I
I
Cerebral Ischemic Events Associated with Endocarditis, Retinal Vascular Disease, and Lupus Anticoagulant JANET M. POPE, M.D., F.R.C.P.C., CHRISTOPHERL. B. CANNY, M.D., F.R.C.S.C., DAVIDA. BELL, M.D., F.R.C.P.C., London, Ontario, Canada
PURPOSE: A group of young patients presenting with cerebral ischemic events, endocardl91_ lesions, and lupus anticoagulant is described in order to highlight the common clinical laboratory features. P A T I E N T S AND METHODS: Fourteen consecutive patients (10 females, age range 17 to 53 years [mean 38 years]) at onset of symptoms of cerebral ischemiA who had evidence of the lupus anticoagulant syndrome and were being followed prospectively are reviewed. All patients had abnormal phospholipid-dependent coagulation test results, and most had anticardiolipin antibody at the time of presentation. Three of 14 had four or more American Rheumatism Association criteria for def'mite systemic lupus erythemAtosus a n d t h e r e m a i n i n g patients were considered to have primary lupus anticoagulant syndrome. RESULTS: The common features among these patients included at least one cerebral ischemic e v e n t at presentation (stroke or transient ischemic attack), or r e c u r r e n t episodes suggesting cerebral ischemia (Amaurosis fugax, r e c u r r e n t severe migraine headaches), livedo reticularis, endocardiai valvular lesions noted on echocardiography (11 mitral, two aortic valve) that were o f t e n associated with discrete vegetations, retinal vascular lesions, and computed tomographic/magnetic resonance imAwing scanning or angiographic evidence of multiple cerebral infarction. Venous thromboembolic events were uncommon (three of 14). Common laboratory studies included thrombocytopenia (10 of 14), positive direct Coombs' test result (11 of 14), a n d hypocomplementemla (11 of 14). Follow-up after initial treatment with either salicylates or anticoagulant therapy (warfarin) for up to 10 years i n d i c a t e d that while many patients had recurrent symptoms suggesting cerebral ischemiA, From the Department of Medicine, University of Western Ontario, University Hospital, London, Ontario, Canada. This work was supported in part from a research grant from the Canadian Arthritis Society. Requests for reprints should be addressed to David A. Bell. M.D., University Hospital, Box 5339, London, Ontario, Canada N6A 5A5. Manuscript submitted June 28, 1990, and accepted in revised form November 19, 1990.
major stroke syndromes did n o t recur nor n e w episodes emerge. CONCLUSION:The combination of multiple cerebral ischemic lesions and endocardiA1 lesions, including valvular vegetations, suggests that these cerebral ischemic events represent cerebral emboli, and that these cerebral embolic events originate from vegetative lesions on t h e mitral or, less commonly, aortic valve, in association with lupus anticoagulant.
ecently, a number of isolated case reports and R reviews have expanded the clinical spectrum seen in patients with serum antiphospholipid and/ or anticardiolipin antibody [1-3]. Traditionally, the lupus anticoagulant syndrome has been associated with arterial and venous thrombosis, an increased incidence of spontaneous abortions [1,4], and, more recently, thrombocytopenia [1,5], recurrent migraine headaches [6], livedo reticularis [1], and amaurosis fugax [7,8] have been noted in these pat i e n t s . T h e r e is also a p o s s i b l e i m p l i c a t e d association of verrucous endocarditis with the lupus anticoagulant [9,10]. The natural history of this syndrome, and the most satisfactory means of managing it, however, are presently not known. We have recently identified a group of patients with many clinical features of the lupus anticoagulant syndrome, whose diagnosis was considered only after they presented with a variety of fixed or episodic central nervous system ischemic events. These patients usually were not among patients known to have had established systemic lupus erythematosus (SLE). The vast majority of them had a fairly readily recognizable and predictable clinical and laboratory presentation; therefore, we considered it worthwhile to document these features. In addition, since we have followed these patients closely since the onset of their neurologic problems, the shortterm course of their disease in response to treatment is also discussed.
