Literature Review on Neuropsychiatric Lupus

Abstract

Neuropsychiatric systemic lupus erythematosus (NPSLE) had been described in several medical literatures These included the pathogenesis, mechanisms and current approach to management and treatment. Although still limited, more information is coming with the advancement of medical knowledge and technology regarding systemic lupus erythematosus and neuropsychiatric involvement. NPSLE remains elusive in the context of outright diagnosis and management. Its manifestations need to be carefully assessed before a final diagnosis is made for the proper treatment. Thus, attribution models were later developed to address these problems. NPSLE will likely develop among lupus patients in the first 5 years from SLE onset. The development and exact pathogenetic mechanisms of the disease also remain controversial but the discovery of the blood-brain barrier injury has given points of clarity. The focus of management is based on the identified etiology. Targets include symptomatic treatment and addressing the underlying SLE process. Likewise, the use of corticosteroids, cyclophosphamide, mycophenolate mofetil, azathioprine, antimalarial agents, warfarin or low dose aspirin depending on the pathways involved is also being utilized with positive results. More researches are being done to better elucidate the complex nature of NPSLE.

1. Introduction

Systemic lupus erythematosus is a complex autoimmune disease that affects virtually all organ systems. The interplay of the cellular make-up of the affected host and environment creates aberrancy that leads to a cascade of inflammatory response causing organ damage. The central and peripheral nervous system similar to other organs impaired by lupus exhibit more complicated manifestations. This in turn has resulted to a difficult timely recognition of the disease for an appropriate management. Within the past 10 years or more, researches worldwide are being done to address this problem. Neuropsychiatric systemic lupus erythematosus (NPSLE) had been described in several medical literatures These included the pathogenesis, mechanisms and current approach to management and treatment. Although still limited, more information is coming with the advancement of medical knowledge and technology regarding systemic lupus erythematosus and neuropsychiatric involvement.

2. Pathogenesis

The pathogenesis of NPSLE is being discussed in several literatures. There is no doubt that multiple pathogenic mechanisms are involved. Likewise, it encompasses a variety of process including cytokines, autoantibodies, and other infiltrating cells (cell-mediated inflammation).

The underlying mechanisms of neurologic manifestations depend on whether the central nervous system (CNS) or the peripheral nervous system (PNS) is being involved. For the CNS manifestations to occur, the primary injury can either be directed towards the vasculature or the brain parenchyma. The vascular injury includes damage to both large and small vessels via thromboembolic events, often as a consequence of antiphospholipid (aPL); a bland vasculopathy of small vessels characterized by vascular hyalinization, perivascular inflammation, and endothelial proliferation; and atherosclerotic lesions [1].

The disruption of BBB is central to the pathophysiology of NPSLE. This serves as the primary factor for the transit of pathogenic proteins towards the CNS. On the other hand, the BBB permeability remains debatable in diffuse NPSLE. Diamond et al. have provided data on studies from murine models supporting the role of circulating anti-DNA autoantibodies in the development of NPSLE thru cross-reactivity with the NR2 subunits of the anti-N-methyl-D-aspartate receptors during inflammation causing damage in the integrity of BBB [2, 3]. Other substances which are also associated with NPSLE include intrathecal IgG production, antiphospholipid (aPL) antibodies, anti-ribosomal P antibodies, anti-endothelial cell antibodies (AECAs), anti-microtubule-associated protein 2 (MAP-2) antibodies, anti-aquaporin 4 antibodies (AQP4) and anti-suprabasin antibodies (Table 1). Meanwhile, intrathecal markers associated with NPSLE include matrix metalloproteinase-9 (MMP-9) and plasma activator inhibitor 1 (PAI-1). CSF levels of interleukin 6 (IL-6) and interleukin 8 (IL-8) are significantly correlated with MMP-9. Aside from BBB, other brain structures which are disrupted enhancing the penetration of the central nervous system (CNS) include the meningeal barrier, and the choroid plexus. Aseptic meningitis occurs as a result of a breached meningeal barrier in SLE patients and also with non-steroidal anti-inflammatory drug (NSAID) use. Similarly, due to immunosuppression, infectious meningitis is likely. Studies involving the choroid plexus have shown that in SLE patients, immune complex deposition is evident but nonspecific [4, 5].

Intrathecal	IgG
	PAI-1
	MMP-9
Antibodies	aPL
	anti-ribosomal P
	AECAs
	anti-MAP2
	anti-AQP4
	anti-NMDAR/NR-2
	anti-suprabasin
CSF	IL-6
	IL-8
	IL-2, IL-10

Table 1.

Antibodies/proteins of interest implicated in NPSLE.

