Neuro‐Sjögren: Uncommon or underestimated problem?

Abstract Objectives Sjögren's syndrome (SS) is a chronic inflammatory disease with an autoimmune background with possible complications from peripheral (PNS) and central nervous system (CNS). The aim of this study was to assess the prevalence and to describe the phenotype of peripheral neuropathies in patients with SS. Materials & Methods We studied fifty patients with primary Sjögren's syndrome for peripheral nervous system involvement. All patients underwent neurological and rheumatological examination followed by nerve conduction studies (NCS) of nine peripheral nerves. Results Thirty‐six patients (72%) fulfilled the criteria for the diagnosis of neuropathy. Carpal tunnel syndrome (54%) and axonal sensorimotor neuropathy (22%) were the most common. Neurological symptoms preceded the diagnosis of SS in eight patients. Conclusions Peripheral neuropathies are frequent in SS patients. Neurologists should be aware of possible autoimmune causes of neuropathies because clinical manifestations of neuropathy may precede the development of other symptoms of the autoimmune disease.

Peripheral neuropathy may be the first symptom, but may also develop later in the course of the disease. The following subtypes of peripheral polyneuropathy in SS were identified: three types of sensory neuropathies (sensory axonal, small-fiber, and sensory ataxic), axonal sensorimotor polyneuropathy, mononeuropathy, multiple mononeuropathy, demyelinating polyradiculoneuropathy, cranial neuropathy, and autonomic neuropathy. The clinical course of the subtypes of peripheral neuropathy vary widely, but all of them may lead to significant pain and deterioration of quality of life in patients with pSS (Padua et al., 2005;Sopacua et al., 2019).
The aim of this study was to assess the prevalence and to describe the phenotype of peripheral nervous system involvement in patients with pSS in tertiary care university hospital in Poland.
To our knowledge, this is the first study in this subject in Polish population.

| Patients
We studied a group of fifty consecutive patients with pSS from the Clinical Alliance from 2012 (Shiboski et al., 2012;Vitali, 2003).
Patients with additional connective tissue diseases (secondary SS), as well as patients with diabetes, were excluded. All patients were evaluated by rheumatologist, and all clinical data were collected during medical history.

| Neurological assessment and nerve conduction studies
All patients underwent clinical neurological examination including conventional investigation of sensory and motor functions, tendon reflexes, cranial nerve investigation, and nerve conduction studies (NCS). Neurological examination and NCS were performed and evaluated by one certified neurologist.
The NCS examinations were conducted following our standard laboratory methods in accordance with the recommended protocol of the American Association of Electrodiagnostic Medicine (AAEM) using Medtronic's Dantec Keypoint G4. Recordings were performed with skin temperature control 32-34°C. NCS was performed on nine peripheral nerves in each patient-peroneal, tibial, sural, median-bilaterally, and ulnar-unilaterally.
In motor NCS, we analyzed the following parameters: conduction velocity, distal latency, M-wave amplitude, F-wave latency, and frequency. In sensory NCS, conduction velocity and amplitude of sensory potentials were analyzed. The lower limits of normal values [sensory nerve and compound muscle action potential (SNAP and CMAP) amplitudes measured peak to peak] were determined from 50 age-matched healthy individuals.
For the classification of polyneuropathies into axonal, demyelinating, or mixed, we used the ESTEEM (European Standardized Telematic tool to Evaluate Electrodiagnostic Methods) guidelines (Tankisi et al., 2005).

| Laboratory data
Routine laboratory tests and following immunological parameters were recorded in all patients: serum protein electrophoresis; monoclonal gammopathy detected by immunofixation; C3 and C4 complement detected by nephelometry; serum cryoglobulins, measured after centrifugation and storage at 4°C for at least 72 hr; rheumatoid factor (RF) measured by nephelometry; ANA tested by IF using HEp-2 cell substrate; antibodies to anti-Ro/SS-A and anti-La/SS-B assayed by ELISA method or detected by Western blot; anti-Ro-52 (detected by Western blot); erythrocyte sedimentation rate (ESR); c-reactive protein (CRP); β-2-microglobulin; vitamin B12; fibrinogen; lactate dehydrogenase (LDH); creatine kinase (CK); and peripheral blood counts.
Normal ranges for local laboratory were taken into account.
Organ complications were mainly diagnosed based on the

