Rachael Gartly¹, Sameer Kalgaonkar², Chris Isles^3*

¹Third year medical student, University of Glasgow, Glasgow G12 8QQ, United Kingdom

²Department of Radiology, Dumfries and Galloway Royal Infirmary, Dumfries DG2 8RX, United Kingdom

³Department of Medicine, Dumfries and Galloway Royal Infirmary, Dumfries DG2 8RX, United Kingdom

^*Corresponding Author: Rachael Gartly, Third year medical student, University of Glasgow, Glasgow G12 8QQ, United Kingdom

Abstract
Background
To investigate how often aortic dissection (AD) may masquerade as pulmonary embolism (PE).

Methods
Retrospective cohort study. The electronic records of 998 consecutive patients who underwent CT imaging of their aorta between 2016 and 2021 were analyzed. Those with aortic dissection were then examined to determine if CTPA had been requested before CT aortography.

Results
Thirty-one patients had AD. Five (16 %) AD patients underwent CTPA as their initial imaging modality mainly because they presented with pleuriticpain. All five had radiological signs suggestive of a false lumen which was confirmed subsequently in each case by CT aortography. All five had type B dissections. Four of the five had no evidence of PE, while one had PE and AD.

Discussion
We have shown that a significant proportion of patients with AD were misdiagnosed as PE initially. clinicians should be aware that aorticdissection may masquerade as pulmonary embolism.

Keywords: aortic dissection, pulmonary embolism, CT pulmonary angiography, CT aortography

Introduction
Acute aortic dissection (AD) is an important cause of cardiac death, with an incidence estimated to lie between 5 and 30 cases per million per year [1]. It occurs when an intimal tear allows blood to penetrate the aortic media or when the rupture of vessels within the media resultsin separating the aortic wall layers [2]. Intramural hematoma, thoughsometimes regarded as a separate entity, is usually considered a formof AD [1]. ADs are generally categorized as type A or type B based on the Stanford classification: Type A dissections involve the ascending aorta, whereas type B dissections do not, regardless of the site of the entry tear [2]. Risk factors associated with dissection are hypertension, smoking, male sex, genetic connective tissue disorders, and positive family history of the aortic disease [1,2].

AD usually presents with the sudden onset of severe anterior chest pain (more common in type A dissection) and back pain (more common in type B) [2,3]. The pain is typically described as ‘tearing’ or ‘ripping’ in character and may migrate as the dissection progresses [2,4]. However, AD may also be painless in 15 % of patients [5]. This and other atypical presentations frequently lead to delays in diagnosis. Spittell and colleagues found that the initial clinical impression was of an alternative diagnosis in up to 38 % of 235 patients with aortic dissection [5]. A Task Force of the European Society of Cardiology reported that up to 30 % of patients later found to have aortic dissection were initially suspected of having other conditions [6]. A retrospective study of 66 consecutive patients with the acute aortic syndrome, which included 43 patients with type A and 20 patients with type B dissection, revealed that 26 (39 %) were incorrectly diagnosed initially [7].

The most common misdiagnosis in patients subsequently found to have AD is an acute coronary syndrome (ACS) [7-9], the pain of which may radiate to the back, and which may also be a consequence of AD if the dissection extends proximally to involve the coronary vessels. Other presentations that may lead to a delayed diagnosis include suspected ischaemic stroke [10,11], acute abdomen [12], and PE [13-17]. Against this background, we undertook a study to investigate how often AD presents as suspected PE, based on the number of patients with AD in whom CTPA was requested before diagnostic CT Aortography.

Methods
This was a retrospective cohort study conducted in Dumfries and Galloway Royal Infirmary, which serves a population of 148,000 in southwest Scotland. Data were collected on 998 consecutive patients having CT aortography between November 2016 and October 2021. These patients underwent 1363 CT aortograms, as many had repeat or follow-up imaging within our timeframe. Each patient was counted once and classified according to their initial scan unless they subsequently developed an AD. We considered the following predefined disease categories: aortic dissection (AD), aortic disease other than dissection, peripheral vascular disease (PVD), and others.