PAI-1: plasma activator inhibitor; MMP-9: matrix metalloproteinase; aPL: antiphospholipid; AECAs: anti-endothelial cell autoantibodies; anti-MAP2: anti-microtubule protein 2; anti-AQP4: anti-aquaporin 4; anti-NMDAR/NR2: glutamate receptor antibodies; and IL: interleukins.

The most recent studies have referred to the proposed two pathologic mechanisms contributing to the development of NPSLE, the autoimmune or inflammatory pathway and the ischemic or thrombotic pathway. The previous pathway includes autoantibodies, proinflammatory cytokines, chemokines, microglia and C1q while the latter involves the immune complexes, complement system and the aPL autoantibodies. Both of which manifest with focal and diffuse neuropsychiatric symptoms [6].

Intrathecal IgG production is found elevated during a central nervous system flare. The aPL antibodies are directly related to focal NPSLE via autoantibody-mediated thrombosis. This predisposes aPL-antibody positive patients for the increased risk of cerebrovascular events such as stroke and transient ischemic attack (TIA). Studies also shown that aPL antibodies hasten the process of atherosclerosis among susceptible individuals. The risk of developing NPSLE is twice as likely among those with aPL antibodies than those who are aPL-antibody negative individuals. Seizures, chorea, cognitive dysfunction and myelopathy were also observed among those with aPL antibodies. On the other hand, greater cognitive impairment manifests with persistently elevated anti-cardiolipin. This consistent finding is also similar among SLE patients with positive lupus anticoagulant. The ischemic events are implicated in the brain regions including the amygdala, frontal cortex and hippocampus. The anti-ribosomal P antibodies have not been associated in coexisting cognitive impairment. These antibodies have an association with NPSLE, particularly psychosis, but are not reliable to make a diagnosis. These are highly specific for SLE and found to be present in up to 46% of patients with SLE. Other NPSLE syndromes including seizure, coma, depression, aseptic meningitis and transverse myelopathy are also associated with these antibodies. MAP-2, a cellular protein, is strictly found in neurons and essential to the cytoskeletal integrity. In one study, involving 100 SLE patients and 74 patients with various neurologic disorders, it was found out that more SLE patients as compared to neurologic disease control patients have presence of anti-MAP-2 antibodies (17% vs. 4%, p = 0.028) AQP4 is a water channel protein expressed on astrocytic foot processes around blood vessels controlling the flow of water into and out of the brain. In one study, anti-AQP4 antibodies were detected in 3% of all patients with NPSLE and 27% of patients with NPSLE who had demyelinating lesions. Suprabasin is a protein used as an epidermal differentiation marker. In one study it was found out that titers of anti-suprabasin antibodies were higher in the patients with NPSLE than patients with non-neuropsychiatric SLE, multiple sclerosis and normal-pressure hydrocephalus. MMP-9 enhances T cell migration through connective tissue. It is secreted by cells found in the walls of the vasculature including macrophages, T lymphocytes, endothelial cells, and smooth muscle. Intrathecal levels of MMP-9 in significant amount are found in all patients with SLE as compared to non-SLE patients including those with NPSLE. Similarly, intrathecal levels of PAI-1 have been found to be significantly elevated in patients with NPSLE [1, 4, 5].

AECAs play a role in the pathogenesis of NPSLE. It recognizes molecules bound to endothelial cells, antigens expressed constitutively or cytokine-induced and adhesion molecules. Psychosis and depression are also implicated with AECAs due to vasculitis via expression of adhesion molecules, cytotoxic effect, induction of apoptosis and the activation of coagulation cascade [7].

Previous studies have revealed the associations of elevated CSF IL-6 levels with seizures and IFN-α with lupus psychosis. Some evidence suggested the roles of other cytokines including IL-2, IL-8 and IL-10. Another interesting study involves microglial cells. They play a fundamental role in regulating BBB function and in shaping brain circuits and development (‘synaptic pruning’) [8].

3. Clinical manifestations

The American College of Rheumatology (ACR) subcommittee categorized NPSLE into 19 distinct syndromes encompassing the CNS and PNS. CNS manifestations include acute confusional state, psychosis, headache, and mood disorders for the diffuse processes (Table 2). On the other hand, seizures, myelopathy and chorea are the CNS focal manifestations [1].

PNS	CNS
Autonomic Disorder	Aseptic meningitis
Cranial neuropathy	Acute confusional state
Guillain-Barre syndrome	Anxiety Disorder
Mononeuropathy, single/multiplex	Cognitive Dysfunction
Myasthenia gravis	Cerebrovascular Disease
Plexopathy	Demyelinating syndrome
Polyneuropathy	Headache
	Mood Disorder
	Movement Disorder
	Myelopathy
	Psychosis
	Seizure

Table 2.

