Small fiber neuropathy in Fabry disease

Molecular Genetics and Metabolism 106 (2012) 135–141

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme

Minireview

Small fiber neuropathy in Fabry disease Marieke Biegstraaten a,⁎, Carla E.M. Hollak a, Mayienne Bakkers b, Catharina G. Faber b, Johannes M.F.G. Aerts c, Ivo N. van Schaik d a

Department of Internal Medicine, Division of Endocrinology and Metabolism, Academic Medical Center, Amsterdam, The Netherlands Department of Neurology, Maastricht University Medical Center, Maastricht, The Netherlands Department of Medical Biochemistry, Academic Medical Center, Amsterdam, The Netherlands d Department of Neurology, Academic Medical Center, Amsterdam, The Netherlands b c

a r t i c l e

i n f o

Article history: Received 31 January 2012 Received in revised form 15 March 2012 Accepted 15 March 2012 Available online 24 March 2012 Keywords: Fabry disease Small fiber neuropathy

a b s t r a c t Previous studies have explicitly shown that small nerve fibers are affected in Fabry disease which is assumed to cause the severe neuropathic pain that patients may have from childhood on. Neuropathic pain and small fiber neuropathy characteristics have therefore been considered as appropriate study endpoints in studies on the efficacy of enzyme replacement therapy. However, the relationship between small fiber neuropathy characteristics and pain, as well as the course of small fiber neuropathy in Fabry disease is still uncertain. In this article a comprehensive overview of the existing literature on small nerve fiber function and structure and the relationship with pain, age and disease severity is presented supplemented with data from the Dutch Fabry cohort, with the aim to identify consensus as well as controversies and to propose a hypothesis on the evolution of neuropathy in Fabry disease. © 2012 Elsevier Inc. All rights reserved.

Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Previous and current data on small nerve fiber structure and function . 3. Diagnosis of small fiber neuropathy in classical and atypical patients 4. The relation of small fiber neuropathy, age and pain in Fabry disease 5. Effect of enzyme replacement therapy . . . . . . . . . . . . . . . 6. Autonomic neuropathy . . . . . . . . . . . . . . . . . . . . . . 7. Pathophysiology . . . . . . . . . . . . . . . . . . . . . . . . . 8. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction

Fabry disease (OMIM 301500) is an X-linked lysosomal storage disease caused by deficient activity of α-galactosidase-A leading to lysosomal accumulation of globotriaosylceramide (Gb3). Affected individuals develop a multi-system disease. Male patients are usually severely affected, but symptoms and signs often appear in female carriers as well [1]. A diagnosis is made if a patient presents with typical Fabry symptoms ⁎ Corresponding author at: Department of Internal Medicine, Division of Endocrinology and Metabolism, F5-169, Academic Medical Center, University of Amsterdam PO Box 22700, 1100 DE Amsterdam, The Netherlands. Fax: +31 20 6917682. E-mail address: [email protected] (M. Biegstraaten). 1096-7192/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2012.03.010

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

135 136 136 137 139 139 140 140 140

and signs in combination with a decreased enzyme activity in males, or a mutation in the α-galactosidase-A gene in females. Typical symptoms and signs include neuropathic pain, intolerance to heat, inability to sweat, angiokeratoma and corneal opacities. Later in life renal failure, stroke and cardiomyopathy may develop. Neuropathic pain and pain attacks are often the presenting symptoms of the disease and start at an average age of 9 years in male patients and 16 years in female patients [2]. They are usually described as burning or shooting pains or painful pins and needles in hands and/or feet. The pathophysiology of pain in Fabry disease is not fully understood, although small fiber neuropathy (SFN) as a result of glycolipid accumulation in the dorsal root ganglia has been proposed as a possible mechanism [3]. In general, somatic small nerve fibers are divided into two groups: small myelinated Aδ-fibers and small unmyelinated C-fibers. Aδ-fibers

