Acute aortic dissection is a potentially life-threatening break in the lining of the main artery leaving the heart.2
It is an uncommon disease with a reported incidence in the range of 3–4 per 100 000 people per year, but remains an important differential in cases of chest pain due to its high mortality in the region of 25–60%.3 4
Diagnosis in the emergency room is often hampered by the absence of clear-cut clinical features, a wide range of common differential diagnoses for chest pain and the lack of simple diagnostic investigations.5
The International Registry of Acute Aortic Dissection (IRAD) has accumulated the largest database of aortic dissections to date in an attempt to characterise the disease further. Common comorbidities and presenting features of patients entered in IRAD are shown in and .6 7
Associated factors of patients presenting with acute aortic dissection in IRAD
Presenting features of patients with dissection in the IRAD.
One of the dilemmas faced by admitting physicians is whether to proceed to definitive imaging with CT or MRI. It is clearly impractical to advise scanning of all patients that present with chest pain, yet chest pain is commonly the sole presenting feature in cases of acute aortic dissection.5
Whereas chest radiograph, electrocardiograph and echocardiography are used as simple non-invasive investigations, they all either lack sensitivity or specificity.
The use of d-dimer as a useful investigation for acute aortic dissection was first reported in a small case series by Weberet al
Subsequent meta-analyses have generally demonstrated a sensitivity of between 95% and 100% with a cut-off value of 500 ng/ml, providing a useful and simple rule-out test for suspected cases of acute aortic dissection.9 10
A recent prospective multicentre substudy of IRAD, IRAD-Bio, which enrolled 220 patients with suspected aortic dissection also confirmed these findings.11
In addition, it demonstrated that greater sensitivity and specificity could be achieved if the sample for analysis was taken within 6 h of onset of symptoms.
One of the drawbacks of d-dimer as a test for aortic dissection is its lack of specificity. However, the IRAD-bio data have shown that if a cut-off value of 1600 ng/ml (normal <500 ng/ml) were used on patients presenting within 6 h of onset of pain, it would generate a positive likelihood ratio of 12.8. Thus, in a group of people with suspected aortic dissection, a high d-dimer could identify those with a high probability of disease. Whether integrating these findings into chest pain algorithms would decrease the need for further imaging or increase correct diagnoses remains to be seen in prospective studies.
Considering the high mortality rate associated with the disease, it is unlikely that physicians will rely on any one test, unless it has very high sensitivity and specificity, to reliably include or exclude acute aortic dissection. However, for a select group of patients the risk of acute aortic dissection can be reduced to virtually zero. For example, in high-risk patients, the probability of dissection can be reduced from 50% to 7% by the absence of tearing or ripping pain, pulse or blood pressure differentials, and mediastinal widening on chest x-ray radiograph.12
If combined with a negative d-dimer, the post-test probability would be negligible obviating the need for further investigations such as CT or MRI.10
- Consider using d-dimer in cases of acute chest pain alongside routine investigations such as trans-thoracic echocardiogram, ECG and chest x-ray radiograph.
- If clinical suspicion of acute aortic dissection is low, d-dimer could be a useful investigation to effectively rule out dissection.
- In suspected cases of aortic dissection, a d-dimer greater than 1600 ng/ml suggests high probability of disease.