PATIENTS AND METHODS Patients T h e 14 patients included in this study were referred for evaluation to one of us ( D A B ) either from March 1991 The American Journal of Medicine Volume 90
299
CEREBRALISCHEMIAAND LUPUSANTICOAGULANT/ POPEET AL
general internists, neurologists, or rheumatologists, within or outside the University Hospital center. All of these patients had some clinical and laboratory features attributed by others to the lupus anticoagulant syndrome [11], and all had specific laboratory evidence of this syndrome with prolongation of the activated partial thromboplastin time (APTT) and elevated anticardiolipin or antiphospholipid antibodies. Since University Hospital is a tertiary referral center, it is possible that these cases represent a caricature of this clinical problem. The patients, for the most part, however, had not been followed in the Lupus Clinic prior to the onset of their symptoms and therefore were not referred initially for that reason. Three of the 14 patients fulfilled American Rheumatism Association (ARA) criteria for SLE [12] at presentation, and none of the. remaining 11 patients developed these features over follow-up for up to 10 years. The tests or procedures performed on most patients with this syndrome and analyzed in this study are described below. Coagulation Studies
ACTIVATED PARTIAL THROMBOPLASTIN TIME (APTT): An A P T T was obtained as previously described [13], using a rabbit brain lipid extract provided commercially (Organon Teknika, Scarborough, Ontario). The reference range in normal subjects was 25 to 42.0 seconds. An unexplained elevated A P T T was attributed to the presence of a lupus anticoagulant after appropriate mixing studies were performed. PROTHROMBINTIME (PT): The PT was measured by the established method of Quick, using Simplastin automated Organon Teknika reagent [14]. The normal reference range was 8.8 to 12.5 seconds. DILUTE RUSSELL VIPER VENOM TIME (DRVVT): DRVVT was measured according to a published method [15]. The reference range was 23 to 29 seconds; with 30 normal control subjects, the mean was 25.5 seconds, and the coefficient of variation was 2.1%. The presence of a lupus anticoagulant was confirmed when the DRVVT was greater than 31 seconds and did not correct with a 50:50 mix of normal control plasma. PLATELET NEUTRALIZATION PROCEDURE: This was performed according to a published method [16]. A greater than 10-second reduction of the A P T T indicated a lupus anticoagulant and the reduction always exceeded the saline control run in duplicate (0.1 mL saline, 0.1 mL plasma, 0.1 mL APTT). ANTIPHOSPHOLIPID ANTIBODYASSAY: This assay measured antibodies to three different phospholipids employing a polyclonal labeled antisera [17]. The first assay was performed on plasma with car300
diolipin, phosphatidylserine, and phosphatidylethanolamine as reagents. This technique has been described elsewhere [18]. This assay was validated using standardized reagents [19].
Immunoserologic Techniques The following immunoserologic techniques were performed on most patients: total hemolytic complement (CHs0) titer; C4 and C3 levels; serum antiDNA antibodies (Farr assay) [20]; and fluorescent antinuclear antibody determination (using Hep-2 cells).
Other Investigations Most patients also underwent the following procedures: chest roentgenography; two-dimensional M-mode echocardiography; 12-lead electrocardiography; four-vessel cerebral angiography; computed tomographic (CT) scan, and magnetic resonance imaging (MRI) scan. In most instances, the investigation was undertaken before the patients received anticoagulation, when this was chosen as the mode of treatment, or therapy was started with high-dose corticosteroids. It should be noted that transesophageal two-dimensional and M-mode echocardiography were routinely undertaken using the transthoracic approach and were interpreted by the same independent investigator who was unaware of the clinical history. In a few instances, a transesophageal echocardiogram was obtained. RESULTS
Table I summarizes the essential principal clinical features in synopsis form for each patient. A detailed clinical and pathologic description of one patient (Patient 13), whose features are typical of many of the patients, is presented below. A 47-year-old woman (Patient 13) presented 2 years previously with an unexplained generalized seizure and a normal echocardiogram. A cardiac murmur had been identified while she was in her 20s; during her 30s, mild mitral stenosis was diagnosed. The patient had experienced an episode of phlebitis in 1959 at age 24, and had had three spontaneous miscarriages in the first trimester of pregnancy. She also described a long history of migraine headaches, and over the last few years before admission, repeated attacks of amaurosis fugax involving the right eye. In 1982, she was admitted to the hospital because of confusion and increasing shortness of breath. She had been examined 3 months previously, and a diagnosis of SLE had been made on the basis of an unexplained seizure episode, false-positive VDRL, and arthralgias. She had a retinal examination prior to initiation of antimalarial drug therapy, and this revealed vaso-oblit-
March1991 The American Journal of Medicine Volume90
CEREBRAL ISCHEMIA AND LUPUS ANTICOAGULANT / POPE ET AL
TABLE I Synopsis of Clinical Features Patient
Age and Sex
Clinical Features
1
38F
Recurrent amaurosis fugax since onset; Raynaud's; livedo reticularis; mild hypertension; mitral regurgitation; MRI ischemic change internal capsule
