Peripheral neuropathy in sarcoidosis
Dr. Jinny Tavee
Division of Neurology, National Jewish Health, Department of Medicine, 1400 Jackson Street, Denver, CO 80209, USA
Dr. Jinny Tavee
Division of Neurology, National Jewish Health, Department of Medicine, 1400 Jackson Street, Denver, CO 80209, USA
Peripheral nerve disorders in sarcoidosis consist of granulomatous neuropathy and non-granulomatous small fiber neuropathy (SFN), which differ in their underlying pathology, diagnostic methods and treatment. While granulomatous nerve involvement is rare in sarcoidosis, SFN is reported in over 40% of systemic cases. Distal symmetric polyneuropathy and asymmetric polyradiculoneuropathy are the most common presentations of granulomatous neuropathy, which typically responds to corticosteroids. In contrast, SFN is often manifested as non-length dependent pain and paresthesias that may improve with intravenous immune globulin or infliximab. Early recognition and treatment of sarcoidosis neuropathy can lead to improved outcomes and patient quality of life.
Sarcoidosis is an immune-mediated multisystemic disorder that is pathologically characterized by noncaseating granulomas. While any portion of the nervous system may be affected, symptomatic polyneuropathy is rare and seen in only 1% to 2% of patients with systemic disease. Axon loss polyneuropathy usually manifests as either a stocking-glove polyneuropathy or subacute non-length-dependent polyradiculoneuropathy in which patients present with asymmetric limb pain and sensorimotor deficits. Other polyneuropathy types include pure sensory or motor neuropathy and, rarely, mononeuritis multiplex.
As the clinical manifestations and electrodiagnostic findings are nonspecific, nerve biopsy and evaluation for systemic sarcoidosis are often required for diagnosis when neuropathy is the presenting symptom. In those with known sarcoidosis, nerve biopsy may still be necessary to exclude other potential causes (eg, infection or malignancy) that may be related to chronic immunosuppression.
Biopsy may demonstrate granulomatous compression or direct infiltration of large nerve fibers as well as vasculitis. Prompt treatment with IV corticosteroids or IVIg can result in improvement, although a number of patients may go on to a relapsing-remitting course or chronic progressive decline. Unlike large fiber disease, small fiber neuropathy is frequently seen in sarcoidosis and has been found in over 30% of patients with systemic disease. Patients often present with migratory burning pain and paresthesia that affect various parts of the face, trunk, and proximal limbs in a non-length-dependent distribution, although a distal painful neuropathy is also common. In one series, autonomic involvement (eg, orthostatic hypotension, sweating abnormalities, and gastrointestinal disturbances) accompanied the somatic symptoms in about half of the cases. Also of note, sarcoidosis-associated small fiber neuropathy was seen more often in whites (87%) than African Americans (10%), which is unusual as the latter group has a higher prevalence of systemic disease.
Sarcoidosis small fiber neuropathy is categorized under ‘‘paraneurosarcoidosis’’ rather than true neurosarcoidosis as the underlying pathophysiology is not granulomatous in nature but likely cytokine mediated.
As with large fiber nerve involvement, the clinical presentation of sarcoidosis small fiber neuropathy is also nonspecific. Diagnosis thus requires histologically confirmed systemic disease in addition to specialized testing (eg, skin biopsy) to detect small fiber involvement
Response to standard immune-modulating therapies such as corticosteroids is poor. In some patients, IVIg and infliximab may be helpful in treating both the somatic and autonomic manifestations. More recently, cibinetide, an experimental erythropoietin agonist that reduces inflammation, was shown to improve nerve fiber density in the skin and corneas of patients with sarcoidosis small fiber neuropathy.
Laboratory Studies
Serologic evaluation: Complete blood count, comprehensive metabolic panel, hemoglobin A1C, thyroid panel, vitamin B12, vitamin B6, homocysteine, sedimentation rate, C-reactive protein, anti-nuclear antibodies, Sjogren’s antibodies (anti-Ro/SS-A, anti-La/SS-B antibodies), serum immunoelectrophoresis, celiac panel, angiotension converting enzyme • Cerebrospinal fluid (CSF) analysis: Cell count with differential, red blood cell count, protein, CSF and serum glucose, CSF and serum oligoclonal bands, CSF and serum IgG indices, cytology, cultures
Neurophysiologic Studies: Electromyography (EMG): NCS, needle electrode examination on both distal and proximal muscles including paraspinals to evaluate for a non-length dependent process.
