EPIDERMAL AXONAL SWELLINGS IN PAINFUL AND PAINLESS DIABETIC PERIPHERAL NEUROPATHY AUDREY CHEUNG, BSc, PETER PODGORNY, BSc, JOSE A. MARTINEZ, MSc, CYNTHIA CHAN, MSc, and CORY TOTH, MD Department of Clinical Neurosciences, HMRB 155, Foothills Hospital, University of Calgary, Hotchkiss Brain Institute, 3330 Hospital Drive NW, Calgary, Alberta T2N 4N1, Canada Accepted 28 July 2014 ABSTRACT: Introduction: The pathophysiology of neuropathic pain (NeP) in diabetic peripheral neuropathy (DPN) is unclear. A potential pathological feature associated with intraepidermal nerve fiber density (IENFD) loss in DPN is axonal swellings. Methods: We determined the prevalence of intraepidermal axonal swellings in DPN patients with or without NeP and compared the findings with diabetes patients without DPN, patients with idiopathic neuropathy with NeP, and control subjects. The primary outcome measure was the ratio of axonal swellings to IENFD. Secondary outcome measures included clinical neuropathy severity and assessment for messenger RNA for voltagegated sodium and calcium channels. Results: IENFD was depressed in DPN (with/without pain) and in idiopathic neuropathy patients. Axonal swelling ratios were similar for DPN subjects with and without pain. There was no overexpression of voltage-gated ion channels in epidermis from DPN patients. Clinical neuropathy severity was only related to IENFD. Conclusions: There was no clinical relationship to pain or clinical neuropathy severity for axonal swellings in DPN. Muscle Nerve 51: 505–513, 2015
Diabetic peripheral neuropathy (DPN) occurs in up to 50% of patients with diabetes mellitus (DM) and leads to sensory, motor, and/or autonomic dysfunction.1 Despite our knowledge of potential causes,2 trials of treatment to date have remained negative.1 As a result, most treatment of DPN centers on management of neuropathic pain (NeP), a prevalent symptom3 that can reduce quality of life severely.4 Patients with DPN present with both positive symptoms (allodynia, paresthesias, and hyperpathia) and negative symptoms (numbness, insensitivAdditional Supporting Information may be found in the online version of this article. Abbreviations: ANOVA, analysis of variance; BM, basement membrane; CaV, voltage-gated calcium channel; DEPC, diethylpyrocarbonate; DN4Q, Douleur Neuropathique 4 Questionnaire; DM, diabetes mellitus; DPN, diabetic peripheral neuropathy; DPN-NoP, diabetic peripheral neuropathy without neuropathic pain; DPN-P, diabetic peripheral neuropathy associated with neuropathic pain; GAP-43, growth-associated protein-43; HbA1C, hemoglobin A1C; HPRT, hypoxanthine–guanine phosphoribosyltransferase; IENF, intraepidermal nerve fiber; IENFD, intraepidermal nerve fiber density; IF, immunofluorescence; NaV, voltage-gated sodium channel; NeP, neuropathic pain; PCR, polymerase chain reaction; PLP, periodatelysine-paraformaldehyde; PGP, protein gene product; qRT-PCR, quantitative real-time reverse transcriptase–polymerase chain reaction; TCSS, Toronto Clinical Scoring System; THSD, Tukey honestly-significantdifference test; UENS, Utah Early Neuropathy Scale; VAS, visual analog scale Key words: axonal swellings; diabetic peripheral neuropathy; epidermal skin biopsy; neuropathic pain; type 2 diabetes mellitus This work was supported by a grant from the Alberta Heritage Medical Research Foundation (to C.T.). Correspondence to: C. Toth; e-mail: [email protected] C 2014 Wiley Periodicals, Inc. V
Published online 1 August 2014 in Wiley Online Library (wileyonlinelibrary. com). DOI 10.1002/mus.24351
Epidermal Axonal Swellings in DPN
ity to temperature, and painful stimuli). These symptoms vary between patients; the presence of NeP also has a variable and inconsistent presentation and is present in up to half of DPN patients who present to tertiary-care centers.3 Other than reporting of symptoms and examination of findings related to NeP, it is not possible to differentiate patients who have DPN associated with NeP (DPN-P) from those with DPN without NeP (DPNNoP). Performance of traditional testing, including nerve conduction studies or assessment of peripheral nerve biopsy, fails to differentiate DPN-P from DPN-NoP. The identification of a biomarker that could differentiate between DPN-P and DPN-NoP is of interest, as it could provide clues regarding the pathophysiology of NeP. In patients with peripheral neuropathy, intraepidermal nerve fiber density (IENFD) is decreased in most cases and correlates with both