Thoracic Aortic Dissection: Avoiding Failure to Diagnose
By Marshall S. Salkin, MD, JD, FACEP, FCLM, Emergency Physician, Northwest Community Hospital, Arlington Heights, Illinois
When a patient enters the ED with chest pain, the reasonable and prudent emergency physician must include in the differential diagnosis the classic six life-threatening causes of this symptom: myocardial infarction, unstable angina, tension pneumothorax, pericarditis with tamponade, pulmonary embolism, and aortic dissection.1 Of all these conditions, aortic dissection may have the highest mortality rate: between 1% and 2% per hour for the first 24-48 hours. In addition, aortic dissection is the most common and the most lethal emergency involving the aorta.2 Approximately 25,000 cases occur annually in the United States, making aortic dissection 2-3 times more common than ruptured abdominal aortic aneurysm.3 Despite major advances in noninvasive diagnosis, the correct antemortem diagnosis is made in less than half the cases.4 This review is focused on improving that statistic.
Pathogenesis
The term "aneurysme dissequant" (i.e., dissecting aneurysm) was first coined by Laennec in 1819 and has been used inaccurately to refer to this condition ever since. Because the affected aorta is infrequently aneurysmal, the term aortic dissection is more appropriate than dissecting aortic aneurysm and will be used in this article.5
The aorta, like most other arteries, consists of three layers: the intima, the media, and the adventitia. Aortic dissection involves separation of the intima from the adventitia within the aortic wall by a dissecting column of blood propagating, via longitudinal cleavage, through the media.6 There are several theories as to the precise initiating event of this entity. Some authorities believe that an intimal tear occurs first, with blood then entering the media through the tear, creating a dissection. An alternative theory suggests that a spontaneous hemorrhage occurs into a previously abnormal media without any initiating tear.7 In either case, once the dissecting hematoma is established in the media of the aorta, migration can occur either in an antegrade or retrograde fashion, creating a false lumen within the outer half of the media. The dissection progresses until it ruptures either back into the true lumen, resulting in a "double barrel" aorta and a rare "spontaneous cure," or, more commonly, the dissection may rupture out of the adventitia into the pericardium or pleural cavity.8 The majority of deaths from aortic dissection are due to rupture of the aorta into the pericardium, with subsequent death due to tamponade, or to rupture into the pleural cavity, with death due to exsanguination. The vigor of the propagation of the dissection is proportional to the blood pressure and the velocity of ventricular contraction referred to as dP/dT.9 This fact is used in the initial medical treatment of aortic dissection.
Classification
The anatomic classification of aortic dissection is relevant not only to diagnosis but to therapy. There are two accepted systems of classification. The DeBakey system divides aortic dissection into three types: Type I dissections begin in the ascending aorta and extend distally to involve the aortic arch and the descending aorta. Type II dissections involve only the ascending aorta. Type III dissections involve the descending aorta, distal to the left subclavian artery. Type IIIA dissections stop above the diaphragm, whereas Type IIIB dissections propagate below the diaphragm.10 Type III dissections may also propagate in a retrograde fashion to involve the proximal aorta.
The Daily (also called the Standford) classification is based on involvement of the ascending aorta. All dissections involving the ascending aorta are Type A. All other dissections are Type B.11 Obviously, DeBakey Types I and II correspond to Daily Type A, whereas DeBakey Type III corresponds to Daily Type B. Proximal aortic dissections (Types I, II, or A) are the most common, accounting for about 70% of cases, with distal aortic dissections (Type III or B) comprising the remaining 30%. Another manner of classifying aortic dissections is by duration at time of first presentation. Aortic dissections are categorized as acute if presentation occurs within two weeks of onset and chronic if more than two weeks have elapsed. Two-thirds of patients with untreated aortic dissections die within this two week period of time; thus, one-third of aortic dissections are chronic.12
Risk Factors
A history of hypertension is the most important risk factor for developing aortic dissection and is documented in 70-90% of all cases.13 Aortic dissection occurs much more commonly in men than in woman and is twice as likely to occur in African-Americans than Caucasians. Although the peak incidence of aortic dissection occurs between the ages of 40 and 80, younger patients may also be affected. In women younger than age 40, 50% of aortic dissections occur during pregnancy, so parturition is also a risk factor. Connective tissue diseases, such as Marfan’s syndrome and Ehlers-Danlos syndrome, which change the structure of the aortic media, have been identified as predisposing to dissection. Aortic valvular disease, such as a congenital biscupid valve, which redirects the stream of left ventricular ejection on the aortic wall, is also a risk factor for dissection. Although arteriosclerosis frequently coexists with aortic dissection, its etiologic role is unclear. One can postulate that arteriosclerosis of the vasovasorum, the small blood vessels nourishing the aorta itself, decreases nutrition to the media, predisposing to dissection. On the other hand, another point of view is that the "bonding" of the three layers of the aorta by the process of arterio- sclerosis actually decreases the likelihood of dissection.14
Clinical Presentation
Acute chest pain is by far the most common presentation of aortic dissection and occurs in more than 90% of patients.15 Typically, the pain occurs quite abruptly and is most severe at onset.16 This sudden chest pain with maximal severity at its inception should be contrasted with acute myocardial ischemia in which the pain is more crescendo-like. There is often a quiescent period in which the pain with aortic dissection may diminish or disappear altogether. This generally is associated with a cessation of the dissection itself; however, the pain inevitably returns after this latent period. Although the literature states the pain of aortic dissection is typically described as "ripping" or "tearing," the description of pain in the actual ED setting is often merely a reflection of the individual patient’s verbal skills and may be characterized differently by different patients. Hence, the lack of the typical adjectives should not dissuade the prudent physician from pursuing the diagnosis.
