Advancing methods for characterizing structure and functions of small nerve fibres in neuropathic conditions

1 Improving diagnostic accuracy of painful distal sensory polyneuropathies

The authors address in this report a very important question: how to improve the diagnostic accuracy of painful distal sensory polyneuropathy (DSP). Painful DSP is worldwide a large problem, affecting many millions of patients. Diabetes mellitus is the most common cause of neuropathy [2] and with a considerable world-wide increase in diabetes, the occurrence of diabetic neuropathy is increasing [2,3]. In addition, other diseases like hypothyroidism [4], HIV [5], connective tissue diseases [6], and a large variety of inherited diseases may be complicated by or manifested by a painful neuropathy [7]. More patients are now surviving cancer, but an increasing number are surviving with chemotherapy-induced peripheral neuropathy [8].

2 Investigations of large fibre neuropathy not useful for small fibre neuropathy

EMG with neurography (nerve conduction study, NCS) is the standard investigation of a neuropathy but would only be applicable for a large-fibre neuropathy. The method will not be suitable for assessment of pure small-fibre neuropathies. There are several methods available for the diagnosis of small-fibre neuropathies.

Intraepidermal nerve fibre density (IENFD) is considered to be the most reliable method, followed by assessment of thermal thresholds [9]. It is however known that up to 12% of patients diagnosed clinically with a small fibre neuropathy will have normal IENFD [9]. In addition, this structural method would not be able to differentiate between an affection of the small myelinated A-delta fibres and the small unmyelinated C fibres. As emphasized by Karlsson and coworkers [1], there is a weak or moderate association between IENFD and spontaneous as well as evoked pain [10,11,12,13,14], i.e. in most cases the major complaints of the patients.

Quantitative sensory testing (QST) with measurements of thermal thresholds, which is not primarily a test for pain (except cold and heat pain), but provides an access to the functional properties of small nerve fibres. The sensation of cold is mediated by A-delta fibres, while the sensation of warmth is mediated by the small unmyelinated C-fibres. The diagnostic efficiency of QST was found to be less than for IENFD in the diagnosis of small-fibre neuropathies [9].

Quantitative sudomotor axon-reflex test [15,16] only assesses the function of the efferent sympathetic sudomotor nerve fibres, and thereby may not specifically be of interest for pain.

Axon-reflex flare is yet another method [17,18,19], testing the function of a subgroup of small nerve fibres, the so-called insensitive (or silent) afferent C nociceptive fibres [20]. These fibres normally do not respond to painful stimuli, but they may be sensitized in cases of disease (also painful neuropathy) (see e.g. [21,22,23,24]). In painful neuropathy, the size of the flare would normally be decreased as a measure of the impairment of these specific nerve fibres [17,18].

However, in a study of patients with neuropathy with and without pain [25], there was a trend towards more patients with pain having normal flare induced by electrical stimulation compared with those with non-painful neuropathy. A possible explanation would be that hyperexcitable C-fibres in patients with pain are producing a bigger flare response by means of reduced electrical excitation thresholds, which would compensate for the C-fibre loss [25].

Heat-evoked potentials may be employed [26], but this is not a frequently employed method for the diagnosis of small-fibre neuropathy.

3 What is the gold-standard for assessment of small fibres?

There is no gold-standard for the choice of small-fibre testing, all methods offer advantages and disadvantages. Therefore, it is of great importance to compare different methods in the same group of patients.This is one of the strengths of the study by Karlsson and coworkers: they compared IENFD, including new structural methods, QST, and capsaicin-induced vasodilatation, thereby combining structural and functional tests of small nerve fibres.

4 Comparison of patients with painful distal sensory polyneuropathies with normal subjects

Karlsson and coworkers studied 17 patients with painful DSP of different etiologies (diabetes, following chemotherapy, connective tissue disease, and idiopathic) and with pain intensities ratings ≥4 on a 0–10 numerical rating scale. They compared those 17 pain patients with 19 healthy control subjects. The authors found decreased sensibility to cold and warmth in patients compared with healthy subjects. They further report a lower pain rating following application of capsaicin and a reduced local vasodilatation response. All variables of the structural studies (IENFD, eNFLD, dNFLD and swelling ratios) differed in the patients compared with healthy subjects (lower nerve fibre densities but higher swelling ratios). Correlations between the various measures were different in healthy subjects compared with the patients, because of pathological changes in the nerve fibres.