the presence and severity of peripheral neuropathy.5,6 Evaluation of epidermal biopsies can identify morphological changes beyond fiber loss, including swellings of axons.5 Although these axonal swellings are observed in normal humans,7 they may also precede or even predict loss of IENFD.8,9 Such axonal swellings demonstrate watery axoplasm, a dilated vesicular appearance, particulate organelles, and granular debris,8 with fewer neurofilaments than intact axons, but continued ability to express protein gene product (PGP) and tubulin. The reasons why axonal swellings develop are unclear, but they may be related to dysfunctional axonal transport or cytoskeletal elements10 that lead to microtubule accumulation. Degradation of these dysfunctional cytoskeletal components may lead to appearance of multiple organelles in axonal swellings.8,11,12 Because axonal swellings are present early in the course of DPN in patients with predominantly positive sensory symptoms, it is possible that they may act as pain generators via ephaptic transmission.13 In addition, heightened expression of voltage-gated ion channels along the length of the axons, which are postulated to play an important role in NeP development, may be present within such axonal swellings.14 As such, the expression of both voltagegated sodium and calcium channels, often implicated in conditions related to presence of peripheral NeP,15,16 was also sought. MUSCLE & NERVE
April 2015
505
We hypothesized that epidermal axonal swellings would be more prevalent in patients with DPN-P when compared with DPN-NoP. We performed a cohort study to quantify epidermal axonal swellings using skin biopsies from patients with either DPN-P or DPN-NoP as a function of IENFD. Results from patients with DPN (with NeP and without) were compared with those from subjects with DM without DPN, subjects diagnosed to have idiopathic peripheral neuropathy with associated NeP, and control subjects without DM or peripheral neuropathy. Concurrently, we analyzed epidermal samples for messenger RNA for sodium and calcium channels, which may play roles in the development of NeP.15,17–20 METHODS Participant Recruitment.
The University of Calgary Centre for Advancement of Health provided ethics approval for this study. Subject recruitment took place between March 2010 and January 2012 using poster recruitment within the Neuromuscular, Neuropathic Pain, and Endocrinology Clinics. All subjects provided informed written consent prior to participation. We assessed subjects with and without DM. Subjects with a diagnosis of pre-existing type 1 or type 2 DM were included after confirmation of the diagnosis as established by Canadian Diabetes Association guidelines. The presence of 2 prior separate positive laboratory results from the following criteria was required: fasting glucose results of 7.1 mmol/L (126 mg/dl) or 2 oral glucose tolerance tests leading to a 2-hour serum glucose of 11.1 mmol/L (200 mg/dl); or a random glucose of 11.1 mmol/L (200 mg/dl) at any time. The age at diagnosis of DM and the duration of symptoms of DPN (if present) were recorded. In subjects with either DPN or idiopathic peripheral neuropathy, assessment for other systemic illnesses, prior or current alcohol dependence (based on criteria from the Diagnostic and Statistical Manual of Mental Disorders, 4th ed.), toxin and medication exposures, and any family history of peripheral neuropathy was performed to exclude other potential causes of peripheral neuropathy. We also recorded use of anti-DM and other medications (including those used for chronic pain management associated with DPN). The presence of diabetic complications other than DPN was also noted. Subjects with possible DPN were excluded if potential causes for peripheral neuropathy other than DM were identified. Subjects with impaired glucose tolerance or impaired fasting glucose only were excluded. Lastly, subjects were excluded if they refused skin biopsy or laboratory testing. Subjects diagnosed with idiopathic peripheral neuropathy were included based on negative labo506
Epidermal Axonal Swellings in DPN
ratory testing for other potential causes of peripheral neuropathy, including, but not limited to, laboratory testing, as described previously.21 Control subjects were age-matched and recruited from friends and relatives of patient subjects without DM or known peripheral neuropathy, as well as from staff working at each of the clinics where the study was conducted.