The location of the pain may help to localize the dissection17 and the location of the dissection determines therapy. When the pain is located in the anterior aspect of the chest, the vast majority of patients have Type I or Type II dissections. Pain in the neck, throat, or jaw is also common with this type of dissection, whereas pain in the posterior aspect of the chest, such as the interscapular area, is most commonly associated with Type III aortic dissections. Pain in the lumbar and abdominal areas is also found primarily with Type III dissections.18 An important distinguishing feature of the pain of aortic dissection is the tendency to migrate from its initial site to other areas as the dissection advances through the aorta.19 In one study, 71% of patients with aortic dissection had this migration of pain.20 Thus, the patient with aortic dissection may present with pain above and below the diaphragm, as the dissection propagates down the aorta. This symptom complex of pain located both above and below the diaphragm is extremely suggestive of the diagnosis of dissection. In addition to pain, patients with aortic dissection may present with neurologic symptoms such as stroke, syncope, ischemic peripheral neuropathy, or paraplegia,21 as well as gastrointestinal symptoms such as abdominal pain and hematemesis.
Physical Examination
It is important to bear in mind that the physical examination of a patient with aortic dissection may be completely normal. Therefore, the absence of physical findings does not rule out the diagnosis. The blood pressure in these patients is elevated in more than half of the cases, especially with distal dissection. True hypotension is much less common and suggests a rupture of the dissection, usually with a proximal dissection. A condition called pseudohypotension, in which the blood pressure in the arms is low or unobtainable while the central arterial pressure is normal or high, may be present. This condition results from the interruption of the blood flow to the subclavian arteries by the dissection. This same mechanism may also result in diminution or absent peripheral pulses that can be found in up to 50% of cases and is especially common with proximal dissections.22 Careful and frequent examination of the pulses is important because the pulse deficits are often transient. The diastolic murmur of aortic insufficiency may be present in 50% of cases, again more commonly with proximal dissections. The presence of jugular venous distention and muffled heart tones signify cardiac tamponade, which is a common cause of death in patients with aortic dissection. Neurologic deficits may be present in the form of ischemic paraparesis, peripheral neuropathy, strokes, and alteration in level of consciousness.
Laboratory
The clinical laboratory is not generally helpful in aortic dissection. An elevation of the white count, secondary to stress, is present in the majority of cases, but this finding is so nonspecific that it rarely is of value in making the diagnosis. A fall in hemoglobin may result from leaking of the dissection or sequestration of blood within the dissection, but this also is a nonspecific finding and is rarely helpful in confirming the diagnosis. ECG and cardiac enzyme tests are helpful in the sense that they are generally normal and thus help to rule out myocardial infarction, which is the most common misdiagnosis.23
Chest X-Ray
Eighty to 90% of patients with aortic dissection have some abnormality suggesting the diagnosis on standard chest films.24 These abnormalities include: mediastinal widening, disparity in size between the ascending and descending aorta, the "calcium sign," which is a separation of greater than 5 mm between the intimal calcium and the outer border of the aorta, depression of the left main stem bronchus, an apical cap, obliteration of the aortic knob, and displacement of the trachea or nasogastric tube to the right by the dissection. Once again, it is important to remember that 10-20% of patients with aortic dissection will have a normal chest x-ray and, therefore, the absence of these abnormalities on standard chest films does not rule out the diagnosis.
Differential Diagnosis
When a patient presents with chest pain, the prudent and reasonable ED physician must consider the six most common life-threatening conditions that can cause this symptom: myocardial infarction, unstable angina, pulmonary embolus, pericarditis with tamponade, tension pneumothorax, and aortic dissection. Frequently, the primary focus is on diagnosing or ruling out acute myocardial ischemia, to the exclusion of these other entities. This is probably why the most common misdiagnosis in aortic dissection cases is myocardial infarction. Obviously, this can be disastrous for two reasons: 1) the diagnosis of the true condition of aortic dissection is missed or delayed with a dramatic increase in mortality rate; and 2) the misdiagnosis of an acute myocardial infarction may lead to the use of thrombolytic agents that greatly exacerbate the condition of aortic dissection.25
Several clinical syndromes suggest the diagnosis of aortic dissection as the cause of chest pain: pain that is acute and maximal at inception; pain that starts in the chest and then migrates to other areas; pain that occurs both above and below the diaphragm; chest pain associated with absent or diminished peripheral pulses; chest pain with neurological findings; chest pain with the murmur of aortic insufficiency; chest pain with a widened mediastinum on chest x-ray. All of these findings should suggest aortic dissection, but the important take-home message is that it is the responsibility of the emergency physician to consider more than just an acute myocardial infarction when presented with a patient complaining of chest pain.
