Sarcoid neuropathy: correlation of nerve ultrasound, electrophysiological and clinical findings.

JNS-13448; No of Pages 8 Journal of the Neurological Sciences xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns

Sarcoid neuropathy: Correlation of nerve ultrasound, electrophysiological and clinical findings A. Kerasnoudis ⁎, D. Woitalla, R. Gold, K. Pitarokoili, M.-S. Yoon Department of Neurology, St. Josef Hospital, Ruhr-University of Bochum, Germany

a r t i c l e

i n f o

Article history: Received 16 August 2014 Received in revised form 15 September 2014 Accepted 19 September 2014 Available online xxxx Keywords: Sarcoid neuropathy Nerve ultrasound Cross sectional area Electrophysiology Functional disability Echogenity

a b s t r a c t Introduction: We present the nerve ultrasound findings in sarcoid neuropathy and examine their correlation with electrophysiology and functional disability. Materials and methods: 40 healthy controls and 13 patients with sarcoid neuropathy underwent clinical, sonographic and electrophysiological evaluation, a mean of 2.1 years (SD ± 0.7) after disease onset. Results: Nerve ultrasound revealed significantly higher cross sectional area (CSA) values of the ulnar (elbow, p b 0.001), fibular (fibular head, p b 0.001), sural (between the lateral and the medial head of the gastrocnemius muscle, p b 0.001) and tibial nerves (ankle and popliteal fossa, p b 0.001), when compared to controls. The electroneurography documented significantly lower values of the 1) compound muscle action potentials (cMAPs) in the median, fibular and tibial nerves (p b 0.001), and 2) sensory nerve action potential (sNAP) in the median, ulnar and sural nerves (p b 0.001). A significant correlation between sonographic and electrophysiological findings in the group with sarcoid neuropathy was found only between cMAP and CSA of the ulnar nerve at the elbow (r = 0.894, p b 0.001). Neither nerve sonography nor electrophysiology correlated with functional disability. Discussion: Sarcoid neuropathy seems to show predominantly CSA enlargement in peripheral nerves of the lower extremities, without any significant correlation to electrophysiological findings. The electroneurography documented signs of sensorimotor axonal loss in various peripheral nerves. Neither nerve sonography nor electrophysiology correlated with functional disability. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Sarcoidosis is a polysystemic, granulomatous disorder mainly showing a pulmonary localization, while the skin, eyes, liver and lymph nodes may be also affected. 5% of patients with systemic sarcoidosis show neurological complications. In 20% of patients with neurosarcoidosis peripheral neuropathy is diagnosed, which is usually asymptomatic [3]. The most common types of sarcoid polyneuropathy are the sensorimotor and pure motor polyneuropathies and to a lesser degree the small fiber neuropathy, the acute inflammatory demyelinating polyradiculoneuritis and the lumbosacral plexopathy [15,24,31]. The definite diagnosis of sarcoid neuropathy is based on the histological demonstration of sarcoid granulomas in nerve biopsy specimens. Utrasound detection of asymmetric, inhomogenous increase in nerve cross sectional area (CSA) of peripheral nerves has been already reported in different immune-mediated neuropathies [6–8,11,19]. Various studies reported the absence of significant correlation between ⁎ Corresponding author at: Department of Neurology, Ruhr University, St. JosefHospital, Gudrunstr, 56, 44791 Bochum, Germany. Tel.: + 49 17662923235; fax: +49 2345093024. E-mail address: [email protected] (A. Kerasnoudis).

sonographic, electrophysiological (mainly compound muscle action potential) and clinical findings in inflammatory neuropathies [6–8,10], showing that each diagnostic method highlights different aspects of peripheral nerve impairment. We aimed to investigate the correlation between nerve sonography, nerve conduction studies, and functional disability in sarcoid neuropathy. 2. Material and methods 2.1. Subjects and patients The ethics committee of the Ruhr University in Bochum, Germany approved our study protocol and all patients with sarcoid neuropathy and healthy subjects signed informed consent. Prior to the study enrolment, all healthy subjects interested in participating in this study underwent nerve conduction studies (median, ulnar, sural, tibial and fibular nerves) on both sides. Healthy subjects having symptoms, clinical or electrophysiological signs referable to peripheral nerve disease were excluded from the study. Healthy subjects or patients having a history of diabetes or alcoholism were also excluded from the study. Overall, 40 healthy controls, aged over 18 years, were recruited in the study. In addition, 13 patients with sarcoid lesions in peripheral

http://dx.doi.org/10.1016/j.jns.2014.09.033 0022-510X/© 2014 Elsevier B.V. All rights reserved.

Please cite this article as: Kerasnoudis A, et al, Sarcoid neuropathy: Correlation of nerve ultrasound, electrophysiological and clinical findings, J Neurol Sci (2014), http://dx.doi.org/10.1016/j.jns.2014.09.033

