A Twist of Fate?
http://www.100md.com
《新英格兰医药杂志》
in this Journal feature, information about a real patient is presented in stages (boldface type) to an expert clinician, who responds to the information, sharing his or her reasoning with the reader (regular type). The authors' commentary follows.
A 39-year-old Sri Lankan male physiotherapist presented to an emergency department in Toronto with a three-day history of headache, chills, diarrhea, nausea, vomiting, and neck stiffness. He reported having had vertigo, left-sided facial paresthesia, incoordination, and dysarthria, which had lasted for several minutes at the onset of his illness. These symptoms disappeared and were followed by the other symptoms noted above. His history was significant only for type 2 diabetes mellitus and chronic neck symptoms following a motor vehicle collision seven years earlier.
The patient reported that his symptoms had begun shortly after he had eaten a dorado fish that was freshly caught and imported from the Dominican Republic by his brother. He was worried about possible poisoning due to "red tide," a contamination of water by toxic algae.
The dorado fish, along with other species of fish from the Caribbean, has been associated with ciguatera poisoning. Although at first glance this may seem unlikely in an urban center in North America, such poisoning is possible outside of regions where ciguatera is endemic if fish that have been contaminated with ciguatoxin are exported or are eaten locally by travelers who then return home. During the past decade, we have seen several cases of this disorder per year at the hospital where the patient presented. The hospital is situated near a large fresh-food market.
The patient has some symptoms that are consistent with ciguatera poisoning, including paresthesia beginning in the face, headache, vertigo, incoordination, diarrhea, nausea, and vomiting. However, other findings make this diagnosis unlikely. Among those findings is the transient nature of the paresthesia. Although the gastrointestinal manifestations of ciguatera may resolve within hours, the neurologic manifestations tend to last for a minimum of several days. The lateralization of the neurologic symptoms is also inconsistent with ciguatera poisoning and raises concern about a focal lesion in the central nervous system that could have been caused by a stroke, a tumor, or an abscess.
The patient's brother was contacted. He confirmed that the fish had been caught the day before its consumption. He had transported it on ice by airplane to Toronto, where he was visiting family. A physician who was contacted in the Dominican Republic reported that the algae containing ciguatoxin were not in bloom at that time.
This new information rules out ciguatera poisoning.
On examination, the patient appeared acutely ill, with a dazed appearance, a heart rate of 96 beats per minute, a respiratory rate of 14 breaths per minute, a blood pressure of 124/80 mm Hg, and a temperature of 35.8°C. The neurologic examination was normal except that the patient was drowsy and had a sustained horizontal nystagmus on leftward gaze. His neck was supple and was negative for Kernig's and Brudzinski's signs. The remainder of the physical examination was normal. The results of initial laboratory tests were as follows: hemoglobin level, 15.2 g per deciliter; white-cell count, 13,100 per cubic millimeter (neutrophilia); platelet count, 267,000 per cubic millimeter; and normal levels of electrolytes, creatinine, glucose, and prothrombin and a normal partial-thromboplastin time.
The combination of acute headache and neck pain with focal neurologic symptoms initially suggests a primary neurologic disorder. The diagnoses that come to mind first are the stroke syndrome, migraine headache, and infections of the central nervous system.
While the patient was in the emergency department, his temperature increased to 38.0°C.
The presence of fever, headache, and drowsiness arouses concern about the possibility of meningoencephalitis. Although I would want to evaluate the patient for infection as quickly as possible, the history of paresthesia, dysarthria, and incoordination at the onset of the illness, plus nystagmus when the patient was examined, calls for neuroimaging studies to rule out a mass lesion before a lumbar puncture is performed. Computed tomographic (CT) scanning would be reasonable to rule out a mass lesion; if no mass effect is found, a lumbar puncture should be performed. The presence of transient focal neurologic deficits provides a rationale for ordering magnetic resonance imaging (MRI) of the brain if the CT scan and lumbar puncture fail to explain the symptoms and signs. MRI is more sensitive than CT for detecting strokes and abnormalities associated with infectious meningoencephalitis.
CT scanning of the head showed no evidence of a space-occupying lesion or an intracranial hemorrhage. An emergency room physician then performed a lumbar puncture, which was uncomplicated and successful on the first attempt. Analysis of the clear and colorless cerebrospinal fluid revealed the following values: in the first tube, 197 leukocytes per cubic millimeter and 7690 erythrocytes per cubic millimeter; in the fourth tube, 237 leukocytes per cubic millimeter (90 percent of which were neutrophils), 4700 erythrocytes per cubic millimeter, 181 mg of protein per deciliter, and 124 mg of glucose per deciliter (6.9 mmol per liter); the serum glucose level was 162 mg per deciliter (9.0 mmol per liter). Staining was negative for bacteria, acid-fast bacilli, and yeast. A rapid latex-agglutination test for cryptococcal antigen was negative.