Definitive Evaluation
Aortic angiography remains the most definitive method for confirming the diagnosis of aortic dissection, with a diagnostic accuracy of 95-99%.26 However, the disadvantages of this procedure are obvious. Aortography is the most invasive of the radiographic techniques and requires the insertion of a catheter into an abnormal aorta. Moreover, this procedure involves the risk of contrast material and requires the patient to be out of the ED for a relatively long period of time.27 For these reasons, other tests are becoming more popular.
Transesophageal echocardiography (TEE), in which the transducer is at the end of a flexible gastroscope, is being used with increasing frequency. The procedure requires topical anaesthesia and intravenous sedation but can be performed at the patient’s bedside in the ED and has been shown to be safe even in critically ill patients.28 The sensitivity of TEE for detecting both proximal and distal dissections is 100%.29 The primary disadvantage of TEE is its lack of widespread availability, but this is changing rapidly.
Computed tomography (CT) is the most widely used noninvasive technique for the diagnosis of aortic dissection. The CT rates of both false-positive and false-negative diagnoses of aortic dissection are less than 5%.30 However, CT scans do require contrast medium for precise delineation and also require the patient to leave the ED, albeit generally for shorter periods of time than for angiography.
Magnetic resonance imaging (MRI) is the newest of the imaging methods for the diagnosis of aortic dissection. Standard techniques do not require the use of contrast material and the sensitivity and specificity are excellent. However, disadvantages include: a longer time than CT to complete the study; greater cost; need to move the patient out of the ED; inability to perform MRI studies on ventilator or monitored patients; and lack of availability.31
Which of these relatively definitive tests should the emergency physician order? The literature suggests that all patients in whom the diagnosis of aortic dissection is suspected, even if the index of suspicion is very low, should undergo one of these procedures. A patient with a moderate to high clinical probability of this condition should undergo a second procedure if the findings of the first are negative. When the probability of aortic dissection is high, it is suggested that the emergency physician should order that diagnostic test that is the most rapidly available.32
Treatment
As with all potentially unstable patients, a patient with aortic dissection should have continuous monitoring of blood pressure, pulse, and rhythm, as well as obtaining a large intravenous access route. Proximal (Types I, II, or A) aortic dissections require surgical therapy with resection of the involved segment and replacement with a graft. While these patients are in the ED, the emergency physician should direct therapy towards eliminating the forces that favor progression of the dissection. This can be accomplished by lowering the systolic blood pressure and by reducing the force and velocity of the arterial pulse wave (dP/dT). Prompt reduction in blood pressure can be accomplished by the use of sodium nitroprusside with the rate adjusted to achieve a systolic blood pressure between 100 and 120 mmHg. However, used alone, sodium nitroprusside will increase the velocity of ventricular contraction so that simultaneous use of a beta-blocker is necessary. Propranolol can be used for this purpose at a rate of 1 mg intravenously every five minutes, with the objective of reducing the pulse to 60-80 beats/minute. Alternatively, an esmolol drip can be started for the same purpose.
As suggested above, all patients with acute dissection involving the ascending aorta (Types I, II, or A) should be considered candidates for surgery. Extensive, irreversible injury to the central nervous system may be the only major contraindication to surgery.33 A number of important technical advances have substantially reduced the risks of operations on the thoracic aorta. These include the development of Dacron grafts that are impregnated with collagen or gelatin, which makes them impervious to blood; improved cardiopulmonary bypass circuits that reduce injury to blood elements; and better intraoperative protection of the myocardium. With these advances, the current early mortality rate of surgery on the ascending aorta is approximately 10%, with a five-year, post-surgical survival of 56-87% and a 10-year survival of 40-60%.34
On the other hand, most patients with distal aortic dissections, that is Type III or B, can be treated medically. Survival rates for the medical therapy of distal dissections is approximately 80%, which is equivalent to the surgical survival rates for this type of dissection.35 Medical therapy consists of the use of sodium nitroprusside and a beta blocker as outlined above. Surgery is indicated for patients with signs of impending rupture such as persistent pain, hypotension, and left-sided hemothorax, and those distal dissection patients with ischemia of the legs or abdominal viscera, renal failure, paraparesis, or paraplegia. Patients who present with chronic dissections, (2 weeks or older) should also be treated medically. The survival figures for both treated proximal and treated distal dissections should be contrasted with the untreated one-year survival rate of aortic dissections of 5%.36
Missed Aortic Aneurysm Litigation
Case #1
The patient was a 58-year-old male who entered the ED with the chief complaint of acute onset of chest pain one hour prior to admission, associated with radiation of the pain into the back, neck, and both arms. The patient also stated he had a transient episode of blindness on the morning of admission. He had no past medical problems. His physical examination was essentially normal. An ECG showed some nonspecific T-wave changes and a chest x-ray was normal, as were the cardiac enzymes.