Subject Assessment and Allocation. All subjects had neurological examinations with scoring for presence and severity of peripheral neuropathy using the Toronto Clinical Scoring System (TCSS) and the Utah Early Neuropathy Scale (UENS) performed by an unblinded neuromuscular disease specialist. After completion of clinical scales, subjects were categorized based on having 1 of the following: absence of peripheral neuropathy (control subject or DM-only subject, both TCSS 3 and UENS 3); or presence of peripheral neuropathy (idiopathic neuropathy subject or DPN subject, both TCSS >5 and UENS >6). Any subjects with a TCSS score of 4 or 5 and/or a UENS score of 4–6 were excluded due to ambiguity about the presence or absence of peripheral neuropathy. In addition, control subjects were excluded from participation if they had TCSS >4 and/or UENS >3. Subjects with DPN and idiopathic neuropathy were categorized further for the presence or absence of NeP using the question “Do you have pain on a daily or near-daily basis?” For subjects answering “yes,” the Douleur Neuropathique 4 Questionnaire (DN4Q) was completed; this categorizes pain to be neuropathic or non-neuropathic in nature with strong specificity (89.9%) and sensitivity (82.9%).22 DPN subjects who scored 4 on the DN4Q were categorized to have DPN with NeP (DPN-P), whereas subjects who denied having pain or discomfort (answering “no”) were categorized as DPN without NeP (DPN-NoP). We only included idiopathic neuropathy subjects who had daily pain and a DN4Q score 4, excluding any subjects with idiopathic peripheral neuropathy without NeP for the study. If DPN or idiopathic neuropathy subjects admitted to having pain, but had a score of 4 and
Diabetic peripheral neuropathy (DPN) occurs in up to 50% of patients with diabetes mellitus (DM) and leads to sensory, motor, and/or autonomic dysfunction.1 Despite our knowledge of potential causes,2 trials of treatment to date have remained negative.1 As a result, most treatment of DPN centers on management of neuropathic pain (NeP), a prevalent symptom3 that can reduce quality of life severely.4 Patients with DPN present with both positive symptoms (allodynia, paresthesias, and hyperpathia) and negative symptoms (numbness, insensitivAdditional Supporting Information may be found in the online version of this article. Abbreviations: ANOVA, analysis of variance; BM, basement membrane; CaV, voltage-gated calcium channel; DEPC, diethylpyrocarbonate; DN4Q, Douleur Neuropathique 4 Questionnaire; DM, diabetes mellitus; DPN, diabetic peripheral neuropathy; DPN-NoP, diabetic peripheral neuropathy without neuropathic pain; DPN-P, diabetic peripheral neuropathy associated with neuropathic pain; GAP-43, growth-associated protein-43; HbA1C, hemoglobin A1C; HPRT, hypoxanthine–guanine phosphoribosyltransferase; IENF, intraepidermal nerve fiber; IENFD, intraepidermal nerve fiber density; IF, immunofluorescence; NaV, voltage-gated sodium channel; NeP, neuropathic pain; PCR, polymerase chain reaction; PLP, periodatelysine-paraformaldehyde; PGP, protein gene product; qRT-PCR, quantitative real-time reverse transcriptase–polymerase chain reaction; TCSS, Toronto Clinical Scoring System; THSD, Tukey honestly-significantdifference test; UENS, Utah Early Neuropathy Scale; VAS, visual analog scale Key words: axonal swellings; diabetic peripheral neuropathy; epidermal skin biopsy; neuropathic pain; type 2 diabetes mellitus This work was supported by a grant from the Alberta Heritage Medical Research Foundation (to C.T.). Correspondence to: C. Toth; e-mail: [email protected] C 2014 Wiley Periodicals, Inc. V
Published online 1 August 2014 in Wiley Online Library (wileyonlinelibrary. com). DOI 10.1002/mus.24351
Epidermal Axonal Swellings in DPN
ity to temperature, and painful stimuli). These symptoms vary between patients; the presence of NeP also has a variable and inconsistent presentation and is present in up to half of DPN patients who present to tertiary-care centers.3 Other than reporting of symptoms and examination of findings related to NeP, it is not possible to differentiate patients who have DPN associated with NeP (DPN-P) from those with DPN without NeP (DPNNoP). Performance of traditional testing, including nerve conduction studies or assessment of peripheral nerve biopsy, fails to differentiate DPN-P from DPN-NoP. The identification of a biomarker that could differentiate between DPN-P and DPN-NoP is of interest, as it could provide clues regarding the pathophysiology of NeP. In patients with peripheral neuropathy, intraepidermal nerve fiber density (IENFD) is decreased in most cases and correlates with both the presence and severity of peripheral neuropathy.5,6 Evaluation of epidermal biopsies can identify morphological changes beyond fiber loss, including swellings of axons.5 Although these axonal swellings are observed in normal humans,7 they may also precede or even predict loss of IENFD.8,9 Such axonal swellings demonstrate watery axoplasm, a dilated vesicular appearance, particulate organelles, and granular debris,8 with fewer neurofilaments than intact axons, but continued ability to express protein gene product (PGP) and tubulin. The reasons why axonal swellings develop are unclear, but they may be related to dysfunctional axonal transport or cytoskeletal elements10 that lead to microtubule accumulation. Degradation of these dysfunctional cytoskeletal components may lead to appearance of multiple organelles in axonal swellings.8,11,12 Because axonal swellings are present early in the course of DPN in patients with predominantly positive sensory symptoms, it is possible that they may act as pain generators via ephaptic transmission.13 In addition, heightened expression of voltage-gated ion channels along the length of the axons, which are postulated to play an important role in NeP development, may be present within such axonal swellings.14 As such, the expression of both voltagegated sodium and calcium channels, often implicated in conditions related to presence of peripheral NeP,15,16 was also sought. MUSCLE & NERVE
April 2015
505
We hypothesized that epidermal axonal swellings would be more prevalent in patients with DPN-P when compared with DPN-NoP. We performed a cohort study to quantify epidermal axonal swellings using skin biopsies from patients with either DPN-P or DPN-NoP as a function of IENFD. Results from patients with DPN (with NeP and without) were compared with those from subjects with DM without DPN, subjects diagnosed to have idiopathic peripheral neuropathy with associated NeP, and control subjects without DM or peripheral neuropathy. Concurrently, we analyzed epidermal samples for messenger RNA for sodium and calcium channels, which may play roles in the development of NeP.15,17–20 METHODS Participant Recruitment.