The elevated red-cell count in the cerebrospinal fluid is probably not an indication of a subarachnoid hemorrhage, given the marked fall in the erythrocyte count between the first and fourth tubes and the absence of xanthochromia. Even though the lumbar puncture was reported as uncomplicated, it was most likely traumatic. The leukocyte count can be elevated with traumatic lumbar puncture, but the ratio of the white-cell count to the red-cell count should be no more than 1:500 in this setting. In the fourth tube, the ratio of the white-cell count to the red-cell count was 1:20, suggesting there is a true leukocytosis. Faced with a fever and leukocytosis in the cerebrospinal fluid, with a high proportion of neutrophils, the first consideration is always bacterial meningitis. However, that diagnosis is unlikely in this case since I would expect the white-cell count in the cerebrospinal fluid to be higher and the glucose level lower, and that, after 72 hours of illness, the patient would be more severely ill. The negative results of Gram's staining are not particularly helpful, given its low sensitivity. Although bacterial meningitis is unlikely, a negative culture is required to exclude the possibility in this case. Viral meningoencephalitis can also cause fever and an elevated white-cell count in the cerebrospinal fluid, but again that diagnosis does not fit well with the findings in this case. Typically, the leukocytosis in the cerebrospinal fluid will have a lymphocytic predominance in viral infections. A final consideration with respect to infectious agents is mycobacterial meningitis, which can also cause focal neurologic symptoms. In the small proportion of patients who present with acute tuberculous meningitis, the findings in the cerebrospinal fluid may include a neutrophil-predominant leukocytosis, often with a count above 1000 per cubic millimeter. However, most patients with tuberculous meningitis present with a longer duration of symptoms and a leukocytosis with a lymphocytic predominance.
While awaiting the results of the cerebrospinal fluid cultures, I would administer broad-spectrum antibiotics to cover meningococcus, pneumococcus, Haemophilus influenzae, and listeria. I would also administer acyclovir to cover meningoencephalitis caused by herpes simplex virus.
Treatment was started with cefotaxime (2 g given intravenously every 6 hours), vancomycin (1 g given intravenously every 12 hours), and acyclovir (700 mg given intravenously every 8 hours).
When questioned further about other antecedent events, the patient reported that several hours before the onset of his symptoms, one of his patients (who may have been a retired chiropractor) had manipulated his neck in an attempt to alleviate his chronic neck pain. On the basis of his description, the manipulation appeared to involve a high-velocity rotary movement.
This new information makes me consider the possibility of a vertebral-artery dissection that resulted in a stroke (either ischemic or hemorrhagic). If the stroke occurred in the cerebellum or brain stem, it could explain the patient's nausea and vomiting on presentation. Ischemic events may also cause fever and abnormalities in the cerebrospinal fluid. The increase in both the neutrophil count and the protein level and the normal glucose level are consistent with the inflammatory and repair processes that have been reported after a stroke. Even though the CT scan did not show a hemorrhage or an infarct, MRI (and possibly magnetic resonance angiography) is warranted, since it is more sensitive than CT is for revealing small strokes and dissection. MRI will also be useful in evaluating the patient for the possibility of certain infections, such as herpes simplex virus, in which inflammation and hemorrhage of the temporal lobe can be found, or meningitis, in which MRI shows enhancement and thickening of the leptomeninges (especially around the cranial nerves), increased fluid in the subarachnoid space, local venous thrombi, or edema of the adjacent brain. These findings associated with meningoencephalitis may be seen on a CT scan but are usually more evident on an MRI scan.
MRI of the brain showed an abnormality in the left cerebellar hemisphere. However, the study was considered nondiagnostic since the images were of poor quality due to movement artifact. Another scan was obtained (Figure 1) in conjunction with three-dimensional magnetic resonance angiography (Figure 2). This study showed increased signal intensity on T2-weighted images, with enhancement after the administration of contrast material, in the posteroinferior arterial territory of the left cerebellar hemisphere. The magnetic resonance angiogram showed irregularities in the left vertebral artery at the C1–2 level, which is consistent with vertebral-artery dissection; there was no evidence of hemorrhage. All the cultures were negative, as were polymerase-chain-reaction assays of the cerebrospinal fluid for herpes simplex virus, cytomegalovirus, and Mycobacterium tuberculosis; a test for serum antibodies against human immunodeficiency virus types 1 and 2; and serologic Venereal Disease Research Laboratory tests for syphilis.
