atraumatic chest pain10 will have AAS but 979 will have other causes. This is important when we think about how we are going to diagnose AAS, because we can't arrange a CT aorta angiogram for everybody who has chest pain. The diagnosis of AAS in the ED is a low signal to high noise ratio problem with the weak signal of AAS being overwhelmed by the background noise of all the patients that we have presenting with complaints that could, but do not have AAS.
Research into the diagnosis of AAS in the ED is also problematic. One of the reasons we miss AAS, is because we don't always think about it. It has multiple presenting complaints, not just chest pain, and the Hawthorne effect means that when we start studying AAS, clinicians’ practice changes. D-dimer and CT aorta angiogram are not tests we routinely do in everybody meaning observational research is difficult, and research consent processes risk not recruiting unwell patients.
NHS Lothian recently teamed up with Frimley Health NHS Foundation Trust to perform a retrospective review of missed AAS cases between 2011 and 2020 to better understand why we miss AAS. Morbidity and Mortality records were used to identify 43 cases (including postmortem reports and complaints), as well as reviewing results of CT scans requested by downstream inpatient teams querying AAS following discharge from ED with a different diagnosis. Electronic patient records were reviewed by two independent reviewers to establish the reason the diagnosis was missed11. Of 43 cases, 22 were type A, 9 were type B with the rest being intramural aortic haematoma, penetrating aortic ulcer or unknown underlying AAS pathology. Chest pain was the presenting complaint in 27 patients (63%), with 28 describing symptoms being of sudden onset. The three commonest alternative diagnoses made were acute coronary syndrome (ACS), pulmonary embolism (PE) and non-specific chest pain.
So why did we miss AAS? In most of the cases, AAS was missed because it was never considered in the differential diagnosis. In some of these cases, the clinician was satisfied by an alternative clinical diagnosis or happy that ACS was excluded. In other cases, AAS was clearly considered but not pursued further with imaging due to the clinician being inappropriately reassured by the absence of certain ‘textbook’ clinical symptoms and signs, by resolved symptoms, or by a normal chest radiograph (CXR).
Lovatt et al1, reviewed 12 studies, including 1663 AAS patients with a misdiagnosis rate of 33.8%. Factors leading to the diagnosis being missed included the symptoms being attributed to other conditions, the reassurance of a normal CXR, patients having walked into the ED and the absence of ‘typical’ AAS symptoms such as tearing or ripping pain, differential upper limb blood pressures, a pulse deficit or acute hypertension. The lack of any of these features does not reliably rule out AAS. For example, if you think a patient has a 50% pre-test (pre-CT scan) chance of having AAS, if you do a CXR and it is totally normal, you only reduce the chance of the patient having AAS to 40%, which is still significant and not sufficient to be able to rule out AAS. So, is there anything that is helpful? Well abrupt onset pain as well as worst ever pain are much more associated with AAS and are a useful start. If a patient doesn't have abrupt onset pain it halves the likelihood that the patient in front of you has AAS, but unfortunately still doesn't rule it out. If someone has abrupt onset pain, we need to be investigating further. Worst ever pain, is AAS until proven otherwise.
The NHS Resolution report released in 20224, looked at clinical negligence fatality claims in English EDs. Of 86 claims worth a total of £5.8 million the most common causes of deaths were related to misdiagnosis of infection/sepsis, PE, suicide, ACS and AAS. The review identified poor awareness and poor recognition of the significance of the presenting symptoms along with evidence of lost opportunities to use information from the ambulance and triage notes. It is vital we improve our use of ambulance and triage notes which always contain helpful and useful information that may be lost in handover. The patient you have in front of you, may be very different to how they were maybe several hours previously when they had their initial symptoms.