2
30F
Migraines; recurrent amaurosis fugax; single TIA; Raynaud's; mitral regurgitation; sister died following mitral valve replacement for Libman-Sacks endocarditis
3
39F
Thrombophlebitis X2 and pulmonary embolus, during first year of onset of SLE at age 28; Raynaud's; single TIA with abnormal cerebral angiogram, age 39, while receiving warfarin
4
40M
Onset of recurrent amaurosis and single transient episode of limb paresis, age 37; well-documented ankylosing spondylitis
5
43F
Recurrent migraines since age 27; amaurosis fugax; hypertension; mitral regurgitation confirmed on echocardiogram; Raynaud's; vertigo and thrombocytopenia at age 27. Occlusive retinal vascular lesions; CSF oligoclonal bands; MRI--multiple ischemic lesions; 3 first-trimester spontaneous abortions
6
36F
Daily amaurosis fugax; transient episode global amnesia; recurrent migraines; recurrent spontaneous abortions; aortic stenosis/insufficiency; retinal vascular occlusions
7
38M
Onset age 28 with phlebitis and single episode of amaurosis; vertigo; vaso-obliterative retinopathy; large sessile mitral valve vegetations; membranoproliferative glomerulonephritis (absence of inflammation and immune complex deposition)
8
51F
Acute psychosis age 48 and aphasia due to hemispheric stroke with recurrence while taking warfarin; spontaneous abortions in 6 of 8 pregnancies; strong family history of stroke and of protein S deficiency
9
23M
Right hemispheric stroke age 17 and left hemispheric stroke age 19 with recurrence after warfarin discontinuation; recurrent seizures; mitral valve prolapse; vaso-obliterative retinal vascular lesions; MRI--multiple areas of ischemia
10
38F
Recurrent amaurosis x l O years; hemiparesis and focal seizures age 38; multiple branch occlusions of middle cerebral artery on angiography; Raynaud's; mitral valve thickening with regurgitation on echocardiogram; no neurologic recurrence with warfarin
11
51F
Headaches; dysarthria with loss of consciousness 13 years after diagnosis of SLE; progressive memory loss since age 47; Raynaud's; chronic cutaneous leg ulceration; transient murmurs of aortic insufficiency
12
17M
Phlebitis age 16; renal failure due to immune complex proliferative glomerulonephritis; acute stroke age 17 while taking 50 mg of prednisone; transient mitral regurgitation; CT left cerebral ischemia; mild hypertension
13
47F
Seizures; recurrent abortions 20 years earlier; phlebitis; recurrent migraine and amaurosis fugax: retinal occlusive vascular disease; cardiac murmur; splinter hemorrhages; renal failure (see detailed clinicopathologic description
14
40F
Seizures age 29; hypertension; recurrent amaurosis fugax for years and stroke age 40; 3 spontaneous abortions age 22-30 years; mitral insufficiency; multiple infarcts on CT scan
MRI = Magneticresonanceimaging scan: TIA = transient ischemic attack: CSF = cerebrospinalfluid: CT = computed Iomographicscan.
erative disease in the right eye only (see below for description). Anti-DNA antibodies were negative, serum complement titers were persistently below normal with low C3 and C4 levels, and a fluorescent antinuclear antibody test was weakly positive. The P T T value was markedly elevated at 84 seconds, and there were reduced levels of factors VIII, IX, and XI, all of which were due to the presence of a lupus anticoagulant. Anticardiolipin antibodies were not tested. During her final admission in 1982, vasculitic lesions were noted in the fingertips, as well as splinter hemorrhages. Right heart failure and renal'failure were present. She was admitted to the intensive care unit, became comatose, and failed to respond to brief courses of high-dose (1 g pulse) steroids or cyclophosphamide therapy. Blood cultures were persistently negative. She died 1 month later. At autopsy, there were widespread areas of in-
farction involving the cerebral hemispheres. In the cerebellum, there was widespread vascular occlusion by bland embolic material, which showed various degrees of organization, and widespread in situ thrombosis in small cerebral vessels. Multiple splenic infarcts were noted, and gangrenous changes were present in the right foot. No signs of acute vasculitis or thickening of vessels was noted in the microscopic sections of blood vessels, except in the renal glomerulus, where there was slight thickening in the basement membrane without immunoglobulin or complement deposition. The lesions in the kidney were thought to be consistent with an embolic process. All four cardiac valves contained vegetations. These were not believed to be typical of LibmanSacks vegetations because of their large size and the fact that they did not extend around the ventricular surface of the mitral valve. The largest vegetations
March 1991 The American Journal of Medicine Volume 90
301
CEREBRAL ISCHEMIA AND LUPUS ANTICOAGULANT / POPE ET AL TABLE II Clinical Thromboembolic Events Cerebral Events Patient
Ageand Sex
Stroke or TIA
1 2 3§ 4 5 6 7 8 9 10 11§ 12§ 13 14
38F 30F 39F 40M 44F 36F 38M 53F 22M 38F 51F 17M 47F 40F
O* TIA CVA CVA TIA 0 0 CVA CVA CVA* CVA CVA CVA' CVS*
Amaurosis
Migraine
DVT
Raynaud's
+ + Jr + + + + + 0 + + 0 + +
+ + 0 + + + 0 + 0 + + 0 + 0
0 0 + 0 0 0 + 0 0 0 0 -t+ 0
+ + + 0 0 + 0 + 0 + + 0 0 0
Recurrent Abortions 1/4t 0 NAt NA 3 5 NA 5/8 NA NA 0 NA 4/4 3/3
TIA = transient ischemic attack; DVT = deepvenousthrombosis; CVA = cerebrovascularaccident. * Patient presentedwith seizures. Indicates one miscarriage and three live births. NA = not applicable(male or no attempts at pregnancy). § Four or more ARAcriteria for SLE.
