Recessive dystrophic epidermolysis bullosa results in painful small fibre neuropathy

Abstract

Small fibres in the skin are vulnerable to damage in metabolic or toxic conditions such as diabetes mellitus or chemotherapy resulting in small fibre neuropathy and associated neuropathic pain. Whether injury to the most distal portion of sensory small fibres due to a primary dermatological disorder can cause neuropathic pain is still unclear. Recessive dystrophic epidermolysis bullosa (RDEB) is a rare condition in which mutations of proteins of the dermo-epidermal junction lead to cycles of blistering followed by regeneration of the skin. Damage is exclusive to the skin and mucous membranes, with no known direct compromise of the nervous system. It is increasingly recognized that most RDEB patients experience daily pain, the aetiology of which is unclear but may include inflammation (in the wounds), musculoskeletal (due to atrophy and retraction scars limiting movement) or neuropathic pain. In this study we investigated the incidence of neuropathic pain and examined the presence of nerve dysfunction in RDEB patients. Around three quarters of patients presented with pain of neuropathic characteristics, which had a length-dependent distribution. Quantitative sensory testing of the foot revealed striking impairments in thermal detection thresholds combined with an increased mechanical pain sensitivity and wind up ratio (temporal summation of noxious mechanical stimuli). Nerve conduction studies showed normal large fibre sensory and motor nerve conduction; however, skin biopsy showed a significant decrease in intraepidermal nerve fibre density. Autonomic nervous system testing revealed no abnormalities in heart rate and blood pressure variability however the sympathetic skin response of the foot was impaired and sweat gland innervation was reduced. We conclude that chronic cutaneous injury can lead to injury and dysfunction of the most distal part of small sensory fibres in a length-dependent distribution resulting in disabling neuropathic pain. These findings also support the use of neuropathic pain screening tools in these patients and treatment algorithms designed to target neuropathic pain.

Keywords: epidermolysis bullosa; neuropathic pain; small fibre neuropathy.

Figure 1

Somatosensory profiles. Somatosensory profiles determined with QST in the dorsum of the foot (S1 dermatome) of patients with RDEB (dark grey) and healthy volunteers (light grey). Data are expressed as mean z-scores with standard deviations and the grey area indicates the normal range of ±2 SD of normative data. (A) Patients with RDEB have a significant loss of function in cold and warm detection thresholds compared to control subjects. They also have a reduced ability to differentiate temperature changes (e.g. thermal sensory limen). A small but significant decrease in mechanical detection thresholds was observed, but no patients present loss or gain in vibration detection thresholds. (B) RDEB patients had increased pain sensibility compared to control participants, which could be seen as an altered heat pain sensitivity, altered mechanical pain sensitivity and altered wind up ratio. No differences in cold pain thresholds were found between both groups. There was also no difference in mechanical and pressure pain thresholds. (C) Percentage of patients with value outside the normal range. To the left side are shown the loss-of-function (values < −2). To the right side are shown the gain of function (values > 2). *P < 0.05, **P < 0.001. CDT = cold detection threshold; CPT = cold pain threshold; HPT = heat pain threshold; MDT = mechanical detection threshold; MPS = mechanical pain sensitivity; MPT = mechanical pain threshold; PPT = pressure pain threshold; TSL = thermal sensory limen; VDT = vibration detection threshold; WDT = warm detection threshold; WUR = wind up ratio.

Figure 2

Correlation of QST findings and severity of RDEB disease. Pearson’s correlation analyses were performed to explore associations between findings on the quantitative sensory profile and the severity of RDEB measured using the Birmingham severity score (BEBS). (A) The correlation coefficient (r = −0.64) showed that there was a negative correlation between cold detection threshold (CDT) and severity of RDEB disease. This correlation was highly significant (P = 0.005), n = 29 patients. (B) The correlation coefficient (r = −0.78) showed that there was a negative correlation between warm detection threshold (WDT) and severity of RDEB disease. This correlation was highly significant (P < 0.001), n = 29 patients. (C) The correlation coefficient (r = −0.58) showed that there was a negative correlation between thermal sensory limen (TSL) and severity of RDEB disease. This correlation was highly significant (P = 0.002), n = 29 patients. (D) The correlation coefficient (r = −0.61) showed that there was a negative correlation between heat pain threshold (HPT) and severity of RDEB disease. This correlation was highly significant (P = 0.001), n = 29 patients. (E and F) There was no significant correlation between the mechanical pain sensitivity and the wind up ratio and the severity of disease (r = 0.003, r = 0.16, respectively) n = 29 patients.

Figure 3

RDEB patients have a reduced IENFD. (A) Representative sections of skin biopsy of two RDEB patients and an age matched control volunteer immunostained with PGP9.5 (a pan neuronal marker). The dashed red line indicates the limit between dermis and epidermis. In the control subject it can be seen that there are several thin fibres crossing the dermo-epidermal border (yellow arrow). Conversely, the skin of RDEB patients shows very few, if any, fibres crossing into the epidermis. Note that there is no active skin blistering observable at the site of the biopsy. Scale bar = 100 μm. (B) The graph shows the mean IENFD expressed as fibres per mm as well as single data points of every subject. RDEB patients have a significantly lower IENFD than the matched control group (P < 0.001). Note that in A and B the controls subjects are from our own Chilean cohort. (C) Pearson’s correlation analysis was performed to explore the association between IENFD and the severity of RDEB measured using the BEBS sore. There was a negative correlation coefficient of r   =  −0.5.This correlation was significant (P = 0.04), n   = 18 patients. In this panel, z-scores were generated using published normative data (Provitera et al., 2016).

Figure 4

Sympathetic skin response and autonomic innervation. The sympathetic skin response measures changes in skin conductance, which depends on the presence of sweat. Sweating is controlled by the sympathetic nervous system, and therefore skin conductance is a surrogate measure of activity of the sympathetic system. In A, representative traces of sympathetic skin response are shown (control and RDEB, respectively). Responses were elicited by suddenly asking the subject to take a deep breath. In B we show the mean sympathetic skin response amplitude of control subjects and RDEB patients. In C we show quantification of PGP 9.5 immunofluorescence signal in sweat glands and in blood vessels in RDEB and control subjects. In D representatives sections of dermis of control and RDEB are shown. These sections were immunostained with CD31 to label endothelial cells (green), PGP9.5 to label nerves (red), and DAPI to show nuclei and the tubular structure of sweat glands (blue) (n = 10 per group, scale bar = 30 μm).