121 Although the diagnostic criteria suggest that definite SFN can be diagnosed with abnormal TTT and/or decreased IENFD, decreased IENFD has been suggested as surrogate gold standard, based on the fact that it is an objective measurement.37 In clinical practice, the majority of SFN diagnosis relies on an abnormal TTT rather than decreased IENFD, revealing the low sensitivity of the latter test.16,38–42 For example, a recent large study described the role of both IENFD and TTT in the diagnostic trajectory of SFN in 243 patients.42 In this cohort, 50% of patients had decreased IENFD, while 90% of patients with SFN showed an abnormal TTT. Given that TTT plays an important role in the diagnosis of SFN in current clinical practice, regardless of the underlying cause, our findings regarding a higher diagnostic yield of TTT in patients reporting continuous length-dependent pain need to be further elucidated. If this is confirmed in other studies, it could be debated whether in-depth phenotyping of pain symptoms should be taken into account when using a diagnostic test such as TTT in the diagnostic trajectory of SFN, especially in patients with inflammatory disease who are likely to have a non-length-dependent phenotype. In addition to distinguishing between length-dependent and non-length-dependent phenotypes, the SFNPQ skin pain data distinguished between continuous and intermittent pain. Of the four different phenotypes, continuous non-length-dependent pain was the only phenotype less reported in patients with probable SSFN. The fact that only continuous length-dependent pain correlates with TTT NOAs and the SFNSL implies that intermittent pain cannot be measured by any diagnostic method. It is important to note that only subjective measures correlated (continuous length-dependent pain, the SFNSL and TTT NOAs), while none of the objective diagnostic methods did. Moreover, intermittent pain likely has different pathological mechanisms than continuous pain, which are relatively unknown and more difficult to understand and measure. The results show that the diagnostic criteria, based on symptoms, clinical signs (dominant peripherally determined) and abnormal QST at the feet and/or reduced IENFD, tend to be more sensitive for patients that report continuous length-dependent pain. The SFNSL questionnaire, although no diagnostic test, was validated against the diagnostic criteria and inherits the same bias.23 Therefore, awareness of this bias should be increased when validating new methods against these diagnostic criteria. A limitation of our study was the lack of 50 µm IENFD for the assessment of SFN according to the EFNS guidelines27 and a limited sample size of patients with SFN. The thin tissue sections result in more fragmented nerve structures rather than entire nerve branches. Furthermore, the counting method was modified to count nerve fragments instead of whole structures. This might have contributed to the fact that no association was found between IENFD and a length dependent phenotype in our cohort. Another limitation is that only CCM was hypothesized to correlate with patient-reported non-lengthdependent pain. For example, a proximal lower extremity biopsy could have been used to diagnose non-length dependent SFN. Consequently, an association between patient-reported non-lengthdependent pain assessed by skin biopsy and the SFNPQ cannot be evaluated. The proximal biopsy was not performed due to unknown added value of the new staining protocol based on 10 µm sections. In addition to proximal skin biopsy, the quantitative sensory axon reflex test (QSART) could have been used to assess non-length-dependent SFN. Like the Sudoscan, QSART assesses the sudomotor function. However, Sudoscan is only performed on the hands and feet, while QSART measures an indirect reflex-mediated sweat response over time at both proximal and distal sites of the limbs.43 Therefore, QSART would have been a useful addition to this study.44 A limitation of the SFNPQ is that it remains unclear whether the symptoms are actually related to SFN or not. Like the general symptoms presented in Table 1, the symptoms are not specific for SFN and 7 127 7
RkJQdWJsaXNoZXIy MjY0ODMw