136

M. Biegstraaten et al. / Molecular Genetics and Metabolism 106 (2012) 135–141

carry cold sensation and mechanical pain sensitivity for pinprick stimuli while warm sensation and pain sensitivity for heat are carried by Cfibers. The relative contribution of C- and Aδ-fiber nociceptors to cold and pressure pain is less clear [4]. Damage to Aδ- and C-fibers often occurs as part of a mixed polyneuropathy, but sometimes small-caliber nerve fibers are preferentially or solely affected, leading to a pure SFN. To diagnose SFN and to get insight in the severity of small fiber damage, two methods are used: quantitative sensory testing (QST) and skin biopsies. QST such as determination of temperature perception and pain thresholds is an established, non-invasive, technique to assess small nerve fiber function [5]. Skin biopsies are used as a method to quantitate the number of Aδ- and C-fibers. Small unmyelinated nerve fibers represent both Aδ-fibers that lose their myelin sheet before entering the dermis and unmyelinated C-fibers [6]. Both are counted per millimeter epidermis resulting in the intraepidermal nerve fiber density (IENFD). The latter has a greater diagnostic efficiency than QST (88 versus 47%) using a ‘gold standard’ based on the presence of at least two abnormal results at clinical, QST and skin biopsy examination [5]. In the past 10 years, enzyme replacement therapy for Fabry disease has become available. Infusions with purified alfa-galactosidase A (Replagal®, Shire Human Genetic Therapies, Boston, MA, USA, and Fabrazyme®, Genzyme Corporation, Cambridge, MA, USA) can result in some clearance of substrate and clinical improvements [7–9]. The pivotal study on the efficacy of enzyme replacement therapy in Fabry disease used a decrease in neuropathic pain as the primary outcome measure. The double-blind placebo-controlled trial of 26 adult male patients showed a decline in Brief Pain Inventory (BPI) neuropathic pain severity score from 6.2 to 4.3 in ERT treated patients versus no significant change in the placebo group [10]. In the same study population, the IENFD did not change after 12–18 months of treatment [11], whereas small nerve fiber function improved upon ERT [12]. The latter was confirmed by a study in 22 Fabry patients: following ERT, these patients developed lower heat pain thresholds than prior to ERT. Moreover, fewer patients had abnormal results of cold detection and heat pain thresholds after than before ERT [13]. Thus, pain and small nerve fiber function improves on therapy while nerve fiber structure does not change. These studies have increased the interest in the relationship between Fabry disease and small fiber neuropathy characteristics. Although QST and IENFD studies showed unequivocal evidence of SFN in Fabry disease, the relationship with age, disease severity and pain is still uncertain probably due to the small number of patients included in these studies. In this article a comprehensive overview of the existing literature – including our study on QST findings in Fabry patients [14] – is presented and supplemented with previously unpublished data from our center on skin biopsy results, with the aim to identify consensus as well as controversies and to propose a hypothesis on the evolution of neuropathy in Fabry disease. 2. Previous and current data on small nerve fiber structure and function Previous studies have explicitly shown that small nerve fibers and their cell bodies are affected in Fabry disease [11,15–25], whereas thick myelinated nerve fibers are spared, at least in patients who have not yet developed renal insufficiency [19]. Neuropathological studies revealed loss of cell bodies in the dorsal root ganglia together with a pronounced reduction of small, thinly myelinated (Aδ) and unmyelinated (C) nerve fibers in sural nerve biopsy specimens [22,23]. Reduced intraepidermal nerve fiber densities were found in 19 out of 20 male patients [24] and in 27–53% of female patients [15,25]. In addition, more than hundred male and female Fabry patients have been studied with QST. These studies have repeatedly shown small

fiber dysfunction with a predilection for cold sensation impairment in Fabry patients [15–21]. Sixty-three to 100% of male patients [16,17,19–21] and 16 to 33% of female patients [15,18] were shown to have an impaired temperature perception. In a previous study from our group, a cohort of 48 patients (15 M/33 F, median age 47, median MSSI 15.5, 63% on ERT) showed solid evidence for the presence of SFN, confirming previous findings [14]. Of these 43 patients agreed to undergo a skin biopsy. These data have not been published before and are therefore described in more detail. One 49-year old male patient with advanced renal insufficiency (mGFR of 12 ml/ min, MSSI sum score of 34, aGalA activity 6.40 μmol/l/h, lyso-Gb3 level of 289 nM) was removed from all analyses as this patient had a history of multiple bilateral thalamic infarctions. It was therefore uncertain to what extent abnormal temperature perception was attributable to SFN or to intracerebral pathology in this patient. Comparable to previously published findings, 100% of male and 57% of female patients had an abnormal IENFD (see Figs. 1a and b). Just like earlier studies, a decreased thermal sensation was found in most Fabry patients [14]. The majority of male patients showed at least one pathologically decreased thermal threshold (Z-score b 1.96) (see Table 1). In females, only half of the patients studied with QST had at least one abnormal thermal threshold at the upper or lower limb (see Table 2). Three quarters of the females who underwent a skin biopsy as well as QST measurements, had either an abnormal IENFD, or a decreased thermal detection thresholds, or both. These studies indicate that almost all male and about half to three quarters of female Fabry patients have diagnostic features compatible with SFN. Since some male and many female patients are non- or oligosymptomatic with respect to renal, cardiac and cerebral manifestations, these estimates suggest that small nerve fibers are affected early during the course of the disease (i.e. before overt multi-system disease) which is not surprising considering the fact that neuropathic pain is one of the first symptoms in most Fabry patients [2]. 3. Diagnosis of small fiber neuropathy in classical and atypical patients Data on differences in ‘classical’ and ‘atypical’ patients have so far received little attention. It is, however, important, as diagnosis of Fabry disease can be difficult. A disease course in a classically affected male patient does not pose any diagnostic dilemmas, as these patients usually exhibit a characteristic combination of symptoms including neuropathic pain, inability to sweat, angiokeratoma and a gradual development of renal insufficiency. This is, however, different for individuals with a single otherwise unexplained non-specific symptom such as proteinuria, left ventricular hypertrophy, or pain in hands or feet. The identification of decreased α-galactosidase A activity or a genetic variation is not necessarily related to the symptoms in these cases. Moreover, these ‘atypical’ patients are biochemically different from ‘classical’ Fabry patients in that they do not show grossly elevated lyso-Gb3 levels. Lyso-Gb3 is the deacylated form of Gb3 (globotriaosylsphingosine) of which increased concentrations are found in plasma of Fabry patients [26]. It has been even questioned whether these patients should be considered as Fabry patients at all [27]. In the AMC cohort, two male patients carried a so called atypical mutation. They had normal thermal detection thresholds, but a decreased IENFD. In these patients, it is not yet clear whether the signs and symptoms are caused by Fabry disease or not. Diagnostic criteria for a diagnosis of small fiber neuropathy have been developed [28], but additional criteria are needed to determine whether the SFN is related to Fabry disease or not. The presence of Fabry diseasespecific SFN features should then be obligatory. Fabry patients usually report burning or shooting pains or painful pins and needles in hands and/or feet, starting at young age and with typical heat-provoked pain crises [29]. We would propose to consider a diagnosis of SFN due to Fabry disease only in patients with (a history of) neuropathic