American college of rheumatology classification of neuropsychiatric syndromes in systemic lupus erythematosus.

PNS: peripheral nervous system; and CNS: central nervous system.

In a meta-analysis of 5057 SLE patients, it was found out that NPSLE prevalence varied from 17.6 to 44.5% in retrospective and prospective studies [8]. NPSLE may be the first manifestation of the disease. CNS syndromes are more common than peripheral [9]. The most frequent NPSLE manifestations are headaches, depression, anxiety and cognitive dysfunction. In one study, ethnicity and older age are factors associated with earlier neuropsychiatric damage [10].

Headaches manifest in more than 50% of SLE patients with both migrainous and tension-type headaches being described [1]. Headache in SLE is not associated with disease activity, treatment, imaging such as MRI and biomarkers including aPL, anti-P, and glutamate receptor antibodies (anti-NR2) or any specific antibody. Seizures occur in about 10 to 20% of SLE patients and associated with increased morbidity and mortality. Generalized tonic-clonic seizure is the most common type. In contrast to headache, seizure tend to be associated with APS, disease activity, severe organ damage and other NPSLE manifestations. It is crucial to rule out other causes of seizures such as infections [9, 10].

Cognitive dysfunction which manifests with deficits in memory, thinking and concentration is increasingly observed among SLE patients. Studies suggested the association of aPL, anti-NMDAR/anti-NR2 antibodies, and anti-Sm antibodies [1, 9].

Psychosis, depression and anxiety can occur in SLE. Postal et al. found that mood disorders were associated with disease activity, high prednisone doses, cutaneous disease, and longitudinal extensive transverse myelitis [11]. Depression is the most common disorder in NPSLE, and its lifetime prevalence may reach 65% [12]. Higher incidence of depression among SLE patients were associated with anti-P and anti-NMDA receptor autoantibodies [9]. Interestingly, anti-P antibody levels were 5- to 30-fold higher during the active phase of SLE psychosis, but not during other SLE manifestations [13]. Also, it is very important to consider the differential diagnoses of psychosis in SLE patients such as CNS infection, primary schizophrenia, metabolic abnormalities and psychosis secondary to glucocorticoid therapy or illicit drugs [1]. Anxiety disorders are common and found in up to 40% of SLE patients [14].

SLE patients are susceptible to developing cerebrovascular events such as ischemic stroke and intracerebral hemorrhage with the previous more common than the latter. The development of cerebrovascular disease in SLE patients can be attributed to accelerated atherosclerosis and inflammatory mediators such as cytokines, aPL antibodies and complement system [1, 9]. In relation to which, one systematic review found a fivefold increase in the risk of ischemic stroke in patients with aPL antibodies compared to controls [15]. Posterior reversible encephalopathy syndrome (PRES) is a known mimic of CNS lupus. It is characterized by headache, seizures, altered consciousness, and visual changes often in a background of hypertension, eclampsia, renal disease and/or immunosuppressive therapies [1].

Chorea is the most common movement disorder observed among SLE patients. It appears in 2–3% of adult patients more common in women. Recent evidence suggests an autoimmune mechanism related to aPL antibodies. Aseptic meningitis is also a manifestation of SLE. It can occur at any time during the disease course presenting with headache and/or altered mental status. Other causes include infections of various etiologies (bacterial, viral, fungal or tuberculosis), immunosuppressants or medications and malignancy [9, 10].

Optic neuropathy and myelopathy rarely occur as part of the spectrum of NPSLE. The manifestations of optic neuropathy include monocular central visual loss, color vision and afferent pupillary problems. It can be caused by thrombotic or inflammatory mechanisms in the setting of lupus. On the other hand, myelopathy is a syndrome affecting the spinal cord presenting with numbness, paresthesia of bilateral lower extremity and weakness that can progress to involve the upper limbs. A characteristic symptom of which is a band-like pain or discomfort around the abdomen [1]. Lupus myelitis occurs in about 1.5% of cases. In nearly half of the patients with SLE, acute transverse myelitis occurs as the first clinical manifestation within the first 5 years of diagnosis. Histopathology findings revealed ischemic or thrombotic myelopathy or a localized inflammation [9]. Several studies link transverse myelitis-SLE with aPL antibodies [16].