Results
Thirty-one of 998 consecutive patients who underwent CT aortography had an aortic dissection (figure 1). Eleven (35 %) of these were type A, and twenty (65 %) were type B. Twenty-four (77 %) were acute, and seven (23 %) were chronic. Five (16 %) patients with AD were initially suspected of having PE and underwent CTPA scans before CT aortography. All five had type B dissections. Their median age was 72 years (range 54 to 76) at the time of diagnosis, four of the five were male, and two were hypertensive. D-dimer values at presentation in these five patients ranged from 1,151 to 11,027ng/ml (normal range < 500ng/ml). The clinical features which led the receiving physician to suspect PE as the most likely diagnosis at presentation were as follows.

Figure 1: Study flow diagram showing the resulting numbers in each predefined disease category for those who underwent CT aortography. CTPA= CT Pulmonary Angiogram.

Case 3. 54-year-old male who complained of sudden onset pleuritic chest pain, which radiated to the upper abdomen and was associated with palpitations and sweating. He was not SOB and had no significant past medical history. The presentation followed a 10hr long-haul flight three days prior in which he hardly moved. BP on admission was 174/100mmHg. ECG was normal. Recent immobilization, pain worse on inspiration, and a D-dimer of 11,027ng/ml led to a request for CTPA. This patient was unusual in that he was found to have both PE and AD [19].

Case 4. 73-year-old male who developed stabbing central chest pain following the death of his mother 24hrs previously. He had a background of PVD and smoked 10-15 cigarettes daily. He was hemodynamically stable with a BP of 136/69mmHg and little difference between his arms. ECG was normal, but D-dimer was raised at 1151ng/ml, hence the referral for CTPA.

Case 5. A 72-year-old female presented with a stabbing back pain that radiated to the chest and was worse on deep inspiration. She was SOB, nauseated, and tender on palpation over the thoracic vertebrae and paravertebral muscles. This was on a background of chronic obstructive pulmonary disease, angina, myocardial infarction, and triple vessel coronary disease. She was an ex-smoker. BP on admission was 133/54mmHg with no difference between arms. Her ECG was expected, but D-dimer was raised at 1690ng/ml, hence the referral for CTPA.

All five AD patients underwent CTPA as their initial imaging modality. All had radiological signs suggestive of a false lumen (figure 2 top panel) which was subsequently confirmed in each case by CT aortography (figure 2 lower panel). Four of the five had no evidence of PE, while one (case 3), somewhat unusually, had both. All five dissections were Stanford type B.

Figure 2: Representative example showing impression of false lumen in descending aorta on CTPA (upper panel) subsequently confirmed as AD by CT aortography (lower panel). CTPA = CT Pulmonary Angiogram. AD = Aortic Dissection.

Discussion
Our study highlights that a proportion of patients who develop AD are first thought to have PE. Misdiagnosis of AD as PE is less well recognized than misdiagnosis as ACS [7-9], although several case reports have shown that AD may masquerade as PE [13-17].

We and others have shown that occasional AD cases may be found in a consecutive series of patients undergoing CTPA for suspected PE. A retrospective review of 465 CTPA scans found AD in two [20], while Lim and colleagues found two cases of AD among 85 patients with very high D-dimer whose CTPA was negative for PE [21]. To our knowledge, there are no previous estimates of the frequency with which patients whose final diagnosis is AD were first considered to have PE. We suspect the finding that 5(16 %) of 31 patients with AD had CTPA before CT aortography is a higher figure than most might have expected.

D-dimer is invariably elevated in PE and AD but lacks specificity for both [21,22]. These observations may nevertheless explain why some of our AD patients were thought to have PE initially. D-dimer can, however, be used as a rule-out test for AD. A prospective multicenter study [23] and a meta-analysis [22] concluded that D-dimer levels of less than 500ng/ml obtained within the first 24 hours of symptom onset could be used to identify patients who do not have AD.