M. Biegstraaten et al. / Molecular Genetics and Metabolism 106 (2012) 135–141

137

Fig. 1. a: IENFD in male patients (n = 14). Columns represent abnormal values for each decade Fig. 1. b: IENFD in female patients (n = 28). Columns represent abnormal values for each decade.

Table 1 Intraepidermal nerve fiber density and quantitative sensory testing results in male Fabry patients (n = 14). Age

IENFD

21 23 25 38 44 46a 47 47 49a 49 52 52 63 66

Abnormal Abnormal Abnormal Abnormal Abnormal Abnormal Abnormal Abnormal Abnormal Abnormal Abnormal Abnormal Abnormal Abnormal

CDT hand

WDT hand

TSL hand

CDT foot

WDT foot

TSL foot

Z-scores −.03 − 1.57 − 2.66 −.71 − 1.56 − 1.62 -.76 − 2.78 − 1.87 − 4.87 − 4.88 − 3.96 − 4.23 − 2.86

−.89 .52 −.37 − 2.97 − 3.85 .00 − 1.05 − 1.50 − 1.33 − 3.79 − 4.45 − 4.20 − 2.81 − 1.41

−.72 − 1.08 − 1.70 − 1.81 − 1.84 −.23 −.75 − 1.34 − 1.64 − 4.16 − 4.25 − 3.65 − 2.69 − 2.09

− 4.12 − 4.01 − 3.45 − 4.46 − 3.23 .66 − 3.23 − 3.14 −.31 − 3.23 − 3.27 − 3.94 − 2.36 − 3.14

− 2.43 −.57 − 1.56 − 1.53 − 2.21 .97 −.99 − 2.01 .88 − 2.23 − 2.01 − 2.54 − 1.19 − 1.18

− 3.67 − 4.20 − 3.27 − 2.22 − 2.48 .46 − 2.30 − 1.84 −.02 − 2.48 − 2.50 − 3.72 − 1.77 − 2.05

IENFD = intraepidermal nerve fiber density, CDT = cold detection threshold, WDT = warm detection threshold, TSL = thermal sensory limen. a ‘atypical’ Fabry patients with normal thermal detection thresholds.

pain in hands and/or feet, with either an onset of pain in childhood or adolescence (i.e. b18 years of age), or a course of pain that is characterized by ongoing pain with exacerbations that are provoked by fever, exercise or heat, or both (see Fig. 2). Patients, who additionally exhibit a decreased cold sensation at the upper or lower limb or have an abnormal IENFD, are then classified as “probable SFN due to Fabry disease”. Patient with typical pain complaints, a decreased cold sensation and an abnormal IENFD have “confirmed SFN due to Fabry disease”. 4. The relation of small fiber neuropathy, age and pain in Fabry disease The relation of small nerve fiber involvement and pain has been studied only scarcely. Moreover, the five studies that have been done showed conflicting results [15–18,24]. One study in 30 male patients showed more severe small nerve fiber impairment with older age, while no correlation between thermal thresholds and pain severity was found [16], another study in 19 females found a positive correlation between age and pain severity but no associations between IENFD, thermal sensation and pain severity [18], a third did not reveal

138

M. Biegstraaten et al. / Molecular Genetics and Metabolism 106 (2012) 135–141

Table 2 Intraepidermal nerve fiber density and quantitative sensory testing results in female Fabry patients (n = 33). Age

IENFD

12 14 15 18 22 23 24 25 25 25 25 26 37 38 39 41 45 47 48 49 49 50 50 51 52 53 55 58 61 62 62 62 73

Normal Normal Abnormal Normal . Abnormal Abnormal . Abnormal Normal . Abnormal Abnormal Normal Normal . Normal Abnormal Abnormal Abnormal Abnormal Abnormal Abnormal Normal Abnormal Normal Abnormal Abnormal Normal Normal Abnormal . Normal

CDT hand

WDT hand

TSL hand

CDT foot

WDT foot

TSL foot

Z-scores −.14 −.25 1.16 −.39 − 2.96 −.03 − 2.81 −.90 −.43 − 2.82 .20 − 1.03 .17 −.01 −.33 .19 − 2.10 −.27 −.10 − 1.44 .67 − 1.91 .51 −.50 − 3.13 −.11 − 3.00 .46 −.48 .40 − 1.03 − 1.00 −.16

−.35 −.09 .33 .62 −.48 − 1.35 − 1.77 − 1.03 −.39 −.39 1.28 −.23 − 1.43 − 1.02 −.74 −.65 − 1.75 −.61 −.06 − 1.21 −.58 −.17 .22 −.80 −.66 .30 − 2.88 .30 −.24 −.42 −.79 −.54 −.22