2

A. Kerasnoudis et al. / Journal of the Neurological Sciences xxx (2014) xxx–xxx

nerve biopsy samples were enrolled in the study. At the time point of evaluation, all patients received as a therapy, either azathioprine (2 mg/kg body weight) or infliximab (5 mg/kg body weight). Patients, that received oral or intravenous corticosteroids during the last 3 months prior evaluation, were excluded from the study. 2.2. Ultrasound examination Ultrasonography was performed from a board certified neurologist (A.K.). All ultrasound studies have been performed with the use of an Aplio® XG ultrasound system (Toshiba Medicals, Tochigi, Japan). For the superficial nerves of the body (median, ulnar, radial, brachial plexus, tibial at the ankle, and sural) a 18-MHz linear array transducer was used, and for the deeper nerves (tibial and fibular in the popliteal fossa) a 12-MHz linear array transducer was used. The transducer was always kept perpendicular to the nerves. No additional force was applied other than the weight of the transducer and the extremities were kept in the neutral position to avoid causing any artificial nerve deformity. Cross sectional area measurements were performed at the inner border of the thin hyperechoic epineural rim by the continuous tracing technique and the average values were calculated after serially measuring three times. All peripheral nerves and brachial plexus were measured bilaterally in all healthy subjects and patients with sarcoid neuropathy at the following sites: the median nerve at the entrance to the carpal tunnel (retinaculum flexorum), forearm (15 cm proximal to the retinaculum flexorum), upper arm (middle of the distance between the medial epicondyle and axillary fossa), ulnar nerve at Guyon's canal, forearm (15 cm proximal to Guyon's canal), elbow (between the medial epicondyle and olecranon), upper arm (middle of the distance between the medial epicondyle and axillary fossa), radial nerve in the spiral groove, tibial nerve in the popliteal fossa and at the ankle, fibular nerve at the fibular head and in the popliteal fossa and sural nerve

(between the lateral and the medial head of the gastrocnemius muscle). The brachial plexus was also assessed in the supraclavicular (next to the subclavian artery) and interscalene spaces. After obtaining the CSA values in each predefined site of clinical interest, we performed an ultrasound scan of the complete course of the median and ulnar nerves from proximal (axillary fossa) to distal (carpal tunnel for the median nerve and Guyon's canal for the ulnar nerve), in order to measure the maximal and minimal cross sectional areas of each nerve. Maximal CSA was defined as: the site in the course of the nerve with the maximal area at the inner border of the thin hyperechoic epineural. Similarly, minimum CSA was defined as: the site in the course of the nerve with the minimal area at the inner border of the thin hyperechoic epineural rim. For the quantification of ultrasound findings we used the recently proposed measures in the literature [10–13,20]. Therefore, we calculated for each peripheral nerve and brachial plexus the intranerve-, internerve-, and intraplexus CSA variabilities and “side to side difference ratio of the intranerve CSA variability”. Due to anatomic limitations in the imaging of the nerves of the lower extremities (short visualizable length of the fibular and tibial nerves) we used for the calculation of the above measures, the CSA values acquired from the predefined sites of interest. All patients underwent sural nerve biopsy, so no documentation of the complete anatomic course of the sural nerve was done. Therefore, only the CSA of the sural nerve at one site is included in this study. Due to the short visualizable course of the brachial plexus as a unique entity, we used for the calculation of this measure the CSA values acquired from the supraclavicular and interscalene spaces. 2.3. Nerve conduction studies All the electrophysiological studies were performed from a board certified neurologist (M.-S. Y.) with the use of a Medtronic 4 canal electromyography device (Medtronic, Meerbusch, Germany). All testing

Table 1 Overview of the sonographic results in the neurosarcoidosis and control groups. Neurosarcoidosis (n = 13)

Age Nerve

Site

Median

Carpal tunnel Forearm Upper arm Intranerve CSA variability SSDIVA Guyon's canal Forearm Elbow Upper arm Intranerve CSA variability SSDIVA Spiral groove Supraclavicular space Interscalene space Intraplexus CSA variability SSDIVA Fibular head Popliteal fossa Intranerve CSA variability SSDIVA Popliteal fossa Ankle Intranerve CSA variability SSDIVA Between the lateral and the medial head of the gastrocnemius muscle Internerve CSA variability

Ulnar

Radial Brachial plexus

Fibular

Tibial

Sural

Controls (n = 40)

Upper cut-off

Mean

SD

Mean

SD

59.8

19.

53.2

12.1

9.78 6.85 8.92 1.56 1.23 5.42 5.57 8.42 7.28 1.80 1.33 4.28 56.92 33.21 2.00 1.40 9.71 9.07 1.52 1.36 13.50 11.42 1.50 1.32 2.42 2.24

4.80 1.79 2.80 0.33 0.44 1.69 1.86 2.59 2.30 0.65 0.44 1.89 19.91 12.22 0.82 0.32 3.64 1.77 0.30 0.25 4.75 3.83 0.43 0.21 0.75 0.49

8.37 6.62 8.28 1.10 1.22 5.12 5.30 5.41 6.60 1.52 1.23 3.21 46.80 30.25 1.59 1.20 6.86 8.49 1.31 1.17 8.40 6.31 1.39 1.29 1.83 1.71

2.10 1.60 2.55 0.15 0.14 1.31 1.22 1.40 1.61 0.48 0.21 1.48 13.27 10.10 0.37 0.15 2.05 2.14 0.32 0.25 2.52 1.42 0.28 0.44 0.41 0.57

p

0.15

12.57 9.82 13.38 1.40 1.50 7.74 7.80 8.21 9.80 2.48 1.65 6.17 73.34 50.45 2.33 1.5 10.96 12.77 1.95 1.67 13.44 9.15 1.95 2.61 2.65 2.85

0.143 0.661 0.442 b0.001 0.890 0.500 0.552 b0.001 0.282 0.097 0.271 0.039 0.047 0.387 0.015 0.003 b0.001 0.381 0.042 0.021 b0.001 b0.001 0.289 0.814 b0.001 b0.001

All values are calculated in mm2. CSA = cross sectional area, SD = standard deviation, SSDIVA = side to side difference ratio of the intranerve cross sectional area variability. p-Values 0.001 were considered as statistically significant and are highlighted with bold. Upper cut-off reference values of the CSA were defined as mean value + 2 standard deviation (SD).