Small fiber testing in patients with normal EMG/NCS
Imaging Studies:
MRI lumbar spine with and without gadolinium for evaluation of nerve roots
MR neurography
Peripheral nerve ultrasound
Fluorodeoxyglucose positron emission tomography (FDG-PET)
Gallium 67 scanning
Biopsy:
Obtain both peripheral nerve and muscle for increased sensitivity (e.g., sural and gastrocnemius; superficial peroneal nerve and peroneus brevis)
CT body with biopsy of abnormal tissue suggestive of sarcoidosis
Bronchoscopy or endobronchial ultrasound-guided transbronchial needle aspiration
Histologic examination
Cytology
Microbial cultures
DNA testing for tuberculosis
Consider meningeal nerve root biopsy for more prominent proximal involvement if distal biopsy is non-diagnostic
Sarcoidosis is an idiopathic multi-organ granulomatous disorder that was first described to involve the nervous system by Winkler in 1905 (Salveson, 1935; Winkler, 1905). While clinically apparent central nervous system (CNS) manifestations have been reported in 5 to 10% of patients with sarcoidosis, granulomatous involvement of the extracranial peripheral nerves is seen in only about 1% of cases (Baughman et al., 2001; Burns et al., 2006; Fritz et al., 2016; Ramos-Casals et al.,2021). This is reflected in the paucity of literature on sarcoidosis neuropathy compared to the numerous reviews on CNS disease. As with other forms of neurosarcoidosis, peripheral nerve disorders, which range from polyradiculopathy to mononeuritis multiplex, may be the presenting symptom of sarcoidosis and occasionally the only manifestation although subclinical involvement of muscle or extraneural organs is typically seen (Mattiassich et al., 2012; Said et al., 2002; Vital et al., 2008; Zuniga et al., 1991). In recent years, the emergence of newer diagnostic techniques has resulted in the recognition of sarcoidosis associated small fiber neuropathy (SSFN), which is non-granulomatous in nature and commonly seen among patients with systemic disease (Bakkers et al., 2009; Hoitsma et al., 2002; Hoitsma et al., 2003; Hoitsma et al., 2004). Classification of peripheral nerve involvement in sarcoidosis into (1) granulomatous neuropathy, which often affects both large and small nerve fibers, and (2) non-granulomatous small fiber neuropathy is clinically useful given the differences in the underlying pathophysiology, method of diagnosis and treatment modalities, all of which are described in this review. Also highlighted are the most recent consensus guidelines on the diagnostic criteria for peripheral nerve sarcoidosis, a clinical approach to evaluation, proposed pathogenic mechanisms of SSFN and recent studies on immunotherapies.
The clinical presentation of sarcoidosis granulomatous neuropathy (SGN) varies widely and is most commonly manifested as a distal symmetric sensorimotor polyneuropathy or asymmetric polyradiculoneuropathy in biopsy-confirmed series (Burns et al., 2006; Ramos-Casals et al., 2021; Said et al., 2002; Vital et al., 2008). Pure sensory or motor neuropathy, mononeuritis multiplex, polyradiculopathy, mononeuropathy and lumbosacral plexopathy have also been reported (Burns et al., 2006; Godwin & Sahn, 1990; Mattiassich et al., 2012; Said et al., 2002; Scott et al., 1993; Zuniga et al., 1991). Less commonly seen are demyelinating presentations, which include chronic inflammatory demyelinating polyneuropathy (CIDP) and multifocal motor neuropathy with conduction block (Kono et al., 2013; Singhal et al., 2015). An acute to subacute polyneuropathy with cranial nerve involvement mimicking Guillain-Barre syndrome (GBS) has also been reported in association with sarcoidosis in a number of case reports and may be axonal or demyelinating in nature (Brier et al., 2020; Fahoum et al., 2009; G´omez et al., 2013; Miller et al., 1989; Said et al., 2002; Zuniga et al., 1991). Although it is possible that the GBS may be an incidental occurrence in patients with sarcoidosis, steroid responsiveness has been observed in many of the cases further strengthening the possibility of a causal link (Brier et al., 2020; Fahoum et al., 2009; Miller et al., 1989; Said et al., 2002; Zuniga et al., 1991).