− 1.00 −.38 −.02 −.45 − 2.15 −.81 − 1.70 −.60 −.70 − 1.34 .13 −.89 −.92 −.64 −.84 −.68 − 1.91 −.57 .07 − 1.38 − 1.09 − 1.19 .91 − 2.09 − 2.16 .61 −.86 .23 −.19 −.45 − 1.01 − 1.31 −.59

− 1.69 −.99 −.40 − 1.47 − 2.49 .73 − 1.54 −.12 − 2.10 − 1.21 .51 − 1.10 .27 − 1.04 − 2.64 − 2.04 − 1.54 −.84 − 3.45 −.58 − 2.08 − 1.88 1.04 − 2.37 − 1.46 −.90 − 3.32 − 1.86 − 1.61 − 1.82 − 3.27 − 1.91 − 3.76

1.06 .67 1.19 .21 −.39 1.77 − 2.25 .83 .32 .34 .10 .34 −.81 .15 −.97 −.78 .46 − 2.09 − 1.71 −.70 − 2.39 −.75 −.05 − 1.77 .15 − 1.78 − 2.27 .34 4.60 − 1.62 .05 − 1.18 −.83

−.07 −.37 −.09 − 1.08 − 1.69 .44 − 1.97 −.91 − 1.51 −.62 −.25 −.48 −.17 −.07 − 1.55 − 1.88 −.80 − 1.93 − 2.53 − 1.11 − 2.20 − 1.94 .22 − 2.56 − 1.23 −.23 − 2.30 − 2.41 −.44 − 1.44 − 2.99 − 1.88 − 1.33

IENFD = intraepidermal nerve fiber density, CDT = cold detection threshold, WDT = warm detection threshold, TSL = thermal sensory limen.

an association between small nerve fiber function and age or disease severity in the 22 male patients studied [17], the fourth study did not demonstrate correlations between age, pain severity and nerve fiber

function in 12 women with Fabry disease, although they did find an association between IENFD and pain intensity [15], and the last study in 20 males could not demonstrate a correlation between disease severity and IENFD [24]. Thus, the relation of SFN, age, and pain in Fabry disease has been unclear so far. These conflicting results are possibly due to the small number of patients included in each of those studies. In the AMC cohort there was a clear association between age and disease severity on the one hand and small nerve fiber function on the other hand: older and more severely affected patients exhibit more severely impaired thermal sensation [14]. The same goes for nerve fiber structure: the older and the more severely affected a patient is, the lower the IENFD (see Table 3). Moreover, more severe loss of intraepidermal nerve fibers turned out to be clearly associated with more severe loss of nerve fiber function (see Tables 4a and 4b). Although the IENFD has been found to be reduced at both the upper and the lower limbs of Fabry patients [25], the SFN in Fabry disease has been demonstrated to be a length-dependent neuropathy by using proximal and distal skin biopsy sites; patients showed a greater proportional loss of innervation at the distal site than at the proximal site [18,24,25]. Findings in the AMC cohort support this. Considering male and female patients separately, both older age and more severe disease were associated with more extensive small nerve fiber dysfunction at the upper limbs in male patients and with the presence of small nerve fiber dysfunction at the lower limbs in female patients. This may be explained as follows: first the nerve fibers at the lower limbs become damaged, leading to loss of nerve fiber function. While small fiber damage at the lower limbs proceeds, the same process starts at the upper limbs. Indeed the relatively older male patients in our study exhibited complete loss of nerve fiber function at the lower limbs and older age and more severe disease in male patients led to more extensive small nerve fiber function loss at the upper limbs. In females, who in general show a more attenuated disease course than male patients, only slight dysfunction at the lower limbs could be demonstrated. This could explain the association between age, disease severity and small nerve fiber dysfunction at the lower limb in females, while small nerve fibers at the upper limbs are still relatively spared. Furthermore, more severe loss of intraepidermal nerve fibers was associated with the presence of small nerve fiber dysfunction at the lower limbs in female patients. Males

Pain in hands and/or feet described as burning/ shooting/pins and needles AND ≥ 1 of the following: -ongoing pain with exacerbations that are provoked by fever, exercise or heat -on set in childhood or adolescence (i.e. < 18 years)

Yes

Tests ≥ 1 of the following: -impaired coldsensation at the upper or lower limb -IENFD < the 5th percentile

Impaired cold sensation AND IENFD < the 5th percentile

Yes

Yes

No

No

No Decreased heat or cold pain threshold

Fabry SFN unlikely

Confirmed Fabry SFN

Probable Fabry SFN

No Yes

Beginning Fabry SFN?

Fig. 2. Flow chart to diagnose Fabry small fiber neuropathy.