The patient was admitted with the diagnosis of chest pain; myocardial infarction was ruled out . After admission, the patient began to complain of abdominal pain, and the chest pain diminished. The patient had an ultrasound of the gallbladder and abdominal aorta, both of which were negative. A CT scan of the abdomen was also normal. Serial evaluation, with ECGs and enzymes, for a myocardial infarction was negative. The patient’s back and abdominal pain persisted and he received approximately 12 injections of morphine for pain. Thirty-six hours after admission the patient was to be transferred to another institution for further evaluation when he arrested. An autopsy revealed a Type I aortic dissection with rupture into the pericardium and pericardial tamponade. The case was mediated, and lifetime benefits (confidential amount) were paid to the widow.36
This patient presented with chest pain radiating into his back, neck, and both arms, with the risk factors of male gender and age of 58. After admission, the chest pain decreased and he began to develop abdominal pain. As mentioned above, there is often a quiescent period with aortic dissection when the pain temporarily decreases or disappears. In addition, the migration of pain, associated with pain both above and below the diaphragm, is very suggestive of aortic dissection with symptoms appearing sequentially as the dissection propagates through the aorta.
In this case, the patient had transient blindness associated with his chest pain. Chest pain associated with neurologic symptoms is commonly seen with aortic dissection. The case stands for the maxim that the absence of proof is not the proof of absence. The fact that he had a normal physical examination does not exclude the possibility of an aortic aneurysm. Similarly, a normal chest x-ray, which is found in up to 20% of patients with aortic dissection, does not rule out this entity. The attending physician here did not seem to integrate the patients admitting complaint of chest pain with his subsequent development of abdominal pain.
Don’t make the same mistake: Chest pain that later migrates to involve the neck, back, arms, or abdomen, and is located above and below the diaphragm, is aortic dissection until proven otherwise.
Case #2
The patient was a 38-year-old male who entered the hospital with the complaints of sharp substernal chest pain that began one hour before his admission. The pain radiated "straight through to the back." There were no prior such episodes, and the patient had no significant past medical history. The patient had a normal physical examination, as well as a normal ECG, cardiac enzymes, and chest x-ray. While in the ED the pain migrated from the chest posteriorly and inferiorly coming to rest in the low back. There was a nursing note that stated: "the pain originally substernal radiating straight through to the back, now centered in the low back area." The patient was discharged with a diagnosis of "acute muscle spasms, chest and back." He returned to the same ED the following day in cardiac arrest and expired from a aortic dissection. Jury verdict for the plaintiff was 1.8 million dollars.38
The fact that this patient was only 38 years old should not have disuaded the emergency physician from considering the diagnosis of aortic dissection when the patient entered with chest pain that radiated posteriorly into the back and then migrated inferiorly into the low back. By the same token, a normal chest x-ray and normal peripheral pulses can be found in a significant percentage of patients with aortic dissection, so that the presence of these normal findings does not by itself eliminate the diagnosis. The location of the pain, as it evolved, both above and below the diaphragm is so strongly suggestive of aortic dissection that more definitive tests, such as a TEE or a CT scan should have been obtained. In addition, the patient presented with chest pain, severe enough to come to the hospital. A diagnosis of "acute muscle spasm" is a soft and dangerous diagnosis unless you have completely eliminated the more life-threatening causes of chest pain. This was not done here, to both the patient’s and the physician’s detriment.
Case #3
The patient was a 35-year-old male with a history of hypertension who presented with the chief complaints of chest pain, left leg numbness, with left leg weakness and pain. On physical examination, the patient had an absent left femoral pulse, absent left leg deep-tendon reflexes, and a paresis of the left leg. His admitting blood pressure was 98/60. Over time, the chest pain diminished and the left femoral pulse returned. He then began to complain of low back pain. A CT of the low back was obtained which showed a protrusion of the L4-L5 disc. A workup for myocardial ischemia was negative. The patient was admitted with the diagnosis of radiculopathy. He was discharged the following day. The patient expired approximately two weeks later. An autopsy revealed a Type I aortic dissection with pericardial tamponade. Settlement was for a high six figure amount.39
This was a patient who presented with chest pain; this symptom seemed to have been ignored as the work-up focused on his low back pain and left leg paresis. According to deposition testimony, the diagnosis of aortic dissection was never considered. The patient had a history of hypertension and presented with chest pain that migrated into the low back, associated with a weak left leg, a somewhat low blood pressure, and an absent femoral pulse. This is a classic scenario for an aortic dissection. In this case, the pain was located above and below the diaphragm. The patient had a risk factor for aortic dissection with a history of hypertension. The absent pulse in these cases is often due to compression of the lumen of a vessel by the expanding hematoma in the aorta. It should be remembered that an absent pulse is often a transient phenomena and thus must be checked for more than once. The physician here did not seem to integrate the chest pain with the other symptoms and findings.
