The nervous system in systemic lupus erythematosus

Introduction

The immune dysregulation observed in lupus can damage every level of the nervous system ( Fig. 49.1 ). To complicate matters, neurological disease in lupus is not limited to direct immunological damage and can arise from comorbid disease, or as a complication of therapy. As a result, neurological disease in lupus is heterogeneous, complex, and poorly understood. “Neurolupus” is perhaps the simplest term to use to capture the broad spectrum of neurological disease caused by lupus, and “neuropsychiatric lupus” (NPSLE) is also used, perhaps giving undue prominence to psychiatric syndromes which are not commonly seen.

Figure 49.1, An overview of nervous system structures which can be targets of lupus-related disease.

This heterogeneity of neurological manifestations was recognized in the first systematic neuropathological analyses of brain disease in lupus from the 1940s. These brain-focused studies highlighted an important observation—that much of the neuropathological damage observed in lupus had a neurovascular origin, with comparatively little active inflammatory cell infiltration. These observations have been confirmed by subsequent neuropathological and imaging studies.

Classification of neurolupus

The first serious effort to classify neurolupus was over two decades ago, with the introduction of the American College of Rheumatology (ACR) nomenclature for neurolupus. These definitions have been helpful in recognizing the breadth of clinical manifestations, subdividing syndromes into those which can affect the central nervous system and those which can affect the peripheral nervous system. As a result the ACR nomenclature has facilitated the epidemiological study of these neurological syndromes and proved an important starting point for the quantified study of neurolupus. Using these criteria in large cohort studies has demonstrated that neurological disease in lupus can occur soon after lupus diagnosis, in individuals with high generalized disease activity.

However, such classification systems for defining brain disease in lupus have distinct problems. First, the ACR nomenclature is largely based upon clinical syndromes. This is in contrast to the field of renal lupus where both classification and therapeutic stratification is based on a strong pathological understanding of the disease. As such the classification of neurolupus is more remote from mechanistic understanding than disease of other organ systems. Second, a number of these clinical syndromes may not exist, and their continued inclusion has the potential to incorrectly suggest a causal relationship with lupus. For example migraines are common in the general population, as well as in people with lupus, and the entity of “lupus headache” does not stand up well to epidemiological scrutiny.

As our understanding of the biological basis of neurolupus improves, we are beginning to see how the heterogeneity of neurolupus can be unpicked. A good example is provided by spinal cord disease in lupus. The ACR nomenclature refers to “lupus myelopathy”—a rather vague descriptive term of spinal cord dysfunction without mechanistic underpinning. Advances in our understanding of how spinal cord disease develops in lupus have identified multiple mechanisms. Many cases of severe spinal cord inflammation in lupus are associated with likely pathogenic autoantibodies directed again the glial water channel Aquaporin 4 (AQP4). It is likely that other cases of inflammatory myelitis are explained by other autoantibodies, whose presence or absence may help guide intensity of immunotherapy. To add to this heterogeneity, some spinal cord disease in lupus can be driven by ischaemia rather than inflammation. It is clear therefore that the term “lupus myelopathy” does not accurately describe spinal cord disease in lupus, and more precise descriptions will be needed to advance clinical trial efforts in this area.

Neurological manifestations can also contribute toward a diagnosis of lupus. The ACR diagnostic criteria for lupus include neurologic manifestations, although only a couple of the rarer neurological manifestations (seizures and psychosis) are represented in these criteria.

Mechanisms of neurolupus

There are many aspects of immunological dysfunction, which drive organ damage in lupus—including cytokines, immune cells and antibodies. All can attack neural structures. A detailed description of preclinical and experimental medicine studies is outlined in Chapter 38 . As well as direct effects from innate and adaptive immune mechanisms, the proinflammatory state creates a vasculopathic environment affecting vessels of all sizes. Furthermore, neurological disease can arise as a consequence of treatment, for example as a direct side effect of the drug or as an infectious complication of iatrogenic immunosuppression. This complexity highlights the need for a carefully considered clinical approach.

Clinical approach

Consensus recommendations for clinical approach to neurological disease in lupus are sparse. The 2010 EULAR recommendations for the management of neurolupus emphasise that neurological symptoms in lupus should be evaluated in the same way as neurological symptoms in people without lupus. As such, neurological symptom evaluation, with detailed history and examination, should be targeted toward the identification of the location of the problem within the nervous system. Sometimes localization is straightforward. However pathology may exist at more than one neuroanatomical level and diffuse brain disease can be more difficult to localize. Carefully selected investigations can provide useful confirmatory evidence of a lesion, and its nature. Within the clinical history, particular attention should be to the severity and activity of the lupus, potential comorbid disease (immunological and/or vascular) and drugs. It is important to bear in mind that neurological infections in immunosuppressed individuals can present in an atypical manner.

Investigations

Routine laboratory tests

Routine tests of renal function, liver function, bone profile are important to exclude metabolic derangements which can contribute to neurological symptomatology. Routine autoantibody (antinuclear antibodies, anti-dsDNA antibodies) and complement testing is helpful to confirm diagnosis, in particular with patients who present with neurological disease in the absence of an established diagnosis of lupus. More extended serological testing is important in the identification of coexisting “secondary” immune diseases such as Sjogren's syndrome and antiphospholipid syndrome (APS), which can contribute to neurological disease in people with lupus.