M. Biegstraaten et al. / Molecular Genetics and Metabolism 106 (2012) 135–141

139

Table 3 Relation between age, disease severity and IENFD, males and females. IENFD simple regression, males

Age MSSI mGFR or eGFR Enzyme activity

IENFD simple regression, females

B (95%CI)

β

p-value

B (95%CI)

β

p-value

− 0.06 − 0.06 0.02 0.15

− 0.57 − 0.58 0.55 0.11

0.033 0.030 0.043 0.723

− 0.06 − 0.09 0.01 − 0.01

− 0.33 − 0.29 0.07 − 0.04

0.087 0.129 0.712 0.855

(− 0.11;−0.01) (− 0.12;−0.01) (0.00;0.03) (− 0.78;1.09)

(− 0.13;0.01) (− 0.21;0.03) (− 0.03;0.05) (− 0.10;0.09)

B = regression coefficient, β = standardized regression coefficient, MSSI = Mainz Severity Score Index, mGFR = measured Glomerular Filtration Rate, eGFR = estimated Glomerular Filtration Rate.

exhibited a more pronounced reduction of IENFD and complete small nerve fiber function loss at the lower limbs. A further decrease of intraepidermal nerve fibers was associated with more severe loss of small nerve fiber function at the upper limbs. These observations point to a length-dependent neuropathy that starts at young age and leads to an abnormal IENFD and complete loss of small nerve fiber function at the lower limbs in male patients in the first two decades. Small nerve fiber function at the upper limbs decreases in the years thereafter. In females this process starts later in life, and most females will never develop complete loss of small nerve fiber function at the lower limbs (see Figs. 3a and b). Just like the observations in other studies, no linear relationship between structural and functional small nerve fiber impairment on the one hand and pain intensity on the other hand was found in the AMC cohort. On the contrary, young patients reported more severe pain. Moreover, young females exhibited slightly positive heat pain threshold Z-scores (positive Z-scores for this item indicate a gain of function or increased sensitivity). We hypothesize that peripheral sensitization underlies this observation in young patients. Future studies using microneurography could support this hypothesis which is currently the only technique able to record and quantify positive sensory phenomena mediated by small nerve fibers [30]. Besides, one-third of females aged 50 years and older reported no pain on the visual analog scale for mean pain in the last 4 weeks which is in line with a natural history study of Fabry disease that showed a decrease in pain in a subset (11%) of patients as they became older [31]. The lack of a linear relationship between QST results, IENFD and pain severity is therefore possibly explained by peripheral sensitization in the youngest patients, and disappearance of pain in a subset of the oldest patients (see Figs. 3a and b). Altogether, the SFN in Fabry disease is a length-dependent neuropathy that presumably progresses with older age and more severe disease while pain takes another course possibly due to peripheral sensitization in earlier stages of nerve fiber damage and disappearance of pain in later stages.

5. Effect of enzyme replacement therapy As outlined in the Introduction section, enzyme replacement therapy has been shown to reduce neuropathic pain [10]. The IENFD did not change after 12–18 months of treatment [11], whereas small nerve fiber function improved upon ERT [12,13]. Possibly, the

structural damage to the small nerve fibers is irreversible, while function of the remaining fibers may improve on ERT. Another hypothesis is that structural nerve fiber damage is only reversible after a longer period of treatment.

6. Autonomic neuropathy Autonomic functions are also carried by small nerve fibers. As a consequence, small fiber damage often results in autonomic dysfunction. In Fabry disease several features, such as an- or hypohydrosis and cardiac rhythm disturbances, have been ascribed to autonomic neuropathy [32–34]. This assumption was largely based on a study in 10 patients who were assessed with several (non-) invasive clinical tests for autonomic dysfunction and were found to have hypohidrosis, impaired pupillary constriction, reduced saliva and tear formation, disordered intestinal motility and diminished cutaneous flare with scratch or intradermal histamine injection [32]. In addition, reduced peripheral blood flow after vasodilatation procedures and abnormal cerebral blood flow velocities have been found in male and female Fabry patients which may point to dysfunction of the autonomic nerve fibers [35–37]. Studies on heart rate variability (HRV) in pediatric Fabry patients revealed significantly different results between Fabry boys and Fabry girls and between Fabry boys and controls, with significant improvement of heart rate variability in Fabry boys upon enzyme replacement therapy [38,39]. However, doubts arose about the existence of autonomic neuropathy in Fabry disease: orthostatic intolerance and male sexual dysfunction that are invariably found in autonomic neuropathies are infrequently reported by Fabry patients. Moreover, the defective sweating which has long been thought to originate from autonomic neuropathy is no longer considered to be caused by autonomic neuropathy as skin biopsy studies did not reveal a decrease in nerve fiber density of sweat gland innervation, but revealed storage of lipids in sweat glands. Also, the nonlength-dependent distribution of the an- or hypohydrosis and the rapid effect of single enzyme infusions, suggested a sweat gland dysfunction rather than an autonomic neuropathy [12]. Possibly, end-organ failure, such as stiffness of vascular smooth muscles and endothelial dysfunction, plays a role in the found abnormalities in the studies on peripheral hemodynamics. A study on vascular hyperreactivity in Fabry disease supports this hypothesis; absence of a difference in plasma epinephrine or norepinephrine

Table 4a Relation between IENFD and QST results at the upper limb, males (n = 14). CDT

IENFD

WDT

TSL

HPT

CPT

B (95% CI)

β

p-value

B (95% CI)

β

p-value

B (95% CI)

β

p-value

B (95% CI)

0.85 (0.32;1.38)

0.71

0.005

0.94 (0.42;1.46)

0.75

0.002

0.71 (0.31;1.12)

0.74

0.003

0.29 0.44 (− 0.09;0.66)

β

β

p-value

B (95% CI)

0.120

− 0.11 − 0.15 (− 0.58;0.36)

p-value 0.617

B = regression coefficient, β = standardized regression coefficient, CDT = cold detection threshold, WDT = warm detection threshold, TSL = thermal sensory limen, HPT = heat pain threshold, CPT = cold pain threshold.

