Chapter 37 The Nervous System in Systemic Lupus Erythematosus
Although neurological (N) and psychiatric (P) events are frequently encountered in patients with systemic lupus erythematosus (SLE), they remain one of the more perplexing aspects of the disease and pose diagnostic and therapeutic challenges for clinicians. The difficulties are due in part to the lack of specificity of the majority of NP manifestations, uncertainty regarding the pathogenic mechanisms, and a paucity of controlled data to support therapeutic strategies.
The ACR classification criteria for SLE include NP manifestations, seizures, and psychosis. However, it is widely acknowledged that a much broader range of NP disease manifestations occur in SLE patients. With a preference for affecting the central nervous system, individual NP events may reflect either a diffuse disease process (e.g., psychosis and depression) or focal process (e.g., stroke and transverse myelitis), depending on the anatomic location of pathology. Several classifications have been developed for NP-SLE,1–3 but until recently most have lacked definitions for individual manifestations and the approach to investigation and diagnosis has been inconsistent.
In 1999, the ACR research committee produced a standard nomenclature and diagnostic criteria for 19 NP syndromes known to occur in SLE patients4 (Table 37.1). For each of the 19 NP syndromes, potential etiologies other than SLE are identified either for exclusion or recognition as an “association” (acknowledging that in some clinical presentations definitive attribution is not possible). The identification of other non-lupus causes for NP events in SLE patients is of critical importance and has not been adequately addressed in previous classification systems. Guidelines for reporting NP events were also developed by the ACR research committee, and specific diagnostic tests were recommended for each syndrome. Although these criteria were developed primarily to facilitate research studies of NP-SLE, they also provide a practical guide to the assessment of individual SLE patients with NP disease.
|Central Nervous System||Peripheral Nervous System|
|Guillain Barré syndrome|
From The American College of Rheumatology nomenclature and case definitions for neuropsychiatric lupus syndromes. Arthritis Rheum 1999;42(4):599–608, with permission.
In a representative selection of studies utilizing the ACR nomenclature and definitions,5–9 the overall prevalence of NP disease has varied between 37 and 95%. The most common of the 19 NP syndromes in each of these five SLE cohorts were cognitive dysfunction (55 to 80%), headache (24 to 72%), mood disorder (14 to 57%), cerebrovascular disease (5 to 18%), seizures (6 to 51%), polyneuropathy (3 to 28%), anxiety (7 to 24%), and psychosis (0 to 8%). Most of the other NP syndromes were infrequent, with a prevalence of less than 1% in most studies, emphasizing the rarity of many of these entities.
The attribution of individual NP events to SLE per se or to an alternative etiology remains a challenge. In the absence of a diagnostic gold standard for most of the NP-SLE syndromes, attribution is determined case by case on the basis of exclusion using the best available clinical, laboratory, and imaging data. The ACR NP-SLE classification4 provides a basis for addressing this issue in a systematic manner because for each NP syndrome there is a comprehensive list of exclusions and associations, the presence of which may indicate an etiology other than SLE.
Utilizing this approach and taking into consideration the temporal relationship between the NP event and the diagnosis of SLE, a recent study has reported that up to 41% of all NP events in SLE patients may be attributed to factors other than lupus.7 This finding is in keeping with data from a Finnish study5 that concluded that five NP syndromes (headache, anxiety, mild depression, mild cognitive impairment, and polyneuropathy, without electrophysiologic confirmation), should not be considered primary manifestations of the disease.
Although the diversity of NP manifestations reported in SLE patients makes it unlikely that there is a single pathogenic mechanism, there are at least three primary immunopathogenic mechanisms implicated in NP-SLE: vasculopathy of predominantly small intracranial blood vessels, autoantibody production, and the generation of inflammatory mediators (Fig. 37.1).
(Modified from Hanly JG. Neuropsychiatric lupus. Current Rheumatology Reports 2001;3:205-212.)
In a limited number of neuropathological studies,10–13 the predominant finding was a bland noninflammatory vasculopathy. In contrast, inflammatory disease of small or large vessels was rare. Brain microinfarcts occurred in close anatomical proximity to the microangiopathy.10 Although instructive, the majority of neuropathologic studies in SLE have significant limitations due to a bias in patient selection, a temporal disconnect between the NP event and tissue sampling, and the potential impact of confounding factors such as infection, hypertension, and corticosteroids on neuropathology.