Figure 1. MRI Scan of the Cerebellum Showing Increased Signal Intensity in the Left Cerebellar Hemisphere.
Acute infarction in the territory of the left posteroinferior cerebellar artery is evident in the sagittal T2-weighted view (Panel A, arrow), the axial view obtained with a fluid-attenuated inversion recovery (FLAIR) sequence (Panel B, arrow), an axial diffusion-weighted image (Panel C, arrow), and an apparent-diffusion-coefficient image (Panel D, arrow).
Figure 2. Magnetic Resonance Angiogram Showing Vertebral-Artery Dissection.
Irregular stenotic segments — a finding consistent with the diagnosis of arterial dissection — can be seen within the distal left vertebral artery in the left anterior oblique view (Panel A, arrow) and the magnified right anterior oblique view (Panel B, arrows). Both images were obtained with the use of an autotriggered gadolinium-enhanced technique. The cervical carotid arteries have been eliminated from the image.
Given the MRI results showing vertebral-artery dissection and the negative results on investigations for infection, I would discontinue all antimicrobial agents and initiate anticoagulation.
Warfarin was administered, and the patient's condition improved, with complete resolution of his fever, nausea, vomiting, and diarrhea. He was discharged with a plan for warfarin to be discontinued after three months if repeated magnetic resonance angiography showed a normal vertebral artery and reestablishment of normal blood flow.
Commentary
In this case, the patient's initial presentation included nausea, vomiting, and transient neurologic symptoms. His rapidly increasing fever and the cerebrospinal fluid abnormalities confused the clinical picture. Since the patient had recently eaten a dorado fish, he was concerned about ciguatera poisoning,1,2,3,4 but this proved to be a red herring. When considering diagnostic possibilities, clinicians must first consider those that have serious consequences if left untreated. However, clinicians must also attend to the patient's concerns. In this case, obtaining more history regarding the fish that the patient had ingested quickly ruled out the diagnosis that was his primary concern.
The diagnosis of vertebral-artery dissection could have been made earlier if MRI had been performed initially instead of CT. However, the decision regarding which neuroimaging technique to use is neither easy nor without controversy. The American College of Emergency Physicians recently considered the question about choice of neuroimaging when it was developing its guidelines for the management of acute headaches.5 After carefully considering the evidence, the group recommended CT without the administration of contrast material for patients with acute headaches and neurologic symptoms. In addition, CT is more sensitive than MRI for detecting hemorrhage in the central nervous system. On a practical level, in many emergency departments such as ours, CT is more readily available and faster to perform than MRI is, and unlike MRI, CT can be used in patients who have metal or implantable devices in their bodies and can be performed more easily in patients who have claustrophobia. These are important considerations, given that time is of the essence when considering a diagnosis of bacterial meningitis. Whereas CT is more sensitive for the detection of acute hemorrhage, MRI provides more detailed images of the brain and may detect strokes and other conditions that are missed by CT, including signs of meningoencephalitis, which was a serious concern in this case. In the end, it seems reasonable to start with CT in order to rule out contraindications for lumbar puncture and then move to MRI if further information is required to explain the focal neurologic deficits.5
Several aspects of this case made it difficult to arrive at a diagnosis immediately. The first was the patient's concern about the possibility of ciguatera poisoning. Although a diagnosis of vertebral-artery dissection would account for fever, nausea, and vomiting, it would not account for diarrhea and chills.
Once the history of neck manipulation came to light, stroke caused by vertebral-artery dissection became an important consideration. Since the CT scan was negative, there was an urgent indication for MRI followed by magnetic resonance angiography, with its increased sensitivity for detecting infarcts and vertebral-artery dissection.6,7
Because of their anatomy, the vertebral arteries are particularly susceptible to injury resulting from head or neck torsion. When dissection occurs, the initial symptoms of neck pain and headache are caused by the direct trauma to the arteries. Subsequently, focal neurologic symptoms may develop as a result of ischemic injury to the cerebellum or brain stem from arterial occlusion by the intimal flap or artery-to-artery thromboembolism.6 The hallmark symptoms are neck pain and headache, which occur in 47 to 79 percent of cases.8,9 Patients tend to describe the headache as very severe, sharp, one-sided, and radiating from the neck,6 and as being different from any other pain they have experienced. The onset of neck pain or headache typically precedes the development of neurologic symptoms. The delay before the neurologic symptoms and signs develop is usually less than 24 hours6,8,9 but may range from several minutes to two months. Symptoms and signs reflect the territory of the brain involved, such as a lateral medullary stroke (i.e., Wallenberg's syndrome) or cerebellar stroke, and include vertigo, nausea, vomiting, diplopia, tinnitus, dysarthria, dysphagia, and hemiparesthesia. In the present case, nausea, vomiting, and nystagmus were consistent with infarction of the cerebellum.