The onset of SGN may be acute or chronic and can present with a monophasic, remitting-relapsing or chronically progressive course. In patients with a polyneuropathy, pain and paresthesias in the affected limbs are the predominant and often most disabling symptoms although negative sensory symptoms, e.g., numbness, and less commonly weakness are also reported (Burns et al., 2006; Said et al., 2002). Focal sensorimotor deficits may also occur due to direct granulomatous compression or infiltration of a single peripheral nerve and can mimic commonly seen mononeuropathies at compressive sites such as an ulnar neuropathy at the elbow (Mattiassich et al., 2012). Thoracic neuropathy or radiculopathy results in paresthesias throughout the affected portion of the chest and abdomen, and in some cases can present with a band or girdle-like sensation similar to that reported by patients with multiple sclerosis lesions in the spinal cord (Burns et al., 2006; Reshad et al., 1993). Additional cranial nerve involvement has been reported in a number of cases (most commonly facial nerve palsy), but concomitant CNS involvement of the brain and spinal cord is rare (Burns et al., 2006; Ramos-Casals et al., 2021; Said et al., 2002; Zuniga et al., 1991). Finally, involvement of the phrenic nerve can lead to hypoventilation and respiratory insufficiency (Robinson et al., 1998).
Small fiber neuropathy (SFN) is a disorder of thinly myelinated A-delta and unmyelinated C nerve fibers, which mediate pain, temperature and autonomic function. Although large fiber involvement is rarely seen in patients with sarcoidosis, SFN has been reported in 32 to 44% of patients with systemic disease (Bakkers et al., 2009; Hoitsma et al., 2002; Hoitsma et al., 2003; Hoitsma et al., 2004). The onset of SSFN in most cases is within two years of the systemic diagnosis and primarily consists of sensory manifestations (e.g., pain, numbness and burning paresthesias) that are initially intermittent and migratory affecting the feet or fingers (Tavee & Culver, 2011; Tavee et al., 2017). Over time, the symptoms gradually progress to where they become constant with patients reporting fixed pain and paresthesias usually in the feet or hands with additional transient symptoms that occur in a non-length dependent distribution throughout the face, trunk or proximal limbs. Less commonly, there may be a subacute onset of symptoms in a stocking-glove distribution that quickly progresses to diffuse paresthesias. Pain and sensory symptoms can markedly affect patient quality of life and at times may be severe enough that they interfere with sleep, ambulation and activities of daily living. Dysautonomia is also common in SSFN and may occur either in conjunction with sensory symptoms or in isolation.
In one series of patients with painful SSFN, autonomic dysfunction was seen in 53% of cases with orthostatic intolerance and palpations as the most frequently reported symptoms followed by gastrointestinal disturbances (mainly bloating), hyperhidrosis and bowel/bladder dysfunction (Tavee et al., 2017). The neurologic examination can be normal in SSFN despite the degree of sensory loss and pain reported by the patient. In some cases, hyperesthesia, allodynia and sensory loss to pinprick and temperature may be seen in the distal legs or a patchy, non-dermatomal distribution. Reduced vibratory sense in the toes and an antalgic gait due to pain in the feet can also be seen, but proprioception, deep tendon reflexes and other large fiber modalities are usually preserved. In patients with autonomic dysfunction, the general examination may demonstrate hair loss (usually in the distal lower extremities where neuropathy symptoms are common), skin abnormalities (color changes or dryness) or orthostatic hypotension.
Neurosarcoidosis in general affects women more than men and is more common among African Americans and individuals of Scandinavian descent with peak incidence in the third to fifth decades (Baughman et al., 2001). In contrast, a number of small series have reported a slightly higher frequency of SGN in men compared to women and older age at presentation with mean ages ranging from 50 to 61 years (Burns et al., 2006; Mattiassich et al., 2012; Ramos-Casals et al., 2021; Said et al., 2002). With respect to SSFN, the age and gender of affected patients are similar to that seen with general neurosarcoidosis. In a large cohort of 115 patients with SSFN, the median age was 46 years with a 2:1 female predominance although Caucasians accounted for 87% of the patients while only 10% were African American (Tavee et al., 2017).
Characteristic non-caseating granulomas (NCG) can be seen in the peripheral nerves, usually in an epineural distribution although perineural and endoneural granulomas may also be present (Burns et al., 2006; Nemni et al., 1981; Oh, 1980; Said et al., 2002; Vital et al., 2008). While infiltration, compression and displacement of nerve fibers by the NCG have been observed in pathologic studies, the exact mechanism of a generalized polyneuropathy or polyradiculopathy remains unclear (Heaney et al., 2004; Oh, 1980; Said et al., 2002; Vital et al., 2008). Necrotizing vasculitis and microvasculitis without associated necrosis have been observed, suggesting nerve ischemia with resultant axonal degeneration as a potential etiology (Vital et al., 2008). Other pathologic findings on autopsy and biopsy studies include macrophage infiltration and perivascular inflammation within the nerves, spinal roots and ganglia that in some cases were seen in the absence of granulomas suggesting a more diffuse inflammatory process (Burns et al., 2006; Galassi et al., 1984; Mattiassich et al., 2012; Said et al., 2002).