Brain-reactive antibodies

Antibodies which react against neuronal and glial cell surface antigens can potentially alter brain function and contribute to neurological disease. While there is a strong preclinical literature about a direct role for brain reactive antibodies in the pathogenesis of brain dysfunction in lupus, the clinical relevance of these tests must be interpreted with caution. Antibodies against neuronal cell-surface proteins such as the NR1/2 subunit of the NMDA-receptor and leucine-rich glioma-inactivated 1 (LGI1) protein are strongly associated with immune-mediated encephalitis, and may play a pathogenic role in these brain diseases, which display distinctive encephalitic phenotypes. However this association is most strongly seen with high titre antibodies, particularly within the cerebrospinal fluid (CSF). These brain-reactive antibodies can be detected at low titre in a significant proportion of lupus patients but the specificity of these antibodies at these lower levels is much more questionable. Strong, clinically relevant elevation of these antibodies in people with lupus is rare. This lack of specificity of these antibodies at lower titres in the serum may explain the conflicting findings of the role of brain-reactive antibodies in neurolupus. It is therefore likely that routine testing of these antibodies may be unhelpful, unless there is a neurological phenotype which is strongly associated with particular antibodies, such as a limbic encephalitis. Autoantibody testing in the context of opticospinal inflammation is important, in particular inflammatory lesions of the spinal cord. Antibodies directed against AQP4 are an important clinical test and their presence signifies a high risk of recurrent disabling neuroinflammatory events.

Conventional neuroimaging

Conventional neuroimaging techniques including computed tomography (CT) and magnetic resonance imaging (MRI) play an important role in the identification of structural brain disease in lupus. While MRI imaging offers greater resolution, CT scanning can play a role, in particular in the acute setting when intracerebral bleeding is suspected. CT angiography can provide accurate noninvasive assessment of intracranial large and medium sized vessels.

Conventional brain MRI is often abnormal in people with lupus, with small white matter hyperintensities, which largely correspond to microvascular damage in paired MRI-neuropathology studies. Additional conventional MRI sequences (fluid-attenuating inversion recovery sequence, diffusion weighted imaging) can add further information and gadolinium enhancement can provide insights relating to blood-brain barrier breakdown, although the use of gadolinium-containing contrast agents carry risks in patients with renal impairment. Both conventional CT and MRI brain imaging modalities offer options to perform noninvasive angiography. The extracranial vessels can also be assessed using Doppler ultrasound, and this is a critical part of the workup of anterior circulation thromboembolic stroke. Advanced MRI techniques such as diffusion tractography and functional MRI are providing interesting experimental insights into the brain in lupus but do not yet play an established role. Equally nuclear medicine techniques such as positron emission tomography and single photon emission CT, while promising, do not yet have a clear role in clinical practice.

MRI of the spinal cord is essential in the evaluation of lupus myelopathy, and imaging of the nerve roots to look for contrast enhancement and nerve root hypertrophy can also aid in the diagnosis of inflammatory neuropathies associated with lupus.

CSF examination

Careful CSF examination plays an important role in the evaluation of patients with brain disease, in particular those who display clinical evidence of meningeal inflammation. CSF examination in people with lupus is frequently nonspecifically abnormal, with mild elevations of protein and white cell count. A primary purpose of CSF examination is often to exclude the presence of infectious agents, especially those which can occur in the context of immunosuppression. Sterile inflammation can also occur and is discussed later in the chapter. Distinguishing infectious from inflammatory meningeal disease is not always straightforward and the two can coexist.

Neurophysiology

Routine electroencephalogram (EEG) evaluation in people with lupus is unlikely to be helpful. However there is a role for EEG in the evaluation of possible seizure activity and to assist in the diagnosis of encephalopathic states. Visual evoked potential can supplement clinical evaluation in the diagnosis of optic neuropathy. Nerve conduction studies and electromyography are important neurophysiological investigations in the assessment of neuromuscular complications of lupus, and can help provide specific diagnoses, particular in the setting of inflammatory neuropathies, described later.

Neuropathology

The most commonly performed neuropathological investigation in neurolupus is nerve biopsy. Many inflammatory neuropathies which occur in the context of lupus can be diagnosed without nerve biopsy. For example inflammatory demyelinating neuropathies are seen in about 20% of individuals and can be diagnosed on the basis of nerve conduction studies with supportive CSF findings. Mononeuritis multiplex can occur and is typically caused by vascular inflammation affecting the nerves. It is in this scenario that nerve biopsy is typically performed. Sural nerve biopsy carries risks of permanent loss of sensation and wound infection but can demonstrate vasculitis and lymphocytic infiltration. Skin biopsy, looking for small fibre neuropathy, can also be helpful.

Brain biopsy is much less frequently performed and much of the neuropathology in lupus comes from evaluation of postmortem material. Care should be taken in extrapolating what is often end-stage disease, but these studies confirm that cerebral microvascular disease is common.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here