In this case, the neck manipulation may have caused the dissection. Twisting of the neck from a wide variety of activities (including coughing, sports, lifting, ceiling painting, yoga, archery, and falls) has been associated with vertebral-artery dissection. Several recent reports have described an association of the condition with chiropractic neck manipulation.7,10,11,12,13 When patients, especially those under 60 years of age, present with an acute onset of a focal neurologic deficit (i.e., a stroke syndrome) involving a part of the brain that is supplied by the posterior circulation, they should be routinely questioned about neck manipulation or other activities that cause rapid neck movements.
The management of vertebral-artery dissection remains controversial and has not been examined in any randomized, controlled trials. Anticoagulant therapy is the current standard of care in the absence of apparent hemorrhage on imaging and is typically continued for three to six months,14 followed by long-term antiplatelet therapy. The rationale is to prevent thromboembolism from the vertebral artery to the brain, which is the most frequent complication of vertebral-artery dissection.6 The main exception is intradural vertebral-artery dissection, in which anticoagulation is contraindicated because of the high incidence of associated subarachnoid hemorrhage. In such cases, antiplatelet agents are sometimes used instead. The optimal time to begin anticoagulation remains uncertain. Recently, interest has been growing in the use of angioplasty and stenting to treat vertebral-artery dissection. However, there is evidence that anticoagulant therapy alone often results in the resolution of neurologic symptoms and may prevent recurrence.15
We are indebted to Dr. Richard Farb for interpretation of the neuroimaging studies and preparation of the images, and to Dr. Hillar Vellend for his insights into the case and years of mentorship.
Source Information
From the Departments of Medicine (M.D.C., A.S.D.) and Health Policy, Management, and Evaluation (A.S.D.), University of Toronto; and the Departments of Medicine at Mount Sinai Hospital and University Health Network (M.D.C., A.S.D.) — all in Toronto.
References
Angibaud G, Rambaud S. Serious neurological manifestations of ciguatera: is the delay unusually long? J Neurol Neurosurg Psychiatry 1998;64:688-689.
Lange WR, Snyder FR, Fudala PJ. Travel and ciguatera fish poisoning. Arch Intern Med 1992;152:2049-2053.
Pearn J. Neurology of ciguatera. J Neurol Neurosurg Psychiatry 2001;70:4-8.
Poli MA, Lewis RJ, Dickey RW, Musser SM, Buckner CA, Carpenter LG. Identification of Caribbean ciguatoxins as the cause of an outbreak of fish poisoning among U.S. soldiers in Haiti. Toxicon 1997;35:733-741.
American College of Emergency Physicians. Clinical policy: critical issues in the evaluation and management of patients presenting to the emergency department with acute headache. Ann Emerg Med 2002;39:108-122.
Sturzenegger M. Headache and neck pain: the warning symptoms of vertebral artery dissection. Headache 1994;34:187-193.
Ernst E. Life-threatening complications of spinal manipulation. Stroke 2001;32:809-810.
Frisoni GB, Anzola GP. Vertebrobasilar ischemia after neck motion. Stroke 1991;22:1452-1460.
Norris JW, Beletsky V, Nadareishvili ZG. Sudden neck movement and cervical artery dissection. CMAJ 2000;163:38-40.
Kapral MK, Bondy SJ. Cervical manipulation and risk of stroke. Can Med Assoc J 2001;165:907-908.
Phillips SJ, Maloney WJ, Gray J. Pure motor stroke due to vertebral artery dissection. Can J Neurol Sci 1989;16:348-351.
Rothwell DM, Bondy SJ, Williams JI. Chiropractic manipulation and stroke: a population-based case-control study. Stroke 2001;32:1054-1060.
Sherman DG, Hart RG, Easton JD. Abrupt change in head position and cerebral infarction. Stroke 1981;12:2-6.
Schievink WI. The treatment of spontaneous carotid and vertebral artery dissections. Curr Opin Cardiol 2000;15:316-321.
Kitanaka C, Tanaka J, Kuwahara M, et al. Nonsurgical treatment of unruptured intracranial vertebral artery dissection with serial follow-up angiography. J Neurosurg 1994;80:667-674.