Obviously, a herniated lumbar disc cannot cause chest pain or an absent femoral pulse, and the chest pain was never adequately explained. Acute leg ischemia has been reported to occur in up to 25% of patients with aortic dissection. In addition, there are numerous studies suggesting that the presence of a bulging, protruding, or herniated lumbar disc is often not associated with any clinical findings. Just because the patient had low back pain and a protruding disc on CT scan, does not infer a causal relationship between the two. The relatively young age of this patient should not eliminate consideration of an aortic dissection. Although the majority of cases are between 40 and 80, aortic dissection, especially with a history of hypertension, can occur in a younger age group. The primary mistake that seems to have been made here is that the emergency physician did not unify all of the patient’s symptoms and the diagnosis was missed. You just can’t pick and choose which symptoms and signs to attribute to your admitting diagnosis; you have to explain them all.
Case #4
The patient was a 23-year-old six-foot female patient who entered the ED with the chief complaint of severe thoracic and low back pain that started suddenly one hour prior to admission. The pain was described as sharp and radiated into the flanks. The patient had one episode of emesis, after the pain started, on the way to the hospital. She also had diarrhea twice in the prior few days. Her admitting blood pressure was 163/120. On physical examination she had micrognathia, pectus excavatum, long, thin fingers and toes, and was extremely myopic. The patient gave a history of an elective abortion three weeks previously. Her pain was severe enough to warrant the use of multiple doses of demerol.
While in the ED, she reported the onset of abdominal pain in addition to the persistent mid and low back pain. Chest x-ray and laboratory tests were normal. The patient was discharged from the ED with the diagnosis of acute gastroenteritis and was later found dead in bed. An autopsy revealed the cause of death to be "pericardial tamponade due to dissecting aortic aneurysm due to Marfan’s syndrome." Settlement was for a six figure amount.40
This patient had the following risk factors for aortic dissection: hypertension, recent pregnancy, and a body habitus suggesting Marfan’s syndrome. In addition, she presented with the acute onset of thoracic and low back pain with radiation into the flanks. Later in the course of her hospital stay, she developed abdominal pain. One should be aware of the fact that thoracic back pain is really posterior chest pain and can occur with aortic dissections that involve the aortic arch or the descending aorta.
Here again, there was evidence of a migration of pain with mid and low back pain, flank pain, and finally abdominal pain; all of which, in the presence of risk factors, suggests aortic dissection. As in the previous cases, the presence of a normal chest x-ray and (except for the stigmata of Marfan’s Syndrome) a normal physical exam, does not rule out aortic dissection. The discharge diagnosis of gastroenteritis is another soft and dangerous diagnosis for the ED physician to make without definitely ruling out aortic dissection, especially in the presence of these symptoms and risk factors and with pain severe enough to require narcotics.
When was the last time you had to give demerol for a patient with gastroenteritis? The take-home message in this case is that you must integrate all the patient’s history and physical findings. The presence of migratory back, flank, and abdominal pain was not integrated with her hypertension, Marfan’s habitus, and recent pregnancy. These factors when unified do not suggest the discharge diagnosis of gastroenteritis. They do suggest aortic dissection.
Case #5
The patient was a 56-year-old male African-American who entered the ED with complaints of right-sided chest pain associated with right shoulder pain, shortness of breath, and diaphoresis, which began 1-2 hours prior to admission. Later, the chart notes the pain "moved" into the abdomen. He had a history of hypertension. The patient’s heart and lung exam was normal, but the abdominal exam revealed mild right abdominal tenderness. An ECG showed left ventricular hypertrophy. The chest x-ray was interpreted by the ED physician as normal, but was later officially interpreted as "moderate dilatation of descending thoracic aorta. The possibility of aneurysmal dilatation should be ruled out." The patient was admitted with the diagnosis of hypertension and abdominal pain of unclear etiology. Later in the day, the patient sustained a cardiac arrest and expired. An autopsy revealed a dissecting aneurysm of the ascending aorta with massive hemopericardium. Settlement was for $500,000.41
This patient had the risk factors of male gender, age of 56, African-American race, and hypertension. The location and migration of pain and the presence of pain above and below the diaphragm, with these risk factors, suggests the presence of an aortic dissection. Left ventricular hypertrophy on the ECG is commonly found with aortic dissection and is a reflection of hypertension, a risk factor for this condition. In the deposition, the physician stated that aortic dissection was considered in the differential diagnosis. Once this entity is considered, with these risk factors and this history, the physician is obligated to rule out aortic dissection by one of the definitive tests suggested above. This was not done here. In addition, the emergency physician misread the chest x-ray that actually showed a dilatation of the thoracic aorta compatible with a dissection. The fact that the pain in this case was right-sided is not relevant, as chest pain with aortic dissection can occur on eithe, or both sides of the chest, and can, in addition, be anterior or posterior or both.