The main histologic nerve feature of SGN is axonal degeneration of large myelinated fibers, which is concordant with EMG findings. (Burns et al., 2006; Nemni et al., 1981; Scott et al., 1993) Demyelinating pathology is usually secondary to axonal atrophy but may be the primary pathology when features such as segmental demyelination and conduction block are seen in the absence of significant axonal degeneration (Burns et al., 2006; Nemni et al., 1981; Vital et al., 1982).
In contrast, SSFN is considered a type of parasarcoidosis as the underlying pathophysiology is not due to granulomatous inflammation and remains unknown. However, a cytokine-mediated mechanism has been proposed based on studies evaluating painful small fiber neuropathy related to other disorders. In a 2010 study by Üceyler evaluating cytokines in idiopathic and diabetic small fiber neuropathy, increased gene expressions of interleukin-6 (IL-6) and IL-8 were seen in the skin of affected patients compared to unaffected skin and controls (Uceyler et al., 2010). Other studies evaluating painful diabetic and HIV neuropathy found that elevated levels of tumor necrosis factor-alpha (TNF-α), interleukin-1 beta (IL1-β) and transforming growth factor-β correlated with progression of nerve degeneration (Edwards et al., 2008; Herder et al., 2017; Kamerman et al., 2012; Mangus et al., 2014). IL-6 and TNF-α in particular may be relevant in the setting of SSFN as both play a critical role in promoting inflammation in systemic disease and are secreted by alveolar cells in higher levels in patients with sarcoidosis as compared to controls (Prior et al., 1996). Although there have been no studies evaluating how these cytokines may actually lead to small nerve fiber pathology specifically in sarcoidosis patients, in vitro studies have that shown that both IL-6 and TNF-α modulate the release of calcitonin gene-related peptide (CGRP), which may sensitize nociceptors to noxious stimuli (Opree & Kress, 2000; Sommer & Kress, 2004). TNF-α is also produced by injured Schwann cells and may directly lead to sensitization of nociceptors via an excitatory effect on A and C-fibers after nerve injury (Sch¨afers et al., 2003; Sommer & Kress, 2004). Further evidence of direct cytokine-mediated neuropathic changes are seen in rat models, in which IL-6 and TNF-α were both found to induce allodynia when directly injected into the intraplantar space, dorsal root ganglion (TNF-α only) and intrathecally (IL-6 only) (Cunha et al., 1992; Sommer & Kress, 2004). Finally, TNF-α antagonists have been shown to help reduce neuropathic pain and autonomic symptoms.
According to the 2018 Neurosarcoidosis Consortium Consensus Diagnostic Criteria, which were built upon Zajicek’s criteria from 1999 and now includes EMG/NCS to account for peripheral nerve involvement, the diagnosis of SGN is classified as definite, probable or possible based on the presence of confirmed granulomatous inflammation in the setting of clinical manifestations and diagnostic findings that are suggestive of neurosarcoidosis
The diagnosis is definite when there is histologic finding of granulomatous inflammation of the affected peripheral nerve, probable with histologic evidence of systemic involvement and possible when there is no histologic confirmation or other supportive evidence.
As the clinical and EMG patterns of nerve involvement in sarcoidosis are not distinct from non-sarcoidosis neuropathies, an extensive diagnostic evaluation including nerve biopsy is often needed to confirm sarcoidosis as the primary etiology and exclude other causes especially in cases with a subacute or rapidly progressive presentation and no prior diagnosis of sarcoidosis. A systemic evaluation should also be performed to look for extraneural involvement of sarcoidosis, which can help confirm the diagnosis if the nerve biopsy is indeterminate. In patients with known systemic disease who subsequently develop a neuropathy, sarcoidosis may be considered the underlying cause after all other etiologies have been reasonably excluded. However, if the neuropathy worsens or fails to respond to treatment, biopsy should be the next step.
Diagnostic Criteria for Central Nervous System and Peripheral Nervous System Neurosarcoidosis
Possible
1. The clinical presentation and diagnostic evaluation suggest neurosarcoidosis, as defined by the clinical manifestations and MRI, CSF, and/or EMG/NCS findings typical of granulomatous inflammation of the nervous system and after rigorous exclusion of other causes.