A humoral immune response directed against neuronal antigens, ribosomes, and phospholipid-associated proteins has been implicated in the pathogenesis of NP-SLE. The data from human studies on antineuronal antibodies is largely circumstantial. This includes the temporal relationship between clinical events and serologic findings,14 the presence of autoantibodies in the cerebrospinal fluid,15 and to a very limited extent their identification in neuronal tissues from patients succumbing to the disease.16 Autoantibodies gain access to the CSF of SLE patients by means of passive transfer from the circulation through a permeabilized blood/brain barrier17,18 and independently by direct intrathecal production.14,17
The fine specificity of antineuronal antibodies has been studied extensively but in general has not resulted in greater diagnostic specificity. Most recently, attention has been focused on anti-NR2 glutamate receptor antibodies as a potentially novel system that could explain some of the complexities of NP-SLE. Although of considerable interest, the findings to date are largely derived from animal studies and require confirmation in human subjects with NP-SLE. The limited studies in human lupus examining the association between this subset of antineuronal antibodies and cognitive impairment have yielded conflicting results.19,20
Anti-ribosomal P (anti-P) antibodies were first described in SLE patients in 1985 and are quite specific for SLE, with a prevalence of 13 to 20% depending on ethnic group.21 In 1987, these autoantibodies were first linked to NP-SLE, in particular psychosis.22 Subsequent work either supported, refuted, or extended this initial observation to include depression.21,23–25 Potential explanations for the differences in study outcomes include variability in diagnostic criteria for psychiatric disease, variance in the temporal relationship between clinical events and serologic testing, and differences in assay technique (particularly antigen preparation and purity).
Autoimmune antiphospholipid antibodies, which are directed against phospholipid-binding proteins such as β2-glycoprotein I and prothrombin,26 are associated with predominately focal manifestations of NP-SLE. The most common neurologic disorders are those of vascular origin (such as transient cerebral ischemia or stroke), but other associations include seizures, chorea, transverse myelitis, and cognitive dysfunction.8 In a review of more than 1000 SLE patients, NP manifestations occurred in 38% of patients with lupus anticoagulant compared to 21% of patients without these antiphospholipid antibodies.27 In a study of 118 SLE patients, 33% of whom were positive for the LA,28 there was a significantly greater proportion of cognitive impairment in LA-positive (50%) compared to LA-negative (25%) patients.
The favored pathogenic mechanism for this subset of autoantibodies in NP-SLE is thrombosis within vessels of different caliber and subsequent downstream cerebral ischemia. A procoagulant state may be induced through acquired resistance to protein C and protein S, platelet aggregation, and direct activation of endothelial cells.26 However, the intrathecal production of antiphospholipid antibodies in patients with NP-SLE,17 their association with diffuse cognitive impairment,29,30 and in vitro evidence indicating modulation of neuronal cell function31 raise the possibility of an alternative pathogenic mechanism.
Studies from Japan were the first to report an association between enhanced intracranial production of interleukin (IL)-6 with seizures32 and interferon-alpha with lupus psychosis.33 Subsequent studies have provided further evidence for the intrathecal production of IL-634–37 and have identified other potential candidate cytokines such as IL-10,37,38 IL-2,39 and IL-8.36 The sources of intrathecal production of these cytokines include neuronal33,35 and glial cells.33 The stimulus for and regulation of this enhanced cytokine response remain to be determined. Although potentially an epiphenomenon, it could be a consequence of cell activation mediated by autoantibodies within the intrathecal space. However, measuring CSF cytokine levels unselectively in patients with any manifestation of NP-SLE is unlikely to be of diagnostic value in individual cases.40
Other potentially important inflammatory mediators are matrix metalloproteinases (MMPs), a family of endoperoxidases that can degrade extracellular matrix components.41 MMP-9 is a gelatinase and is secreted by a variety of cells (including macrophages, T lymphocytes, and endothelial and smooth muscle cells) in the blood vessel wall.42 Implicated in the pathogenesis of plaque rupture,43 elevated levels have also been associated with other conditions [including multiple sclerosis (MS),44 Guillain-Barré syndrome,45 rheumatoid arthritis,46 and SLE].47 A recent study48 has examined the association between circulating levels of MMP-9 and NP-SLE. Although there was no difference in the levels of MMP-9 between SLE patients and healthy population controls, elevated levels of MMP-9 were associated with NP-SLE and in particular with cognitive impairment. It is of interest that increased expression of MMP-9 is found in the disrupted blood/brain barrier following cerebral ischemia and may facilitate lymphocyte migration into and possibly through the arterial wall.49
Elevated MMP-9 levels have also been detected in CSF samples of patients with NP-SLE compared to SLE patients without NP manifestations and normal controls.50 Furthermore, the positive correlation among CSF MMP-9 levels, proinflammatory cytokines, and biomarkers of neuronal and glial degradation50 supports the suggestion that the enhanced production of MMP-9 is under cytokine control and is responsible for central nervous system damage.