140

M. Biegstraaten et al. / Molecular Genetics and Metabolism 106 (2012) 135–141

Table 4b Relation between IENFD and QST results at the lower limb, females (n = 28). CDT B (95% CI) IENFD

WDT β

0.02 0.04 (− 0.14;0.17)

TSL

HPT

CPT

p-value

B (95% CI)

β

p-value

B (95% CI)

β

p-value

B (95% CI)

β

p-value

B (95% CI)

β

0.850

0.22 (0.04;0.40)

0.44

0.021

0.14 (0.03;0.26)

0.45

0.018

0.21 (0.00;0.41)

0.37

0.052

0.10 0.26 (− 0.05;0.24)

p-value 0.188

B = regression coefficient, β = standardized regression coefficient, CDT = cold detection threshold, WDT = warm detection threshold, TSL = thermal sensory limen, HPT = heat pain threshold, CPT = cold pain threshold.

levels between patients and controls suggested that the altered vessel response in Fabry disease may be attributed to vasogenic and not to neurogenic factors [40]. Likewise, cardiac pathology (i.e. left ventricular hypertrophy and/or conduction system pathology) could have influenced the difference in heart rate variability between boys and girls with Fabry disease [38,39]. Observations in the AMC cohort are in line with previous findings; there was a low prevalence of orthostatic intolerance and male sexual dysfunction, normal cardiovascular autonomic control in most patients, and a low resting heart rate making it unlikely that Fabry patients suffer from severe autonomic neuropathy. This seemingly contradiction may be explained by the difference between nerve fiber types involved in the autonomic control of organs and the nerve fiber type affected by Fabry disease; preganglionic autonomic fibers consist of small myelinated B-fibers and postganglionic autonomic fibers are small unmyelinated C-fibers [41], whereas Fabry disease causes relatively selective damage to Aδ-fibers [15–21]. Possibly, the selective damage to ‘non-autonomic’ Aδ-fibers in Fabry disease leads to the preservation of some of the autonomic functions in Fabry

A

Pain severity

Loss of function hand Loss of function foot &Hypersensitivity hand Hypersensitivity foot Normal sensation

pain

disease. Although this is no more than a hypothesis, we found some support for this thought by similar findings, i.e. no evidence for autonomic failure, in hereditary sensory and autonomic neuropathy type 5 that is also known to cause selective loss of Aδ-fibers [42]. 7. Pathophysiology Despite more extensive knowledge about the course and consequences of SFN in Fabry disease, the underlying pathophysiological mechanism remains speculative. It is uncertain whether the neuropathy arises from storage of lipids in ganglia leading to a so-called dying-back neuropathy, or from direct damage to small nerve fiber axons. The preference for Aδ-fibers suggests that the ganglia or axons of small myelinated nerve fibers are more vulnerable to lipid accumulation than small unmyelinated fibers. We looked for associations between lyso-Gb3 and small fiber neuropathy in our cohort. The lifetime exposure to lyso-Gb3 was calculated for each patient with ‘classical’ Fabry disease which was estimated as concentration of plasma lyso-Gb3 ⁎age. In male patients, simple regression analyses revealed significant correlations between lyso-Gb3 exposure and the cold detection threshold (β =−0.786, p = 0.007) and thermal sensory limen (β= −0.752, p =0.012) at the upper limb. How this finding should be interpreted remains speculative. Lyso-Gb3 has shown to be capable of promoting proliferation of smooth muscle cells in vitro, suggesting a role in the pathogenesis of vascular pathology in Fabry disease. Possibly, vascular pathology at the level of the vasa nervorum plays a role in the development of SFN although the preference for Aδ-fibers cannot be explained in this way. Alternatively, lysoGb3 itself exerts a direct pathological effect on the nervous system. Interestingly, very recently hereditary sensory neuropathy type 1 (HSN1) has been shown to be caused by plasma accumulation of two sphingoid bases, 1-deoxy-sphinganine and 1-deoxymethyl-sphinganine, that highly resemble lyso-Gb3 [43]. The neuropathy in HSN1 and Fabry disease show clear similarities, supporting the assumption of neurotoxicity of lyso-Gb3 [44].

Age Disease severity

8. Conclusions

B

Loss of function foot &Hypersensitivity hand Hypersensitivity foot Normal sensation

pain

Pain severity

Loss of function hand

Age Disease severity Fig. 3. a) Pain and nerve fiber dysfunction in males with Fabry disease. Red area = course of pain. Green line = small nerve fiber dysfunction; with older age and more severe disease, nerve fiber dysfunction increases. b) Pain and nerve fiber dysfunction in females with Fabry disease. Red area = course of pain. Green line = small nerve fiber dysfunction; with older age and more severe disease, nerve fiber dysfunction increases.