I've painted a bleak picture here of how we diagnose, or fail to diagnose AAS in the ED, so how can we do better? There have been campaigns and great educational resources such as those by The Aortic Dissection Charitable Trust (TADCT). Think Aorta have also attempted to improve awareness of AAS in the ED. However, these have not led to improvements in mortality. Awareness is extremely important, and it is vital that we have campaigns like these, but they are not the only answer. A clinician survey of practice across the UK showed that only 12 of 56 EDs have a formal pathway for working up patients with potential AAS and no particular guideline predominated, probably due to none being particularly simple to use in the ED, and there being no robust evidence-based method of ruling out AAS12.
The diagnostic challenge is that Aorta CT Angiogram (CTA) has high sensitivity and specificity to diagnose AAS. Locally in NHS Lothian we scan around 300 patients per year looking for AAS, and about 5-6% of these scans are positive for AAS. However, over testing leads to diagnostic yields as low as 2%3,13, significant costs and resource implications, ionising radiation risks, CT delays for non-AAS patients and the burden of ‘incidentalomas’. Clinicians therefore need to use CTA selectively, yet despite several being proposed, there is no validated clinical decision tool for this scenario,14-16 and none that has been studied in undifferentiated ED populations. All clinical decision tools have low diagnostic yield for AAS, modest specificity and lead to higher rates of CTA. D-Dimer has been suggested as a rule-out biomarker in low pre-test probability patients17-18 and is part of the ADD-RS clinical decision tool, but it is currently unclear whether any AAS clinical decision tools have sufficient sensitivity to be acceptable to clinicians, which is the most accurate, and whether AAS clinical decision tools are likely to lead to over-investigation with CTA and D-Dimer.
Even the best risk guidance or clinical decision tool is meaningless unless applied to the correct patients. Clinicians must understand which of the myriad of presentations could be the ‘needle in a haystack’ manifestation of AAS and any diagnostic work-up must be able to be applied to an all-comer population of patients with potential AAS symptoms. With these challenges in mind, our group has recently completed the DAShED study19, aiming to describe the characteristics of ED attendances with possible AAS, and to assess clinical decision tools in our undifferentiated ED population with a view to informing future research in this area and hopefully help ED clinicians to make this most difficult and most terrifying of diagnoses.
Summary
AAS is rare, devastating and often misdiagnosed, missed, or delayed in diagnosis, due to lack of consideration, lack of awareness, atypical presentations, mimics of other disease, clinicians being falsely reassured by a normal CXR, or lack of typical clinical signs. Its clinical features are highly unreliable but sudden and/or severe pain must always be taken seriously. AAS is a dynamic process and symptoms may come and go. We can improve things with better education, i.e., who to consider AAS in, and better detection strategies, i.e., once AAS is a consideration, how do you rule it out without investigating everyone with a CTA. The current focus20-21 will hopefully do this and mean that fewer AAS patients suffer misdiagnosis, missed or delayed diagnosis in future.
References:
[1] Lovatt S et al. Misdiagnosis of aortic dissection: A systematic review of the literature. Am J Emerg Med 2022; 53: 16-22
[2] Harris KM et al. Correlates of delayed recognition and treatment of acute type A aortic dissection the International Registry of Acute Aortic Dissection (IRAD). Circulation 2011; 124: 1911-8
[3] Lovy AJ et al. Preliminary development of a clinical decision rule for acute aortic syndromes. Am J Emerg Med. 2013; 31: 1546 – 1550.