2. There is no pathologic confirmation of granulomatous disease.
Probable
1. The clinical presentation and diagnostic evaluation suggest neurosarcoidosis, as defined by the clinical manifestations and MRI, CSF, and/or EMG/NCS findings typical of granulomatous inflammation of the nervous system after rigorous exclusion of other causes.
2. There is pathologic confirmation of systemic granulomatous disease consistent with sarcoidosis.
Definite
1. The clinical presentation and diagnostic evaluation suggest neurosarcoidosis, as defined by the clinical manifestations and MRI, CSF, and/or EMG/NCS findings typical of granulomatous inflammation of the nervous system after rigorous exclusion of other causes.
2. The nervous system pathology is consistent with neurosarcoidosis.
Type a. Extraneural sarcoidosis is evident.
Type b. No extraneural sarcoidosis is evident (isolated CNS sarcoidosis).
Unlike affected CNS tissue, which is often difficult to obtain, peripheral nerves are readily accessible for biopsy. In patients with prominent lower extremity involvement, the sural or superficial peroneal nerve is typically sampled. Concomitant muscle biopsy of the peroneus brevis (when obtaining the peroneal nerve) or gastrocnemius (when obtaining the sural nerve) helps to increase the yield of tissue diagnosis as granulomas in-between muscle fibers tend to be larger and more developed than those found within the epineurium or endoneurium (Said et al., 2002; Vital et al., 2008). In the largest histopathological review of SGN, which included 38 patients, the sensitivity of nerve biopsy in detecting characteristic non-caseating granulomatous inflammation was 66% (n = 25) (Vital et al., 2008). This increased to 79% when muscle sampling was included (Vital et al., 2008). Muscle tissue can also be used to exclude mimics such as leprosy as nerve granulomas can be seen in both sarcoidosis and leprosy but muscle involvement is not seen in the latter (Said et al., 2002).
For more proximal nerve lesions affecting the nerve roots in isolation, the meninges over the roots and sometimes the nerve root itself may be sampled, although the risk of permanent neurologic deficits may be similar to that seen with obtaining CNS tissue. Histologic examination, cytology and microbial cultures should be performed on the biopsy specimens. On finding granulomatous inflammation in the nerve specimen, leprosy, syphilis, granulomatosis polyangiitis and lymphomatoid granulomatosis should be excluded. Additional DNA testing to search for evidence of tuberculous infection should also be considered.
Diagnostic testing ||| Measurement ||| Sensitivity in sarcoidosis
Skin biopsy ||| Nerve fiber quantification (skin) ||| 28 to 84%
Corneal confocal microscopy ||| Nerve fiber morphology (cornea) ||| 45%
Quantitative sensory testing (multiple modalities) ||| Sensory function (pain,heat, vibratory, other) ||| 69 to 84%
Pain-related evoked potentials Sensory function (pain) N/A*
Laser evoked potentials Sensory function (heat) N/A
Contact-heat evoked potentials Sensory function (heat) N/A
Thermoregulatory sweat test Sudomotor function N/A
Quantitative sudomotor axon reflex test (Tavee et al., 2017) ||| Sudomotor function ||| 69%
Sudoscan Sudomotor function N/A
Tilt table testing with cardiovagal maneuvers ||| Cardiac autonomic function ||| N/A
Cardiac scintigraphy with iodine-123 ||| MIBG** (Hoitsma et al., 2005) ||| Cardiac autonomic function ||| 75%***
For patients presenting with peripheral nerve symptoms, one of the most important tests is the EMG, which serves many purposes including characterization of fiber type involvement, determination of underlying nerve pathophysiology and lesion localization. Most cases of SGN are axonal, which may be manifested as reduced or absent sensory and motor nerve amplitudes on NCS and active denervation of affected muscles on needle examination. Demyelinating lesions may result in slowed conduction velocities, prolonged latencies, absent late responses and/or conduction block. In all cases, careful needle EMG of proximal muscles including the paraspinal muscles should be performed to evaluate for nerve root involvement and overall distribution (lengthdependent versus non-length dependent) of abnormalities. However, as noted above there is no specific EMG finding to suggest the diagnosis of sarcoidosis. As the EMG is designed to evaluate large nerve fibers, cases of SSFN will demonstrate no abnormalities.
