ARTHRITIS & RHEUMATISM
Vol. 48, No. 10, October 2003, pp 2881–2887
DOI 10.1002/art.11279
© 2003, American College of Rheumatology
Neuronal and Astrocytic Damage in
Systemic Lupus Erythematosus Patients With
Central Nervous System Involvement
Estelle Trysberg, Karin Nylen, Lars E. Rosengren, and Andrej Tarkowski
Objective. Symptoms originating from the central
nervous system (CNS) frequently occur in patients with
systemic lupus erythematosus (SLE). CNS involvement
in lupus is associated with increased morbidity and
mortality. Currently, reliable markers for activity in
this condition are absent. The goal of this study was to
determine the level of the light subunit of the neurofila-
ment triplet protein (NFL) and that of glial fibrillary
acidic protein (GFAP) in the cerebrospinal fluid of SLE
patients with clinically verified CNS involvement and
compare them with the levels in SLE patients without
CNS involvement.
Methods. We assessed cerebrospinal fluid ob-
tained from 99 patients with SLE and 99 age-matched
controls for the presence of soluble molecules indicating
neuronal destruction and astrogliosis—NFL and GFAP,
respectively. Patients were evaluated clinically, with
magnetic resonance imaging (MRI) of the brain, cere-
brospinal fluid analyses, and neuropsychiatric tests.
Results. In the group of lupus patients with CNS
involvement, intrathecal levels of NFL and GFAP were
increased an average of 7-fold (P < 0.0001) and 3-fold
(P < 0.05), respectively, compared with the levels in
SLE patients without overt CNS disease. Intrathecal
levels of NFL correlated significantly with cerebrospinal
fluid levels of interleukin-6 (IL-6) (P < 0.005), IL-8 (P
< 0.005), pleocytosis (P < 0.05), the albumin ratio (P <
0.0005), and the presence of oligoclonal IgG bands (P <
0.005). Cerebrospinal fluid levels of both NFL and
GFAP also showed a significant correlation with MRI
abnormalities ( P < 0.001). Successful cyclo-
phosphamide treatment of CNS lupus resulted in sig-
nificantly decreased levels of both proteins; levels of
GFAP reached those observed in healthy subjects.
Conclusion. This study is the first to show bio-
chemical signs of neuronal and astrocytic damage in
patients with neuropsychiatric lupus. It is suggested
that biochemical markers of brain damage should be
used as a followup tool in this patient group.
Central nervous system (CNS) involvement has
been reported to occur in 14–75% of patients with
systemic lupus erythematosus (SLE) (1–3). The large
differences in frequency relate to the diagnostic criteria
applied. CNS lupus can occur at any time during the
course of SLE, and the symptoms are extremely diverse
(4). The features of this condition may include seizures,
stroke, depression, psychosis, and disordered mentation.
Minor symptoms such as headache, poor concentration,
and mood swings may be attributable to other intercur-
rent conditions, and differentiating the origin of such
symptoms may be difficult. In addition, other diseases
capable of causing CNS symptoms, such as infections,
severe hypertension, metabolic derangements (associat-
ed, for instance, with renal failure), or drug toxicity may
coexist in patients with SLE (5). Without treatment,
CNS lupus is one of the major causes of morbidity and
mortality in SLE. In recent years, CNS lupus has been
treated with cytotoxic drugs, which improve the disease
outcome (6). The availability of beneficial treatment
remedies increases the need for early recognition of
CNS manifestations in lupus, as well as an evaluation of
local (i.e., intrathecal) responses to lupus medication.
Supported by the Go ¨teborg Medical Society, Swedish Asso-
ciation Against Rheumatism, King Gustaf V Foundation, Swedish
Medical Research Council, Nanna Svartz’ Foundation, Bo ¨rje Dahlin’s
Foundation, Swedish National Inflammation Network, Swedish Na-
tional Infection and Vaccination Network, AME Wolff Foundation,
and the University of Go ¨teborg.
Estelle Trysberg, MD, Karin Nylen, MD, Lars E. Rosengren,
MD, PhD, Andrej Tarkowski, MD, PhD: University of Go ¨teborg,
Go ¨teborg, Sweden.
Address correspondence and reprint requests to Estelle Trys-
berg, MD, Department of Rheumatology and Inflammation Research,
University of Go ¨teborg, Guldhedsgatan 10, S-413 46 Go ¨teborg, Swe-
den. E-mail: estelle@immuno.gu.se.
Submitted for publication November 14, 2002; accepted in
revised form June 6, 2003.
2881