The association between SLE and headaches, including migraine, is controversial. The reported prevalence of headache has varied widely (between 24 and 72%5–9), but the prevalence of headache in the general population is also high (with up to 40% of individuals reporting a severe headache at least once per year).51 Two of the most recent studies52,53 found no increase in the prevalence of headache in SLE, which was also the conclusion of a recent metanalysis.54 Furthermore, there is only one study55 reporting an association between headache and other clinical features of active lupus.
Aseptic meningitis is a relatively uncommon but well documented cause of headache in SLE56,57 and requires confirmation by analysis of CSF. Other potential causes must be considered, including infection and idiosyncratic reactions to medications such as antibiotics and nonsteroidal anti-inflammatory drugs.58–60 Thus, although headache may be a component of active SLE in rare individual patients it is more likely that the majority of headaches in SLE patients are due to non-SLE causes.
Psychosis is reported in up to 8% of SLE patients5–9,61 and is characterized by either the presence of delusions (false belief despite evidence to the contrary) and/or hallucinations (perceptual experiences occurring in the absence of external stimuli). The latter are most frequently auditory. Psychosis is a rare but dramatic manifestation of NP-SLE and when present it must be distinguished from other causes, including high doses of corticosteroids, non-prescribed drug abuse, schizophrenia, and depression.
Depression and anxiety are common symptoms in lupus and occur in 24 to 57% of patients.5–9,25 However, as there are no features of these syndromes that are unique to SLE patients there is often uncertainty about the etiology and attribution in individual cases. The association between psychosis depression and anti-P antibodies in SLE is supported by some but not all studies.21,22,24,25
The many forms of cerebrovascular disease are reported in 5 to 18% of SLE patients5–9 and are likely multifactorial in etiology. Accelerated atherosclerosis is well recognized in SLE, particularly in relation to coronary heart disease (which is 5 to 10 times more frequent in SLE patients compared to control populations).62 This also contributes to the increased rate of cerebrovascular events in SLE. An additional etiologic factor is the prothrombotic state as a consequence of antiphospholipid antibodies,26 which provides a rationale for therapeutic intervention with anticoagulants in selected cases.
Generalized and focal seizures are reported in 6 to 51%5–9 of SLE patients and may occur either in the setting of active generalized multisystem lupus or as isolated neurologic events. Their occurrence is frequently associated with the presence of antiphospholipid antibodies,8 which co-occur with microangiopathy, arterial thrombosis, and subsequent cerebral infarction.
These are rare manifestations of central nervous system disease in SLE and occur no more frequently than in 1 to 3% of patients.5–9,63 Clinical and neuroimaging evidence of demyelination has been described and is frequently indistinguishable from multiple sclerosis.64 Thus, this particular syndrome may represent a concordance or overlap of two autoimmune conditions. Transverse myelopathy65 and chorea66 present acutely and are frequently associated with antiphospholipid antibodies.65,66 Although an arterial thrombotic event is a likely contributory mechanism for transverse myelopathy, the cause of chorea is less clear and there has been speculation that it may be a consequence of a direct interaction of antiphospholipid antibodies with neuronal structures in the basal ganglia.67
A sensorimotor neuropathy is the most common neuropathy and has been reported in up to 28%5–9 of SLE patients. It frequently occurs independently of other disease characteristics.68 The abnormalities are persistent, but in one study 67% of patients had no change in their neuropathy over a 7-year period.68 Other less frequent forms of neuropathy include cranial neuropathy,7 autonomic neuropathy,69,70 plexopathy,71 mononeuritis multplex,72 and Guillain-Barré syndrome.73,74 Myasthenia gravis has been reported in SLE but is rare.75,76
This term has replaced what was previously called “organic brain syndrome” and is synonymous with “encephalopathy” and “delirium.” It encompasses a state of impaired consciousness or level of arousal that can progress to coma. Characteristics include reduced ability to focus, disturbed mood, and impaired cognition. It has been reported in 4 to 7% of SLE patients5,7,8 and must be distinguished from other causes, including metabolic abnormalities and hypertensive encephalopathy.
Cognition is the sum of intellectual functions that result in thought. It includes reception of external stimuli, information processing, learning, storage, and expression. Disturbance of even one of these functions can result in disruption of normal thought production and present as cognitive dysfunction.
No simple screening test for cognitive dysfunction in SLE patients is currently available for use in the clinical setting. Subjective reports of cognitive dysfunction lack reliability, and neuropsychologic test batteries are long and cumbersome and require expert administration and interpretation. Although easily administered, the Modified Mini Mental Status exam is not very sensitive for detecting mild (albeit clinically significant) cognitive dysfunction (especially problems with executive function, which are common in SLE patients). Self-report of cognitive difficulties is currently the only means by which clinicians can screen patients who may have significant cognitive dysfunction.