-


3

was done while maintaining the skin temperature at 35.5 °C. Bilateral motor and sensory studies were performed in the median and ulnar nerves. In addition, bilateral motor studies were performed in the fibular and tibial nerves and unilateral sensory studies in the sural nerve. 2.4. Statistics The Anderson–Darling Normality Test was performed in both groups to check the distribution pattern of the samples. Both study groups showed a normal distribution pattern according to the Anderson–Darling Normality Test (sarcoid neuropathy p = 0.236 and control group p = 0.215), so that statistical comparison of groups and correlation analysis were performed with the help of Student's t test and Pearson's product-moment correlation coefficients for independent samples, using SPSS 17.0 for Windows. Due to the small study sample and the large number of sonographic and electrophysiological measurements performed in this study, a Bonferroni correction was performed, so that only p b 0.001 values were accepted as statistically significant. Upper cut-off reference values of the CSA were defined as mean value + 2 standard deviation (SD). Pathological CSA enlargement was defined as CSA N upper cut-off reference value (Table 1). 2.5. Study protocol The study was divided in two phases. On the first phase, all 13 patients with sarcoid neuropathy and healthy controls underwent a double blinded ultrasound and electrophysiological examination of their peripheral nerves. In addition, a documentation of the muscle strength, functional disability and fatigue severity of patients with sarcoid neuropathy was done from an experienced neurologist (K.P.) using the Medical Research Council (MRC) sum score [14], Rasch-built Overall Disability Scale (R-ODS) [28] and the modified Rasch-built Fatigue Severity Scale respectively [27]. On the second phase, a statistical comparison of the ultrasound and electrophysiological data acquired from the two groups was performed, in order to detect statistical significant differences.

Fig. 1. Overview of the three possible ultrasound patterns in patients with sarcoid neuropathy. (A) Axial scan of the median nerve in the upper arm showing a normal cross sectional area and echogenity of the nerve (BA = brachial artery, n.med re = right median nerve), (B) axial scan of the median nerve showing a cross sectional area enlargement with hypo- and hyperechoic patterns of edema (n.med re = right median nerve), (C) axial scan of the ulnar nerve in the upper arm showing a cross sectional area enlargement with normal echogenity (“hypertrophic nerve remodeling”) (uln re = right ulnar nerve).

3. Results

3.2. Intranerve-, internerve- and intraplexus cross sectional area variability

A total of 40 healthy subjects (mean age 53.2, SD +/− 12.1, 23 women) and 13 patients with sarcoid neuropathy (mean age 59.8, SD +/− 19.2, 8 women) underwent clinical, sonographic and electrophysiological evaluation, a mean of 2.1 years (SD +/− 0.7, min 6 months, max 5 years) after onset of symptoms indicating peripheral nerve disease. The two groups were matched for age (p = 0.150), height (p = 0.163) and weight (p = 0.239). The patients with sarcoid neuropathy showed a mean MRC Sum score of 38 (SD +/− 2.8), Rasch-built Overall Disability Scale score of 28.1 (SD +/− 4.2), and modified Rasch-built Fatigue Severity Scale of 17.1 (SD +/− 2.7).

The group with sarcoid neuropathy showed significantly higher intranerve cross sectional area variability values only of the median nerve (p b 0.001), compared to controls. In addition, higher values of the internerve CSA variability were documented (p b 0.001) (Table 1). No highly significant changes were noted in the intraplexus CSA variability (p = 0.015).

3.1. Cross sectional area and echogenity The group with sarcoid neuropathy showed significantly higher cross sectional area values of the ulnar (elbow, p b 0.001), fibular (fibular head, p b 0.001), tibial (ankle and popliteal fossa, p b 0.001) and sural nerves (p b 0.001), but not of the brachial plexus (supraclavicular space, p = 0.047; intrascalene space, p = 0.387), compared to controls (Table 1). Considering the echogenity pattern at anatomical sites of pathological peripheral nerve enlargement in neurosarcoidosis, we documented two different types: Type 1: increased CSA, due to edema, with a partial or complete loss of the fascicular echostructure (iso, hypo- or hyperechoic) (Fig. 1B), and Type 2: increased CSA, due to an increase of the size of the fascicles, but normal echogenity (hypertrophic remodeling, Fig. 1C).

3.3. Side to side difference ratio of the intranerve cross sectional area variability Between the sarcoid neuropathy group and the control group no significant changes in the “side to side difference ratio of the intranerve cross sectional area variability” of the various peripheral nerves or brachial plexus were documented (Table 1). 3.4. Nerve conduction studies The results of the nerve conduction studies are summarized in Table 2. Compared to controls, the sarcoid neuropathy group showed: 1) significantly lower values of the motor conduction velocity and compound muscle action potentials (cMAP) in the median (p b 0.001), fibular (p b 0.001) and tibial nerves (p b 0.001), 2) significantly increased values of the F-wave latency in the median nerve (p b 0.001), 3) significantly lower values of the sensory nerve action potentials (sNAPs) in the median (p b 0.001), ulnar (p b 0.001) and


4


sural nerves (p b 0.001), and 4) no differences in distal motor latency, and sensory and motor conduction velocities in all peripheral nerves examined.