TABLE III Abnormal Findings on Clinical Examination and Investigation
Patient
Livedo Reticularis
Hypertension
Cardiac Murmur
Retinal Abnormality
Abnormal Echo
1
Jr
+
+
0
2 3
0 Jr
0 Jr
+ +
0 0
MR MR MR
4
0
0
+
0
MVP
5 6 7 8 9 10 11 12 13 14
+ + + + 0 + -I+ + +
+ 0 + + + 0 0 + + +
+ + + + + + + + + +
+ + + 0 + 0 0 0 + 0
MT AI/AS MT MR MVP/MR MT/MR AI MR* N MR
Abnormal CT or MRI NI 0 INF
0 INFS N INFS INFS INFS INFS INFS INFS INFS INFS
Abnormal Cerebral Angiogram NP NP + NP N N NP + + + N NP NP +
CT = computed tomographic scan; MRI = magnetic resonance imagingscan; Echo = echocardiogram; + = present; 0 = absent; MR = mitral regurgitation; MVP = mitral valve prolapse; MT = mitral vane thickening;AS = aortic stenosis;AI = aortic insufficiency; NI = noninterpretableabnormality; INF = infarct; INFS= multiple infarcts; NP = not performed; N = normal. • Cardiaccatheterization revealedmild mitral stenosisand aortic insufficiency.
were on the mitral and aortic valves. The mitral valve was stenotic and friable vegetations were present along the closure line on the atrialsurface measuring up to 0.8 c m in diameter. At the aortic valve, there was thickening of the cusps and fusion of the anterior commissure measuring 0.6 × 0.4 cm and vegetations measuring up to 0.6 c m in diameter on the posterior and left cusps. Microscopically, the valve lesions showed sclerosis fibrin and platelet vegetations with very little evidence of inflammation and no bacterial or fungal colonies. In the cerebral cortex, no evidence of vasculitis was seen in or near the areas of cerebral infarction. Most vessels showed areas of thromboembolic material without infarctions. Small arterial vessels in the sclera and choroid contained granular thromboembolic material that, in most cases, showed some evidence of 302
organization. The vessels themselves were unremarkable. A small vessel in the left eye (ganglion cell layer of the retina) was partially occluded by thromboembolic material showing early endothelialization.Thus, the autopsy findings were not considered to show evidence of SLE, but did demonstrate nonbacterial thrombotic endocarditis with large friable vegetations occurring in the setting of valves previously damaged by rheumatic heart disease. Cerebral infarcts were widespread and of varying ages, related to the presence of widespread thromboembolic lesions. S u m m a r y of C l i n i c a l Procedures
Features and Investigative
Tables II and III summarize and compare the clinical and some investigative procedures identi-
March 1991 The American Journal of Medicine Volume 90
CEREBRAL ISCHEMIA AND LUPUS ANTICOAGULANT / POPE ET AL
TABLEIV
Laboratory Findings P'I'I"
Anti-PPL Antibody
Thrombo.