An important finding was that while sensitivity and specificity of IENFD alone was 82.4% (related to the clinical diagnosis of small fibre involvement in DSP), the sensitivity increased to 100% when combining IENFD and eNFLD (epidermal nerve length density). Previous studies have shown that a combination of IENFD and dermal NFLD (nerve fibre length density) increased diagnostic specificity and sensitivity in diagnosing small fibre neuropathy. Combining other test results or adding a third measurement to IENFD and eNFLD did not result in a higher specificity, and surprisingly cold detection thresholds and cold pain thresholds did not correlate with the morphological small fibre variables.

Combination of IENFD with other measures from QST (warmth sensitivity index, WSI) would also increase the diagnostic sensitivity to 94.1%, but with no alteration of diagnostic specificity (84.2%). This would possibly imply that the correlation between morphology and function is more direct in heat-mediating C-fibres than in the cold-mediating A delta fibres. The WSI, representing the range in which non-noxious heat is perceived, may be an important measure for diagnosis of small fibre neuropathy, as also described previously [27].

5 The importance of Karlsson and coworkers study [1]

The results of Karlsson and coworkers’ study are interesting and increase our knowledge on possible improvements of small fibre diagnosis in painful neuropathy. There are several previous studies reporting correlations between various methods (among them [25,28,29]). In the present study, the authors have compared results from patients having painful neuropathy with healthy subjects. Although there are clear differences between these two groups, the results do not necessarily reflect pain. In order to test the importance of the results for pain, the authors would have had to include also patients with neuropathies without pain, in addition to correlating the findings with pain intensities.

However, as pointed out by Schley et al. [29], in a larger study of 36 patients with neuropathic pain, the severity of small fibre neuropathy did not correspond to clinical pain intensity, and they could not identify a specific variable or pattern that would predict pain intensity in peripheral neuropathy. Schley et al. [29] therefore concluded that skin innervations (here also at different depths) correlate with small fibre function but not with pain. This implies, that although we have an increasing amount of methods for assessing small fibres (structural and functional), we still do not know their importance for pain.

6 When large fibres are damaged, small-fibres are also damaged: small fibre testing is superfluous

One interesting finding in Karlsson and coworkers’ study is that the patients with normal results from large fibre nerve conduction study (NCS), in 7 of 13 patients tested, had higher IENFD values compared with patients with an NCS-diagnosed mild to moderate sensorimotor axonal polyneuropathy. These data confirm previous studies [30], and as stated by Karlsson and coworker the data indicate that detectable damage to the larger nerve fibres occurs first when there is pronounced damage to the nociceptors. This is important for the following reason: many patients with a known polyneuropathy as shown by NCS are admitted to small fibre testing with the question of small fibre damage. Small fibre testing is whatever method employed time-consuming and with of ten long waiting-lists. Knowing that small fibres almost always will be damaged in a concomitant large-fibre neuropathy, there is no indication for additional small fibre testing in these patients.

7 Importance of silent C-nociceptors in dermis that mediate axon reflex flare

Another interesting finding is the lack of correlation between vasodilatation (size of flare) and IENFD. Karlsson and coworkers employed capsaicin to induce flare. Axon reflex flare is, as stated above, mediated by the mechano-insensitive (silent) C-nociceptors [20]. It is assumed that these fibres are located nearer to the blood vessels than to the epidermal surface [28], thereby deeper in the dermis than what is evaluated by IENFD. It has also previously been shown that the size of axon reflex flare correlates with dermal fibres, but not with epidermal fibres, supporting such an assumption [28]. A lack of correlation with IENFD would suggest that IENFD, possibly in combination with WSI, would assess the spatial- temporal (discriminative) aspects of nociception, while an axon-reflex flare mediated by deeper located nerve fibres, could possibly be more relevant for clinical neuropathic pain, although reduced flare areas cannot differentiate between painful and non-painful neuropathy [25].

To summarize, an increasing amount of tests will assess various subgroups of small nerve fibres. The relevance to the clinical neuropathic pain is still uncertain. We still need many studies correlating the different methods and in such a context, the contribution by Páll Karlsson and coworkers [1], is highly appreciated.