The Cognitive Symptoms Inventory is a 21-item self-report questionnaire designed for patients with rheumatic disease to assess self-perception of one’s ability to perform everyday activities.77 Although it has not been validated against neuropsychological testing, a recent report suggests that this instrument may be useful as a bedside screening tool to identify SLE patients at risk for cognitive impairment.78 However, formal testing of cognitive function is the only definitive way of diagnosing cognitive impairment and applicable patients should be referred to a neuropsychologist.
Cognitive dysfunction, assessed using neuropsychological assessment techniques, has been reported in up to 80% of SLE patients,5 although most studies have found a prevalence between 17 and 66%.79,80 Many individual patients have subclinical deficits. For example, a review of 14 cross-sectional studies of cognitive function in SLE revealed subclinical cognitive impairment in 11 to 54% of patients.79 In non-SLE populations there are several potential causes of cognitive dysfunction, many of which may also be present in and exacerbated by SLE (Table 37.2). Whether they contribute to or even cause cognitive dysfunction in SLE patients requires individual consideration. However, the prevalence, severity, course, and impact of SLE-associated cognitive dysfunction are greater than that of both healthy controls81–83 and several non-CNS disease populations,83–85 strongly suggesting that SLE-specific events may play a significant role in the development of cognitive dysfunction.
|Direct CNS disease or injury|
|Psychological or psychiatric disturbance|
Modified from Hanly JG, Harrison MJ. Management of neuropsychiatric lupus. Bailliere’s Best Practice & Research Clinical Rheumatology 2005;19(5):799-821.
A single pattern of SLE-associated cognitive dysfunction has not been found, but commonly identified cognitive abnormalities include overall cognitive slowing, decreased attention, impaired working memory, and executive dysfunction (e.g., difficulty with multitasking, organization, and planning). As the majority of SLE patients with cognitive impairment have relatively mild deficits, the careful selection and assessment of cognitive performance in control groups is of critical importance in defining expected levels of function in healthy individuals and those with other chronic diseases. Although cognitive impairment may be viewed as a distinct subset of NP-SLE, it can also serve as a surrogate of overall brain health in SLE patients who may be affected by a variety of factors including other NP syndromes.
The association between cognitive impairment in SLE and other clinical variables has been examined in a number of studies. It is intuitive that certain clinically overt NP-SLE events such as stroke and antiphospholipid antibody-associated multifocal infarction would likely be accompanied by cognitive dysfunction. Thus, it is not surprising that the prevalence of cognitive dysfunction in patients with past or current clinically overt NP-SLE is usually greater than in those with no such history.81,82,86,87
Mood and psychological distress, even in the absence of a frank psychiatric diagnosis, are known to influence cognitive functioning as well as alter performance on neuropsychologic tests. The high prevalence of psychological disturbance in SLE has led some to hypothesize that SLE-associated cognitive dysfunction is primarily due to the psychological impact of the underlying disease. Studies to date have both supported88 and refuted89 this hypothesis. Longitudinal data suggest that SLE patients with psychiatric involvement experience an improvement in cognition with improved psychiatric status at one-year follow-up that is not observed in patients with persistent psychiatric disorders.90
Most studies, using several methods to measure global SLE disease activity, have detected no association between the presence of cognitive impairment and disease activity.88,90,91 Furthermore, no association has been found between cognitive dysfunction and the use88,90,92,93 or dose of corticosteroid88,90,93 in patients with SLE.
The outcome of cognitive impairment in SLE patients has been examined in a number of studies. For example, in a five-year prospective study of 70 SLE patients using a standardized panel of neuropsychological tests87 the prevalence of overall cognitive impairment in SLE patients fell from 21 to 13% over the period of study. Five patterns of cognitive performance were observed over the five-year period (Fig. 37.2).
[Derived from Hanly JG, Cassell K, Fisk JD. Cognitive function in systemic lupus erythematosus: Results of a 5-year prospective study. Arthritis Rheum 1997;40(8):1542–1543 (reference 87).]
Eighty-three percent of patients were either never impaired or had resolution of cognitive impairment without specific therapeutic interventions. An additional 13% of patients demonstrated either an emerging or fluctuating pattern of impairment, and only 4% (2 patients) showed persisting deficits which were stable over time. Similar benign changes in cognitive performance over time have been reported by Waterloo and colleagues94 in 28 patients over five years, by Hay and colleagues95 in a two-year prospective study, and by Carlomagno and colleagues.96