Coombs'
VDRL
Patient
Abnormal
Positive
cytopenia
Positive
Positive
1 2 3 4 5t 6~ 7t 8 9t I0 1H 12t~ 13t 14
+ + -t0" + + + + + + + + + +
+ 0 + 0 + + NT + NT + + + NT +
+ + -I0 + + + 0 + 0 + + + 0
+ + + 0 0 + + + + 0 + + + +
NT + 0 0 + NT NT + + 0 + + + +
Anti-DNA (%)
I
Cerebral Ischemic Events Associated with Endocarditis, Retinal Vascular Disease, and Lupus Anticoagulant JANET M. POPE, M.D., F.R.C.P.C., CHRISTOPHERL. B. CANNY, M.D., F.R.C.S.C., DAVIDA. BELL, M.D., F.R.C.P.C., London, Ontario, Canada
PURPOSE: A group of young patients presenting with cerebral ischemic events, endocardl91_ lesions, and lupus anticoagulant is described in order to highlight the common clinical laboratory features. P A T I E N T S AND METHODS: Fourteen consecutive patients (10 females, age range 17 to 53 years [mean 38 years]) at onset of symptoms of cerebral ischemiA who had evidence of the lupus anticoagulant syndrome and were being followed prospectively are reviewed. All patients had abnormal phospholipid-dependent coagulation test results, and most had anticardiolipin antibody at the time of presentation. Three of 14 had four or more American Rheumatism Association criteria for def'mite systemic lupus erythemAtosus a n d t h e r e m a i n i n g patients were considered to have primary lupus anticoagulant syndrome. RESULTS: The common features among these patients included at least one cerebral ischemic e v e n t at presentation (stroke or transient ischemic attack), or r e c u r r e n t episodes suggesting cerebral ischemia (Amaurosis fugax, r e c u r r e n t severe migraine headaches), livedo reticularis, endocardiai valvular lesions noted on echocardiography (11 mitral, two aortic valve) that were o f t e n associated with discrete vegetations, retinal vascular lesions, and computed tomographic/magnetic resonance imAwing scanning or angiographic evidence of multiple cerebral infarction. Venous thromboembolic events were uncommon (three of 14). Common laboratory studies included thrombocytopenia (10 of 14), positive direct Coombs' test result (11 of 14), a n d hypocomplementemla (11 of 14). Follow-up after initial treatment with either salicylates or anticoagulant therapy (warfarin) for up to 10 years i n d i c a t e d that while many patients had recurrent symptoms suggesting cerebral ischemiA, From the Department of Medicine, University of Western Ontario, University Hospital, London, Ontario, Canada. This work was supported in part from a research grant from the Canadian Arthritis Society. Requests for reprints should be addressed to David A. Bell. M.D., University Hospital, Box 5339, London, Ontario, Canada N6A 5A5. Manuscript submitted June 28, 1990, and accepted in revised form November 19, 1990.
major stroke syndromes did n o t recur nor n e w episodes emerge. CONCLUSION:The combination of multiple cerebral ischemic lesions and endocardiA1 lesions, including valvular vegetations, suggests that these cerebral ischemic events represent cerebral emboli, and that these cerebral embolic events originate from vegetative lesions on t h e mitral or, less commonly, aortic valve, in association with lupus anticoagulant.
ecently, a number of isolated case reports and R reviews have expanded the clinical spectrum seen in patients with serum antiphospholipid and/ or anticardiolipin antibody [1-3]. Traditionally, the lupus anticoagulant syndrome has been associated with arterial and venous thrombosis, an increased incidence of spontaneous abortions [1,4], and, more recently, thrombocytopenia [1,5], recurrent migraine headaches [6], livedo reticularis [1], and amaurosis fugax [7,8] have been noted in these pat i e n t s . T h e r e is also a p o s s i b l e i m p l i c a t e d association of verrucous endocarditis with the lupus anticoagulant [9,10]. The natural history of this syndrome, and the most satisfactory means of managing it, however, are presently not known. We have recently identified a group of patients with many clinical features of the lupus anticoagulant syndrome, whose diagnosis was considered only after they presented with a variety of fixed or episodic central nervous system ischemic events. These patients usually were not among patients known to have had established systemic lupus erythematosus (SLE). The vast majority of them had a fairly readily recognizable and predictable clinical and laboratory presentation; therefore, we considered it worthwhile to document these features. In addition, since we have followed these patients closely since the onset of their neurologic problems, the shortterm course of their disease in response to treatment is also discussed.
