Case Series: Neuropsychiatric Symptoms With Pediatric

Systemic Lupus Erythematosus

SUSAN BECKWITT TURKEL, M.D., JOHN H. MILLER, M.D., AND ANDREAS REIFF, M.D.

ABSTRACT
Objective: To describe the psychiatric presentation, serologic findings, and neuroimaging patterns in children and adoles-
cents with central nervous system involvement with systemic lupus erythematosus (CNS-SLE). Method: Pediatric patients
with psychiatric symptoms who fulfilled the 1997 revised diagnostic American College of Rheumatology criteria for SLE
were studied. Complement levels, and anti-double-stranded DNA, anti-Smith, anti-phospholipid, and anti-neuronal anti-
bodies were evaluated. Computed tomography, magnetic resonance imaging, and single photon emission computed
tomography (SPECT) neuroimaging studies were reviewed. Results: The 10 patients with CNS-SLE presented with psy-
chosis, mood disturbance, or confusion, and 8 patients had concomitant neurologic symptoms.The 8 girls and 2 boys ranged
in age from 7.5 to 17 years. Serum anti-neuronal antibodies were positive with onset of symptoms and declined with improve-
ment. Initial SPECT was abnormal in all 10 patients and remained abnormal. Conclusion: SPECT and anti-neuronal anti-
bodies help confirm CNS involvement in patients with SLE and neuropsychiatric symptoms. J. Am. Acad. Child Adolesc.
Psychiatry, 2001, 40(4):482–485. Key Words: systemic lupus erythematosus, neuropsychiatric symptoms.

Systemic lupus erythematosus (SLE) is an autoimmune METHOD

disorder with variable presentation and course. The We reviewed the clinical, serologic, and neuroimaging findings in
American College of Rheumatology 1982 and 1997 10 pediatric SLE patients who presented with psychiatric symptoms
to the Psychiatric Consultation-Liaison service and the Division of
revised criteria for SLE include “neurologic disorder” as Rheumatology at Childrens Hospital Los Angeles from January 1992
one of the diagnostic criteria for SLE and define it by the through July 1997. All children fulfilled the 1982 revised criteria for
presence of either seizures or psychosis (Hochberg, 1997; classification of SLE and the 1997 update of these criteria (Hochberg,
Tan et al., 1982). For the most recent definition of neu- 1997; Tan et al., 1982).
All patients underwent thorough evaluations, which included phys-
ropsychiatric lupus, a multidisciplinary committee rec- ical and psychiatric examinations. The patients were evaluated by elec-
ommended using DSM-IV criteria to diagnose psychosis, troencephalography (EEG) and neuroimaging, routine hematologic
mania, depression, and delirium (American College of and chemistry studies, serum serology, lumbar puncture, and routine
cerebrospinal fluid studies.
Rheumatology Committee, 1999). Multiple serologic tests included anti-double-stranded DNA (anti-
This study was undertaken to describe psychiatric ds) antibody by a modified Farr assay, anti-Smith (anti-Sm) antibody
symptoms in children and adolescents with acute SLE, to using ribonuclease-treated extractable nuclear antigen, complement C3
and C4 levels by radial immunoelectrophoresis, anti-phospholipid
determine whether there is any clinically relevant associa- (APL) antibodies by IgG anti-cardiolipin antibodies using enzyme-
tion between the initial psychiatric presentation, sero- linked immunoadsorbent assay, and anti-neuronal antibodies (ANAB)
logic findings, and neuroimaging patterns. The findings by flow cytometry using human neuroblastoma cell line SK-N-SH
could assist psychiatrists and rheumatologists in making (obtained from American Type Culture Collection, Rockville, MD).
Neuroimaging studies included computed tomography (CT), mag-
the diagnosis rapidly and providing appropriate therapy. netic resonance imaging (MRI), and single photon emission computed
tomography (SPECT). CT scans were done in five patients, and MRI
Accepted September 22, 2000. studies in nine. SPECT scans were performed on all the children in the
From Childrens Hospital Los Angeles and the University of Southern study using either intravenous injection of I-123-labeled iodoamphe-
California Keck School of Medicine, where Dr. Turkel is with the Department of tamine (IMP 0.100 mCi/kg) or 99m-technicium coupled to demethy-
Psychiatry, Dr. Miller is with the Department of Radiology, and Dr. Reiff is lated propylene-oxide (Tc99 HMPAO, 0.280 mCi/kg). The same
with the Division of Rheumatology of the Department of Pediatrics. isotope was used serially for each patient. Both tracers produced com-
Correspondence to Dr. Turkel, Childrens Hospital Los Angeles, 4650 Sunset parable results. An LFOV digital gamma camera was used for SPECT
Blvd. #82, Los Angeles, CA 90027. acquisition. A parallel-hole, low-energy, all-purpose LEAP collimator
0890-8567/01/4004-0482䉷2001 by the American Academy of Child was applied for IMP, and a high-resolution, low-energy collimator was
and Adolescent Psychiatry. applied for HMPAO. A 64 ⫻ 64 image matrix was used for both