Related Letters:
Ciguatera Poisoning
Selcer U. M., Christian M. D., Detsky A. S.(Michael D. Christian, M.D)
A 39-year-old Sri Lankan male physiotherapist presented to an emergency department in Toronto with a three-day history of headache, chills, diarrhea, nausea, vomiting, and neck stiffness. He reported having had vertigo, left-sided facial paresthesia, incoordination, and dysarthria, which had lasted for several minutes at the onset of his illness. These symptoms disappeared and were followed by the other symptoms noted above. His history was significant only for type 2 diabetes mellitus and chronic neck symptoms following a motor vehicle collision seven years earlier.
The patient reported that his symptoms had begun shortly after he had eaten a dorado fish that was freshly caught and imported from the Dominican Republic by his brother. He was worried about possible poisoning due to "red tide," a contamination of water by toxic algae.
The dorado fish, along with other species of fish from the Caribbean, has been associated with ciguatera poisoning. Although at first glance this may seem unlikely in an urban center in North America, such poisoning is possible outside of regions where ciguatera is endemic if fish that have been contaminated with ciguatoxin are exported or are eaten locally by travelers who then return home. During the past decade, we have seen several cases of this disorder per year at the hospital where the patient presented. The hospital is situated near a large fresh-food market.
The patient has some symptoms that are consistent with ciguatera poisoning, including paresthesia beginning in the face, headache, vertigo, incoordination, diarrhea, nausea, and vomiting. However, other findings make this diagnosis unlikely. Among those findings is the transient nature of the paresthesia. Although the gastrointestinal manifestations of ciguatera may resolve within hours, the neurologic manifestations tend to last for a minimum of several days. The lateralization of the neurologic symptoms is also inconsistent with ciguatera poisoning and raises concern about a focal lesion in the central nervous system that could have been caused by a stroke, a tumor, or an abscess.
The patient's brother was contacted. He confirmed that the fish had been caught the day before its consumption. He had transported it on ice by airplane to Toronto, where he was visiting family. A physician who was contacted in the Dominican Republic reported that the algae containing ciguatoxin were not in bloom at that time.
This new information rules out ciguatera poisoning.
On examination, the patient appeared acutely ill, with a dazed appearance, a heart rate of 96 beats per minute, a respiratory rate of 14 breaths per minute, a blood pressure of 124/80 mm Hg, and a temperature of 35.8°C. The neurologic examination was normal except that the patient was drowsy and had a sustained horizontal nystagmus on leftward gaze. His neck was supple and was negative for Kernig's and Brudzinski's signs. The remainder of the physical examination was normal. The results of initial laboratory tests were as follows: hemoglobin level, 15.2 g per deciliter; white-cell count, 13,100 per cubic millimeter (neutrophilia); platelet count, 267,000 per cubic millimeter; and normal levels of electrolytes, creatinine, glucose, and prothrombin and a normal partial-thromboplastin time.
The combination of acute headache and neck pain with focal neurologic symptoms initially suggests a primary neurologic disorder. The diagnoses that come to mind first are the stroke syndrome, migraine headache, and infections of the central nervous system.
While the patient was in the emergency department, his temperature increased to 38.0°C.
The presence of fever, headache, and drowsiness arouses concern about the possibility of meningoencephalitis. Although I would want to evaluate the patient for infection as quickly as possible, the history of paresthesia, dysarthria, and incoordination at the onset of the illness, plus nystagmus when the patient was examined, calls for neuroimaging studies to rule out a mass lesion before a lumbar puncture is performed. Computed tomographic (CT) scanning would be reasonable to rule out a mass lesion; if no mass effect is found, a lumbar puncture should be performed. The presence of transient focal neurologic deficits provides a rationale for ordering magnetic resonance imaging (MRI) of the brain if the CT scan and lumbar puncture fail to explain the symptoms and signs. MRI is more sensitive than CT for detecting strokes and abnormalities associated with infectious meningoencephalitis.
CT scanning of the head showed no evidence of a space-occupying lesion or an intracranial hemorrhage. An emergency room physician then performed a lumbar puncture, which was uncomplicated and successful on the first attempt. Analysis of the clear and colorless cerebrospinal fluid revealed the following values: in the first tube, 197 leukocytes per cubic millimeter and 7690 erythrocytes per cubic millimeter; in the fourth tube, 237 leukocytes per cubic millimeter (90 percent of which were neutrophils), 4700 erythrocytes per cubic millimeter, 181 mg of protein per deciliter, and 124 mg of glucose per deciliter (6.9 mmol per liter); the serum glucose level was 162 mg per deciliter (9.0 mmol per liter). Staining was negative for bacteria, acid-fast bacilli, and yeast. A rapid latex-agglutination test for cryptococcal antigen was negative.