There are currently no specific serologic markers indicating sarcoidosis as the cause of peripheral neuropathy. An elevated angiotension converting enzyme (ACE) level has been reported in only 18 to 28% of patients with biopsy confirmed SGN (Burns et al., 2006; Said et al., 2002). Conversely, in a study of 195 patients with idiopathic small fiber neuropathy in which 44% had an abnormal serum ACE level, no evidence of sarcoidosis was seen on additional testing that included CT of the chest (Lang et al., 2016). Given these findings and the lack of data to support its use in monitoring disease progression, the ACE is currently of limited clinical utility for sarcoidosis-related peripheral nerve involvement. However, an elevated level in subacute or atypical presentations of large fiber neuropathy may help raise the possibility of sarcoidosis in patients with no prior history of systemic disease.
On the other hand, serologic studies evaluating for more common etiologies of neuropathy in patients with or without a known diagnosis of sarcoidosis is crucial for proper diagnosis and treatment. In particular, evaluation for diabetes mellitus should be performed due to the widespread use of corticosteroids in this population. Screening for B12 deficiency should also be done as this is another possible cause of (or contributor to) neuropathy in sarcoidosis patients, especially those taking proton pump inhibitors and histamine-2 blockers, which can lead to increased gastric pH levels and reduced B12 absorption. In one series of 143 patients with small fiber neuropathy and a known diagnosis of systemic sarcoidosis, 28 (20%) were found to have other potential causes that included diabetes mellitus (most common), B12 deficiency and Sjogren’s syndrome (Tavee et al., 2017).
More recently, neural autoantibodies have been reported in patients with sarcoidosis neuropathy, although their clinical or pathologic significance is currently unknown. The most frequently reported are the anti-ganglionic acetylcholine receptor antibodies, which were found in patients with sarcoidosis and painful neuropathy or autonomic dysfunction (Lower et al., 2019; Oishi et al., 2021). Although some of the tests results may have been false positives, one study found autoantibodies in 22% of SSFN patients compared to only 6.5% of those with similar symptoms but no diagnosis of sarcoidosis (Oishi et al., 2021).
Also, anti-neurofascin (NF)-155 and NF-140 antibodies were reported in a patient who presented with a subacute polyradiculopathy and multiple cranial neuropathies mimicking GBS that was subsequently found tohave sarcoidosis. Further investigation may help to elucidate whether or not these auto-antibodies play a role in the development of neuropathy in patients with sarcoidosis.
In patients with proximal nerve root involvement, gadolinium enhanced MRI of the spine can sometimes reveal patchy, nodular enhancement or enlargement of nerve roots (Brier et al., 2020; Burns et al., 2006). With more distal involvement, e.g., lumbosacral plexopathy or sciatic neuropathy, MR neurography with high resolution T1-weighted sequences and fat-suppression can be helpful in delineating abnormalities of the nerve along its course through the limb (Moore et al., 2001). Similarly, peripheral nerve ultrasound may be used to detect more distal nerve abnormalities by demonstrating focal enlargements with a hypo-echoic appearance at affected sites of granulomatous compression or inflammation (Kerasnoudis et al., 2014; Kitaoji et al., 2021).
CSF findings in SGN are non-specific and may include a number of abnormalities. Most commonly reported is an elevated CSF protein, which was seen in 46 to 75% of cases in small case series (Burns et al.,2006; Said et al., 2002; Vital et al., 2008). Other abnormalities include a lymphocytic pleocytosis, low CSF glucose, an elevated IgG and the presence of oligoclonal bands (Burns et al., 2006; Oh, 1980; Said et al., 2002; Scott et al., 1993; Vital et al., 2008). The diagnostic value of a CSF ACE level in patients with SGN is unknown but is likely of low yield as has been noted with neurosarcoidosis in general.
Although there is not yet a gold standard for assessing small fiber neuropathy due to sarcoidosis or other disorders, there are a number of diagnostic tests of varying sensitivities that can be utilized to help to confirm the disorder. The most widely used test for diagnosing SSFN is a skin biopsy of the leg with quantification of intraepidermal nerve fiber density (IENFD) (Hoitsma et al., 2002). The sensitivity of IENFD in SSFN ranges from 28% when only the distal leg is sampled to 84% when the biopsy is obtained from both the distal leg and thigh (Bakkers et al., 2009; Oudejans et al., 2017; Tavee et al., 2017). This is likely due to the patchy, non-length dependent distribution frequently seen in SSFN. Quantitative sensory testing (QST) includes various modalities of testing (the temperature threshold test is most common) and is a simple but time-consuming technique that relies on subjective reporting of symptoms by the patient. With a sensitivity of 69 to 84% in SSFN, QST is a valuable diagnostic tool but is mainly used in research studies as it is rarely available outside of academic institutions and not reimbursable in the U.S. by many insurance companies (Hoitsma et al., 2003; Lefaucheur et al., 2015; Oudejans et al., 2017; Raasing et al., 2021). Corneal confocal microscopy is a non-invasive study that quantifies c0orneal fiber density, which has been found to correlate with IENFD and neuropathy symptoms (Chen et al., 2015; Oudejans et al., 2017; Tavakoli et al., 2010). The test is easy to perform and has been used in clinical trials for SSFN, but is still considered experimental and requires further investigation in the setting of small fiber neuropathy (Brines MS et al., 2013).