Case #6
The patient was an 81-year-old female with a history of hypertension who entered the hospital with a history of the "rapid" onset of midsternal chest pain of sufficient severity that the patient became profoundly weak and fell to the ground. The pain was described as "sharp in her chest yet now going through to her back and dull constant ache in her right upper back and over her right scapula." The patient’s physical examination was essentially negative except for her color, which was described as "ashen," and the presence of a hematest positive stool. The cardiovascular and abdominal examinations were completely normal. The patient had a negative cardiac workup recently, including a negative thallium stress test. ED ECGs were read as nonspecific T-wave changes, and the cardiac enzymes were normal, as was a chest x-ray.
While in the ED, the patient had a single hematest-negative emesis. She was admitted with the diagnosis of chest pain, rule-out infarct, and upper GI bleed. Over the course of the next day, her pain localized in the upper and lower back and the interscapular area. An ultrasound of her gallbladder revealed gallstones, without thickening of the gallbladder wall. Thirty-six hours after admission, the patient was seen by a consultant who suggested the possibility of aortic dissection. A CT scan of the chest was subsequently done that revealed an aortic dissection. The patient was transferred to a tertiary care center where an aortogram revealed a Type I dissection with antegrade progression down to the suprarenal area. While she was discussing her case with the physicians, she arrested and died. Settlement in this case was confidential.42
This patient had at least two risk factors for aortic dissection: advanced age and a history of hypertension. Her complaints were of the sudden onset of chest pain radiating into the back and right shoulder, and then migrating into the lower back. This migration of pain, and its presence both above and below the diaphragm, is much more suggestive of an aortic dissection than either myocardial ischemia or a gastrointestinal bleed, which were her admitting diagnoses. With a known recent negative exercise thallium scan, and normal enzymes and ECGs during the current hospitalization, the likelihood of acute myocardial ischemia is remote. Moreover, the patient had no abdominal findings and had a hematest negative emesis, both of which decrease the likelihood of a significant acute upper GI bleed. The presence of gallstones does not mean the patient had symptomatic cholecystitis. Most gallstones are asymptomatic.
In this case, valuable time was lost while diagnostic efforts were being erroneously focused on ruling out myocardial ischemia and a GI bleed. As the mortality rate of aortic dissection is 1-2% per hour for the first 48 hours, time is of the essence when attempting to diagnose or rule-out this condition. Don’t be lead astray. Evaluate the patient for the most lethal and likely condition first or you will wind up with a dead patient in whom you have carefully ruled-out less life-threatening illnesses.
Case #7
The patient was a 46-year-old hypertensive male who was admitted to the hospital with the chief complaint of severe abdominal pain The admitting diagnosis was pancreatitis. While in the hospital, the patient began to experience chest pain. He died while hospitalized of an undiagnosed aortic dissection. Jury ruled on a settlement of $457,588.43
This is another case where the patient experienced pain both above and below the diaphragm. This history is relatively specific and sensitive for aortic dissection in the presence of risk factors such as hypertension. In this case, the pain occurred first below the diaphragm and later above the diaphragm in the chest. If the patient has a dissection of the descending aorta, which then propagates in a retrograde manner toward the aortic valve, the pain will be experienced first below and later above the diaphragm. Hence, the fact that the pain began below the diaphragm first is not as important as the fact that the pain migrated from one area to another and eventually appeared both below and above the diaphragm. This is aortic dissection until disproved by appropriate diagnostic procedures.
Litigation Summary
In attempting to manage risk with any clinical entity, it is important to review the case law in order to understand why physicians are sued. An overview of missed aortic dissection cases reveals several common patterns. First, is the missed diagnosis, followed by a discharge from the ED. As you can see from the cases, the patient often receives a diagnosis of musculoskeletal pain, radiculopathy, gastroenteritis, or some other clinical entity that partially fits the patient’s presentation. The second common pattern is the hospital admission with a diagnosis such as "rule out myocardial infarction," or GI bleed. Often, as with the cases discharged from the ED, the admitting diagnosis is usually not completely consistent with the patient’s presentation. Again, delay in making the diagnosis with subsequent injury is typically the basis for this suit.