The SFN in Fabry disease is a length-dependent and Aδ-fiber preferential neuropathy that is likely to progress with older age and more severe disease while pain takes another course possibly due to peripheral sensitization in earlier stages of nerve fiber damage and disappearance of pain in later stages. Enzyme replacement therapy probably should start before structural damage of nerve fibers has taken place. Finally, the relatively sparing of C-fibers may underlie the absence of evidence for autonomic neuropathy in Fabry patients. References [1] W.R. Wilcox, J.P. Oliveira, R.J. Hopkin, et al., Females with Fabry disease frequently have major organ involvement: lessons from the Fabry Registry, Mol. Genet. Metab. 93 (2008) 112–128. [2] A. Mehta, R. Ricci, U. Widmer, et al., Fabry disease defined: baseline clinical manifestations of 366 patients in the Fabry Outcome Survey, Eur. J. Clin. Invest. 34 (3) (2004) 236–242. [3] R. Schiffmann, Neuropathy and Fabry disease: pathogenesis and enzyme replacement therapy, Acta Neurol. Belg. 106 (2) (2006) 61–65.

M. Biegstraaten et al. / Molecular Genetics and Metabolism 106 (2012) 135–141 [4] R. Rolke, R. Baron, C. Maier, et al., Quantitative sensory testing in the German Research Network on Neuropathic Pain (DFNS): standardized protocol and reference values, Pain 123 (3) (2006) 231–243. [5] G. Devigili, V. Tugnoli, P. Penza, et al., The diagnostic criteria for small fibre neuropathy: from symptoms to neuropathology, Brain 131 (2008) 1912–1925. [6] V. Provitera, M. Nolano, A. Pagano, G. Caporaso, A. Stancanelli, L. Santoro, Myelinated nerve endings in human skin, Muscle Nerve 35 (6) (2007) 767–775. [7] B. Hoffmann, M. Schwarz, A. Mehta, S. Keshav, Fabry Outcome Survey European Investigators. Gastrointestinal symptoms in 342 patients with Fabry disease: prevalence and response to enzyme replacement therapy, Clin. Gastroenterol. Hepatol. 5 (12) (2007) 1447–1453. [8] R. Schiffmann, H. Askari, M. Timmons, et al., Weekly enzyme replacement therapy may slow decline of renal function in patients with Fabry disease who are on long-term biweekly dosing, J. Am. Soc. Nephrol. 18 (5) (2007) 1576–1583. [9] D.A. Hughes, P.M. Elliot, J. Shah, et al., Effects of enzyme replacement therapy on the cardiomyopathy of Anderson–Fabry disease: a randomised, double-blind, placebo-controlled clinical trial of agalsidase alfa, Heart 94 (2) (2008) 153–158. [10] R. Schiffmann, J.B. Kopp, H.A.I. Austin, et al., Enzyme replacement therapy in Fabry disease: a randomized controlled trial, JAMA 285 (2001) 2743–2749. [11] R. Schiffmann, P. Hauer, B. Freeman, et al., Enzyme replacement therapy and intraepidermal innervation density in Fabry disease, Muscle Nerve 34 (1) (2006) 53–56. [12] R. Schiffmann, M.K. Floeter, J.M. Dambrosia, et al., Enzyme replacement therapy improves peripheral nerve and sweat function in Fabry disease, Muscle Nerve 28 (2003) 703–710. [13] M.J. Hilz, M. Brys, H. Marthol, B. Stemper, M. Dütsch, Enzyme replacement therapy improves function of C-, Ad, and Ab-nerve fibers in Fabry neuropathy, Neurology 62 (2004) 1066–1072. [14] M. Biegstraaten, A. Binder, R. Maag, C.E.M. Hollak, R. Baron, I.N. van Schaik, The relation between small nerve fibre function, age, disease severity and pain in Fabry disease, Eur. J. Pain (2011), doi:10.1016/j.ejpain.2011.01.014. [15] S.M. Laaksonen, M. Roytta, S.K. Jaaskelainen, I. Kantola, M. Penttinen, B. Falck, Neuropathic symptoms and findings in women with Fabry disease, Clin. Neurophysiol. 119 (6) (2008) 1365–1372. [16] M. Dütsch, H. Marthol, B. Stemper, M. Brys, T. Haendl, M.J. Hilz, Small fiber dysfunction predominates in Fabry neuropathy, J. Clin. Neurophysiol. 19 (6) (2002) 575–586. [17] C.A. Luciano, J.W. Russell, T.K. Banerjee, et al., Physiological characterization of neuropathy in Fabry's disease, Muscle Nerve 26 (5) (2002) 622–629. [18] A. Torvin Møller, F. Winther Bach, U. Feldt-Rasmussen, et al., Functional and structural nerve fiber findings in heterozygote patients with Fabry disease, Pain 145 (1–2) (2009) 237–245. [19] R. Maag, A. Binder, C. Maier, et al., Detection of a characteristic painful neuropathy in Fabry disease: a pilot study, Pain Med. 9 (8) (2008) 1217–1223. [20] S.H. Morgan, P. Rudge, S.J.M. Smith, et al., The neurological complications of Anderson– Fabry disease (alpha-galactosidase A deficiency)—investigation of symptomatic and presymptomatic patients, Q. J. Med. 75 (2) (1990) 491–507. [21] M. Low, K. Nicholls, N. Tubridy, et al., Neurology of Fabry disease, Intern. Med. J. 37 (2007) 436–447. [22] F. Gemignani, A. Marbini, M.M. Bragaglia, E. Govoni, Pathological study of the sural nerve in Fabry's disease, Eur. Neurol. 23 (3) (1984) 173–181. [23] A. Onishi, P.J. Dyck, Loss of small peripheral sensory neurons in Fabry disease. Histologic and morphometric evaluation of cutaneous nerves, spinal ganglia, and posterior columns, Arch. Neurol. 31 (2) (1974) 120–127.

