Faculty Bibliography

OBJECTIVE. Neuropsychiatric manifestations are common in patients with systemic lupus erythematosus (SLE), but accurate diagnosis is often difficult. We conducted a prospective study to determine the utility of neurometric quantitative electroencephalography (QEEG) as an indicator of cerebral dysfunction in SLE patients. METHODS. Fifty-two SLE patients were divided into 4 groups based on the results of neuropsychiatric evaluations. These included patients with objective evidence of neuropsychiatric SLE (NPSLE), patients with neuropsychiatric symptoms, patients with no evidence of NPSLE, and patients with a prior history of NPSLE. All QEEG findings were compared with data in an age-regressed normative database and with findings in an independent sample of normal subjects. RESULTS. QEEG sensitivity was 87%, and specificity was 75%. QEEG results were abnormal in 74% of the SLE patients with neuropsychiatric symptoms and in 28% of the patients with no evidence of active NPSLE. QEEG profiles varied as a function of the severity and type of neuropsychiatric manifestation present. Within this patient population, QEEG was more sensitive than magnetic resonance imaging, computed tomography scanning, or conventional EEG. CONCLUSION. Neurometric QEEG may be a sensitive indicator of cerebral dysfunction in patients with NPSLE and can differentiate patients with diverse neuropsychiatric manifestations. When combined with a careful clinical history and evaluation, QEEG provides information that may be useful for the early detection of NPSLE and for serial evaluation of disease activity and treatment efficacy

OBJECTIVE. To determine whether increased levels of the complement split products generated in the activation of the alternative or classical pathway accompany more severe disease activity in patients with systemic lupus erythematosus (SLE) and whether these measurements are useful in predicting flares of disease. METHODS. Levels of Ba, Bb, SC5b-9, and C4d were measured in 380 plasma samples obtained from 86 SLE patients who were prospectively followed up for 15 months. RESULTS. In the 20 patients who had inactive disease at the initiation of the study, the mean values of all of the complement split products at entry were within the normal range. In the 47 patients with stable or moderate disease activity, levels of Ba were significantly increased, while the mean values for Bb, SC5b-9, and C4d did not differ significantly from those in patients with inactive disease. The mean entry value of each analyte was highest in the group of 19 patients who had the most severe disease activity at initial evaluation. Traditional measurements of complement, i.e., C3, C4, and CH50, followed similar trends, but did not discriminate between the 3 groups of patients as well as did measurements of the split products. Analysis of the disease course in the patients with inactive or stable/moderate disease revealed that an elevated level of C4d had the most sensitivity with regard to subsequent flare, while an elevated Bb level had the highest specificity and the greatest predictive value. CONCLUSION. These data suggest that elevated levels of complement split products, particularly products of alternative and terminal pathway activation, more accurately reflect disease activity than do conventional measurements of complement in SLE and may be useful in the prediction of impending disease flares

OBJECTIVE: To determine the frequency of unexplained reversible hypoxemia in patients with systemic lupus erythematosus and to assess the relation between hypoxemia and elevated plasma levels of complement split products. DESIGN: Cohort study. SETTING: Inpatient and outpatient facilities of the New York University Medical Center/Bellevue Hospital and the Hospital for Joint Diseases. PATIENTS: Case patients were 22 patients hospitalized with disease exacerbation and no evidence of parenchymal lung disease on chest roentgenogram. Four patients with stable disease were followed in the outpatient clinic, and five healthy normal volunteers served as controls. MEASUREMENTS: Plasma levels of complement split products (C3a, factor Bb fragment), alveolar-arterial (A-a) Po2 gradients, and pulmonary function were measured. MAIN RESULTS: Nine episodes of hypoxemia or hypocapnia (mean A-a gradient, 30.4 +/- 4.8 mm Hg) or both (despite normal chest roentgenogram results) were noted in six hospitalized patients (group 1). Gas exchange improved within 72 hours of steroid therapy (mean A-a gradient, 11.6 +/- 4.3 mm Hg; P less than 0.01). These patients had an elevated initial mean C3a level (938.4 +/- 246.8 ng/mL) that decreased within 72 hours (407.8 +/- 80.9 ng/mL; P less than 0.01), concomitant with improved oxygenation. Ventilation-perfusion scans, obtained for four of six group 1 patients, excluded pulmonary emboli. Four hospitalized patients (group 2) had a normal A-a gradient (mean, 7.5 +/- 2.7 mm Hg). The mean C3a level of this group (358.3 +/- 39.2 ng/mL) was lower than that of group 1 (P less than 0.05). Four patients with stable disease (group 3) had a mean A-a gradient and a mean C3a level of 3.3 +/- 2.7 mm Hg and 237.8 +/- 105.7 ng/mL, respectively, similar to values found in five normal volunteers, in whom the mean A-a gradient was 3.7 +/- 1.7 mm Hg and the mean C3a level was 124.8 +/- 9.2 ng/mL. CONCLUSION: A syndrome of reversible hypoxemia, unassociated with parenchymal lung disease, is unexpectedly common in acutely ill, hospitalized patients with systemic lupus erythematosus. The pathogenesis of this syndrome is unclear, although the data are compatible with the hypothesis that hypoxemia may be related to pulmonary leukoaggregation