Recent advances in CT imaging likely reduce the risk that AD is misdiagnosed as PE. Multidetector CT angiography is increasingly employed when evaluating the vascular system using higher quality and more efficient fast scanners in managing radiation dose [24]. Traditional CTPA protocols involve a bolus tracking technique with the region of interest placed on the pulmonary artery [25]. This gives excellent contrast within the pulmonary artery but limited evaluation of the aorta. A triple rule-out protocol involving an ECG-gated study can assess the pulmonary artery, aorta, and coronary vessels when the region of interest is placed on the ascending aorta [24]. Based on recent experience and review of these recent advances, radiologists in Dumfries and Galloway Royal Infirmary have adopted a non-ECG gated double rule-out protocol, requiring a slight increase in the radiation dose and volume of contrast, to image the pulmonary arterial tree and aorta simultaneously.

Strengths and limitations
This study included every CT aortogram performed within the specified time frame. The case information was available electronically for every patient. This ensured that no dissection was missed and avoided selection bias. There is controversy surrounding intramural hematomas and whether they result from dissection or develop into dissection [1]. We classified an intramural hematoma as a dissection for this study. Patients with dissection were demographically like those reported by others in that the majority were male and of older age [1-4]. It is likely that our results are not only representative of AD generally but also valid and applicable to the UK population.

A potential limitation of this study is that it was retrospective and observational. General physicians evaluated presenting symptoms subjectively rather than cardiologists with a particular interest in aortic diseases. The dissections which were first thought to be PE might have been diagnosed more rapidly had there been a greater awareness of the atypical presentations of AD [5-8]. It is also possible that clues to AD were present but not recognized or documented in the case record.

Conclusion
We have shown that a significant proportion of patients with AD were misdiagnosed as PE initially, mainly because they presented with pleuritic pain. Clinicians should be aware that aortic dissection may masquerade as pulmonary embolism.

Funding: None declared.

Competing interests: None declared.