4. Discussion We present the possible distribution pattern of pathological ultrasound changes in 13 patients with sarcoid neuropathy and its correlation with electrophysiology and most importantly with functional disability. First of all, the mean Rasch-built Overall Disability Scale score of 28.1 documented in patients with sarcoid neuropathy highlights the functional impairment that can occur in this type of immune-mediated peripheral nerve injury [1,3]. In addition, the mean Rasch-built Fatigue Severity Scale score of 17.1 depicts that fatigue may be a highly debilitating symptom in sarcoid neuropathy [2,4]. In addition, the reduction in the MRC sum score documented in the group with neurosarcoidosis may be a result of the significant mean disease duration (2.1 years)

3.5. Correlation of sonography, electrophysiology and functional disability A significant correlation between sonographic and electrophysiological findings in the sarcoid neuropathy group was only found between the cMAP and CSA of the ulnar nerve at the elbow (r = 0.894, p b 0.001). Neither sonography nor electrophysiology correlated with the MRC sum score, Rasch-built Overall Disability Scale score or Rasch-built Fatigue Severity Scale (Table 3).

Table 2 Overview of the electrophysiological findings in the sarcoid neuropathy and control groups. Site

Neurosarcoidosis (n = 13) Mean

Latency (ms) Median nerve Ulnar nerve

Fibular nerve

Tibial nerve

cMAP (mV) Median nerve Ulnar nerve

Fibular nerve

Tibial nerve

mCV (m/s) Median nerve

Wrist to APB Upper arm to APB Wrist to ADM b. elbow to ADM a. elbow to ADM Ankle — EDB Fib. head — EDB Popl. fossa — EDB Ankle — AH Popl. fossa — AH Wrist to APB Upper arm to APB Wrist to ADM b. elbow to ADM a. elbow to ADM Ankle — EDB Fib. head — EDB Popl. fossa — EDB Ankle — AHB Popl. fossa — AH

3.3 7.8 2.9 8.8 10.8 3.9 8.9 11.3 4.2 13.2

4.3 3.8 6.8 6.5 6.2 5.95 5.1 4.9 8.82 7.7

SD

Controls (n = 40) (min, max)

0.5 2.2 0.3 1.6 1.4 0.7 2.4 2.2 0.5 3.1

0.8 1.2 1.1 0.9 1.4 2.17 2.4 2.2 3.1 3.4

(1.5, 5.3) (1.4, 5.1) (4.5, 8.1) (4.9, 8.5) (3.9, 9.1) (1.3, 8.7) (0.7, 9.1) (0.5, 8.7) (2.1, 13.8) (1.3, 13.5)

Mean

p SD

3.7 7.2 2.7 9.4 10.7 3.6 8.5 10.2 3.8 12.5

0.28 1.2 0.5 1.7 1.6 0.8 2.1 1.7 0.9 1.5

0.316 0.214 0.418 0.267 0.864 0.290 0.567 0.065 0.645 0.248

7.1 6.7 7.8 7.1 6.9 10.1 9.5 8.1 14.1 12.5

2.4 1.5 3.2 2.9 2.5 3.1 2.1 1.9 7.4 1.8

b0.001 b0.001 0.298 0.469 0.365 b0.001 b0.001 b0.001 b0.001 b0.001

10.2 7.5 14.5 8.6 6.1 5.2 5.9 6.2 9.1 13.8

0.329 0.317 0.756 0.916 0.233 0.330 0.679 0.195 0.392 0.561

Wrist to APB Upper arm to APB Wrist to ADM b. elbow to ADM a. elbow to ADM Ankle — EDB Fib. head — EDB Popl. fossa — EDB Ankle — AHB Popl. fossa — AH

53.2 46.2 58.3 57.4 54.0 51.6 47.3 45.2 46.2 44.7

13.7 5.8 10.6 7.8 4.2 10.3 6.4 7.2 10.5 10.4

(27.4, 65.8)

56.7 48.5 59.8 57.7 56.2 49.5 48.1 47.9 48.8 47.5

F-wave (s) Median nerve Ulnar nerve Fibular nerve Tibial nerve

Wrist Wrist Ankle Ankle

32.3 24.5 46.3 46.7

7.5 2.1 5.6 7.3

(21.8, 43.1) (20.4, 29.7) (34.7, 57.8) (35.5, 59.2)

27.3 26.2 45.1 44.2

3.8 4.7 3.2 5.1

b0.001 0.083 0.339 0.273

sNAP (μV) Median nerve Ulnar nerve Sural nerve

Dig I — wrist Dig V — wrist Calf — lateral malleolus

6.3 5.2 5.4

4.46 1.7 2.9

(0, 15.1) (1.5, 7.8) (0, 12.5)

13.7 11 17.2

3.5 2.4 7.1

b0.001 b0.001 b0.001

sCV (m/s) Median nerve Ulnar nerve Sural nerve

Dig I — wrist Dig V — wrist Calf — lateral malleolus

49.7 48.4 41.7

(0, 55.4) (35.2, 60.1) (0, 58.5)

54.2 53.8 51.1

8.2 7.4 6.2

0.063 0.019 0.053

Ulnar nerve

Fibular nerve

Tibial nerve

21.9 5.5 28.6

(33.7, 60.1) (40.1, 69.4) (45.1, 65.3) (34.7, 61.8) (32.5, 63.1)

DML = distal motor latency, cMAP = compound muscle action potential, mCV = motor conduction velocity, sNAP = sensory nerve action potential, sCV = sensory conduction velocity, APB = abductor pollicis brevis, ADM = abductor digiti minimi, EDB = extensor digitorum brevis, AH = abductor hallucis. p-Values b 0.001 were considered as statistically significant and are highlighted with bold.



