A Systematic Review of the Diagnostic Methods of Small Fiber Neuropathies in Rehabilitation

This systematic review describes the several methods to diagnose and measure the severity of small fiber neuropathies and aims to guide the physician to define all the diagnostic approaches for adopting the best strategies described in the current literature. The search was conducted in PubMed, EMBASE, Cochrane Library and Web of Science. Two reviewers independently reviewed and came to consensus on which articles met inclusion/exclusion criteria. The authors excluded all the duplicates, animals’ studies, and included the English articles in which the diagnostic measures were finalized to assess the effectiveness of rehabilitation and pharmacologic treatment of patients with small fiber neuropathies. The search identified a total of 975 articles with the keywords “small fiber neuropathy” AND “rehabilitation” OR “therapy” OR “treatment”. Seventy-eight selected full-text were analyzed by the reviewers. Forty-one publications met the inclusion criteria and were included in the systematic review. Despite the range of diagnostic tools for the assessment of small fiber neuropathy, other robust trials are needed. In addition, always different diagnostic approaches are used, a unique protocol could be important for the clinicians. More research is needed to build evidence for the best diagnostic methodologies and to delineate a definitive diagnostic protocol.


Small Fibre Neuropathy
Small fiber neuropathy (SFN) is caused by impairment of unmyelinated C and thinly myelinated Aδ fibers. The symptoms are characterized by sensory symptoms, pain and autonomic symptoms, such as palpitations, gastrointestinal disturbances, and orthostatic dizziness. The symptoms and signs can be present as spontaneous (e.g., burning, deep, itching and paroxysmal) or evoked (e.g., thermal allodynia, light tough allodynia and hyperalgesia) pain.
The diagnostic criteria for small fiber neuropathy are not established, influencing the approach to patients in clinical practice [1].
Nerve biopsy, genetic testing and quantitative sensory testing (QST) permit the definitive diagnosis, due to limitation of clinical examination, needle electromyography and nerve conduction.
The sensitivity (78-92%) and specificity (65-90%) of skin biopsy for diagnosing a SFN is high [2]. The punch skin biopsies could identify decreased intra-epidermal nerve-fiber density (IENFD) of unmyelinated nerves. QST could be a useful screening test for small and large fiber neuropathies [3]. The combination of clinical signs and abnormal QST and/or IENFD findings is a more reliably diagnostic Our systematic review defined the several methods to assess SFN and to guide the physician to delineate a diagnostic protocol adopting the best strategies described in the current literature. Our guide could help the multidisciplinary team to measure, objectively and easily, the severity of SFN and to assess the disorder. The current literature did not describe a unique diagnostic protocol and use arbitrarily, several methods. A diagnostic protocol should make this more objective, reproducible, and repeatable by the multidisciplinary team.

Search Strategy
The search was carried out on the following medical electronic databases: PubMed, EMBASE, Cochrane Library and Scopus Web of Science. The review was conducted from 22 May to 1 July 2020.

Selection Criteria and Data Extraction
Studies considered for this review have to include the diagnostic methods in patients with SFN. We included English original articles about diagnostic tools useful to determine the severity of SFN after therapy. We excluded animal studies, participants with other neuropathies. We also excluded all of the remaining duplicates ( Figure 1).
Diagnostics 2020, 10, x FOR PEER REVIEW 3 of 19 Our systematic review defined the several methods to assess SFN and to guide the physician to delineate a diagnostic protocol adopting the best strategies described in the current literature. Our guide could help the multidisciplinary team to measure, objectively and easily, the severity of SFN and to assess the disorder. The current literature did not describe a unique diagnostic protocol and use arbitrarily, several methods. A diagnostic protocol should make this more objective, reproducible, and repeatable by the multidisciplinary team.

Search Strategy
The search was carried out on the following medical electronic databases: PubMed, EMBASE, Cochrane Library and Scopus Web of Science. The review was conducted from 22 May to 1 July 2020.