482 J . A M . A C A D . C H I L D A D O L E S C . P S YC H I AT RY, 4 0 : 4 , A P R I L 2 0 0 1
 NEUROPSYCHIATRIC SYMPTOMS IN CNS-SLE

agents. All SPECT scans were interpreted by an experienced nuclear eight performed, including in those patients with clinical
radiologist (J.H.M.) using visual grading and without knowledge of delirium.
the patients’ clinical symptoms. Normal findings were defined as
homogenous brain perfusion without focal uptake defects or visible At the time of the initial CNS event, all nine children
asymmetry. Perfusion deficits were defined as at least two color steps tested were positive for serum ANAB. Determination of
difference compared with the unaffected contralateral hemisphere. ANAB in CSF was positive in two and negative in two
patients. Three patients were positive for APL antibodies
RESULTS and four for anti-Sm antibodies. Serum C3 and C4 com-
There were eight girls and two boys in the study group, plement levels were depressed in seven patients.
who ranged in age from 7.5 to 17 years. There were five During the treatment phase of CNS disease, comple-
Hispanic, two Asian, two African-American, and one ment levels markedly improved or normalized concomi-
Caucasian child in the group. There were no previous or tant with improvement in anti-Sm and anti-dsDNA
family histories of mental illness in any of these patients, antibodies, and serum ANAB resolved.
although one mother had a history of substance abuse. CT scan revealed abnormalities in one of five done.
Psychiatric consultation was requested because of clini- MRI findings were normal in seven of nine patients and
cally observed changes in behavior, mood, or personality. mildly abnormal in two. The most serious lesion on MRI
Three patients were seen by the psychiatrist because of was a pontine infarct found in a male patient with neuro-
acute onset of psychiatric symptoms before the diagnosis pathy symptoms.
of SLE by the rheumatologist. Six psychiatric consulta- Initial SPECT findings were abnormal in all 10 pa-
tions were requested by the rheumatologist because of tients and remained abnormal on all subsequent SPECT
changes in mental status in patients with known SLE for studies. Repeated SPECT scans were done in all 10 pa-
6 months to 4 years, and one was seen the same day the tients, usually several months after the first, and a slightly
diagnosis of SLE was confirmed. All 10 patients presented improved perfusion was noted in 3, including in 1
with psychiatric symptoms and behavioral changes, patient who continues to have severe cognitive impair-
including withdrawal (2), confusion (4), noncompliance ment. SPECT abnormalities were found in areas that
(2), bizarre and disorganized behavior (3), and abnormal appeared intact when conventional imaging modalities
sleep (4). The majority of the patients were psychotic (7), were used.
with auditory (7), visual (4), or tactile (1) hallucinations Typically multiple small superficial cortical defects
and persecutory or bizarre delusions (4). At psychiatric indicative of reduced or absent cerebral perfusion were
interview, the mood of the patients was described as noted on SPECT in the cerebrum (Table 1). Diffuse rather
depressed (6), apathetic (1), irritable (1), guarded (1), anx- than patchy abnormal perfusion was noted in some, pre-
ious (1), or manic (1); affect was described as labile (2), dominantly over the convexities. Cerebellar involvement
constricted or blunted (5), incongruent (1), or flat (2). was not observed.
The patients exhibited poor attention and concentration. Of the four patients who had positive APL antibodies,
Suicidal ideation was present in half the patients, and one all had normal MRI results and abnormal SPECT results,
patient had made several suicide attempts prior to psychi- and the patient with the pontine infarct on MRI did not
atric evaluation. have detectable APL antibodies.
The psychiatric diagnoses were classified according to
DSM-IV as mental disorder due to general medical con- TABLE 1
Psychiatric Symptoms and SPECT Findings
dition (SLE), with depressed (4) or manic (1) mood, with
psychosis (3), or as delirium (5) when more severe global Age CNS Onset 12.5 16 16 17 16 16 13 11.5 7.5 14
Sex F F M F F M F F F F
encephalopathic dysfunction was present (American
Psychiatric Association, 1994). Neurologic symptoms, Confusion + + + + + + + +
present in 8 patients, included seizures (7) with status epi- Psychosis + + + + + + +
Frontal lobe + + + + + + + + +
lepticus (1), optic neuritis (2) with transverse myelitis (1), Parietal lobe + + + + + + + +
peripheral neuropathy (2), and headache (2). Temporal lobe + + + + +
All patients had a lumbar puncture and analysis of CSF, Convexities + + +
Diffuse + + + + +
which was normal except in two patients with increased
opening pressure. EEG findings were unremarkable in the Note: SPECT = single photon emission computed tomography.