Autonomic studies may be used to evaluate for the presence and severity of autonomic dysfunction and can also serve as a complementary test alongside the skin biopsy to help establish the diagnosis of SSFN. These tests include quantitative sudomotor axon reflex testing (QSART), which measures sweat output and latency in response to acetylcholine iontophoresed onto the skin; sudoscan, which measures electrochemical skin conductance; tilt table testing with cardiovagal maneuvers; and cardiac scintigraphy with iodine-123 metaiodobenzylguanidine (I-123 MIBG), which assesses myocardial sympathetic nerve function (Hoitsma et al., 2005).
Finally, the Small Fiber Neuropathy Screening List, is a survey that was specifically developed and validated for SSFN for use in both the research setting and clinical practice (Hoitsma et al., 2011; Voortman et al., 2018).
As with CNS sarcoidosis, there are no prospective randomized controlled trials evaluating the efficacy of immunosuppressive therapies in SGN. Due to the rarity of the disease, treatment regimens are based on anecdotal reports and small case series and are not unique to SGN. The choice of medications is also guided by the clinical and electrodiagnostic pattern of the neuropathy, especially in the setting of atypical presentations such as multifocal motor neuropathy with conduction block. In those cases, the treatment is similar to those without sarcoidosis.
As reported anecdotally and in most series, the first line of treatment for SGN is corticosteroid therapy (Burns et al., 2006; Said et al., 2002; Scott et al., 1993; Tavee & Stern, 2015). Starting doses of oral prednisone tange from 40 to 60 mg daily although some clinicians prefer dosing regimens of 0.5 to 1.0 mg/kg per day. In cases with significant deficits or an acute to subacute onset of symptoms, pulse dose IV methylprednisolone 1 g daily for three to five days may be considered followed by maintenance therapy with oral prednisone. After one month, the dose may be tapered by 5 mg every 2 weeks (or 10 mg per month) as the patient’s functional status improves. However, in patients with chronic manifestations such as CIDP or refractory symptoms in the setting of active disease, low dose maintenance therapy with doses of 5 to 10 mg per day may be needed for prolonged periods of time.
Methotrexate, azathioprine, mycophenolate mofetil and other steroid-sparing agents used in CNS sarcoidosis may also be of benefit although there is very little data on specific treatments for SGN outside of corticosteroids. In those with mononeuritis multiplex or acute motor involvement refractory to corticosteroids, cyclophosphamide alone or in combination with corticosteroids can result in clinical improvement (Marques et al., 2014; Vital et al., 2008). Another option may be etanercept, a TNF-α antagonist, which in combination with corticosteroids resulted in complete motor recovery in one patient with recurrent mononeuritis multiplex and multisystemic sarcoidosis (Marques et al., 2014). In contrast to the favorable responses seen with SSFN, the effects of IVIG in the setting of SGN are mixed. Immediate improvement of neurologic symptoms was seen after IVIG treatment in one patient with sarcoidosis who had a progressive sensory axonal neuropathy although she required ongoing intermittent treatment over the next few years (Heaney et al., 2004). Another case report of a patient with multifocal motor nerve conduction block associated with sarcoidosis showed a beneficial response to IVIG (Kono et al., 2013). However, no change was seen in a patient with sarcoidosis and a sensory-dominant, vasculitic neuropathy characterized by mixed axonal and demyelinating features (Levy et al., 2005). As noted above, in cases of SGN with a GBS-like presentation, patients tended to respond to corticosteroids as well as IVIG (Fahoum et al., 2009).