5

Table 3 Overview of the correlation of ultrasound and electrophysiology with the functional disability in patients with sarcoid neuropathy. Nerve

Ultrasound

Electrophysiology

Median Ulnar Radial Brachial plexus Fibular Tibial Median Ulnar Fibular Tibial

Measurement

CSA carpal tunnel CSA Guyon's canal CSA spiral groove CSA supraclavicular space CSA fibular Head CSA ankle DML cMAP DML cMAP DML cMAP DML cMAP

MRC Sum score

R-ODS

R-FSS

r

p

r

p

r

p

0.351 0.221 0.168 0.245 0.456 0.152 0.179 0.354 0.235 0.158 −0.171 −0.426 0.232 0.120

0.119 0.234 0.291 0.209 0.058 0.310 0.279 0.117 0.219 0.303 0.288 0.073 0.222 0.348

0.261 0.421 −0.156 0.181 0.271 0.145 −0.362 0.411 0.231 0.247 0.323 0.165 0.252 0.393

0.194 0.075 0.305 0.277 0.185 0.318 0.112 0.081 0.213 0.197 0.129 0.286 0.203 0.092

0.341 −0.211 0.161 0.351 0.154 0.453 0.345 −0.361 0.187 0.217 0.423 0.152 0.188 0.155

0.127 0.244 0.299 0.119 0.307 0.060 0.124 0.112 0.270 0.238 0.074 0.310 0.269 0.306

p-Values b 0.001 were considered as statistically significant and are highlighted with bold. CSA = cross sectional area, MRC = Medical Research Council, R-ODS = Rasch-built Overall Disability Scale, R-FSS = Rasch-built Fatigue Severity Scale, DML = distal motor latency, cMAP = compound muscle action potential.

and decreased CMAP-amplitude noted in several peripheral nerves [16] (Table 3). 4.1. Cross sectional area and echogenity To our knowledge, the significantly higher cross sectional area values of the ulnar (elbow, p b 0.001), fibular (fibular head, p b 0.001), tibial (ankle and popliteal fossa, p b 0.001) and sural nerves (p b 0.0001) documented in the neurosarcoidosis group summarize the first report of such ultrasound findings in the literature (Fig. 2). Peripheral nerve enlargement of the median nerve has been only reported in a case of neurosarcoidosis studied with magnetic resonance neurography [17]. The detection of CSA enlargement during axonal loss has been already reported in patients with ulnar neuropathy at the elbow and different types of polyneuropathy [6,23,25]. Although at a first glance, the detection of peripheral nerve enlargement during axonal loss seems to be an unexpected finding, this aspect could be explained through the pathophysiology of Wallerian degeneration. According to this process, a highly ordered pattern of release of pro- and antiinflammatory cytokines [such as tumor necrosis factor-a (TNF-α)] from resident Schwann cells, fibroblasts and macrophages is followed by the removal of toxic debris from the injury site (especially the myelin debris of the damaged

Fig. 2. Ultrasound scan of the fibular and tibial nerves in a patient with sarcoid neuropathy. Axial scan of the fibular (A) and tibial (B) nerves in the popliteal fossa showing a pathological cross sectional area enlargement of both nerves, due to an increase of the nerve fascicles (“hypertrophic nerve remodeling”). In both nerves no signs of nerve edema could be detected (tib. li = left tibial nerve, fib. li = left fibular nerve).

Schwann cells and oligodendrocytes). This orchestrated production of cytokines is serving to provoke a prolonged inflammatory response. This response could possibly lead to nerve edema and consecutive cross sectional area enlargement of the peripheral nerves, which can be detected with the help of nerve ultrasound [18,26]. Concerning the echogenity of the nerves at sites of CSA enlargement, we documented two different types: Type 1: increased CSA due to edema, with a partial or complete loss of the fascicular echostructure (iso, hypo- or hyperechoic) (Fig. 1B), and Type 2: increased CSA due to a fascicular cross sectional area increase, but normal echogenity (hypertrophic remodeling, Fig. 1C). Although nerve edema is a common finding in demyelinating polyneuropathies and seems to be the result of repeated episodes of demyelination–remyelination [21], the same finding has never been reported in the literature in cases of immunemediated axonal loss. To our experience, nerve edema may be found during axonal loss in neurosarcoidosis and is mainly located under the perineurium. A possible explanation for this finding could derive from the necrotizing type of vasculitic injury expected in neurosarcoidosis. According to pathological studies, granuloma composition, transmural infiltration and thrombosis occurred during necrotizing vasculitic injury and lead mainly to thick edema under the perineurium [5]. On the other hand, Palumbo et al. were the first to describe the molecules, that are involved during peripheral nerve remodeling in cases of Charcot– Marie–Tooth type I disease [22]. The pathophysiological cascade leading though to the “hypertrophic nerve remodeling” during immunemediated injury remains unanswered in the literature. A possible explanation could derive from the physiology of nerve conduction in peripheral nerves. The “nature” has two ways to improve the conduction velocity of the peripheral nerves. The first way is the building of the myelin sheath, which leads to an increase of the membrane resistance. This process seems to act as an art of “cable insulation” and increased the conduction velocity. During axonal loss though, not only a collapse of the membrane resistance, but also an increase of the membrane capacity and consecutive electrical power loss occurs, so that peripheral nerves seek an alternative way to compensate this injury. The second way is the enlargement of the fascicular diameter, which leads to nerve hypertrophy. This hypertrophy results in a decrease of the internal membrane resistance and therefore improvement not only of the current flow in axoplasma, but also of the electric impulse conduction. Another important aspect of this study is that the pathological cross sectional area changes were detected mainly at distal sites in the anatomic course of the peripheral nerves (e.g., ankle in the case of the tibial nerve and gastrocnemius muscle in the case of the sural nerve). This finding may reflect the “dying back” pattern of immune-mediated injury occurring in sarcoid neuropathy. It is interesting, that according


