Background Since the N-methyl-D-aspartate receptors (NMDARs) are most densely populated in the hippocampi and amygdala and are involved in mediating learning and working memory, the discordant activities in the hippocampus and amygdala in lupus patients we previously found on functional magnetic resonance imaging (fMRI) of the brain are postulated to be related to the anti-NR2 antibodies (1). Additionally, because our previous fMRI studies provided objective evidence of brain involvement in subclinical impairment in learning and memory, fMRI brain signals may offer a reasonably sensitive surrogate to examine if serum anti-NR2 antibodies are clinically useful to detect subclinical cognitive impairment in patients with SLE. Currently, the ability of serum anti-NR2 antibodies in detecting and predicting subclinical cognitive dysfunction is not well defined.
Objectives To evaluate if serum level of anti-NR2 antibodies in SLE patients is higher than that of healthy individuals and whether the antibodies correlate with hippocampal-amygdala activities in lupus patients assessed by real-time fMRI of the brain.
Methods Sixty-four SLE patients and 67 age- and sex-matched healthy controls were studied, 12 of the 64 SLE patients without neuropsychiatric manifestation underwent a computer-based neuropsychological test while their real-time brain activities were concomitantly recorded by a 1.5-Tesla fMRI scanner. Serum anti-NR2 antibodies were measured by an indirect ELISA using SVSYDDWDYSLEARV peptide as the substrate. Serum anti-NR2 levels were compared between SLE patients and healthy controls and correlated with clinical and serological features of SLE and real-time fMRI brain signals.
Results In the lupus group, the mean±SD serum C3, C4 and anti-dsDNA levels and SLEDAI were 77.86 ± 31.0 mg/dl, 17.05 ± 11.5 mg/dl, 101.46 ± 85.9 IU and 6.70 ± 5.7 respectively. Among the 64 SLE patients, 6 had history of neuropsychiatric features (4 had psychosis, 1 had seizures and 1 had stroke). The 12 lupus patients who underwent fMRI scans had no significant difference in terms of demographics, clinical and serological characteristics when compared with the rest of the 52 SLE patients. SLE patients had significantly higher serum levels of anti-NR2 antibodies than the healthy controls (OD450 value: 0.147 ± 0.095 vs. 0.106 ± 0.080, P = 0.001). Using two times of the SD above the mean anti-NR2 level of the healthy controls as the cut-off value, 6 (9.4%) SLE patients had high anti-NR2 levels but none of them had clinically-overt neuropsychiatric manifestation. No significant correlation was found between serum anti-NR2 antibody and serum C3, C4 and anti-dsDNA levels and SLEDAI. While abnormal fMRI brain signals were detected in the hippocampus and amygdala in the 12 SLE patients, no significant correlation was found between serum anti-NR2 antibodies and the abnormal functional hippocampal-amygdala fMRI signals.
Conclusions Although SLE patients had significantly higher serum anti-NR2 levels than healthy controls, serum anti-NR2 antibody level did not appear to be an ideal marker to detect subclinical impairment in learning and memory.
Ren T, Ho RC, Mak A. Dysfunctional cortico-basal ganglia-thalamic circuit and altered hippocampal-amygdala activity on cognitive set-shifting in non-neuropsychiatric systemic lupus erythematosus. Arthritis Rheum 2012;64:4048-59
Disclosure of Interest None Declared