Selection Criteria and Data Extraction
Studies considered for this review have to include the diagnostic methods in patients with SFN. We included English original articles about diagnostic tools useful to determine the severity of SFN after therapy. We excluded animal studies, participants with other neuropathies. We also excluded all of the remaining duplicates ( Figure 1). Two reviewers (C.R. and V.M.) independently screened the titles and abstracts from the initial search to identify relevant records and to identify eligible studies based on title and abstract. Selected full texts were then reviewed and included in the systematic review, following the PRISMA protocol [25] and in accordance with the PICOS criteria [26] (population, intervention, comparison, outcome, and study design) shown in Table 1: Participants were all patients affected by SFN; intervention was based on rehabilitation therapy or pharmacological approaches; the comparator was any comparator; the outcomes included clinical assessments, diagnostic scales, electromyography and nerve conduction, and biopsy; the study design was randomized controlled trial (RCTs), case series and case report retrospective studies. Two reviewers (C.R. and V.M.) independently screened the titles and abstracts from the initial search to identify relevant records and to identify eligible studies based on title and abstract. Selected full texts were then reviewed and included in the systematic review, following the PRISMA protocol [25] and in accordance with the PICOS criteria [26] (population, intervention, comparison, outcome, and study design) shown in Table 1: Participants were all patients affected by SFN; intervention was based on rehabilitation therapy or pharmacological approaches; the comparator was any comparator; the outcomes included clinical assessments, diagnostic scales, electromyography and nerve conduction, and biopsy; the study design was randomized controlled trial (RCTs), case series and case report retrospective studies. One patient showed no response to IVIG but good response to prednisolone. One patient had no significant improvement with prednisolone. One patient had gradual spontaneous recovery , deep breathing heart rate variability (DB-HRV), intraepidermal nerve fiber density (IENFD), corneal nerve fiber density (CNFD), corneal nerve branch density (CNBD), corneal nerve fiber length (CNFL), subcutaneous insulin infusion (CSII), daily insulin injection (MDI), complex regional pain syndrome (CRPS), plasma exchange (PE), enzyme replacement therapy (ERT), daily sleep interference score (DSIs), total symptom score (TSS), temperature threshold testing (TTT), neuropathy impairment score NIS, neuropathy impairment score-lower limbs (NIS-LL), chronic inflammatory demyelinating polyneuropathy (CIDP), pulsed electromagnetic field (PEMF), intravenous (iv), small-fiber neuropathy screening list (SFNSL), quantitative sensory testing (QST), burning mouth syndrome (BMS), verbal rating scale (VRS), insulin-like growth factor-I (IGF-I), sudomotor axon reflex testing (QSART), Michigan diabetic neuropathy screening instrument (MNSI).

Description of the Studies
From 1984 to 2019, the database searched of 975 articles with the keywords "small fiber neuropathy" AND "rehabilitation" OR "therapy" OR "treatment", whose titles and abstracts were screened by the reviewers. The papers remained for full text screening were 78 and the eligibility of the study inclusion was assessed independently. Forty-one publications met the inclusion criteria and were included in the systematic review. Thirty-seven were excluded for the following reasons: 18 involved individuals with different disorders from SFN, 7 examined different topics from our aim, 12 did not present any therapeutic procedure (Figure 1).
The qualitative information synthesis for each parameter was attributed to the following evidence levels according to the recommendations of the Oxford Centre for Evidence-Based Medicine: evidence from systematic review of randomized controlled trials (1a), clinical controlled studies (2a), case-control-studies (3a) and from non-systematic reviews [4] (Table 1).

Variations of Experimental Conditions across the Studies
The selected 41 articles were described on the basis of the several diagnostic methods used in each study for the assessment of SFN. Characteristics of the studies are shown in Table 1.
All study groups were not homogeneous for relevant general clinical features as clinical presentation, duration of disease and of the symptoms, kinds of diagnostic measures, severity of symptoms, time of starting therapy, duration of treatment, the follow-up period at the end of the therapy (Table 1).

Diagnostic Examination
We showed all the methods used for the diagnosis of SFN, found in the current literature. Most of the selected articles for the review were used skin biopsy for the definitive diagnosis and/or genetic tests [54,57] (Table 1). The skin biopsy was used alone [32,48,62] or in the most cases together with other diagnostic procedure, as nerve conduction examinations [33,41,56] or scales to assess the severity of the neuropathic symptoms (Table 1). Quantitative sensory testing (QST) with vibratory (VDT), cold (CDT), and heat-pain (HP) detection threshold testing were added for the specific diagnosis [30,40,61].

