25 Since no single method is sensitive enough to confirm or exclude SFN, a combination of multiple methods seems to be the best alternative. The more abnormal test results, the more secure the diagnosis will be. A recent study investigated six different methods and suggested even a combination of four methods (skin biopsy, Electrochemical Skin Conductance (ESC), Laser Evoked Potentials (LEP) and QST) for a definite diagnosis.53 In order to classify SFN, the following definitions are preferably used.54–56 1. Possible SFN: symptoms or clinical signs of small fiber damage 2. Probable SFN: symptoms or clinical signs of small fiber damage and normal sural nerve conduction studies 3. Definite SFN: symptoms or clinical signs of SFN-damage, normal sural nerve conduction studies and decreased intra-epidermal nerve fiber density (IENFD) and/or abnormal quantitative sensory testing (QST) thermal thresholds Diagnosis To diagnose isolated SFN, physicians should be aware of the broad range of symptoms that may exhibit. During physical examination, tendon reflexes should be normal, and there should be no signs of muscle weakness. Vibration sense and proprioception may vary, appearing either normal or abnormal.57 QST or skin biopsy are recommended for a definite diagnosis.50,55 However, these methods are time-consuming and are not widely available in every hospital. Objectives Clarify which diagnostic methods are available Define diagnostic accuracy of each method Present a clinical diagnostic workflow Methods First, the huge amount of available diagnostic methods will be clarified to show an organized overview. Next, a systematic search is performed to develop an overview of diagnostic accuracy (AUC-values) for each method. The results of a systematic literature search between 2000-2019 are presented focusing on the sensitivity and specificity of all diagnostic methods. It is important to state that since no gold standard is available, the AUC-values are relative measures, based on an imperfect standard. Literature search is performed on 19 August 2019 in PubMed and Embase. Exclusion criteria were: Publication date <2000 Case reports Language other than English Animal study Large fiber neuropathy AUC-values are calculated based on published sensitivity and specificity values. Articles were included when sensitivity and specificity were clearly published or could be calculated if an overview of test results for all participants was available. The classification of definite SFN is used, in order to determine sensitivity and specificity. Only articles with isolated SFN were included. Figure 3 shows the search results and exclusion criteria. Review articles, animal models and case reports are labeled as “wrong study design”. In the end, several phenotypes of SFN are described based on different symptoms. Depending on the phenotypes, optimal diagnostic methods are suggested. 2 27 2
RkJQdWJsaXNoZXIy MjY0ODMw