[4] NHS Resolution 2022; Report 1 of 3: High value and fatality related claims https://resolution.nhs.uk/wp-content/uploads/2022/03/1-NHS-Resolution-ED-report-High-value-and-fatalities.pdf
[5] Pape LA et al. Presentation, diagnosis, and outcomes of acute aortic dissection: 17-year trends from the International Registry of Acute Aortic Dissection. J Am Coll Cardiol 2015; 66: 350-8
[6] Howard DP, Banerjee A, Fairhead JF, Perkins J, Silver LE, Rothwell PM; Oxford Vascular Study. Population-based study of incidence and outcome of acute aortic dissection and premorbid risk factor control: 10-year results from the Oxford Vascular Study. Circulation. 2013; 127(20): 2031-7. doi: 10.1161/CIRCULATIONAHA.112.000483 (accessed 27th April 2023)
[7] Gouveia e Melo R, Machado C, Caldeira D, Alves M, Lopes A, Serrano M, Fernandes R, Pedro LM. Incidence of acute aortic dissections in patients with out of hospital cardiac arrest: A systematic review and meta-analysis of observational studies. IJC Heart & Vasculature 2022; 38: 100934
[8] Erbel R et al. 2014 ESC guidelines on the diagnosis and treatment of aortic diseases: document covering acute and chronic aortic diseases of the thoracic and abdominal aorta of the adult. The task force for the diagnosis and treatment of aortic diseases of the ESC. Eur Heart J. 2014; 35: 2873–2926
[9] NHS Digital A&E 2021; https://digital.nhs.uk/data-and-information/publications/statistical/hospital-accident--emergency-activity/2021-22 (accessed 27th April 2023)
[10] Alter SM, Eskin B, Allegra JR. Diagnosis of Aortic Dissection in Emergency Department Patients is Rare. West J Emerg Med. 2015; 16(5): 629-31
[11] McLatchie R, Wilson S, Reed MJ, Ticehurst F, Easterford K, Alawiye S, Cowan A, Gupta A, on behalf of the Aortic Dissection Diagnosis Research Group. Why do Emergency Department Clinicians Miss Acute Aortic Syndrome? A Case Series
and Descriptive Analysis. ECJ 2023; 19: 11153; https://doi.org/10.4081/ecj.2023.11153 (accessed 27th April 2023)
[12] McLatchie R, Aakash Gupta A, Wilson S, Reed MJ on behalf of the Aortic Dissection Diagnosis Research Group. A sinister needle in an enormous haystack: A clinician survey regarding Acute Aortic Syndrome (AAS) diagnostic practice in United Kingdom Emergency Departments. Emergency Care Journal 2022; 18 (10758); https://www.pagepressjournals.org/index.php/ecj/article/view/10758/10550 (accessed 27th April 2023)
[13] Ohle R et al. Variation in emergency department use of computed tomography for investigation of acute aortic dissection. Emerg Radio 2018; 25: 293–298.
[14] Rogers AM et al. Sensitivity of the aortic dissection detection risk score, a novel guideline-based tool for identification of acute aortic dissection at initial presentation: results from the international registry of acute aortic dissection. Circulation. 2011; 123: 2213–2218
[15] Ohle R et al. Diagnosing acute aortic syndrome: a Canadian clinical practice guideline. CMAJ. 2020 Jul 20;192(29): E832-E843
[16] Morello F et al. Development & Validation of a Simplified Probability Assessment Score Integrated With Age-Adjusted d-Dimer for Diagnosis of Acute Aortic Syndromes. J Am Heart Assoc 2021; 10(3): e018425
[17] Yao J, Bai T, Yang B, Sun L. The diagnostic value of D-dimer in acute aortic dissection: a meta-analysis. J Cardiothorac Surg 2021; 27; 16(1):343.
[18] Bima P et al. Systematic review of aortic dissection detection risk score plus d-dimer for diagnostic rule- out of suspected acute aortic syndromes. AcadEmerg Med 2020; 27:1013-1027
[19] The DAShED (Diagnosis of Aortic Syndrome in the ED) Study (DAShED); https://clinicaltrials.gov/ct2/show/NCT05582967
[20] Aortic Dissection: Patient Pathways and Research Funding; House of Commons; Tuesday 13 December 2022; https://hansard.parliament.uk/Commons/2022-12-13/debates/EDA03A26-D578-48F2-986A-8A4D6569C4E6/AorticDissectionPatientPathwaysAndResearchFunding#contribution-ADAA995D-FE14-4E02-A66C-C91C46AC711F (accessed 27th April 2023)
[21] https://www.jla.nihr.ac.uk/priority-setting-partnerships/emergency-medicine-refresh/top-10-priorities.htm (accessed 27th April 2023)
Published by: