AUTHOR AND EDITOR INFORMATION

Section 1 of 8  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• References

INTRODUCTION

Section 2 of 8  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• References

Background

Endocarditis can have profound and devastating neurologic consequences. More than a century ago, Osler described the infective endocarditis (IE) triad of fever, heart murmur, and hemiplegia. In 10-15% of patients with IE, the nervous system yields the first clinical signs. The incidence of central nervous system (CNS) complications in IE is approximately 30%.

The Massachusetts General Hospital experience from 1964-1973 yielded 218 patients with bacterial endocarditis (BE), 84 of whom had neurologic complications (38%). Another series of 133 patients demonstrated a similar proportion (41%) of patients with IE with neurologic problems. The neurologic manifestations of IE are often the first sign of illness (47% of the time in one series), and they are present before initiation of antimicrobial therapy in most cases (76%). The incidence of CNS complications varies by organism (Staphylococcus species, 54%; Enterococcus species or Streptococcus viridans, >19%) and by location (mitral valve with Staphylococcus aureus, 87%).

Pathophysiology

The incidence of neurologic complications of BE is dependent on the organism and valvular location; the highest incidence is 87% with S aureus vegetations on the mitral valve. Mitral valve prolapse occurs in a significant proportion of patients with IE. Mitral valve involvement is most common (28-45%), followed by aortic valve (5-36%) and a combination thereof (as much as 56%). Mitral valve endocarditis carries a higher risk of CNS embolization than aortic valve involvement (17% vs 9%). Interestingly, at least one case report exists of IE arising from an endovascularly placed closure device for patent foramen ovale with atrial septal aneurysm.

Approximately 90% of native valve endocarditis (NVE) is caused by the following 3 organisms:

• S viridans (60%)
• S aureus (20%)
• Enterococcus species (10%)

Gram-negative bacteria and fungi make up the remainder, comprising a smaller percentage.

Staphylococcus species classically leads to embolization earlier than other organisms (ie, <2 wk) and often causes hemorrhage within the first 48 hours or even prior to other symptomatic presentation. S viridans subacute BE (SBE) is more indolent, usually with a clinical course of 1-3 months.

Partially treated BE usually has a slower time course as well. Pseudomonas species and other coliform bacteria, including Haemophilus influenzae and group B streptococci, produce large vegetations. Aspergillus and Candida species endocarditis also have a high incidence of embolic phenomena as a result of large valvular vegetations.

Stroke

A 0.5% risk of occurrence of stroke per day exists in patients with IE. One series showed that of patients with CNS involvement, four fifths had ischemic strokes and one fifth had hemorrhages, which is a distribution similar to that of strokes in general. Approximately 15-50% of the CNS manifestations are due to embolic occlusion and/or stroke. In 1965, McDevitt reported that SBE accounted for 3% of all cerebral emboli. In addition, one recent study showed that 71% of all embolic events in patients with IE involved the CNS.

The rate of major embolic events was 17% in one series (reported range, 6-31%). Of 38 events in 37 patients, 34 involved the middle cerebral artery (MCA) or its branches. If stroke recurs, it usually indicates that the infection is not controlled or has recurred. Embolization occurring greater than 2 months after presentation is uncommon. Strokes with IE can be subtle or even subclinical.

One series showed 23 patients with multiple microscopic infarcts at autopsy that had been clinically silent. Microorganisms rarely were confirmed in the emboli. One study showed that embolic events decreased from 13 per 1000 patient-days in the first week of antibiotic therapy to 1.2 per 1000 patient-days after 2 weeks of antibiotic therapy. Approximately one half of patients with cerebral emboli also have systemic emboli. Systemic emboli (non-CNS) occur equally from involvement of mitral and aortic valves.

Infectious aneurysms

Osler coined the term mycotic aneurysm to describe a mushroom-shaped aneurysm associated with SBE. Originally referring to any infectious etiology, it still is used as a general term and is not specific for fungal etiology (responsible for 3-6% of all intracerebral aneurysms), and case reports exist of at least 26 extracranial and 19 intracavernous carotid artery aneurysms associated with endocarditis.

Mycotic aneurysms occur in as many as 15% of patients with IE, and aneurysms of arteries supplying the brain comprise approximately 15% of the aneurysms occurring in IE. Most IE aneurysms occur at sites outside of the CNS. The proximal aorta, visceral arteries, and limb arteries account for 25%, 24%, and 22% of sites, respectively.

When aneurysms form, the most likely mechanism is bacterial induced weakening (originating from the adventitial side of the vessel wall and apparently after organisms traverse the vasa vasorum). According to animal studies, the pathology is not via direct endoluminal damage to the intima. Controversy exists as to whether the more virulent organisms are associated more commonly with aneurysms. Mycotic aneurysms are less common with acute BE (ABE) than with SBE. However, when present, mycotic aneurysms become symptomatic earlier in the disease course of ABE than in SBE. Intracardiac tumors (eg, left atrial myxomas, metastatic choriocarcinoma) also can embolize fragments and cause aneurysms resembling mycotic aneurysms.

The incidence of clinically diagnosed intracranial mycotic aneurysm in patients with IE is approximately 2% (compared with a 5-10% prevalence postmortem as determined at autopsy). The MCA territory is involved 4 times more commonly with mycotic aneurysms than either the anterior or posterior cerebral arteries. In contrast to berry aneurysms, which occur in or near the circle of Willis, mycotic aneurysms occur at secondary branches and bifurcations, usually in the lateral fissure near the trifurcation of the MCA.

Intracranial hemorrhage

The overall prevalence of hemorrhage in CNS involvement of IE is 3-7%. One series showed that nearly 30% of these cases are due to staphylococci. The etiology of CNS hemorrhage in IE is not completely clear. It once was believed to be due exclusively to aneurysmal rupture. This was not supported by angiographic and autopsy studies that did not reveal a consistent aneurysmal source of bleeding in all cases. In one series, 14 patients had CNS hemorrhage. Four definite mycotic aneurysms were identified, 6 were presumptive, and the other 4 were not felt to be aneurysm related. In another series, the source of subarachnoid hemorrhage in 6 out of 8 patients with IE could not be identified by cerebral angiography.

Other etiologies include nonaneurysmal vascular wall necrosis and hemorrhagic transformation of ischemic strokes. Intraventricular hemorrhage from a mycotic aneurysm rupture is usually fatal. Subarachnoid hemorrhage in IE is often focal and has been shown (in a recent series) to occur more frequently in the perisylvian and/or perirolandic regions.

Meningeal processes

In the Massachusetts General Hospital series of 218 patients, 33 had meningeal symptoms (15%). Of patients with CNS involvement from IE, one third had completely normal cerebrospinal fluid (CSF), one fourth had a slightly abnormal CSF but an aseptic culture, and one fourth had frankly purulent CSF. Noteworthy as well is that one sixth had hemorrhagic fluid. Purulent meningitis or abscess occurred in more than one half of patients with acute IE, whereas this occurred in fewer than 5% of patients with SBE.

When abscess occurs, the CSF can be nearly normal with a mildly increased protein, modest number of cells, normal glucose, and negative Gram stain. Conversely, S aureus abscesses usually are associated with purulent CSF containing polymorphonuclear neutrophilic leucocytes. If the CSF is abnormal, it usually represents an inflammatory consequence (meningeal reaction) rather than a primary infection. This is especially true in patients with acute staphylococcal BE.

A leak of a mycotic aneurysm or an underlying focal lesion can produce meningeal irritation and cause a secondary aseptic meningitis. Cases of true meningeal infection can occur with IE, but pneumococci are the usual culprits in these cases. Of all patients with IE and concurrent bacterial meningitis, 70% have pneumococci as the infecting organisms (although pneumococci account for only 1-2% of IE cases). In patients without a history of trauma, neurosurgery, or CSF leak who have staphylococcal meningitis, a 33% chance exists that they have concurrent staphylococcal endocarditis.

Spinal epidural abscess is also an important complication to recognize in IE, and it most commonly presents with motor deficits or back pain. One study found that IE was the cause in 10 out of 48 patients admitted with spinal epidural abscess (only slightly behind intravenous drug use as a leading cause).

Seizures

Seizures were part of the presenting complex in 5 of 218 patients in the Massachusetts General Hospital series and were present at some time in the course in 11% (24/218). Focal seizures may indicate an embolic etiology, whereas generalized seizures can result from meningitis, pharmacologic adverse effects, or systemic conditions such as uremia or hypoxia.

Nonfocal symptoms and/or encephalopathy

Nonfocal symptoms or encephalopathy can be the result of microscopic emboli and subsequent ischemia. This scenario is present in as many as 25% of patients with IE. Approximately one half of these cases are due to evident metabolic derangements. The other one half was believed to be related to the IE since no other etiology could be identified. Most of the nonfocal symptoms, such as headache or encephalopathy, usually resolve shortly after instituting antibiotics.

Immune phenomena

Late proliferative endarteritis in cerebral blood vessels with thrombotic occlusion and subsequent stroke can be due to immune phenomena. In one case report of streptococcal endocarditis with immune complex vasculitis, episodes of neurologic deterioration improved after dexamethasone therapy after a lack of prior improvement with antibiotic therapy.

Nonbacterial thrombotic endocarditis

The previous term for nonbacterial thrombotic endocarditis (NBTE) was marantic endocarditis, from the Greek marantikos, meaning "wasting away." NBTE is found in 1.6% of all autopsies. An estimated 10% of cerebral emboli occur in the setting of NBTE. A Japanese series of 86 strokes of cardiac origin showed 11% were related to NBTE.

Risk factors for NBTE include disseminated intravascular coagulation (DIC), abnormal heart valves, and mucin-producing tumors (leading to fibrin deposition and subsequent embolic phenomena). Approximately 80% of NBTE cases are associated with known carcinomas (eg, pancreas, colon, lung). Classically described with metastatic mucin-producing adenocarcinomas, it also occurs in localized cancers as well as in noncancerous states such as pregnancy, drug overdose, cirrhosis, vasculitis, and rheumatic heart disease.

Patients with NBTE who do not have cancer often have rheumatic or arteriosclerotic heart disease. NBTE also occurs in patients with AIDS. Hypercoagulable indices suggest a diagnosis of NBTE. One study reports that 30% of patients with NBTE who have CNS manifestations have nonfocal findings. Another series showed 18% NBTE in autopsied patients with cancer and associated neurologic complications.

Frequency

United States

Incidence of IE is 2-4 cases per 100,000 population.

International

Frequency is similar in published data from the United Kingdom and is unchanged when compared to the preantibiotic era.

In patients with prosthetic mitral valves, the risk of endocarditis is 0.5% per year.

Mortality/Morbidity

• In the preantibiotic era, the mortality of endocarditis was nearly 100%. Currently, it is 20-50% overall. Death is often the result of congestive heart failure or acute valvular insufficiency. Studies show that the mortality of patients with BE with CNS involvement is much greater (58-74%) than that of patients who do not have CNS involvement (20-56%). In one study, during the acute phase of IE, 24% of patients with neurologic complications died while only 10% of those without neurologic complications died. Among the patients with neurologic involvement, mortality was 25% in those treated medically, and 20% in those treated surgically.
• In one series, 38 major embolic events occurred in 37 patients. Thirty of these 37 patients died (81% mortality).

Sex

• The male-to-female ratio is 1.7:1 (average of several series).
• In the presence of IE and aortic valve rheumatic disease alone, the male-to-female ratio is 4:1.
• If the mitral valve alone is involved as a sequela to rheumatic fever, the male-to-female ratio is 1:2.

Age

• Patients younger than 30 years account for 26%.
• Patients aged 31-60 years account for 54%.
• Patients older than 60 years account for 21%.

CLINICAL

Section 3 of 8  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• References

History

Nonspecific symptoms such as fatigue, malaise, anorexia, and weight loss are common.

• Symptoms are sometimes nonfocal, and even amnesia can herald a diagnosis of IE.
• Mental status changes (without focal findings) have been attributed to microemboli (11% in the Massachusetts General Hospital series).
• Cancer patients who appear to have reactive or behavioral (psychiatric) symptoms actually may have organic embolic disease. If fever, seizures, and/or DIC are present in a patient with cancer, culture-negative BE may be misdiagnosed as psychiatric disease.
• Transient ischemic attack-like (TIA-like) presentations also can occur with fluctuating focal neurologic signs. Presumably, the transient nature (resolution) results from the disintegration and dislodgement of small emboli.
• Lumbar pain and fever are reported to herald the diagnosis of IE in approximately 3% of patients.

Physical

• Fever and heart murmurs are often present. However, the valvular vegetations are usually not large enough to produce murmurs of their own accord. Thus, as many as 25% of patients with acute IE do not have a cardiac murmur.
• Stigmata of systemic emboli and cardiac dysfunction are not always clinically apparent. When found, they occur in different body regions.
• • Immune phenomena occur in IE with increases in polyclonal immunoglobulin (Ig)G and immune complexes. When deposited in the skin, they produce the characteristic Osler node (ie, painful finger nodules).
  • Janeway lesions are erythematous macules found on the palms and soles.
  • Roth spots are retinal hemorrhages.

Causes

• Those who abuse intravenous (IV) drugs have a slightly different profile, with 60% of all cases secondary to S aureus. This increases to 80% when the tricuspid valve is affected.
• High-risk groups include the following:
• • In the mid-1980s, rheumatic disease was felt to be a disease of the past since it accounted for less than 25% of cases, but since 1985, a resurgence has been reported in some parts of the United States.
  • Presently, the 3 main high-risk groups are (1) illicit drug users, (2) patients with prosthetic cardiac valves, and (3) patients with nosocomial-related endocarditis.
  • IE and other forms of endocarditis such as NBTE can affect patients with normal native heart valves. Differences in opinion remain as to whether patients with NVE or prosthetic valve endocarditis (PVE) are at greater risk of neurologic complications. A recent study suggested that IE occurred more often in patients with mechanical valves than in those with bioprosthetic valves.
• Dental procedure prophylaxis concerns are as follows:
• • Recently, a retrospective study was performed that questioned the use of antibiotic prophylaxis prior to dental work.
  • The only dental procedure with an increased risk of endocarditis was tooth extraction within 2 months after hospitalization.
• Patients with a history of valve abnormalities who were given antibiotics were not protected against endocarditis.
• Guidelines for antibiotic prophylaxis for endocarditis prevention are available from the American Heart Association.

DIFFERENTIALS

Section 4 of 8  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• References

Other Problems to be Considered

Computed tomography in neurovascular diseases
Depression
EEG in neurologic infections
Epidural abscess
Gastrointestinal discomfort
Heart failure
Orbital cellulitis
Psychosis
Weight loss
Carotid disease and stroke

WORKUP

Section 5 of 8  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• References

Lab Studies

• Blood cultures
• • Unless the patient is overtly toxic, the current recommendation is to obtain 3 blood cultures in the first 24 hours and prior to starting antibiotics.
  • In two thirds of patients, all culture samples are positive.
  • Approximately 5% are culture negative (50% of fungal BE is culture negative).
• Cerebrospinal fluid examination: See Meningeal processes in the Pathophysiology section for an outline of CSF findings in IE.
• Complete blood cell count
• • Anemia is present in 70-90% of patients and is usually normochromic and normocytic.
  • The peripheral white blood cell count is elevated in a minority of patients with SBE.
  • Marked leukocytosis is not uncommon in acute IE.
• Erythrocyte sedimentation rate
• • The erythrocyte sedimentation rate (ESR) is elevated in 90% of patients with IE, with a median ESR on admission of 65 mm/h.
  • The C-reactive protein is more helpful to assess therapeutic efficacy since it falls much sooner than the ESR, which can stay elevated for 3-6 months following successful therapy.
• Rheumatoid factor: The rheumatoid factor is more commonly positive in SBE (approximately 50%) than in ABE.
• Urinalysis
• • Urinalysis demonstrates either increased protein or red blood cells (RBCs) in 50% of patients.
  • Immunoglobulin complexes also can deposit in the kidney and cause nephropathy.

Imaging Studies

• Echocardiography
• • Transthoracic echocardiography identifies 60-77% of valvular vegetations; transesophageal echocardiography identifies approximately 96% of vegetations.
• Echocardiography may not be positive, since vegetations are often less than 2 mm in diameter.
• In a series of 212 patients, 113 underwent transthoracic echo, and 53% had visualized vegetations.
• At least one study suggests that vegetations confirmed by echocardiography do not correlate significantly with an increased risk of embolism. However, vegetation size greater than 10 mm does correspond with increased risk of embolism.
• Angiography
• • For patients with IE and focal neurologic deficits, the Cleveland Clinic has recommended 4-vessel angiography at a point between 2 days and 2 weeks from the onset of symptoms.
  • Others do not pursue routine angiography to search for occult aneurysms but reserve it for those with established subarachnoid hemorrhage or persistent headache after the infection is controlled.
• Brain imaging
• • Computed tomography (CT) of the brain may be normal in IE with CNS involvement, but magnetic resonance imaging (MRI) usually reveals multiple focal areas of ischemia even in the absence of clinical manifestations of a CNS disorder (in 33 out of 35 patients in one study).
  • In one neuroimaging report, 2 patients were presented who were transferred with mental status changes. Both had normal contrast-enhanced CT scans, but subsequent cranial MRI scans were abnormal. One patient was agitated and confused but had no neurologic focal or lateralized deficit. The echocardiogram was normal, but CSF subsequently was abnormal. The second patient was hemiparetic, but contrast and noncontrast CT scans were normal. The ESR was normal despite demonstrable mitral valve vegetations by echocardiogram. Thus, a normal ESR does not completely exclude IE.
  • Regions are as follows:
    • A recent series showed that of 12 patients with endocarditis and neurologic dysfunction, 10 had multiple brain lesions. These most frequently were found in the distal MCA territory and at the gray-white junction.
    • Most were less than 1 cm² in diameter and enhanced with contrast.
    • Orbital cellulitis was reported in 2 patients in this series.
  • Diffusion-weighted MRI imaging (DWI) is very sensitive in revealing evidence of ischemic cerebral lesions in patients with IE. In one study, DWI images revealed ischemic lesions in 33 of 35 patients with IE and neurologic symptoms.

Histologic Findings

Macroscopic brain abscesses are rare in patients with SBE (0.5%). Microscopic abscesses (also termed focal cerebritis) are more common (4%) and often are discovered at autopsy. For mass lesions (abscess and/or focal cerebritis), surgery is rarely necessary since these lesions usually improve and occasionally resolve after antibiotic therapy.

In one series, 9 abscesses were found, with staphylococci the predominant organisms (8 of 9 abscesses were <1 cm). Abscesses larger than 1 cm usually are not associated with endocarditis. Epidural abscesses can occur in patients with IE. One such case in the Massachusetts General Hospital series was unsuspected and was discovered at lumbar puncture.

TREATMENT

Section 6 of 8  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• References

Medical Care

All medical care should be directed by an infectious disease specialist and the primary care physician.

• Antibiotics
• • Embolization primarily occurs prior to the initiation of antibiotics or within the first week of starting antibiotic therapy.
  • Antibiotics resolve 50% of formed mycotic aneurysms. They also decrease the risk of bleeding and delay bleeding from mycotic aneurysms from approximately 2-10 days.
• Anticoagulants
• • Anticoagulants are contraindicated in individuals with NVE because of the increased risk of CNS hemorrhage. Some studies show that, even after an embolic event, anticoagulants should not be administered. They often are not warranted because with adequate control of infection within the first week, the risk of CNS embolization declines sharply.
  • If not anticoagulated, patients with PVE are more likely to have embolic phenomena (50-70% risk). The risk of neurologic complications is not increased in patients with IE who are on anticoagulants at the time of onset of IE. Therefore, unless neurologic complications exist, anticoagulation therapy is continued in patients with prosthetic valves but with a lower goal international normalized ratio (INR) of 1.5. Anticoagulation increases the risk of bleeding from mycotic aneurysms. Early angiography should be performed to guide therapy if aneurysms are present or suspected.

Surgical Care

Whether mycotic aneurysms should be observed, clipped, or treated endovascularly is controversial and depends on the need for cardiac surgery. Forty percent of patients with IE will need cardiac surgery, primarily for valve repair or replacement. Indications for surgery include hemodynamic instability, uncontrolled infection, and peripheral embolic events, and these are described in detail in guidelines published by the American College of Cardiology and the American Heart Association in 1998. Timing of surgery remains controversial.

• One study evaluated patients with ruptured and unruptured mycotic aneurysms treated medically or surgically. Of patients with ruptured aneurysms, 10 were treated medically and 6 surgically; 8 of the medically treated aneurysms resolved (the other 2 patients had poor outcomes), and all of the surgically treated patients fared well. Of patients with unruptured aneurysms, 4 were treated medically and 1 surgically; all had good outcomes.
• If cardiac valve replacement is planned, an accessible cerebral aneurysm is often clipped or occluded endovascularly prior to surgery to avoid the risk of hemorrhage with the subsequent cardiac surgery and its inherent anticoagulation perioperatively.
• Timing of surgery in patients with IE and embolic stroke remains controversial, but a recent report suggested that surgery can be performed relatively safely within 3 days of stroke if heart failure is severe; otherwise, a delay of 2-4 weeks is preferable. In patients with associated hemorrhage, a delay of at least 4-6 weeks is preferred.
• • Operative mortality is variable but has been reported as 7.6%, with risk factors for death being cardiogenic shock, insidious illness, and increased age. The 9-year survival rate has been reported to be 71%; risk factors for death include preoperative neurologic complications.
  • Risk of neurologic deterioration after valve replacement for IE is 20% in the first 72 hours, 20-50% 4-14 days postoperatively, less than 10% beyond 14 days postoperatively, and less than 1% after 4 weeks.

Consultations

• Infectious disease specialist, internist
• Neurologist, neurosurgeon
• Cardiologist, cardiothoracic surgeon
• Psychiatrist, drug rehabilitation specialist, social services counselor

MEDICATION

Section 7 of 8  

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• References

See Infective Endocarditis article for treatment of underlying infection and Prosthetic Heart Valves article for anticoagulation therapy.

REFERENCES

Section 8 of 8

• Authors and Editors
• Introduction
• Clinical
• Differentials
• Workup
• Treatment
• Medication
• References

• Alexander W. Thurston's The Heart. 9th ed. 1998:2220-2222.
• Aminoff MJ. Neurology and General Medicine. New York, NY:. Churchill Livingstone;1995:97-117.
• Anderson DJ, Goldstein LB, Wilkinson WE, et al. Stroke location, characterization, severity, and outcome in mitral vs aortic valve endocarditis. Neurology. Nov 25 2003;61(10):1341-6. [Medline].
• Angstwurm K, Borges AC, Halle E, et al. Timing the valve replacement in infective endocarditis involving the brain. J Neurol. Oct 2004;251(10):1220-6. [Medline].
• Bakshi R, Wright PD, Kinkel PR, et al. Cranial magnetic resonance imaging findings in bacterial endocarditis: the neuroimaging spectrum of septic brain embolization demonstrated in twelve patients. J Neuroimaging. Apr 1999;9(2):78-84. [Medline].
• Bertorini TE, Laster RE Jr, Thompson BF, Gelfand M. Magnetic resonance imaging of the brain in bacterial endocarditis. Arch Intern Med. Apr 1989;149(4):815-7. [Medline].
• Bonow RO, Carabello B, De Leon AC. ACC/AHA Guidelines for the Management of Patients With Valvular Heart Disease:A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients With Valvular Heart Disease). JACC. 1998;32:1486-1588.
• Bradley W, Daroff R, Fenichel G. Neurology in Clinical Practice. Vol 2. Butterworth-Heinemann;1996:908-9, 1018-9.
• Chamoun AJ, Conti V, Lenihan DJ. Native valve infective endocarditis: what is the optimal timing for surgery?. Am J Med Sci. Oct 2000;320(4):255-62. [Medline].
• Chapot R, Houdart E, Saint-Maurice JP, et al. Endovascular treatment of cerebral mycotic aneurysms. Radiology. Feb 2002;222(2):389-96. [Medline].
• Chukwudelunzu FE, Brown RD, Wijdicks EF, Steckelberg JM. Subarachnoid haemorrhage associated with infectious endocarditis: case report and literature review. Eur J Neurol. Jul 2002;9(4):423-7. [Medline].
• Curry WT, Hoh BL, Amin-Hanjani S, Eskandar EN. Spinal epidural abscess: clinical presentation, management, and outcome. Surg Neurol. Apr 2005;63(4):364-71; discussion 371. [Medline].
• Dajani AS, Taubert KA, Wilson W, et al. Prevention of bacterial endocarditis. Recommendations by the American Heart Association. JAMA. Jun 11 1997;277(22):1794-801. [Medline].
• Goldstein JA, Beardslee MA, Xu H, et al. Infective endocarditis resulting from CardioSEAL closure of a patent foramen ovale. Catheter Cardiovasc Interv. Feb 2002;55(2):217-20; discussion 221. [Medline].
• Gray IR. Infective endocarditis 1937-1987. Br Heart J. Mar 1987;57(3):211-3. [Medline].
• Groothuis DR, Mikhael MA. Focal cerebral vasculitis associated with circulating immune complexes and brain irradiation. Ann Neurol. Jun 1986;19(6):590-2. [Medline].
• Heiro M, Nikoskelainen J, Engblom E, et al. Neurologic manifestations of infective endocarditis: a 17-year experience in a teaching hospital in Finland. Arch Intern Med. Oct 9 2000;160(18):2781-7. [Medline].
• Jault F, Gandjbakhch I, Rama A, et al. Active native valve endocarditis: determinants of operative death and late mortality. Ann Thorac Surg. Jun 1997;63(6):1737-41. [Medline].
• Kanter MC, Hart RG. Neurologic complications of infective endocarditis. Neurology. Jul 1991;41(7):1015-20. [Medline].
• Kassai B, Gueyffier F, Cucherat M, Boissel JP. Comparison of bioprosthesis and mechanical valves, a meta-analysis of randomised clinical trials. Cardiovasc Surg. Oct 2000;8(6):477-83. [Medline].
• Kaye D. Changing pattern of infective endocarditis. Am J Med. Jun 28 1985;78(6B):157-62. [Medline].
• Le Cam B, Guivarch G, Boles JM, et al. Neurologic complications in a group of 86 bacterial endocarditis. Eur Heart J. Oct 1984;5 Suppl C:97-100. [Medline].
• Mackenzie TB, Popkin MK. Psychological manifestations of nonbacterial thrombotic endocarditis. Am J Psychiatry. Aug 1980;137(8):972-3. [Medline].
• Mugge A, Daniel WG, Frank G, Lichtlen PR. Echocardiography in infective endocarditis: reassessment of prognostic implications of vegetation size determined by the transthoracic and the transesophageal approach. J Am Coll Cardiol. Sep 1989;14(3):631-8. [Medline].
• Oppenheimer SM, Lima J. Neurology and the heart. J Neurol Neurosurg Psychiatry. Mar 1998;64(3):289-97. [Medline].
• Osler W. Gulstonian lectures on malignant endocarditis. Lancet. 1885;1:415, 459, 505.
• Piper C, Wiemer M, Schulte HD, Horstkotte D. Stroke is not a contraindication for urgent valve replacement in acute infective endocarditis. J Heart Valve Dis. Nov 2001;10(6):703-11. [Medline].
• Pruitt AA, Rubin RH, Karchmer AW, Duncan GW. Neurologic complications of bacterial endocarditis. Medicine (Baltimore). Jul 1978;57(4):329-43. [Medline].
• Reagan TJ, Okazaki H. The thrombotic syndrome associated with carcinoma. A clinical and neuropathologic study. Arch Neurol. Dec 1974;31(6):390-5. [Medline].
• Roder BL, Wandall DA, Espersen F, et al. Neurologic manifestations in Staphylococcus aureus endocarditis: a review of 260 bacteremic cases in nondrug addicts. Am J Med. Apr 1997;102(4):379-86. [Medline].
• Salgado AV, Furlan AJ, Keys TF. Mycotic aneurysm, subarachnoid hemorrhage, and indications for cerebral angiography in infective endocarditis. Stroke. Nov-Dec 1987;18(6):1057-60. [Medline].
• Shapiro SM, Young E, De Guzman S, et al. Transesophageal echocardiography in diagnosis of infective endocarditis. Chest. Feb 1994;105(2):377-82. [Medline].
• Singhal AB, Topcuoglu MA, Buonanno FS. Acute ischemic stroke patterns in infective and nonbacterial thrombotic endocarditis: a diffusion-weighted magnetic resonance imaging study. Stroke. May 2002;33(5):1267-73. [Medline].
• Venkatesh SK, Phadke RV, Kalode RR, et al. Intracranial infective aneurysms presenting with haemorrhage: an analysis of angiographic findings, management and outcome. Clin Radiol. Dec 2000;55(12):946-53. [Medline].
• Vlessis AA, Khaki A, Grunkemeier GL, et al. Risk, diagnosis and management of prosthetic valve endocarditis: a review. J Heart Valve Dis. Sep 1997;6(5):443-65. [Medline].
• Ziment I. Nervous system complications in bacterial endocarditis. Am J Med. Oct 1969;47(4):593-607. [Medline].

Article Last Updated: Dec 8, 2006