The elevated red-cell count in the cerebrospinal fluid is probably not an indication of a subarachnoid hemorrhage, given the marked fall in the erythrocyte count between the first and fourth tubes and the absence of xanthochromia. Even though the lumbar puncture was reported as uncomplicated, it was most likely traumatic. The leukocyte count can be elevated with traumatic lumbar puncture, but the ratio of the white-cell count to the red-cell count should be no more than 1:500 in this setting. In the fourth tube, the ratio of the white-cell count to the red-cell count was 1:20, suggesting there is a true leukocytosis. Faced with a fever and leukocytosis in the cerebrospinal fluid, with a high proportion of neutrophils, the first consideration is always bacterial meningitis. However, that diagnosis is unlikely in this case since I would expect the white-cell count in the cerebrospinal fluid to be higher and the glucose level lower, and that, after 72 hours of illness, the patient would be more severely ill. The negative results of Gram's staining are not particularly helpful, given its low sensitivity. Although bacterial meningitis is unlikely, a negative culture is required to exclude the possibility in this case. Viral meningoencephalitis can also cause fever and an elevated white-cell count in the cerebrospinal fluid, but again that diagnosis does not fit well with the findings in this case. Typically, the leukocytosis in the cerebrospinal fluid will have a lymphocytic predominance in viral infections. A final consideration with respect to infectious agents is mycobacterial meningitis, which can also cause focal neurologic symptoms. In the small proportion of patients who present with acute tuberculous meningitis, the findings in the cerebrospinal fluid may include a neutrophil-predominant leukocytosis, often with a count above 1000 per cubic millimeter. However, most patients with tuberculous meningitis present with a longer duration of symptoms and a leukocytosis with a lymphocytic predominance.
While awaiting the results of the cerebrospinal fluid cultures, I would administer broad-spectrum antibiotics to cover meningococcus, pneumococcus, Haemophilus influenzae, and listeria. I would also administer acyclovir to cover meningoencephalitis caused by herpes simplex virus.
Treatment was started with cefotaxime (2 g given intravenously every 6 hours), vancomycin (1 g given intravenously every 12 hours), and acyclovir (700 mg given intravenously every 8 hours).
When questioned further about other antecedent events, the patient reported that several hours before the onset of his symptoms, one of his patients (who may have been a retired chiropractor) had manipulated his neck in an attempt to alleviate his chronic neck pain. On the basis of his description, the manipulation appeared to involve a high-velocity rotary movement.
This new information makes me consider the possibility of a vertebral-artery dissection that resulted in a stroke (either ischemic or hemorrhagic). If the stroke occurred in the cerebellum or brain stem, it could explain the patient's nausea and vomiting on presentation. Ischemic events may also cause fever and abnormalities in the cerebrospinal fluid. The increase in both the neutrophil count and the protein level and the normal glucose level are consistent with the inflammatory and repair processes that have been reported after a stroke. Even though the CT scan did not show a hemorrhage or an infarct, MRI (and possibly magnetic resonance angiography) is warranted, since it is more sensitive than CT is for revealing small strokes and dissection. MRI will also be useful in evaluating the patient for the possibility of certain infections, such as herpes simplex virus, in which inflammation and hemorrhage of the temporal lobe can be found, or meningitis, in which MRI shows enhancement and thickening of the leptomeninges (especially around the cranial nerves), increased fluid in the subarachnoid space, local venous thrombi, or edema of the adjacent brain. These findings associated with meningoencephalitis may be seen on a CT scan but are usually more evident on an MRI scan.
MRI of the brain showed an abnormality in the left cerebellar hemisphere. However, the study was considered nondiagnostic since the images were of poor quality due to movement artifact. Another scan was obtained (Figure 1) in conjunction with three-dimensional magnetic resonance angiography (Figure 2). This study showed increased signal intensity on T2-weighted images, with enhancement after the administration of contrast material, in the posteroinferior arterial territory of the left cerebellar hemisphere. The magnetic resonance angiogram showed irregularities in the left vertebral artery at the C1–2 level, which is consistent with vertebral-artery dissection; there was no evidence of hemorrhage. All the cultures were negative, as were polymerase-chain-reaction assays of the cerebrospinal fluid for herpes simplex virus, cytomegalovirus, and Mycobacterium tuberculosis; a test for serum antibodies against human immunodeficiency virus types 1 and 2; and serologic Venereal Disease Research Laboratory tests for syphilis.