Diagnostic Guide and Clinical Consequences
SFN can be idiopathic or associated with other disorders. The symptoms worsen over time, but the progression is typically slow. The diagnostic process is often complex, also due to the differential diagnosis that pathology requires. According to our experiences, and supported by the literature (Table 2), specific scales are essential for quantifying the impairment and assessing the response to therapy and symptom modifications during follow up. QSART and sudoscan are very useful tools, especially at the beginning of the evaluation to evaluate the autonomic symptoms, which are very often present. In the general evaluation, it always seems extremely useful to include threshold and peripheral nerve conduction studies, to better define the characteristic of the SFN and exclude other concomitant causes. Genetic testing and corneal confocal microscopy are often used for diagnostic confirmation. Skin biopsy, simpler than nerve biopsy, is necessary for a definitive diagnosis.
Treatment of SFN certainly depends on the underlying cause, when detectable, but it is often limited to symptomatic therapy, which is also essential for improving adherence to rehabilitation treatment. The duration of treatment is based on the severity of the symptoms and the progression of the disease. It seems important to understand the complexity of this pathology in order to follow an adequate diagnostic procedure and to find the best therapeutic management to limit the progressive worsening of symptoms, which although generally slow is often present, and consequently the reduction in the quality of life.

Discussion
Our systematic review focused on the several measures useful for the examination of SFN severity after pharmacological or rehabilitative therapy. We realized a comprehensive overview to give a guide to ease the collaboration of a multidisciplinary team.

Comparing Studies: Diagnostic Tools
The definitive diagnosis is based on biopsy. Nerve conduction reveals no abnormality, but is mandatory as exclusion criteria (Table 1).
A blood investigation and the electrophysiological studies have the role to exclude other disorders. Anderson et al. [27] examined in their case report the complete blood count, electrolytes, calcium, magnesium, creatine kinase, thyroid-stimulating hormone, vitamin B12, hemoglobin A1c, fasting glucose, creatinine, urea, serum protein electrophoresis. The blood examination was normal.
Favoni et al. [37] assessed the role of antiganglioside antibodies in SFN. Van Velzen et al. [61] and Hilz et al. [40] used QST. It consists of a battery of psychophysical tests and the patient respond to a specific sensory stimulus to the skin [61]. The tests include cold and arm detection threshold (WDT), cold and warm pain threshold, paradoxical heat sensation, allodynia, and vibration detection threshold. Loss of function (i.e., an increased response threshold) for cold and WDT are indicative of SFN. More objective QST measures include laser-evoked potentials and contact heat-evoked potentials where a short stimulus result in activation of thermo-nociceptive cutaneous nerve fibers [61]. Hilz et al. [40] found that vibratory (VDT), cold (CDT), and heat-pain (HP) detection threshold testing adequately characterized Aß-, Aδ-, and C-fiber dysfunction in Fabry patients. Fewer patients had abnormal results of VDT, CDT, HP, and HP after and before therapy with ERT. The most had always had normal threshold. Van Velzen et al. [61] showed that ARA 290 increases sensory pain thresholds, cold pain threshold and warm pain threshold. Azmi et al. [30] assessed the severity of SFN with vibration perception threshold (VPT), cold threshold (CT), warm threshold (WT), neurophysiology, deep breathing heart rate variability (DB-HRV), intraepidermal nerve fiber density (IENFD), and corneal nerve fiber density (CNFD), branch density (CNBD), and fiber length (CNFL). Gaillet et al. [38] used a quantitative sensory testing at the four extremities with measurement of the average warm detection threshold (WDT) [68]. Namer et al. [51] used temperature thresholds and the genetic examination of the mutation of SFN. Hoitsma et al. [69] used the temperature threshold testing (TTT) for sensory fibers and cardiovascular autonomic testing for autonomic fibers, that resulted abnormal in their case report. In the study of Schiffmann et al. [57], the thermal thresholds remained unchanged after enzyme replacement therapy.

Conclusions
The diagnosis and the follow up of SFN is indispensable for the improvement of quality of life of the individuals with neuropathic symptoms. SFN has a negative psychosocial impact in the lives of the patients and of their families.
We performed a systematic review of the several methods present in the current literature for an accurate examination of SFN. We showed all the diagnostic methods described in the current literature to diagnose and follow the subjects with SFN. On the basis of the diagnostic methods, the physicians could obtain a guide and a common protocol for a multidisciplinary team. The accurate and repeatable assessments could improve the outcome of therapy approaches too. Our guide should help the multidisciplinary team to collaborate, to compare their own assessments with those of other members of the team, and to have more complete examinations. Despite the range of diagnostic tools for SFN, robust trials miss, and thus, different diagnostic approaches are to be used. More research is needed to build evidence for the best diagnostic methodologies and to delineate a definitive diagnostic protocol.