J . A M . A C A D . C H I L D A D O L E S C . P S YC H I AT RY, 4 0 : 4 , A P R I L 2 0 0 1 483
TURKEL ET AL.

Once the diagnosis of CNS-SLE was made, all children changes, and the concomitant appearance of psychiatric
were treated with daily intravenous methylprednisolone symptoms with confusion or seizures led to the suspicion
and then monthly intravenous pulse cyclophosphamide of an underlying medical condition. The psychiatric con-
for at least 6 months. Five children were treated with anti- sultant’s recommendations for a thorough medical eval-
psychotic medications, including haloperidol, risperidone, uation, including serologic studies to investigate the
and trifluoperazine. Dextroamphetamine was used in one possibility of SLE, led to referral to the rheumatologist and
child, and methylphenidate in another to treat inattention collaborative confirmation of the diagnosis of CNS-SLE.
and apathy. One patient was treated with nortriptyline Serologic studies were not as definitive as SPECT in
and then venlafaxine to control depressive symptoms, confirming the diagnosis of CNS-SLE, including dimin-
and another required bupropion later when depression ished serum complement or with the presence of anti-Sm,
recurred. Carbamazepine was effective in controlling APL, or anti-dsDNA antibodies. The activity of ANAB
mood symptoms in two patients, and another with manic seemed to parallel the resolution of clinical CNS symp-
symptoms responded to valproate. toms better than other parameters measured.
Neuropsychiatric symptoms improved or resolved in SPECT is a highly sensitive but not specific test for
most children within 6 months. Several children were left CNS-SLE. SPECT findings in our patients are typical of
with chronic sequelae, including residual neuropathy and reports: generalized diffuse or patchy hypoperfusion or
cognitive problems. Recurrence of psychiatric symptoms multiple small focal perfusion defects mainly in the parie-
was noted in two patients, one with psychosis 2 years after tal and frontal lobes, which is characteristic of conditions
the initial episode and another with depression 5 years later. associated with cerebral vasoconstriction or vasculitis
(Hanly et al., 1992; Lin et al., 1997; Reiff et al., 1997;
DISCUSSION Russo et al., 1998; Sibbit et al., 1999; Szer et al., 1993).
In summary, this study describes the neuropsychiatric
SLE is the prototypal autoimmune disease, with vari-
symptoms in 10 children with active CNS-SLE. The
able clinical features affecting all organ systems. CNS
presence of delirium, psychosis, confusion, depression,
involvement occurs in more than half of children and ado-
or mania in a patient with SLE suggests primary CNS
lescents with SLE, usually early in the course of the disease
involvement with the disease. The diagnosis of CNS-
and usually with both neurologic and psychiatric symp-
toms (Rosenberg, 1994; Silber et al., 1984; Steinlin et al., SLE should be suspected when neuropsychiatric symp-
1995). Approximately 100 children and adolescents with toms occur in patients known to have SLE and should be
psychiatric symptoms associated with CNS-SLE have investigated in children and adolescents with the acute
been reported in the literature in the past 30 years. They onset of delirium or profound psychiatric symptoms
represent about one third of young patients reported in with or without neurologic symptoms. SPECT and
these series; they presented with psychosis (32), depression ANAB may help confirm the diagnosis, and the impact
(25), change in behavior (23), change in personality (1), of pediatric CNS-SLE merits further study.
and delirium (7) (Fish et al., 1977; How et al., 1985;
Meislin and Rothfield, 1968; Parikh et al., 1995; Press REFERENCES
et al., 1996; Szer et al., 1993; Yancey et al., 1981). American College of Rheumatology Ad Hoc Committee on Neuropsychiatric
Psychosis is generally considered a primary manifesta- Lupus Nomenclature (1999), The American College of Rheumatology
nomenclature and case definitions for neuropsychiatric lupus syndromes.
tion of SLE and is part of the criteria for “neurologic dis- Arthritis Rheum 42:599–608
order.” Mood, cognitive, or behavior changes may also American Psychiatric Association (1994), Diagnostic and Statistical Manual of
Mental Disorders, 4th edition (DSM-IV). Washington, DC: American Psy-
indicate active CNS disease (Denburg et al., 1997; Utset chiatric Association
et al., 1994; Wyckoff et al., 1995). Eight of our 10 patients Denberg SD, Carbotte RM, Denburg JA (1997), Psychological aspects of sys-
temic lupus erythematosus: cognitive function, mood, and self report. J
had neurologic as well as psychiatric symptoms, which is Rheumatol 24:998–1003
typical of previous reports (Parikh et al., 1995). Fish AJ, Blau EB, Westberg NG, Burke BA, Verner RL, Michael AF (1977),
Our study reinforces the classical psychiatric precept Systemic lupus erythematosus within the first two decades of life. Am J
Med:62:99–117
that a primary medical condition should be considered Hanly JG, Walsh NMG, Sangalang V (1992), Brain pathology in systemic
first with any acute psychiatric presentation, not only with lupus erythematosus. J Rheumatol 19:732–741
Hochberg MC (1997), Updating the American College of Rheumatology
acute delirium. Three of our 10 patients were evaluated revised criteria for the classification of systemic lupus erythematosus.
first by the psychiatrist because of mood or behavioral Arthritis Rheum 40:1725