Demyelination in NPSLE is also seen in about 0.3% of cases. It can be an isolated syndrome but can overlap with multiple sclerosis [17]. SLE patients also present with peripheral neuropathy. One study had demonstrated a 6% prevalence of peripheral neuropathy, 67% of which were attributable to SLE. Sensorimotor axonal polyneuropathy was the most common type. Similarly, it is crucial to rule out other causes of peripheral neuropathy such as diabetes mellitus, hypothyroidism,, infections, vitamin deficiency, malignancy and drugs [1].

4. Diagnosis

The approach to diagnosis of NPSLE remains challenging as clinicians need to be thorough when it comes to history taking, physical examination and other methodologies including serologic examination, imaging and medical procedures (e.g. lumbar puncture) that can facilitate accurate diagnosis. Primarily, the arduous task begins with identifying whether the manifestations are secondary to the disease activity or not. These steps and processes will help in identifying the pathogenetic pathway involved and the proper course of management at the end (Figure 1).

Figure 1.

NPSLE in kaleidoscope. NPSLE in summary featuring the basic diagnostic approach and management.

Based on the presentation of the neuropsychiatric signs and symptoms, the assessment process should be tailored-fit accordingly. Likewise, localizing the areas of the CNS involved also has its limitations. In example, the focal neurologic symptoms correlate well with the conventional magnetic resonance imaging (MRI) but abnormalities associated with altered perfusion or neurometabolite changes can be best demonstrated by functional imaging techniques [4]. However, more than half of patients diagnosed with NPSLE have a normal MRI of the brain [18]. Single photon emission computed tomography (SPECT) provides an estimate of regional cerebral blood flow and thought to be more sensitive than MRI for the evaluation of NPSLE. But studies were inconsistent [19].

Computerized tomography (CT) is being used to exclude focal abnormalities such as hemorrhage, infarcts, tumors, cortical atrophy and calcifications. On the other hand, metabolic neuroimaging such as positron emission tomography/PET, MR spectroscopy) and perfusion imaging such as single photon emission computer tomography/SPECT) can detect abnormalities in patients with psychiatric manifestations but otherwise have normal MRI studies [4]. Functional MRI (fMRI) also is being used to assess for cognitive function in SLE.

CSF analysis is utilized in cases of ruling-out infection. Likewise, it becomes as important in some cases such as aseptic meningitis, transverse myelitis and vasculitis.

Biomarkers are also used to better screen and monitor treatment for NPSLE. At present, the autoantibodies used in the diagnosis and therapeutic decisions include antineuronal, anti-ribosomal P, and anti-NR2 antibodies. Meanwhile, other than autoantibodies, chemokines and cytokines, intra-thecal levels of PAI-1 and MMP-9 are used for screening & monitoring purposes.

Magro-Checa et al. proposed a diagnostic approach based on the clinical presentation of patients with NPSLE manifestations. This consisted of matched diagnosis and work-up, procedure or imaging to be done. On the other hand, there have been attribution models proposed in order to strengthen the diagnosis. These were developed very carefully and limitations were also determined accordingly [17].

In an international inception cohort study Hanly et al. described an attribution model in which the level of stringency was based on three simple rules that take into account the temporal relationship between the neuropsychiatric (NP) event and the diagnosis of SLE, the type of NP event and a comprehensive list of exclusions or associations according to the American College of Rheumatology (ACR) nomenclature. In this study, they concluded that 28% of SLE patients experienced at least 1 NP event around the time of diagnosis of SLE, of which only a minority were attributed to SLE [20].

Bortoluzzi et al. developed a new algorithm for attribution of neuropsychiatric events in SLE in 2015. This enabled identification of which NP events have a high probability of being or not being attributed to the disease among SLE patients in a more standardized and reproducible manner. When compared with expert clinical judgment, it demonstrated a good performance in sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) with a confident assumption of correctness at 33%. However, this was not intended as a substitute for clinical judgment which remains as the cornerstone of the diagnosis and management of NPSLE [21].

Magro-Checa et al. in 2017 provided a prospective date from the Leiden NPSLE cohort and determined the value of multidisciplinary reassessment in attribution of neuropsychiatric events to SLE. This model has showed that each NP event was attributed to one of the following groups: NPSLE or NP events directly related to SLE, undefined NPSLE, and non-NPSLE or NP events better explained by other etiology. Non-NSPLE events were divided further into subgroups: due to primary NP disease, due to medication or drugs, due to a complication of SLE and due to other concomitant disease. This model showed reassessment of NP symptoms in SLE and re-classified a total of 13.8% of NP events. Furthermore, the percentage of NP events attributed to SLE was 31.3% [22].

Individualizing the approach to each patient presenting with NPSLE is also applicable since there is no gold standard in the approach to diagnosis. The complexity of the signs and symptoms necessitate a diagnostic algorithm applicable for the most number of cases.

