Abstract Small fiber neuropathy is a common complication in patients with sarcoidosis and its prevalence is estimated at 40–86%. The underlying mechanism influences the presentation of small fiber neuropathy. For example, patients with metabolic diseases are often associated with a classic length-dependent small fiber neuropathy pattern, while patients with inflammatory diseases are more often present with a non-length-dependent small fiber neuropathy. Detailed phenotyping may be useful to improve diagnostic efficiency, as a clue to underlying mechanisms and as a precondition for personalized medicine. This study examined four phenotypes distinguishing between length-dependent and non-length-dependent presentation with a new subdivision for continuous and intermittent presentation. Forty-eight sarcoid patients with symptoms and at least two clinical signs of small fiber neuropathy and normal nerve conduction studies were classified as having probable small fiber neuropathy. A new small fiber neuropathy phenotyping questionnaire has been developed that allows patients to mark the anatomical locations of pain at three different levels: the skin, muscles, and joints. The location of symptoms was used to define length dependence, and two colors were used to distinguish continuous (red) from intermittent (blue) symptoms. In addition, skin biopsy, corneal confocal microscopy, Sudoscan and water immersion skin wrinkling were used to investigate a correlation between the four phenotypes, sensory function, nerve fiber density, and autonomic nerve function. Overall, 35% of patients with probable small fiber neuropathy showed length-dependent symptoms and 44% showed non-length-dependent symptoms while 21% suffered from non-neuropathic musculoskeletal pain. The distinction between intermittent and continuous symptoms showed significantly less continuous than intermittent non-length-dependent symptoms (odds ratio = 0.3, P = 0.01). Moreover, continuous length-dependent symptoms were the only phenotype that correlated with thermal threshold testing (R = 0.3; P = 0.02) and the small fiber neuropathy screening list (R = 0.3; P = 0.03). In addition, thermal threshold testing (TTT) also correlated with the small fiber neuropathy (SFN) screening list (R = 0.3; P = 0.03). Other diagnostic methods showed no correlation with any of the four defined phenotypes. A novel finding is that TTT is only associated with continuous length-dependent pain, suggesting that TTT could result in more false negatives in patients with other pain phenotypes. Determining the pathophysiologic mechanisms could help develop new diagnostic methods. If patients suspected of SFN show symptoms without a length-dependent continuous presentation, the diagnosis should focus less on the diagnostic methods used.