Figure 1. MRI Scan of the Cerebellum Showing Increased Signal Intensity in the Left Cerebellar Hemisphere.
Acute infarction in the territory of the left posteroinferior cerebellar artery is evident in the sagittal T2-weighted view (Panel A, arrow), the axial view obtained with a fluid-attenuated inversion recovery (FLAIR) sequence (Panel B, arrow), an axial diffusion-weighted image (Panel C, arrow), and an apparent-diffusion-coefficient image (Panel D, arrow).
Figure 2. Magnetic Resonance Angiogram Showing Vertebral-Artery Dissection.
Irregular stenotic segments — a finding consistent with the diagnosis of arterial dissection — can be seen within the distal left vertebral artery in the left anterior oblique view (Panel A, arrow) and the magnified right anterior oblique view (Panel B, arrows). Both images were obtained with the use of an autotriggered gadolinium-enhanced technique. The cervical carotid arteries have been eliminated from the image.
Given the MRI results showing vertebral-artery dissection and the negative results on investigations for infection, I would discontinue all antimicrobial agents and initiate anticoagulation.
Warfarin was administered, and the patient's condition improved, with complete resolution of his fever, nausea, vomiting, and diarrhea. He was discharged with a plan for warfarin to be discontinued after three months if repeated magnetic resonance angiography showed a normal vertebral artery and reestablishment of normal blood flow.
Commentary
In this case, the patient's initial presentation included nausea, vomiting, and transient neurologic symptoms. His rapidly increasing fever and the cerebrospinal fluid abnormalities confused the clinical picture. Since the patient had recently eaten a dorado fish, he was concerned about ciguatera poisoning,1,2,3,4 but this proved to be a red herring. When considering diagnostic possibilities, clinicians must first consider those that have serious consequences if left untreated. However, clinicians must also attend to the patient's concerns. In this case, obtaining more history regarding the fish that the patient had ingested quickly ruled out the diagnosis that was his primary concern.
The diagnosis of vertebral-artery dissection could have been made earlier if MRI had been performed initially instead of CT. However, the decision regarding which neuroimaging technique to use is neither easy nor without controversy. The American College of Emergency Physicians recently considered the question about choice of neuroimaging when it was developing its guidelines for the management of acute headaches.5 After carefully considering the evidence, the group recommended CT without the administration of contrast material for patients with acute headaches and neurologic symptoms. In addition, CT is more sensitive than MRI for detecting hemorrhage in the central nervous system. On a practical level, in many emergency departments such as ours, CT is more readily available and faster to perform than MRI is, and unlike MRI, CT can be used in patients who have metal or implantable devices in their bodies and can be performed more easily in patients who have claustrophobia. These are important considerations, given that time is of the essence when considering a diagnosis of bacterial meningitis. Whereas CT is more sensitive for the detection of acute hemorrhage, MRI provides more detailed images of the brain and may detect strokes and other conditions that are missed by CT, including signs of meningoencephalitis, which was a serious concern in this case. In the end, it seems reasonable to start with CT in order to rule out contraindications for lumbar puncture and then move to MRI if further information is required to explain the focal neurologic deficits.5
Several aspects of this case made it difficult to arrive at a diagnosis immediately. The first was the patient's concern about the possibility of ciguatera poisoning. Although a diagnosis of vertebral-artery dissection would account for fever, nausea, and vomiting, it would not account for diarrhea and chills.
Once the history of neck manipulation came to light, stroke caused by vertebral-artery dissection became an important consideration. Since the CT scan was negative, there was an urgent indication for MRI followed by magnetic resonance angiography, with its increased sensitivity for detecting infarcts and vertebral-artery dissection.6,7
Because of their anatomy, the vertebral arteries are particularly susceptible to injury resulting from head or neck torsion. When dissection occurs, the initial symptoms of neck pain and headache are caused by the direct trauma to the arteries. Subsequently, focal neurologic symptoms may develop as a result of ischemic injury to the cerebellum or brain stem from arterial occlusion by the intimal flap or artery-to-artery thromboembolism.6 The hallmark symptoms are neck pain and headache, which occur in 47 to 79 percent of cases.8,9 Patients tend to describe the headache as very severe, sharp, one-sided, and radiating from the neck,6 and as being different from any other pain they have experienced. The onset of neck pain or headache typically precedes the development of neurologic symptoms. The delay before the neurologic symptoms and signs develop is usually less than 24 hours6,8,9 but may range from several minutes to two months. Symptoms and signs reflect the territory of the brain involved, such as a lateral medullary stroke (i.e., Wallenberg's syndrome) or cerebellar stroke, and include vertigo, nausea, vomiting, diplopia, tinnitus, dysarthria, dysphagia, and hemiparesthesia. In the present case, nausea, vomiting, and nystagmus were consistent with infarction of the cerebellum.