References

1. Cavalcanti LRP, Sá MPBO, Campos JCS, Braga PGB, Perazzo AM, et al. (2020) Acute Aortic Dissection: an Update. Curr Emerg Hosp Med Rep. 8(3): 90-102.
2. Erbel R, Aboyans V, Boileau C, Bossone E, Bartolomeo RD, et al. (2014) for the Task Force for the Diagnosis and Treatment of Aortic Diseases of the European Society of Cardiology. 2014 ESC Guidelines on the diagnosis and treatment of aortic diseases. Eur Heart J. 35(41): 2873-2926.
3. Pape LA, Awais M, Woznicki EM, Suzuki T, Trimarchi S, et al. (2015) Presentation, Diagnosis, and Outcomes of Acute Aortic Dissection: 17-Year Trends from the International Registry of Acute Aortic Dissection. Am J Cardiol. 66(4): 350-358.
4. Bonaca MP, O'Gara PT (2014) Diagnosis and Management of Acute Aortic Syndromes: Dissection, Intramural Haematoma, and Penetrating Aortic Ulcer. Curr Cardiol Rep. 16(10): 536.
5. Spittell PC, Spittell JA Jr, Joyce JW, Tajik AJ, Edwards WD, et al. (1993) Clinical features and differential diagnosis of aortic dissection: experience with 236 cases (1980 through 1990). Mayo Clin Pro. 68(7): 642-51.
6. Erbel R, Alfonso F, Boileau C, Dirsch O, Eber B, et al. (2001) Diagnosis and management of aortic dissection. Eur Heart J. 22(18): 1642-1681.
7. Hansen MS, Nogareda GJ, Hutchison SJ (2007) Frequency of and Inappropriate Treatment of Misdiagnosis of Acute Aortic Dissection. Am J Cardiol. 99(6): 852-856.
8. Kurabayashi M, Miwa N, Ueshima D, Sugiyama K, Yoshimura K, et al. (2011) Factors leading to failure to diagnose acute aortic dissection in the emergency room. Journal of Cardiology. 58(3): 287-293.
9. Jackson E, Stewart M. Thoracic aortic dissection presenting as acute coronary syndrome. BMJ Case Rep. 2013: bcr2013201904.
10. Uchino K, Estera A, Calleja S, Alexandrov AV, Garami Z (2005) Aortic dissection presenting as an acute ischemic stroke for thrombolysis. Journal of Neuroimaging. 15(3): 281-283.
11. Grupper M, Eran A, Shifrin A (2007) Ischemic Stroke, Aortic Dissection, and Thrombolytic Therapy—the Importance of Basic Clinical Skills. J Gen Intern Med. 22(9): 1370-1372.
12. Scaglione M, Salvolini L, Casciani E, Giovagnoni A, Mazzei MA, et al. (2008) The many faces of aortic dissections: Beware of unusual presentations.  Eur J Radiol. 65(3): 359-364.
13. Hado HS, Scarpello JH, Barton T, Scarborough H, Elhadd TA (2005) A case of thoracic aortic dissection presenting as lateral pleuritic chest pain. Emerg Med J. 22(3): 229-230.
14. Stougiannos PN, Mytas DZ, Pyrgakis VN (2007) The changing faces of aortic dissection: an unusual presentation mimicking pulmonary embolism. Heart. 93(11): 1324.
15. Taylor RC, Vedutla TS, Myint PK, Meenakshisundaram SK (2010) Diagnosis of pleuritic chest pain query PE: Beware of dissecting thoracic aneurysm of aorta. Int J Cardiol. 140(3): E56- E58.
16. Sinha Y, Saleh M, Weinberg D (2013) Use of heparin in aortic dissection: beware the misdiagnosis of acute pulmonary embolism. BMJ Case Rep;2013: bcr2013009367.
17. Berkowitz EJ, Kronzon I, Whyburn J, Perk G (2016) Aortic dissection mimicking a massive pulmonary embolism. Eur Heart J Cardiovasc Imaging. 17(10): 1145.
18. Abishek J, Akintayo R, Isles C (2022) Unexplained chest pain in a patient with giant cell arteritis? Think aortic dissection. J R Coll Physicians Edinb. 52(2): 124-127.
19. Chaulagai B, Acharya D, Poudel S, Puri P (2021) Simultaneous Aortic Dissection and Pulmonary Embolism: A Therapeutic Dilemma. Cureus. 13(1): e12952.
20. Ahmad DS, Esmadi M, Todd A, Kavanagh K, Ahsan H (2013) Incidental vascular findings on CT pulmonary angiography (CTPA). Egypt J Radiol Nucl Med. 44(2): 175-182.
21. Lim J, Cardle C, Isles C (2021) Patients with markedly elevated D-dimer who do not have pulmonary embolism. Postgrad Med J. 97(1144): 77-82.
22. Shimony A, Filion KB, Mottillo S, Dourian T, Eisenberg MJ (2011) Meta-Analysis of Usefulness of D-Dimer to Diagnose Acute Aortic Dissection. Am J Cardiol. 107(8): 1227-1234.
23. Suzuki T, Distante A, Zizza A, Trimarchi S, Villani M, et al. (2009) Diagnosis of Acute Aortic Dissection by D-Dimer. Circulation. 119(20): 2702-2707.
24. Frauenfelder T, Appenzeller P, Karlo C, Scheffel H, Desbiolles L, et al. (2009) Triple rule-out CT in the emergency department: protocols and spectrum of imaging findings. Eur Radiol. 19(4): 798-799.
25. Murphy A, Weerakkody Y, Knipe H, et al. (2022) CT pulmonary angiogram (protocol).