The University of Calgary Centre for Advancement of Health provided ethics approval for this study. Subject recruitment took place between March 2010 and January 2012 using poster recruitment within the Neuromuscular, Neuropathic Pain, and Endocrinology Clinics. All subjects provided informed written consent prior to participation. We assessed subjects with and without DM. Subjects with a diagnosis of pre-existing type 1 or type 2 DM were included after confirmation of the diagnosis as established by Canadian Diabetes Association guidelines. The presence of 2 prior separate positive laboratory results from the following criteria was required: fasting glucose results of 7.1 mmol/L (126 mg/dl) or 2 oral glucose tolerance tests leading to a 2-hour serum glucose of 11.1 mmol/L (200 mg/dl); or a random glucose of 11.1 mmol/L (200 mg/dl) at any time. The age at diagnosis of DM and the duration of symptoms of DPN (if present) were recorded. In subjects with either DPN or idiopathic peripheral neuropathy, assessment for other systemic illnesses, prior or current alcohol dependence (based on criteria from the Diagnostic and Statistical Manual of Mental Disorders, 4th ed.), toxin and medication exposures, and any family history of peripheral neuropathy was performed to exclude other potential causes of peripheral neuropathy. We also recorded use of anti-DM and other medications (including those used for chronic pain management associated with DPN). The presence of diabetic complications other than DPN was also noted. Subjects with possible DPN were excluded if potential causes for peripheral neuropathy other than DM were identified. Subjects with impaired glucose tolerance or impaired fasting glucose only were excluded. Lastly, subjects were excluded if they refused skin biopsy or laboratory testing. Subjects diagnosed with idiopathic peripheral neuropathy were included based on negative labo506
Epidermal Axonal Swellings in DPN
ratory testing for other potential causes of peripheral neuropathy, including, but not limited to, laboratory testing, as described previously.21 Control subjects were age-matched and recruited from friends and relatives of patient subjects without DM or known peripheral neuropathy, as well as from staff working at each of the clinics where the study was conducted.
Subject Assessment and Allocation. All subjects had neurological examinations with scoring for presence and severity of peripheral neuropathy using the Toronto Clinical Scoring System (TCSS) and the Utah Early Neuropathy Scale (UENS) performed by an unblinded neuromuscular disease specialist. After completion of clinical scales, subjects were categorized based on having 1 of the following: absence of peripheral neuropathy (control subject or DM-only subject, both TCSS 3 and UENS 3); or presence of peripheral neuropathy (idiopathic neuropathy subject or DPN subject, both TCSS >5 and UENS >6). Any subjects with a TCSS score of 4 or 5 and/or a UENS score of 4–6 were excluded due to ambiguity about the presence or absence of peripheral neuropathy. In addition, control subjects were excluded from participation if they had TCSS >4 and/or UENS >3. Subjects with DPN and idiopathic neuropathy were categorized further for the presence or absence of NeP using the question “Do you have pain on a daily or near-daily basis?” For subjects answering “yes,” the Douleur Neuropathique 4 Questionnaire (DN4Q) was completed; this categorizes pain to be neuropathic or non-neuropathic in nature with strong specificity (89.9%) and sensitivity (82.9%).22 DPN subjects who scored 4 on the DN4Q were categorized to have DPN with NeP (DPN-P), whereas subjects who denied having pain or discomfort (answering “no”) were categorized as DPN without NeP (DPN-NoP). We only included idiopathic neuropathy subjects who had daily pain and a DN4Q score 4, excluding any subjects with idiopathic peripheral neuropathy without NeP for the study. If DPN or idiopathic neuropathy subjects admitted to having pain, but had a score of 4 and