In this case, the neck manipulation may have caused the dissection. Twisting of the neck from a wide variety of activities (including coughing, sports, lifting, ceiling painting, yoga, archery, and falls) has been associated with vertebral-artery dissection. Several recent reports have described an association of the condition with chiropractic neck manipulation.7,10,11,12,13 When patients, especially those under 60 years of age, present with an acute onset of a focal neurologic deficit (i.e., a stroke syndrome) involving a part of the brain that is supplied by the posterior circulation, they should be routinely questioned about neck manipulation or other activities that cause rapid neck movements.
The management of vertebral-artery dissection remains controversial and has not been examined in any randomized, controlled trials. Anticoagulant therapy is the current standard of care in the absence of apparent hemorrhage on imaging and is typically continued for three to six months,14 followed by long-term antiplatelet therapy. The rationale is to prevent thromboembolism from the vertebral artery to the brain, which is the most frequent complication of vertebral-artery dissection.6 The main exception is intradural vertebral-artery dissection, in which anticoagulation is contraindicated because of the high incidence of associated subarachnoid hemorrhage. In such cases, antiplatelet agents are sometimes used instead. The optimal time to begin anticoagulation remains uncertain. Recently, interest has been growing in the use of angioplasty and stenting to treat vertebral-artery dissection. However, there is evidence that anticoagulant therapy alone often results in the resolution of neurologic symptoms and may prevent recurrence.15
We are indebted to Dr. Richard Farb for interpretation of the neuroimaging studies and preparation of the images, and to Dr. Hillar Vellend for his insights into the case and years of mentorship.
Source Information
From the Departments of Medicine (M.D.C., A.S.D.) and Health Policy, Management, and Evaluation (A.S.D.), University of Toronto; and the Departments of Medicine at Mount Sinai Hospital and University Health Network (M.D.C., A.S.D.) — all in Toronto.
References
Angibaud G, Rambaud S. Serious neurological manifestations of ciguatera: is the delay unusually long? J Neurol Neurosurg Psychiatry 1998;64:688-689.
Lange WR, Snyder FR, Fudala PJ. Travel and ciguatera fish poisoning. Arch Intern Med 1992;152:2049-2053.
Pearn J. Neurology of ciguatera. J Neurol Neurosurg Psychiatry 2001;70:4-8.
Poli MA, Lewis RJ, Dickey RW, Musser SM, Buckner CA, Carpenter LG. Identification of Caribbean ciguatoxins as the cause of an outbreak of fish poisoning among U.S. soldiers in Haiti. Toxicon 1997;35:733-741.
American College of Emergency Physicians. Clinical policy: critical issues in the evaluation and management of patients presenting to the emergency department with acute headache. Ann Emerg Med 2002;39:108-122.
Sturzenegger M. Headache and neck pain: the warning symptoms of vertebral artery dissection. Headache 1994;34:187-193.
Ernst E. Life-threatening complications of spinal manipulation. Stroke 2001;32:809-810.
Frisoni GB, Anzola GP. Vertebrobasilar ischemia after neck motion. Stroke 1991;22:1452-1460.
Norris JW, Beletsky V, Nadareishvili ZG. Sudden neck movement and cervical artery dissection. CMAJ 2000;163:38-40.
Kapral MK, Bondy SJ. Cervical manipulation and risk of stroke. Can Med Assoc J 2001;165:907-908.
Phillips SJ, Maloney WJ, Gray J. Pure motor stroke due to vertebral artery dissection. Can J Neurol Sci 1989;16:348-351.
Rothwell DM, Bondy SJ, Williams JI. Chiropractic manipulation and stroke: a population-based case-control study. Stroke 2001;32:1054-1060.
Sherman DG, Hart RG, Easton JD. Abrupt change in head position and cerebral infarction. Stroke 1981;12:2-6.
Schievink WI. The treatment of spontaneous carotid and vertebral artery dissections. Curr Opin Cardiol 2000;15:316-321.
Kitanaka C, Tanaka J, Kuwahara M, et al. Nonsurgical treatment of unruptured intracranial vertebral artery dissection with serial follow-up angiography. J Neurosurg 1994;80:667-674.
Related Letters:
Ciguatera Poisoning
Selcer U. M., Christian M. D., Detsky A. S.(Michael D. Christian, M.D)