Infliximab (IFX), another TNF-α antagonist, has been found to be helpful in SSFN (see below) and is one of the most promising therapies for CNS sarcoidosis. However, it has also been reported to cause a demyelinating polyneuropathy in a number of patients including one case with longstanding CNS sarcoidosis who developed a demyelinating neuropathy shortly after initiation of IFX that resolved only after its withdrawal (Alshekhlee et al., 2010; Lineback et al., 2019; Lozeron et al., 2009; Richette et al., 2004; Tektonidou et al., 2007). Similarly, thalidomide and leflunomide, which are used to treat pulmonary sarcoidosis, can cause a painful sensory neuropathy and should be avoided in patients with known peripheral nerve involvement.
Unlike systemic and CNS sarcoidosis, SSFN does not respond to commonly used immunosuppressants such as methotrexate or azathioprine, which may be related to its non-granulomatous pathology. Even corticosteroids are of limited benefit and may only be helpful in isolated cases such as those with an acute presentation (Saito et al., 2015). However, a number of studies have shown effectiveness of IVIG and TNF-α antagonists in reducing symptoms of pain and dysautonomia in patients with SSFN (Hoitsma et al., 2006b; Parambil et al., 2011; Tavee et al., 2017). In a retrospective series of SSFN (n = 115), subjective improvement of neurologic symptoms was reported by 47 of 62 (76%) patients who received IVIG, 8 of 12 (67%) patients who received anti-TNF antagonists and 10 of 14 (71%) patients who received combination therapy (Tavee et al., 2017). For most of the patients, the beneficial effects were noted in the first two weeks of therapy. Results of the study, however, were limited by the retrospective nature of the study, absence of a standardized assessment for treatment response, patient selection bias and lack of a placebo group.
A new compound that may be another option for treating SSFN is ARA-290 (Cibinetide®), an experimental erythropoietin derivative that activates the innate repair receptor and is involved in anti-inflammatory pathways and tissue repair (Brines & Cerami, 2005; Brines et al., 2015; Swartjes et al., 2014). Three randomized, double-blinded controlled trials evaluating the use of ARA-290 (cibenitide) in SSFN showed a reduction in SSFN-associated symptoms and increase in IENFD and corneal nerve fiber area (CNFA) after 28 days of treatment compared to placebo although the CNFA was reduced at follow-up (Culver et al., 2017; Dahan et al., 2013; Heij et al., 2012).
Supportive therapy is imperative for improving quality of life issues for both SSFN and SGN patients. Management of chronic neuropathic pain is the same as that for non-sarcoidosis neuropathy and may consist of medications such as antiepileptics (e.g., gabapentin, pregabalin and topiramate), antidepressants (e.g., amitriptyline, nortriptyline and duloxetine) and topical analgesics (e.g., capsaicin and lidocaine cream).
While these medications may provide some degree of pain relief, they have little to no effect on autonomic dysfunction and are often associated with adverse effects. Other options for pain control include cannabidiol products (non-inhaled forms are recommended for sarcoidosis patients) and over-the-counter supplements such as acetyl-L-carnitine and alpha lipoic acid, which are supported by limited data in the setting of chemotherapy and diabetic-induced neuropathy (Hoggart et al., 2015; Tavee, 2018; Ward et al., 2014). Integrative holistic treatments such as yoga and tai chi as well as psychological support may also be considered. For patients with weakness, appropriate bracing and physical and occupational therapy can optimize mobility and patient function.
Some patients with SGN may have a monophasic course with complete recovery after treatment (Ferriby et al., 2001). However, a protracted course with residual sensory deficits is seen in most cases although there is usually some degree of improvement with immunosuppressive therapy. Remitting and relapses cases have also been reported. Variables that have been associated with long term disability in SGN include number of CSF cells, time to treatment and severity of disability at presentation (Burns et al., 2006).
SSFN tends to be chronic with very few experiencing a complete recovery, while some patients report a remitting and relapsing course. SSFN can also persist for years despite remission of systemic disease. In one cohort of patients with SSFN who were followed for a mean of 31 months, 78% of those who received no treatment reported subjective worsening of symptoms, and only 19% improved or remained stable (Tavee et al., 2017).
Peripheral neuropathy in sarcoidosis may result in a number of clinical presentations depending on location of the lesion and fiber type involved, but the predominant symptoms in most cases are pain and paresthesias, which can greatly affect patient quality of life. In those with suspected sarcoidosis neuropathy, EMG and possibly nerve biopsy should be performed to confirm the disease. However, if the EMG is normal, the patient should undergo further evaluation for small fiber neuropathy with specialized testing to confirm the diagnosis as the treatment of SSFN differs from that in systemic and general neurosarcoidosis.
Finally, chronic pain, paresthesias and reduced mobility require supportive management in addition to immune-modulating therapies.