A third common fact pattern occurs when the emergency physician recognizes the underlying pathology as a dissection of the aorta and admits, or attempts to admit, the patient to the hospital. In this group of cases, the plaintiff alleges failure of the emergency physician to expedite patient evaluation and definitive management. In these cases, the patient sits waiting for admission in the ED or is admitted to the inpatient unit but decompensates or dies waiting for coordination of the OR team, cardiovascular consultation, or transfer to a tertiary care facility, etc. Obviously, this is a seconds to minutes’ type emergency, and, once recognized, the emergency physician and the entire ED staff should use every effort to expedite management and carefully document any cause for delays, the timing and substance of communications with other physicians, and the timing of transfer of responsibility for care to the private or consulting physician, transfer team, etc. Other cases involve allegations that the emergency physician did not properly manage the elevated blood pressure, causing a worsening of the dissection, rupture, and permanent injury or death.
Causation
It is important for the emergency physician to fully understand the components of a malpractice suit. The basic medical malpractice action consists of four elements: duty; breach of duty; damages; and proximate cause. The plaintiff must establish all four of these elements in order to prevail. If there is no breach in a standard of the care the lawsuit must fail as a matter of law. By the same token, if the plaintiff cannot establish proximate cause, the suit also fails.
Duty is generally assumed in the ED setting. Breach of duty and damages must be established by expert witness testimony. The concept of proximate cause is central in many thoracic aortic aneurysm lawsuits. The plaintiff must show that the breach in the standard of care resulted in the patient’s injuries. If the patient’s injury would have occurred even in the absence of any breach, then the element of "causation" is not present, and the suit will fail.
For example let us assume the emergency physician diagnosed an aortic dissection, started initial therapy, called the cardiovascular surgeon, and admitted the patient to the appropriate service in a timely manner. If the patient was not seen on the floor by the attending physician for a prolonged period of time and subsequently expired of the aortic dissection before surgical intervention, there would be great difficulty establishing proximate cause attributed to the emergency physician, as the intervening breach by the attending physician would obviate causation on the part of the emergency physician. However, this would only be the case if the emergency physician documented carefully the times of consultation, admission, and assumption of care by the attending physician.
Another example of lack of proximate cause would be a patient with an aortic dissection who entered the ED in extremis with no obtainable blood pressure and expired shortly thereafter. Even if the emergency physician breached a standard of care by failing to make the diagnosis, the terminal condition of the patient would have obviated salvage even if the correct diagnosis had been made; hence no proximate cause and no viable malpractice action. A defendant physician’s strongest position would obviously be in those cases where there is neither a breach nor causation.
Summary
Every patient who presents with chest pain is a candidate for aortic dissection and, therefore, the prudent and reasonable emergency physician must include this entity in the differential diagnosis and not merely focus on the possibility of myocardial ischemia in a patient who enters with the chief complaint of chest pain. In addition, upper back pain is, in reality, posterior chest pain and the diagnosis of aortic dissection must also be considered in patients who present with this symptom.
Other clues in the history of patients with aortic dissection are: the presence of pain both above and below the diaphragm; the presence of neurologic symptoms or extremity weakness in a patient presenting with chest pain; chest pain that is maximal at inception; pain that starts in one area and then migrates to other locations. It is important to keep in mind that the pain in aortic dissection will not necessarily be described by the patient as ripping or tearing and, hence, the absence of these adjectives in the history does not rule out the diagnosis.
Several physical findings also suggest this diagnosis: absent pulses in a patient with chest pain (remember that the pulse deficit may be evanescent) and chest pain in a patient with the murmur of aortic insufficiency. However, it is most important to remember that the physical examination in this condition may be completely normal and the diagnosis is often driven and made by the history alone.
The laboratory also may not be helpful in the diagnosis of aortic dissection, and, although a standard chest x-ray will be abnormal 80% of the time, 20% or one of five patients will have a completely normal chest film. Therefore, this venue may not be of significant value in the emergency physician’s evaluation of a patient with possible aortic dissection. To validate or rule out the diagnosis of aortic dissection, the emergency physician must obtain a TEE, CT scan, MRI, or aortogram; anything less than one of these definitive tests will get you in trouble. The real key to the diagnosis is to maintain a high index of suspicion for aortic dissection when presented with a history similar to those outlined above. If you don’t consider the diagnosis you will not make the diagnosis, and failure to diagnose a patient with aortic dissection, as illustrated by the case studies presented, often leads to regrettable consequences both for the patient and the emergency physician.
References
1. Zappa MJ. Recognition and management of acute aortic Dissection and thoracic aortic aneurysm. Emerg Med Report 1993;14;1-8.
2. Bourland MD. Aortic dissection. In: Rosen P, Barkin R, et al. Emergency Medicine-Concepts and Clinical Practice. 3rd ed. St. Louis: Mosby Yearbook; 1992:1384.
3. Feldman JA. Thoracic and abdominal aortic aneurysms. In: Tintinalli JE, Krome RL, et al. Emergency Medicine-A Comprehensive Study Guide. 3rd ed. New York: McGraw Hill; 1992:247.