ESSDAI.
For the needs of this study, some organ or system involvement, not included in ESSDAI, was defined as follows. Gastrointestinal tract involvement was defined as chronic atrophic gastritis, autoimmune hepatitis, or malabsorption syndrome after exclusion of accompanying celiac disease. Cardiovascular system involvement was defined as pericarditis or pulmonary arterial hypertension. CNS involvement was confirmed in imaging studies (computed tomography, magnetic resonance imaging, single-photon emission computed tomography) or cerebrospinal fluid analysis.
For the assessment of clinical disability due to peripheral neuropathy, the Overall Disability Sum Score (ODSS) scale was used (Merkies et al., 2002). The ODSS scale consists of a checklist for interviewing the patient and focuses on limb functions. The score ranges from 0 to 5 in the upper limb and from 0 to 7 in the lower limb. A score of 0 indicates no limitations, and highest scores (5 or 7 depending on section) indicate inability to perform a purposeful movement.
The Ethics Committee of the Medical University of Gdansk approved the study, and all patients gave their written informed consent to participate in the study.

| RE SULTS
Fifty subjects were included in the study, of which 48 were females aged 33 to 69 (average 53.6 ± 10.5 years) and two were males aged 42 and 67. The mean age at diagnosis of SS was 50.4 ± 14 years. 40 patients (80%) reported subjective symptoms indicative of peripheral nervous system involvement such as paresthesia or neuropathic pain. 36 patients (72%) met the criteria for the diagnosis of neuropathy. The distribution of subtypes of neuropathy is presented in Table 1.
For the needs of the present study, we included 23 of 50 (46%) patients with neuropathies other than carpal tunnel syndrome (CTS) or patients with additional CTS into the group with peripheral nervous system involvement (further referred to as PNS+).

| 13 of 50 (26%) patients had only CTS, and 14 of 23 (61%) PNS+ patients had additional CTS
A pure small-fiber neuropathy was suspected in three patients, but we did not have the possibility to conduct additional neurophysiological testing. All of them reported neuropathic pain present in lower and upper extremities. Two rated the pain at the level of 6 points on the VAS and one at the level of 7.
In one patient with normal NCS, the diagnosis of SFN was established based on abnormal clinical and quantitative sensory testing (QST) performed in another neurological laboratory.
The most common clinical presentations were paresthesias and neuropathic pain beginning in distal parts of extremities, sensory deficits in a "glove-stocking" distribution, and mild-to-moderate muscle weakness reported by patients. The onset was usually chronic or subacute, and the course was predominantly slowly progressive.
The neurological examination revealed sensory deficits (superficial or deep), as well as diminished or absent tendon reflexes in the affected limbs. Muscle weakness was usually mild with a distal-to-proximal pattern.
In one patient, ataxic sensory neuropathy led to severe loss of proprioception and kinesthesia resulting in significant disability.
CNS involvement was confirmed in 12 patients (24%), and the following clinical presentations were observed: multiple sclerosis-like syndrome, cerebral vasculitis, myelitis, meningoencephalitis, and cognitive impairment with abnormalities confirmed in the SPECT examination.