141

[24] L.J.C. Scott, J.W. Griffin, C. Luciano, et al., Quantitative analysis of epidermal innervation in Fabry disease, Neurology 52 (6) (1999) 1249–1254. [25] R. Liguori, V. Di Stasi, E. Bugiardini, et al., Small fibre neuropathy in female patients with Fabry disease, Muscle Nerve 41 (406) (2010) 412. [26] J.M. Aerts, J.E. Groener, S. Kuiper, et al., Elevated globotriaosylsphingosine is a hallmark of Fabry disease, Proc. Natl. Acad. Sci. U. S. A. 105 (8) (2008) 2812–2817. [27] S.M. Rombach, N. Dekker, M.G. Bouwman, et al., Plasma globotriaosylsphingosine: diagnostic value and relation to clinical manifestations of Fabry disease, Biochim. Biophys. Acta 1802 (9) (2010) 741–748. [28] S. Tesfaye, A.J. Boulton, P.J. Dyck, et al., Diabetic neuropathies: update on definitions, diagnostic criteria, estimation of severity, and treatments, Diabetes Care 33 (10) (2010) 2285–2293. [29] R. Schiffmann, L.J.C. Scott, Pathophysiology and assessment of neuropathic pain in Fabry disease, Acta Paediatr. 439 (2002) S48–S52. [30] G. Cruccu, C. Sommer, P. Anand, et al., EFNS guidelines on neuropathic pain assessment: revised 2009, Eur. J. Neurol. 17 (2010) 1010–1018. [31] K.D. MacDermot, A. Holmes, A.H. Miners, Anderson–Fabry disease: clinical manifestations and impact of disease in a cohort of 98 hemizygous males, J. Med. Genet. 38 (11) (2001) 750–760. [32] W.J.L. Cable, E.H. Kolodny, R.D. Adams, Fabry disease: impaired autonomic function, Neurology 32 (5) (1982) 498–502. [33] M.J. Hilz, Evaluation of peripheral and autonomic nerve function in Fabry disease, Acta Paediatr. Suppl. 439 (2002) 38–42. [34] E.H. Kolodny, G.M. Pastores, Anderson–Fabry disease: extrarenal, neurological manifestations, J. Am. Soc. Nephrol. 13 (2002) S150–S153. [35] Y. Seino, J.K. Vyden, M. Philippart, H.B. Rose, K. Nagasawa, Peripheral hemodynamics in patients with Fabry's disease, Am. Heart J. 105 (5) (1983) 783–787. [36] A. Torvin Møller, U. Feldt-Rasmussen, A.K. Rasmussen, et al., Small-fibre neuropathy in female Fabry patients: reduced allodynia and skin blood flow after topical capsaicin, J. Peripher. Nerv. Syst. 11 (2) (2006) 119–125. [37] M.J. Hilz, E.H. Kolodny, M. Brys, B. Stemper, T. Haendl, H. Marthol, Reduced cerebral blood flow velocity and impaired cerebral autoregulation in patients with Fabry disease, J. Neurol. 251 (5) (2004) 564–570. [38] C. Kampmann, C.M. Wiethoff, C.M. Whybra, F.A. Baehner, E. Mengel, M. Beck, Cardiac manifestations of Anderson–Fabry disease in children and adolescents, Acta Paediatr. 97 (4) (2008) 463–469. [39] M. Ries, J.T.R. Clarke, C. Whybra, et al., Enzyme-replacement therapy with agalsidase alfa in children with Fabry disease, Pediatrics 118 (2006) 924–932. [40] G. Altarescu, D.F. Moore, R. Pursley, et al., Enhanced endothelium-dependent vasodilation in Fabry disease, Stroke 32 (7) (2001) 1559–1562. [41] A.J. McDougall, J.G. McLeod, Autonomic neuropathy, I. Clinical features, investigation, pathophysiology, and treatment, J. Neurol. Sci. 137 (2) (1996) 79–88. [42] M. Donaghy, R.N. Hakin, J.M. Bamford, et al., Hereditary sensory neuropathy with neurotrophic keratitis. Description of an autosomal recessive disorder with a selective reduction of small myelinated nerve fibres and a discussion of the classification of the hereditary sensory neuropathies, Brain 110 (3) (1987) 563–583. [43] A. Penno, M.M. Reilly, H. Houlden, et al., Hereditary sensory neuropathy type 1 is caused by the accumulation of two neurotoxic sphingolipids, J. Biol. Chem. (2010), doi:10.1074/jbc.M109.092973. [44] M. Auer-Grumbach, Hereditary sensory neuropathy type I, Orphanet J. Rare Dis. 3 (2008) 7.