6


to previous literature reports structural sural nerve changes have been only detected in primary axonal polyneuropathies, such as uremic ones [30] or secondary, such as in the case of CIDP [8,9]. On the other hand, no such findings have been reported in classical demyelinating polyneuropathies, such as GBS [10] or MMN [7] (Fig. 3). In view of our findings and previous ones, the detection of pathological ultrasound changes in the sural nerve may help in differentiating demyelinating from axonal polyneuropathies. In a retrospective study of 34 patients with sarcoidosis undergoing sural nerve biopsy 29% showed vasculitic changes [29]. Histologically, vasculitic neuropathy in patients with sarcoidosis is of the epineural necrotizing type, with transmural infiltrates and thrombosis [29]. In addition, sarcoid granulomata in the epineurial and perineurial spaces and thick edema under the perineurium may be also observed in the sural nerve of patients with neurosarcoidosis [5]. In our opinion, this finding may explain the observed cross sectional area enlargement of the sural nerve.

4.2. Intranerve-, internerve- and intraplexus cross sectional area variability The intranerve CSA variability may help in differentiating focal (higher intranerve cross sectional area values) from diffuse nerve enlargement (lower intranerve cross sectional area values) [10,11,13,20]. In addition, higher internerve cross sectional area values may reflect the predominance of the immune-mediated neuropathies to distinct peripheral nerves. Systemic data on the applicability and sensitivity of this measure fail in the literature.

Hence, it is of interest that the neurosarcoidosis group showed statistically higher values for the intranerve CSA variability of the median nerve, when compared to the reference values of our lab [13]. This finding may reflect the focal pattern of peripheral nerve affection in this type of immune-mediated injury. In addition, the higher values of the internerve cross sectional area variability highlight well in our opinion the striking predilection of sarcoid neuropathy to distinct peripheral nerves. 4.3. Side to side difference ratio of the intranerve cross sectional area variability The possible value of this measure in detecting any lateralization of pathological changes in immune-mediated neuropathies has been recently proposed in the literature [10,11,13]. Applying the latter ratio in immune-mediated neuropathies (GBS, CIDP, MMN) detects the lateralization of pathological changes in MMN (higher side to side difference ratio of the intranerve cross sectional area variability values) [11]. Therefore, the absence of significant changes of this measure in various peripheral nerves, when compared to controls may reflect the symmetry of pathological changes in neurosarcoidosis. 4.4. Nerve conduction studies The statistically significant lower values of the cMAP and sNAP noted in the neurosarcoidosis group reflect in our opinion the high prevalence

Fig. 3. Ultrasound scan of the sural nerve in axonal and demyelinating polyneuropathies. Axial scan of the sural nerve in patients with (A) sarcoid neuropathy, (B) chronic inflammatory demyelinating polyneuropathy with secondary axonal loss, (C) acute inflammatory demyelinating polyneuropathy, (D) multifocal motor neuropathy. In images (A) and (B) a pathological cross sectional area enlargement during axonal loss is documented. On the other hand, in images (C) and (D) no pathological ultrasound changes are noted during demyelinating lesion of the nerve (sur. re = right sural nerve, sur. li = left sural nerve, CSA A = 0.08 cm2, B = 0.05 cm2, C = 0.02 cm2, D = 0.03 cm2).



of sensorimotor axonal loss occurring during vasculitic lesions. To a lesser degree, an increase in the f-wave latency of the median nerve was documented, a finding which is consistent with axonal loss. 4.5. Correlation of sonography, electrophysiology and functional disability A significant correlation between sonographic and electrophysiological findings in the group with sarcoid neuropathy was found only between the cMAP and CSA of the ulnar nerve at the elbow (r = 0.894, p b 0.001). Although, no systematic reports of such findings in neurosarcoidosis exist in the literature, to our experience, electrophysiology and nerve ultrasound highlight different aspects of peripheral nerve impairment, so that no significant correlation is expected. It is interesting that according to our study neither electrophysiology nor nerve ultrasound seemed to correlate with the functional disability and clinical picture of the patients with neurosarcoidosis. This aspect confirms previously reported findings in other immune-mediated neuropathies, such as CIDP, MMN and GBS [7,8,11]. Some limitations of this study have to be addressed. The mean time point of the clinical, sonographic and electrophysiologic evaluation of the patients with sarcoid neuropathy was 2.1 years after onset of symptoms referring to peripheral nerve disease, therefore the reproducibility of these findings at the beginning of this type of neuropathy remains unknown. In addition, the several new ultrasound measurements proposed and used to quantify pathological changes, require a reliable and precise visualization to the greatest possible extent in the anatomic course of peripheral nerves. Therefore, possible anatomical limitations, especially in the imaging of the lower extremity nerves, may produce difficulties in obtaining such reliable values. Systematic, multicentre, prospective studies are therefore required to examine the value of the previously reported findings. 5. Conclusions To summarize, this study presents a possible distribution pattern of pathological ultrasound changes in 13 patients with sarcoid neuropathy and examines their correlation with nerve conduction studies and functional impairment. Although the detected ultrasound changes are not so prominent, as in other reported immune-mediated neuropathies, this is the first report of such findings in the literature [6–8,11,19]. We also found no correlation between nerve ultrasound, neurophysiology and functional impairment in patients with neurosarcoidosis. This aspect highlights the need for more sufficient monitoring methods of response to immune-therapy. Nerve conduction studies reveal a primary axonal impairment in all sarcoid patients compatible to vasculitic neuropathies. In addition, according to our experience the detection of cross sectional area enlargement in the sural nerve may point out to an axonal neuropathy of immune-mediated injury, so that nerve biopsy and further etiological classification may be undertaken. Contributions of the authors Antonios Kerasnoudis acquired, analyzed and interpreted the data of this study. D. Woitalla made the critical revision of this study. R. Gold made the critical revision of this study. K. Pitarokoili acquired and analyzed the data of this study. M.-S. Yoon acquired, analyzed and interpreted the data and made the critical revision of this study. Conflict of interest Antonios Kerasnoudis reports no disclosures. D. Woitalla reports no disclosures. R. Gold has received consultation fees and speakers honoraria from BayerSchering, BiogenIdec, MerckSerono, Novartis, Sanofi-Aventis and