| Comparison of clinical and laboratory data between patients with and without peripheral nervous system involvement in pSS
Clinical data were compared between patients with peripheral nervous system involvement (PNS+, n = 23) and the remaining 27 patients (PNS−). Table 2 summarizes the comparison of the groups, and Table 3 presents the laboratory data.
The mean time between the onset of sicca symptoms and the diagnosis of pSS was 7,7 ± 5,16 years and 6,81 ± 7,2 years in the PNS+ and PNS− groups, respectively (p = .229).
Those patients were also more often treated with cyclophosphamide (p < .05).
Considering the laboratory results, the groups differed only in the incidence of hypocomplementemia (26% vs. 7% p < .05). There was no statistically significant difference between the two groups in involvement of other organs, laboratory data, or disease duration.
Although lymphadenopathy was comparable between the subgroups, of the two patients with lymphoma (4% of our group), both had sensorimotor neuropathy. Our study shows a high prevalence (46%) of peripheral nervous system involvement in a cohort of patients with pSS in a single hospital. Among polyneuropathies, sensorimotor neuropathies were the most common. SS develops in 0.2 to 1% of the population, and it seems to be as frequent as rheumatoid arthritis. In previous studies, the peripheral nervous system involvement was observed in 2%-60% of SS patients and sensory neuropathies were most common ( Andonopoulos et al., 1990;Delalande et al., 2004;Grant, Hunder, Homburger, & Dyck, 1997;Mori et al., 2005).

| D ISCUSS I ON
It is worth noting that 80% of our subjects reported subjective symptoms suggestive of polyneuropathy, while final diagnosis was made only in 46%. That means that in clinical settings, diagnosis of SS-related polyneuropathy should not be based solely upon history or questionnaires.
Some of the previous studies suggest that nervous system involvement can be the presenting feature of SS, while others report that it is rather a late finding in the course of the disease (Govoni et al., 1999;Grant et al., 1997;Griffin et al., 1990;Malinow et al., 1986;Mori et al., 2005;Vrethem et al., 1990 Previous studies suggest an association between neurological involvement and vasculitis, lung involvement, and constitutional symptoms such as fever and fatigue (Binder, Snaith, & Isenberg, 1988;Gono et al., 2011;Miwa et al., 2003;Teixeira et al., 2013).
Unlike other authors (Jamilloux et al., 2014;Miwa et al., 2003), we did not find specific immunological markers of B-cell activation or monoclonal proliferation between PNS+ and PNS− group. Our hypothesis is that sensorimotor neuropathies are the next step in the development of neuropathy in pSS, preceded by pure sensory neuropathies, and therefore, they are associated with a higher incidence of chronic B-cell activation corresponding to the duration of the disease.

| Treatment
Currently, there are no guidelines based on randomized trials regarding the treatment of patients with SS with nervous system involvement. Patients with oligosymptomatic and self-limiting course of the disease do not require pharmacological treatment.
However, in case of high disease activity and the detected progression of lesions, immunosuppressive treatment is applied. The use of corticosteroids and intravenous cyclophosphamide pulses is most often recommended. Cyclosporine, azathioprine, and methotrexate also seem to be effective (Govoni, Padovan, Rizzo, & Trotta, 2001;Soliotis, Mavragani, & Moutsopoulos, 2004).
In the treatment of neuropathic pain, antiepileptic drugs, tricyclic antidepressants, and opioids are used (Hoitsma et al., 2004).
Our study has some limitations. Main limitation is the small number of patients with pSS. Secondly, this study was conducted in a tertiary care university hospital; thus, referral bias may have played a role in patient selection. Additionally, we did not have the possibility to conduct neurophysiological testing in cases where a pure small-fiber neuropathy was suspected.
The advantage is that our study, unlike many others, is a prospective study with a thorough neurological and rheumatological assessment performed by one certified specialist. We also excluded patients with additional connective tissue diseases, as well as patients with diabetes, which may be a potential complication of steroid therapy.
In conclusion, involvement of the PNS in SS patients seems frequent but remains underestimated. Rheumatologists should be more aware of the risk of neurological complications in pSS patients.
Neurologists should be aware of possible autoimmune causes of neuropathies because clinical manifestations of neuropathy may precede the development of other symptoms of the autoimmune disease.
NCS is a noninvasive procedure, which should be used in diagnostics and follow-up in this group of patients.
Guidelines for the diagnostics and treatment of peripheral neuropathies in patients with pSS are needed.

ACK N OWLED G EM ENTS
As corresponding author, I confirm that the manuscript has been read and approved for submission by all authors and all authors meet the journal's criteria for authorship.

CO N FLI C T O F I NTE R E S T
None of the authors has any conflict of interest to disclose.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data are available on request due to privacy/ethical restrictions.