PATIENTS AND METHODS Patients T h e 14 patients included in this study were referred for evaluation to one of us ( D A B ) either from March 1991 The American Journal of Medicine Volume 90
299
CEREBRALISCHEMIAAND LUPUSANTICOAGULANT/ POPEET AL
general internists, neurologists, or rheumatologists, within or outside the University Hospital center. All of these patients had some clinical and laboratory features attributed by others to the lupus anticoagulant syndrome [11], and all had specific laboratory evidence of this syndrome with prolongation of the activated partial thromboplastin time (APTT) and elevated anticardiolipin or antiphospholipid antibodies. Since University Hospital is a tertiary referral center, it is possible that these cases represent a caricature of this clinical problem. The patients, for the most part, however, had not been followed in the Lupus Clinic prior to the onset of their symptoms and therefore were not referred initially for that reason. Three of the 14 patients fulfilled American Rheumatism Association (ARA) criteria for SLE [12] at presentation, and none of the. remaining 11 patients developed these features over follow-up for up to 10 years. The tests or procedures performed on most patients with this syndrome and analyzed in this study are described below. Coagulation Studies
ACTIVATED PARTIAL THROMBOPLASTIN TIME (APTT): An A P T T was obtained as previously described [13], using a rabbit brain lipid extract provided commercially (Organon Teknika, Scarborough, Ontario). The reference range in normal subjects was 25 to 42.0 seconds. An unexplained elevated A P T T was attributed to the presence of a lupus anticoagulant after appropriate mixing studies were performed. PROTHROMBINTIME (PT): The PT was measured by the established method of Quick, using Simplastin automated Organon Teknika reagent [14]. The normal reference range was 8.8 to 12.5 seconds. DILUTE RUSSELL VIPER VENOM TIME (DRVVT): DRVVT was measured according to a published method [15]. The reference range was 23 to 29 seconds; with 30 normal control subjects, the mean was 25.5 seconds, and the coefficient of variation was 2.1%. The presence of a lupus anticoagulant was confirmed when the DRVVT was greater than 31 seconds and did not correct with a 50:50 mix of normal control plasma. PLATELET NEUTRALIZATION PROCEDURE: This was performed according to a published method [16]. A greater than 10-second reduction of the A P T T indicated a lupus anticoagulant and the reduction always exceeded the saline control run in duplicate (0.1 mL saline, 0.1 mL plasma, 0.1 mL APTT). ANTIPHOSPHOLIPID ANTIBODYASSAY: This assay measured antibodies to three different phospholipids employing a polyclonal labeled antisera [17]. The first assay was performed on plasma with car300
diolipin, phosphatidylserine, and phosphatidylethanolamine as reagents. This technique has been described elsewhere [18]. This assay was validated using standardized reagents [19].
Immunoserologic Techniques The following immunoserologic techniques were performed on most patients: total hemolytic complement (CHs0) titer; C4 and C3 levels; serum antiDNA antibodies (Farr assay) [20]; and fluorescent antinuclear antibody determination (using Hep-2 cells).
Other Investigations Most patients also underwent the following procedures: chest roentgenography; two-dimensional M-mode echocardiography; 12-lead electrocardiography; four-vessel cerebral angiography; computed tomographic (CT) scan, and magnetic resonance imaging (MRI) scan. In most instances, the investigation was undertaken before the patients received anticoagulation, when this was chosen as the mode of treatment, or therapy was started with high-dose corticosteroids. It should be noted that transesophageal two-dimensional and M-mode echocardiography were routinely undertaken using the transthoracic approach and were interpreted by the same independent investigator who was unaware of the clinical history. In a few instances, a transesophageal echocardiogram was obtained. RESULTS
Table I summarizes the essential principal clinical features in synopsis form for each patient. A detailed clinical and pathologic description of one patient (Patient 13), whose features are typical of many of the patients, is presented below. A 47-year-old woman (Patient 13) presented 2 years previously with an unexplained generalized seizure and a normal echocardiogram. A cardiac murmur had been identified while she was in her 20s; during her 30s, mild mitral stenosis was diagnosed. The patient had experienced an episode of phlebitis in 1959 at age 24, and had had three spontaneous miscarriages in the first trimester of pregnancy. She also described a long history of migraine headaches, and over the last few years before admission, repeated attacks of amaurosis fugax involving the right eye. In 1982, she was admitted to the hospital because of confusion and increasing shortness of breath. She had been examined 3 months previously, and a diagnosis of SLE had been made on the basis of an unexplained seizure episode, false-positive VDRL, and arthralgias. She had a retinal examination prior to initiation of antimalarial drug therapy, and this revealed vaso-oblit-
March1991 The American Journal of Medicine Volume90
CEREBRAL ISCHEMIA AND LUPUS ANTICOAGULANT / POPE ET AL
TABLE I Synopsis of Clinical Features Patient
Age and Sex
Clinical Features
1
38F
Recurrent amaurosis fugax since onset; Raynaud's; livedo reticularis; mild hypertension; mitral regurgitation; MRI ischemic change internal capsule