7

TEVA. He also acknowledges grant support from BayerSchering, BiogenIdec, MerckSerono, Sanofi-Aventis and TEVA all not related to this manuscript. K. Pitarokoili reports no disclosures. Min-Suk Yoon has received speakers honoraria from CSL Behring, all not related to this manuscript.

Acknowledgments and funding The study was not industry-sponsored.

References [1] Burns TM, Dyck PJ, Aksamit AJ, Dyck PJ. The natural history and long-term outcome of 57 limb sarcoidosis neuropathy cases. J Neurol Sci 2006;244(1–2):77–87 [15]. [2] De Kleijn WP, Drent M, Vermunt JK, Shigemitsu H, De Vries J. Types of fatigue in sarcoidosis patient. J Psychosom Res Dec 2011;71(6):416–22. [3] Dreyer M, Vucic S, Cros DP, Chong PS. Multiple painful sensory mononeuropathies (MPSM), a novel pattern of sarcoid neuropathy. J Neurol Neurosurg Psychiatry 2004;75(11):1645–6. [4] Fleischer M, Hinz A, Brähler E, Wirtz H, Bosse-Henck A. Factors associated with fatigue in sarcoidosis. Respir Care 2014;59(7):1086–94. [5] Galassi G, Gibertoni M, Mancini A, Nemni R, Volpi G, Merelli E, et al. Sarcoidosis of the peripheral nerve: clinical, electrophysiological and histological study of two cases. Eur Neurol 1984;23(6):459–65. [6] Grimm A, Heiling B, Schumacher U, Witte OW, Axer H. Ultrasound differentiation of axonal and demyelinating neuropathies. Muscle Nerve 2014. http://dx.doi.org/10. 1002/mus.24238. [7] Kerasnoudis A, Pitarokoili K, Behrendt V, Gold R, Yoon MS. Correlation of nerve ultrasound, electrophysiological and clinical findings in multifocal motor neuropathy. J Peripher Nerv Syst May 23 2014. http://dx.doi.org/10.1111/jns5.12067. [8] Kerasnoudis A, Pitarokoili K, Behrendt V, Gold R, Yoon MS. Correlation of nerve ultrasound, electrophysiological and clinical findings in chronic inflammatory demyelinating polyneuropathy. J Neuroimaging 2014. http://dx.doi.org/10.1111/jon.12079. [9] Kerasnoudis A, Pitarokoili K, Behrendt V, Gold R, Yoon MS. Nerve ultrasound score in distinguishing chronic from acute demyelinating polyneuropathy. Clin Neurophysiol 2014;125(3):635–41. [10] Kerasnoudis A, Pitarokoili K, Behrendt V, Gold R, Yoon MS. Correlation of nerve ultrasound, electrophysiological and clinical findings in post Guillain–Barre syndrome. J Peripher Nerv Syst 2013;18(3):232–40. [11] Kerasnoudis A, Klasing A, Behrendt V, Gold R, Yoon M-S. Intra- and internerve crosssectional area variability: new ultrasound measures. Muscle Nerve 2013;47:146–7. [12] Kerasnoudis A, Pitarokoili K. Ulnar nerve reference values for cross-sectional area, intranerve cross sectional area variability and side to side difference ratio. Rheumatol Int 2014;34(4):551–2. [13] Kerasnoudis A, Pitarokoili K, Behrendt V, Gold R, Yoon MS. Cross sectional area reference values for sonography of peripheral nerves and brachial plexus. Clin Neurophysiol 2013;124(9):1881–8. [14] Kleyweg RP, van der Meche FG, Schmitz PI. Interobserver agreement in the assessment of muscle strength and functional abilities in Guillain–Barre syndrome. Muscle Nerve 1991;14:1103–9. [15] Koffman B, Junck L, Elias SB, et al. Polyradiculopathy in sarcoidosis. Muscle Nerve 1999;22:608–13. [16] Marcellis RG, Lenssen AF, de Vries J, Drent M. Reduced muscle strength, exercise intolerance and disabling symptoms in sarcoidosis. Curr Opin Pulm Med 2013;19(5): 524–30. [17] Nakatani-Enomoto S, Aizawa H, Koyama S, Kuroda K, Kimura T, Kikuchi K. Transient swelling of peripheral nerves in a case of neurosarcoidosis. Intern Med 2004;43(11): 1078–81. [18] Oppenheim JJ, Feldman M. Introduction to the role of cytokines in innate and defense and adaptive immunity. In: Oppenheim JJ, Feldman M, editors. Cytokines reference. New York: Academic; 2001. p. 3–20. [19] Padua L, Granata G, Sabatelli M, Inghilleri M, Lucchetta M, Luigetti M, et al. Heterogeneity of root and nerve ultrasound pattern in CIDP patients. Clin Neurophysiol 2014;125(1):160–5. [20] Padua L, Martinoli C, Pazzaglia C, Lucchetta M, Granata G, Erra C, et al. Intra- and internerve cross-sectional area variability: new ultrasound measures. Muscle Nerve 2012;45:730–3. [21] Padua L, Granata G, Sabatelli M, Inghilleri M, Lucchetta M, Luigetti M, et al. Heterogeneity of root and nerve ultrasound pattern in CIDP patients. Clin Neurophysiol Oct 4 2013. http://dx.doi.org/10.1016/j.clinph.2013.07.023 [pii: S1388-2457(13) 00996-6]. [22] Palumbo C, Massa R, Panico MB, Di Muzio A, Sinibaldi P, Bernardi G, et al. Peripheral nerve extracellular matrix remodeling in Charcot–Marie–Tooth type I disease. Acta Neuropathol 2002;104(3):287–96. [23] Rajabally YA, Morlese J, Kathuria D, Khan A. Median nerve ultrasonography in distinguishing neuropathy sub-types: a pilot study. Acta Neurol Scand 2012; 125(4):254–9. [24] Said G, Lacroix C, Plante-Bordeneuve V, et al. Nerve granulomas and vasculitis in sarcoid peripheral neuropathy: a clinicopathological study of 11 patients. Brain 2002;125:264–75.