5. Treatment

Besides the general treatment which includes control or correction of aggravating factors, nonpharmacological interventions and symptomatic therapy for the different syndromes associated with NPSLE, the approach also depends on which pathway is primarily involved. Govoni et al. had illustrated the current treatments on NPSLE. In this study, it had been emphasized that the identification of the most likely cause and contributing factors to the NP event is determined by careful assessment and utility of diagnostic tests that are deemed appropriate. Similarly, it is important to determine whether the disease activity is reversible or irreversible by treatment as this can provide the framework for the appropriate modality in each patient [8].

Magro et al. provided a detailed review on the therapeutic strategies in NPSLE from the general treatment to therapies specific for the pathway that is involved (ischemic vs. inflammatory). In this perspective, it is pointed out that a combination of immunosuppressive therapy and secondary prevention may be used in the same patient when both ischemic and inflammatory pathogenic mechanisms are suspected [22].

In clinical practice, therapy will be directed at inflammation or at prevention of ischemic events upon confirmation of the most likely process involved. Psychotherapy had a beneficial effect on anxiety, depression and quality of life in a controlled clinical trial in 80 SLE patients [23]. Positive outcome with antidepressants had been reported in some observational studies [24]. The use of antiseizure drugs in SLE has also not been subjected to controlled clinical trials while antipsychotic medications are used in the majority of patients with lupus psychosis [25, 26]. Among medications used for generalized seizures include barbiturates and phenytoin while for partial complex seizures may include clonazepam, valproic acid and carbamazepine. Nonsteroidal anti-inflammatory drugs (NSAIDs) are used for pain relief such as headache [22]. Systematic studies in SLE patients are lacking focusing on behavioral rehabilitation of cognitive dysfunction [8].

The presence of aPL antibodies predisposes to the development of thrombotic events such as stroke and prior ischemic events. Primary prevention in APS does not support the use of low-dose of aspirin or warfarin according to current evidence and still necessitates large and well-designed clinical trials [27]. SLE patients with focal manifestations attributed to aPL antibodies requires lifelong anticoagulation [28]. Warfarin is used in the prevention of recurrent thrombosis. Other adjunctive therapies of use are antimalarials including hydroxychloroquine, statins, and antiplatelet agents. Antimalarials exert beneficial effects both as anti-inflammatory and antithrombotic.

In NPSLE patients, the use of glucocorticoids is based on clinical experience. Methylprednisolone pulse therapy consisting of 1 g intravenously for three consecutive days followed by tapering doses of oral prednisolone depending on the severity of NPSLE manifestation is a usual practice. This is acknowledging that the mechanism responsible is inflammatory in nature similar to when other organ systems are affected during lupus flares. Positive effects of cyclophosphamide treatment had been described in several case series [29, 30]. One retrospective study involving 31 NPSLE patients suggested benefit from glucocorticoid use and monthly intravenous cyclophosphamide (250–1000 mg/m²) [31]. Furthermore, Stojanovich et al. concluded in a study involving 60 NPSLE patients that patients treated with cyclophosphamide showed more clinical and eletrophysiological improvement on cerebral function [32].

Due to mild side effects, azathioprine is frequently used for maintenance therapy or as a steroid-sparing agent. There are very few data on the effects of azathioprine in NPSLE. In a study by Ginzler et al. including 68 SLE patients with poor prognosis due to renal or NP events, 54 patients treated with azathioprine had improved significantly on long-term survival and fewer hospitalization. In clinical practice, similar with previous medications, although not yet supported by current evidence, azathioprine is widely used as maintenance therapy after cyclophosphamide in patients with severe NSPLE manifestations and even as an option in mild NPSLE [33].

The effect of mycophenolate mofetil in NPSLE patients is described in very few cases and not conclusive. Similarly, the use of methotrexate is very rarely use in NPSLE and evidence is limited to several case series via intrathecal route. Meanwhile, the efficacy of plasma exchange or cyclosporin A remains unknown because of concomitant use [34].

The effect of biologic therapies remains limited including rituximab, belimumab, and anifrolumab. The use of rituximab alone or in combination with other immunosuppressives like cyclophosphamide have reported positive effects but needed more studies [35, 36, 37].

6. Conclusion

NPSLE remains an area of great interest focusing further on the pathogenetic mechanisms, approach to diagnosis and treatment. There have been a lot of limitations described in the previous studies including the lack of well controlled clinical trials, biomarkers and novel therapies. These are the challenges in which current and future researches revolve for the benefit of SLE patients.

Funding

No funding was received from or in any form.

Disclosure statement

The author has declared no conflicts of interest.