2
30F
Migraines; recurrent amaurosis fugax; single TIA; Raynaud's; mitral regurgitation; sister died following mitral valve replacement for Libman-Sacks endocarditis
3
39F
Thrombophlebitis X2 and pulmonary embolus, during first year of onset of SLE at age 28; Raynaud's; single TIA with abnormal cerebral angiogram, age 39, while receiving warfarin
4
40M
Onset of recurrent amaurosis and single transient episode of limb paresis, age 37; well-documented ankylosing spondylitis
5
43F
Recurrent migraines since age 27; amaurosis fugax; hypertension; mitral regurgitation confirmed on echocardiogram; Raynaud's; vertigo and thrombocytopenia at age 27. Occlusive retinal vascular lesions; CSF oligoclonal bands; MRI--multiple ischemic lesions; 3 first-trimester spontaneous abortions
6
36F
Daily amaurosis fugax; transient episode global amnesia; recurrent migraines; recurrent spontaneous abortions; aortic stenosis/insufficiency; retinal vascular occlusions
7
38M
Onset age 28 with phlebitis and single episode of amaurosis; vertigo; vaso-obliterative retinopathy; large sessile mitral valve vegetations; membranoproliferative glomerulonephritis (absence of inflammation and immune complex deposition)
8
51F
Acute psychosis age 48 and aphasia due to hemispheric stroke with recurrence while taking warfarin; spontaneous abortions in 6 of 8 pregnancies; strong family history of stroke and of protein S deficiency
9
23M
Right hemispheric stroke age 17 and left hemispheric stroke age 19 with recurrence after warfarin discontinuation; recurrent seizures; mitral valve prolapse; vaso-obliterative retinal vascular lesions; MRI--multiple areas of ischemia
10
38F
Recurrent amaurosis x l O years; hemiparesis and focal seizures age 38; multiple branch occlusions of middle cerebral artery on angiography; Raynaud's; mitral valve thickening with regurgitation on echocardiogram; no neurologic recurrence with warfarin
11
51F
Headaches; dysarthria with loss of consciousness 13 years after diagnosis of SLE; progressive memory loss since age 47; Raynaud's; chronic cutaneous leg ulceration; transient murmurs of aortic insufficiency
12
17M
Phlebitis age 16; renal failure due to immune complex proliferative glomerulonephritis; acute stroke age 17 while taking 50 mg of prednisone; transient mitral regurgitation; CT left cerebral ischemia; mild hypertension
13
47F
Seizures; recurrent abortions 20 years earlier; phlebitis; recurrent migraine and amaurosis fugax: retinal occlusive vascular disease; cardiac murmur; splinter hemorrhages; renal failure (see detailed clinicopathologic description
14
40F
Seizures age 29; hypertension; recurrent amaurosis fugax for years and stroke age 40; 3 spontaneous abortions age 22-30 years; mitral insufficiency; multiple infarcts on CT scan
MRI = Magneticresonanceimaging scan: TIA = transient ischemic attack: CSF = cerebrospinalfluid: CT = computed Iomographicscan.
erative disease in the right eye only (see below for description). Anti-DNA antibodies were negative, serum complement titers were persistently below normal with low C3 and C4 levels, and a fluorescent antinuclear antibody test was weakly positive. The P T T value was markedly elevated at 84 seconds, and there were reduced levels of factors VIII, IX, and XI, all of which were due to the presence of a lupus anticoagulant. Anticardiolipin antibodies were not tested. During her final admission in 1982, vasculitic lesions were noted in the fingertips, as well as splinter hemorrhages. Right heart failure and renal'failure were present. She was admitted to the intensive care unit, became comatose, and failed to respond to brief courses of high-dose (1 g pulse) steroids or cyclophosphamide therapy. Blood cultures were persistently negative. She died 1 month later. At autopsy, there were widespread areas of in-
farction involving the cerebral hemispheres. In the cerebellum, there was widespread vascular occlusion by bland embolic material, which showed various degrees of organization, and widespread in situ thrombosis in small cerebral vessels. Multiple splenic infarcts were noted, and gangrenous changes were present in the right foot. No signs of acute vasculitis or thickening of vessels was noted in the microscopic sections of blood vessels, except in the renal glomerulus, where there was slight thickening in the basement membrane without immunoglobulin or complement deposition. The lesions in the kidney were thought to be consistent with an embolic process. All four cardiac valves contained vegetations. These were not believed to be typical of LibmanSacks vegetations because of their large size and the fact that they did not extend around the ventricular surface of the mitral valve. The largest vegetations
March 1991 The American Journal of Medicine Volume 90
301
CEREBRAL ISCHEMIA AND LUPUS ANTICOAGULANT / POPE ET AL TABLE II Clinical Thromboembolic Events Cerebral Events Patient
Ageand Sex
Stroke or TIA
1 2 3§ 4 5 6 7 8 9 10 11§ 12§ 13 14
38F 30F 39F 40M 44F 36F 38M 53F 22M 38F 51F 17M 47F 40F
O* TIA CVA CVA TIA 0 0 CVA CVA CVA* CVA CVA CVA' CVS*
Amaurosis
Migraine
DVT
Raynaud's
+ + Jr + + + + + 0 + + 0 + +
+ + 0 + + + 0 + 0 + + 0 + 0
0 0 + 0 0 0 + 0 0 0 0 -t+ 0
+ + + 0 0 + 0 + 0 + + 0 0 0
Recurrent Abortions 1/4t 0 NAt NA 3 5 NA 5/8 NA NA 0 NA 4/4 3/3
TIA = transient ischemic attack; DVT = deepvenousthrombosis; CVA = cerebrovascularaccident. * Patient presentedwith seizures. Indicates one miscarriage and three live births. NA = not applicable(male or no attempts at pregnancy). § Four or more ARAcriteria for SLE.