4. Spittell PC, Spittell JA, et al. Clinical features and differential diagnosis of aortic dissection: Experience with 236 Cases. Mayo Clinic Proc 1993;68:642-651.
5. DeSanctis RW, Doroghazi RM, et al. Aortic dissection. N Engl J Med 1987;317:1060-1067.
6. Zappa MJ. Op. cit.
7. Ibid.
8. Bourland MD. Op. Cit.
9. Wheat MW Jr, Palmer RF. Dissecting aneurysms of the aorta monograph. Curr Probl Surg 1971;1:43.
10. DeBakey ME, Henly WS, et al. Surgical management of dissection aneurysms of the aorta. J Thorac Cardiovasc Surg 1965;49:130-149.
11. Daily PO, Trueblood W, et al. Management of acute aortic dissections. Ann Thorac Surg 1970;10:237-247.
12. Miller DC. Surgical management of aortic dissections: Indications, perioperative management, and long-term results. In: Doroghazi RM, Slatter EE. Aortic Dissection 1983: 193-243.
13. Crawford ES. The diagnosis and management of aortic dissection. JAMA 1990;264:2537-2541.
14. Larson EW, Edwards WD. Risk factors for aortic dissections: A necropsy study of 161 cases. Am J Cardiol 1984; 53:849-855.
15. DeSanctis RW. Op. cit.
16. Bourland MD. Op. cit.
17. Ibid.
18. Spittell PC. Op. cit.
19. DeSanctis RW. Op. cit.
20. Slater EE, DeSanctis RW. The clinical recognition of dissecting aortic aneurysm. Am J Med 1976;60:625-633.
21. Ibid.
22. Ibid.
23. Feldman JA. Op. cit.
24. Slatter EE. Op. cit.
25 Zappa MJ. Op. cit.
26. Eagle KA, Quertermous T, et al. Spectrum of conditions initially suggesting aortic dissection but with negative aortograms. Am J Cardiol 1986;57:322-326.
27. Bourland MD. Op. cit.
28. Kouchoukos NT, Dougenis DD. Surgery of the thoracic aorta. N Engl J Med 1997;336:1876-1886.
29. Nienaber CA, Spielmann RP, et al. Diagnosis of thoracic aortic dissection: Magnetic resonance imaging vs. transesophageal echocardiography. Circulation 1992;85:434-447.
30. Moncada R, Demos TC, et al. Detecting disease of the aorta by computerized tomography. J Cardiovasc Med 1983;8:186.
31. Kouchoukos NT. Op. cit.
32. Sarasin FP, Louis-Simonet M, et al. Detecting acute thoracic aortic dissection in the emergency department: Time constraints and choice of the optimal diagnostic test. Ann Emerg Med 1996;28:278-288.
33. Kouchoukos NT. Op. cit.
34. Ibid.
35. Glower DD, Fann JI, et al. Comparison of medical and surgical therapy for uncomplicated descending aortic dissection. Circulation 1990;82 (suppl IV):39.
36. Glower DD, Spier RH, et al. Management and long-term outcome of aortic dissection. Ann Surg 1991;214:31-41.
37. South Dakota Case, 1996. Personal communication.
38. Estate of Kenneth Ruscheinsky, deceased v. Lutheran General Hospital, et al. Illinois Case. 89L-13121 Tried 1994.
39. Hawaii Case, 1997. Personal communication.
40. Michigan Case, 1996. Personal communication.
41. Texas Case, 1996. Personal communication.
42. West Virginia Case. Personal communication.
43. Estate of Norbert Jaroszewski, deceased v. Olympia Fields Osteopathic Medical Center. 87L-20543 Tried 1993.
Physician CME Questions
1. True or false? The physical examination in aortic dissection can be completely normal.
a. True
b. False
2. Which of these conditions is a common misdiagnosis of aortic dissection?
a. Unstable angina
b. Myocardial infarction
c. Pulmonary embolus
d. Pericarditis with tamponade
e. Tension pneumothorax
3. Which of the following is not a risk factor for aortic dissection?
a. Marfan’s Syndrome
b. Ehlers-Danlos Syndrome
c. Hypertension
d. Male gender
e. Pregnancy
f. Eisenmenger complex
4. What is the most common presenting symptom of aortic dissection?
a. Syncope
b. Paralysis
c. Abdominal pain
d. Headache
e. Chest pain
5. What are some common characteristics of the pain with aortic dissection?
a. Migration of pain from one area to another
b. Presence of pain both above and below the diaphragm
c. Pain of sudden, acute onset
d. Pain maximal at inception
e. Pain may remit into a latent period
f. all of the above
6. What percentage of standard chest x-rays are completely normal with aortic dissection?
a. 0
b. 5%
c. 10-20%
d. 20-30%
e. more than half
7. What is the most important risk factor for aortic dissection?
a. Hypertension
b. Pregnancy
c. Congenital bicuspid aortic valve
d. Male gender
e. African-American race
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content