Aortic Dissection in the Elderly: Raise That Clinical Index of Suspicion

Case Presentation

An 84-year-old Japanese-American man with a history of dementia, cerebrovascular accident, hypertension, chronic obstructive pulmonary disease (COPD), and cirrhosis presented to the emergency room for chest pain in his mid-sternum after coughing. Symptom history was obtained with his daughter’s assistance and language interpretation because of his impaired cognitive status. Pain was described as a “pulling” type of “dull” pain that could be neither quantified nor further qualified. There was no apparent radiation of the pain or associated nausea or vomiting. Pain was relieved by sitting up but worsened by lying down. His blood pressure on arrival was 202/102 mm Hg. He was awake, alert, and in mild distress. His heart rate and rhythm were regular. Lungs were clear, and abdominal examination showed mid-epigastric tenderness but no guarding or rebound. Bowel sounds were normal, and abdomen was soft without organomegaly. There was trace pedal edema. An electrocardiogram (ECG) showed normal sinus rhythm with left ventricular hypertrophy and some repolarization abnormality. Both creatine kinase (CK)-MB and troponin I were normal. A chest x-ray showed COPD with mild basilar parenchymal scarring, mild cardiomegaly, and aortic and pulmonary vessels of normal size. He was given calcium carbonate, belladonna alkaloid combinations and phenobarbital, nitroglycerin sublingual, and morphine sulfate 2 mg. Pain resolved. The patient’s attending physician was called to admit the patient. On arrival of the attending physician, the emergency room nurse casually mentioned that when the patient arrived in the emergency room, there was something about his facial expression that made her think of aortic dissection. Formally, however, his facial expression was not described in the nursing notes, as this may have been interpreted as being on the verge of subjective gut feeling. Heeding the astute nurse’s observation, however, the clinical index of suspicion was raised and an additional test was requested—computed tomography (CT) of the chest. The CT showed a type III aortic dissection extending just above the renal arteries.  

History

Of historical interest is that aortic dissection is an old disease. An aneurysm was described by Galen (ad 130-200) in the second century, but Morgagni (1682-1771) first described a dissecting aneurysm in the 18th century.^1,2 He described a situation involving one patient as “…blood had burst forth from the great artery that was eroded at the part where, being about to descend to the vertebrae, it is inflected…”² In 1760 at the age of 76, King George II is said to have died from aortic dissection while straining to have a bowel movement.¹ In this generation, the actor John Ritter recently succumbed to aortic dissection.

Demographics

The incidence of aortic dissection is estimated at 5-30 cases per million persons per year.³ Over the age of 40, aortic dissection is more common in men than in women, whereas below the age of 40, the gender distribution is equal.⁴ The peak incidence is between 50 and 70 years of age.³

Diagnostic Challenge

Although it is an old disease, the diagnosis of aortic dissection, a serious and life-threatening disorder, can be easily missed even with the advent of modern diagnostic tools. Maintaining a high index of suspicion remains the key to its diagnosis; this is especially important in elderly patients with atypical presentations of disease, and even more so in elderly patients with dementia whose expressive capacity may be limited. In one population-based longitudinal study by Meszaros et al,⁵ the major cause of high mortality (68.2% in 48 hours) was failure to recognize and treat dissection appropriately. In this study, only 13 (15%) of the 84 patients with aortic dissection were suspected of having aortic dissection.⁵ In a retrospective study by Sullivan et al⁶ on the frequency of diagnosis of aortic dissection by emergency physicians, it was found that emergency physicians suspected the diagnosis of aortic dissection in 43% of cases. In another study, Rosman et al⁷ performed a retrospective chart review with findings showing that physicians suspected aortic dissection 65% of the time after initial clinical evaluation.

Mortality

Older studies in the 1950s such as those by Hirst et al1 discuss a mortality rate for aortic dissection of over 90% at one year. Both medical and surgical treatments were limited in that decade.¹ Currently, mortality rates from the International Registry of Acute Aortic Dissection are 26% for type A dissections managed surgically and 58% for those managed medically due to various circumstances such as contraindications to surgery. For those with type B dissections managed medically, mortality rate was 10.7%, and for those managed surgically, 31.4%.⁸ Overall, treated aortic dissections appear to have a 10-year mortality rate of approximately 55-56%.⁹ Chances for survival are improved with an early diagnosis, as blood pressure and heart rate can be controlled and surgical repair performed if indicated.^6,10

Classification

Aortic dissection can be classified by the location of dissection and its duration. When it has been present for less than 2 weeks, aortic dissection is considered acute, and when present for more than 2 weeks it is considered chronic. In the Stanford classification, type A or proximal aortic dissections involve the ascending aorta no matter where the entry site is located.³ Type B or distal aortic dissections involve dissection distal to the origin of the left subclavian artery or the descending aorta. In the DeBakey system of classification, type II dissections involve the ascending aorta and type I dissections involve both the ascending and descending aorta. Type III involves the descending aorta where it originates. Type III can be further classified as type IIIa if it remains confined to the thoracic aorta or as type IIIb if it extends into the abdominal aorta.¹¹