TABLE III Abnormal Findings on Clinical Examination and Investigation
Patient
Livedo Reticularis
Hypertension
Cardiac Murmur
Retinal Abnormality
Abnormal Echo
1
Jr
+
+
0
2 3
0 Jr
0 Jr
+ +
0 0
MR MR MR
4
0
0
+
0
MVP
5 6 7 8 9 10 11 12 13 14
+ + + + 0 + -I+ + +
+ 0 + + + 0 0 + + +
+ + + + + + + + + +
+ + + 0 + 0 0 0 + 0
MT AI/AS MT MR MVP/MR MT/MR AI MR* N MR
Abnormal CT or MRI NI 0 INF
0 INFS N INFS INFS INFS INFS INFS INFS INFS INFS
Abnormal Cerebral Angiogram NP NP + NP N N NP + + + N NP NP +
CT = computed tomographic scan; MRI = magnetic resonance imagingscan; Echo = echocardiogram; + = present; 0 = absent; MR = mitral regurgitation; MVP = mitral valve prolapse; MT = mitral vane thickening;AS = aortic stenosis;AI = aortic insufficiency; NI = noninterpretableabnormality; INF = infarct; INFS= multiple infarcts; NP = not performed; N = normal. • Cardiaccatheterization revealedmild mitral stenosisand aortic insufficiency.
were on the mitral and aortic valves. The mitral valve was stenotic and friable vegetations were present along the closure line on the atrialsurface measuring up to 0.8 c m in diameter. At the aortic valve, there was thickening of the cusps and fusion of the anterior commissure measuring 0.6 × 0.4 cm and vegetations measuring up to 0.6 c m in diameter on the posterior and left cusps. Microscopically, the valve lesions showed sclerosis fibrin and platelet vegetations with very little evidence of inflammation and no bacterial or fungal colonies. In the cerebral cortex, no evidence of vasculitis was seen in or near the areas of cerebral infarction. Most vessels showed areas of thromboembolic material without infarctions. Small arterial vessels in the sclera and choroid contained granular thromboembolic material that, in most cases, showed some evidence of 302
organization. The vessels themselves were unremarkable. A small vessel in the left eye (ganglion cell layer of the retina) was partially occluded by thromboembolic material showing early endothelialization.Thus, the autopsy findings were not considered to show evidence of SLE, but did demonstrate nonbacterial thrombotic endocarditis with large friable vegetations occurring in the setting of valves previously damaged by rheumatic heart disease. Cerebral infarcts were widespread and of varying ages, related to the presence of widespread thromboembolic lesions. S u m m a r y of C l i n i c a l Procedures
Features and Investigative
Tables II and III summarize and compare the clinical and some investigative procedures identi-
March 1991 The American Journal of Medicine Volume 90
CEREBRAL ISCHEMIA AND LUPUS ANTICOAGULANT / POPE ET AL
TABLEIV
Laboratory Findings P'I'I"
Anti-PPL Antibody
Thrombo.
Coombs'
VDRL
Patient
Abnormal
Positive
cytopenia
Positive
Positive
1 2 3 4 5t 6~ 7t 8 9t I0 1H 12t~ 13t 14
+ + -t0" + + + + + + + + + +
+ 0 + 0 + + NT + NT + + + NT +
+ + -I0 + + + 0 + 0 + + + 0
+ + + 0 0 + + + + 0 + + + +
NT + 0 0 + NT NT + + 0 + + + +
Anti-DNA (%)