484 J . A M . A C A D . C H I L D A D O L E S C . P S YC H I AT RY, 4 0 : 4 , A P R I L 2 0 0 1
 NEUROPSYCHIATRIC SYMPTOMS IN CNS-SLE

How A, Dent PB, Liao S-K, Denberg JA (1985), Antineuronal antibodies in Sibbit WL, Sibbit RR, Brooks WM (1999), Neuroimaging in neuropsychiat-
neuropsychiatric systemic lupus erythematosus. Arthritis Rheum 28:789–795 ric systemic lupus erythematosus. Arthritis Rheum 42:2026–2038
Lin WY, Lan JL, Yen TC, Wang SJ (1997), Technetium-99m-HMPAO brain Silber TJ, Chatoor I, White PH (1984), Psychiatric manifestations of SLE in
SPECT in systemic lupus erythematosus with CNS involvement. J Nucl children and adolescents. Clin Pediatr (Phila) 23:331–335
Med 38:1112–1115 Steinlin MI, Blaser SI, Gilday DL et al. (1995), Neurologic manifestations of
Meislin AG, Rothfield N (1968), Systemic lupus erythematosus in childhood: pediatric systemic lupus erythematosus. Pediatr Neurol 13:191–197
analysis of 42 cases, with comparative data on 200 adult cases followed Szer IS, Miller JH, Rawlings D, Shaham B, Bernstein B (1993), Central per-
concurrently. Pediatrics 42:37–49 fusion abnormalities in children with central nervous system manifesta-
Parikh S, Swaiman KF, Kim Y (1995), Neurologic characteristics of child- tions of lupus detected by single photon emission computed tomography.
hood lupus erythematosus. Pediatr Neurol 13:198–201 J Rheumatol 20:2143–2148
Press J, Palayew K, Laxer RM et al. (1996), Antiribosomal P antibodies in Tan EM, Cohen AS, Fries JF et al. (1982), The 1982 revised criteria for the
pediatric patients with systemic lupus erythematosus and psychosis. classification of SLE. Arthritis Rheum 25:1275
Arthritis Rheum 39:671–676 Utset TO, Golden M, Siberry G, Kiri N, Crum RM, Petri M (1994), Depressive
Reiff A, Miller J, Shaham B, Bernstein B, Szer IS (1997), Childhood central symptoms in patients with systemic lupus erythematosus: association with cen-
nervous system lupus: longitudinal assessment using single photon emis- tral nervous system lupus and Sjogren’s syndrome. J Rheumatol 21:2039–45
sion computed tomography. J Rheumatol 24:2461–2465 Wyckoff PM, Miller LC, Tucker LB, Schaller JG (1995), Neuropsychological
Rosenberg AM (1994), Systemic lupus erythematosus in children. Semin assessment of children and adolescents with systemic lupus erythemato-
Immunopathol 16:261–279 sus. Lupus 4:217–220
Russo R, Gilday D, Laxer RM, Eddy A, Silverman ED (1998), Single photon Yancey CL, Doughty RA, Athreya BH (1981), Central nervous system involve-
emission computed tomography scanning in childhood systemic lupus ment in childhood systemic lupus erythematosus. Arthritis Rheum
erythematosus. J Rheumatol 25:576–582 24:1389–1395

J . A M . A C A D . C H I L D A D O L E S C . P S YC H I AT RY, 4 0 : 4 , A P R I L 2 0 0 1 485