We measured levels of complement anaphylatoxin split products, C3a and C5a, in the circulation of patients with systemic lupus erythematosus (SLE). In 23 SLE patients who were followed serially, the mean C3a value was 179 ng/ml during stable disease and 550 ng/ml during a disease flare. In 10 patients, C3a levels predicted disease activity, with the C3a value rising from a mean of 183 ng/ml at a time of stable disease to a mean of 242 ng/ml 1-2 months prior to a clinical exacerbation of disease. The mean C3a level in 5 patients with acute dysfunction of the central nervous system (CNS) was 1,297 ng/ml, which is significantly higher than that observed in patients with active disease but without CNS involvement (P less than 0.01). C5a levels were also significantly elevated in 4 patients with acute CNS disease. Pathologic specimens from 2 patients who died during an acute lupus flare revealed neutrophils occluding the cerebral and intestinal vessels. Fluorescein angiography in a patient with CNS lupus revealed vasoocclusive retinopathy. In 5 of 7 SLE patients who were pregnant, C3a levels were elevated, with a group mean value of 310 ng/ml. There was a negative correlation (r = -0.59) between C3a and C3 levels in pregnant patients with SLE, and this finding is consistent with complement activation as the cause of decreasing C3 levels. We suggest that serial measurements of C3a can predict flares of disease in lupus patients and can demonstrate complement activation during pregnancy in women with SLE. In addition, release of C3a and C5a (mediators of inflammation) into the circulation may elicit vascular injury, particularly in patients with lupus cerebritis

The deposition of immune complexes within blood vessel walls results in the potential for complement activation and the release of chemotactic factors, such as fragments of C5 (C5fr). The generation of C5fr results in the intravascular aggregation of neutrophils with subsequent leukostatic occlusion of the pulmonary arterioles. The generation of C5fr may contribute to the pathogenesis of adult respiratory distress syndrome and other diseases. Studies were undertaken to determine the role of circulating complement derived peptides and intravascular neutrophil activation in systemic lupus erythematosus

To determine whether activated complement components appear in the circulation of patients with systemic lupus erythematosus (SLE), we measured C5a and C3a by radioimmunoassay. Mean C5a concentration in the plasma of acutely ill SLE patients was 46.0 ng/ml, compared with 17.1 ng/ml in normal controls (P less than 0.01). Mean C3a concentration in patients with severe disease was 526 ng/ml, compared with 134 ng/ml in controls (P less than 0.01). In patients with moderately active SLE, the mean C3a concentration, but not the mean C5a concentration, was also elevated. In addition, C3a was elevated in 15 or 21 patients with active SLE, whereas low levels of C3 or C4 were noted in only 7 of these 21 patients. We conclude that the measurement of complement-derived anaphylatoxins may be useful in the management of patients with SLE. In addition, we suggest that these circulating mediators may contribute to the pathogenesis of vascular injury in patients with the disease