Risk Factors and Symptomatology

Hypertension has been noted to be the most common underlying risk factor, advanced age another.⁶ Other predisposing risk factors include Marfan syndrome, giant cell arteritis, bicuspid arotic valve, coarctation of the aorta, and trauma.⁹ Pain is the most common initial symptom, but in the study by Meszaros et al,⁵ pain was not present in 6 (8%) of 72 patients. Classically described, the pain of aortic dissection is chest pain that is sudden in onset and of severe intensity, eliciting a tearing or ripping sensation radiating to the back or interscapular area.⁹ Epigastric pain may be another presentation; in the study by Meszaros et al,⁵ 9.7% of the patients presented with epigastric pain. However, in a study of aortic dissection in the emergency room by Sullivan et al,⁶ epigastric pain or abdominal pain was the presentation of approximately 30% of patients, but aortic dissection was rarely suspected in these cases. Abdominal pain was characterized as being intense and associated with cramps. Migration and radiation of pain following the path of the dissection is highly suggestive of aortic dissection. Aortic dissection pain may also mimic the pain of myocardial infarction with potentially catastrophic consequences if thrombolytic agents are administered. Neurological involvement may also occur such as syncope, transient ischemic attack, stroke, or spinal cord involvement. Pulse deficits and aortic regurgitation murmurs, although described, are not uniformly present. Rarer clinical symptoms have included dysphagia, superior vena cava syndrome, and bilateral testicular tenderness.⁹ One reported case presented with syncope associated with neck swelling and suprasternal bruising, and another presented with hoarseness of voice.^12,13 Classically associated physical examination findings are reported less than 50% of the time.⁹

Other Tests

Chest x-rays and ECGs have not been uniformly helpful in clinching the diagnosis of aortic dissection. Classical signs of widening of the mediastinum and obliteration of the aortic knob on chest x-ray are often not present. An ECG may show signs of an acute myocardial infarction related to the extension of the dissection into a coronary ostium.¹⁴ In this case, if aortic dissection is not considered as a possible cause of the ischemia, thrombolytic agents may be administered with potentially catastrophic consequences.¹⁴ Some studies have examined blood tests that may assist in the diagnosis of acute aortic dissection. In a study by Weber et al,¹⁵ for example, D-dimer was found to be positive in 100% of patients, leading to a sensitivity of 100% but a specificity of only 68.6%; this suggests that perhaps a negative D-dimer test could be helpful in excluding acute thoracic aortic dissection.¹⁵ Further studies are needed to validate its role in the clinical setting.

Raising the Index of Suspicion in the Elderly

With such protean symptoms, Sir William Osler aptly stated, “There is no disease more conducive to clinical humility than aneurysm of the aorta.”¹⁶ In geriatric medicine, atypical presentations of common diseases have been accepted as an elderly patient’s prerogative. For those patients with an added medical burden that may hinder their ability to communicate, such as those with significant dementia, raising the clinical index of suspicion for aortic dissection would be appropriate. It has been suggested that verbalization of pain in addition to expressions of pain in response to questioning tends to decrease in those with cognitive impairment.¹⁷ Pain in these patients may be expressed by facial expression such as in the case study described previously. In this case, something about the patient’s facial expression suggested “aortic dissection” to the astute nurse. A study by Manfredi et al¹⁸ suggests that clinicians’ observations of facial expressions and vocalizations are a useful means of assessing pain, although they are not as intense in those who are unable to communicate verbally. A larger scale study further qualifying and quantifying clinicians’ observations of facial expressions and their utility would be helpful.

Diagnostic Testing

Suspecting the diagnosis of aortic dissection is the most critical step. Once suspected, aortic dissection may be confirmed by CT scan, magnetic resonance imaging (MRI), or transesophageal echocardiography (TEE). Institutional availability and regional customary practice will probably influence the choice of diagnostic testing used. Helical CT is believed better than standard CT. Of the three techniques, MRI has the highest sensitivity and specificity at almost 100%, CT has a sensitivity of greater than 90% and specificity greater than 85%, and TEE has a specificity and sensitivity of just under 90%.¹⁴ Aortography has been used less frequently for first-line rapid diagnosis.

Treatment

Once diagnosed, aortic dissection should be classified by type. Generally, type A dissections have been managed with surgery, whereas type B have been managed medically short of persistent or recurrent pain, aortic expansion, periaortic hematoma, or mediastinal hematoma.¹⁴ Initial stabilization of aortic dissection should include the control of blood pressure to a low level without compromising perfusion. The goal would be to decrease the aortic pulse wave or force of ventricular contractions. Beta blockers are a class of drugs well suited for this task. Pain should also be relieved and appropriate consultations sought. Surgical intervention should be considered in those with proximal dissections unless there are contraindications.

Summary

Aortic dissection can be easily missed. Classical findings are present less than half of the time and symptoms are protean in nature. The clinical index of suspicion for aortic dissection should be raised in those whose expressive capacity may be impaired and in the elderly, as illnesses often present atypically in older persons. Diagnosed early, treatment would not be delayed and outcome for aortic dissection improved.