8


[25] Scheidl E, Böhm J, Farbaky Z, Simó M, Bereczki D, Arányi Z. Ultrasonography of ulnar neuropathy at the elbow: axonal involvement leads to greater nerve swelling than demyelinating nerve lesion. Clin Neurophysiol 2013;124(3): 619–25. [26] Shamash S, Reichert F, Rotshenker S. The cytokine network of Wallerian degeneration: tumor necrosis factor-alpha, interleukin-1alpha, and interleukin-1beta. J Neurosci 2002;22:3052–60. [27] van Nes SI, Vanhoutte EK, Faber CG, Garssen M, van Doorn PA, Merkies IS, PeriNomS Study Group. Improving fatigue assessment in immune-mediated neuropathies: the modified Rasch-built Fatigue Severity Scale. J Peripher Nerv Syst 2009;14: 268–78.

[28] van Nes SI, Vanhoutte EK, van Doorn PA, Hermans M, Bakkers M, Kuitwaard K, et al. Rasch-built Overall Disability Scale (R-ODS) for immune-mediated peripheral neuropathies. Neurology 2011;76:337–45. [29] Vital A, Lagueny A, Ferrer X, Louiset P, Canron MH, Vital C. Sarcoid neuropathy: clinico-pathological study of 4 new cases and review of the literature. Clin Neuropathol 2008;27(2):96–105. [30] Wei M, Zhu J, Wang N, Xue Q, Liu F, Hu B, et al. Preliminary study of sural nerve morphological changes in uremic hemodialysis patients using 22-MHz high-frequency ultrasound. Muscle Nerve 2012;46(6):937–42. [31] Zuniga G, Ropper AH, Frank J. Sarcoid peripheral neuropathy. Neurology 1991;41: 1558–61.


Multifocal motor neuropathy: correlation of nerve ultrasound, electrophysiological, and clinical findings.

Correlation of nerve ultrasound, electrophysiological and clinical findings in chronic inflammatory demyelinating polyneuropathy.

Correlation of electrophysiological and quantitative histological findings in the sural nerve of man. Studies on alcoholic neuropathy.

Clinical, electrophysiological, and molecular findings in early onset hereditary neuropathy with liability to pressure palsy.

Polyneuritis cranialis: clinical and electrophysiological findings.

Cisplatin neuropathy: clinical course and neurophysiological findings.

Acute ataxic neuropathy: a clinical, electrophysiological and morphological study.

Metal-on-metal hip resurfacing: correlation between clinical and radiological assessment, metal ions and ultrasound findings.

Imaging Findings in Sarcoid of the Humerus.

Postmortem chorionic villus sampling: correlation of cytogenetic and ultrasound findings.

Analysis of clinical and electrophysiological findings in Jakob-Creutzfeldt disease.

[Diagnostic and clinical aspects of neuropathy of small nerve fibers].

Clinical and Electrophysiological Findings in Two Siblings with Familial Hyperekplexia.

Hypothyroid neuropathy and myopathy: clinical and electrodiagnostic longitudinal findings.

Electrophysiological profile in arsenic neuropathy.

Phenelzine associated peripheral neuropathy--clinical and electrophysiologic findings.

Correlation between distribution of muscle weakness, electrophysiological findings and CTG expansion in myotonic dystrophy.

A clinical, electrophysiological, morphological and immunological study of chronic sensory neuropathy with ataxia and paraesthesia.

Hypoxic neuropathy versus diabetic neuropathy. An electrophysiological study in rats.

Sarcoid-like disorder of the intracranial optic nerve.

Renal sarcoid: Pseudotumoral radiologic manifestations and pathologic correlation.

Graves orbitopathy: correlation of CT and clinical findings.

The choledochocele: correlation of radiological, clinical and pathological findings.

Ledderhose Disease: Clinical, Radiological (Ultrasound and MRI), and Anatomopathological Findings.