Pattern of neuropathic pain induced by topical capsaicin application in healthy subjects

Human experimental pain models are widely used to study drug effects under controlled conditions, but they require further optimization to better reflect clinical pain conditions. To this end, we measured experimentally induced pain in 110 (46 men) healthy volunteers. The quantitative sensory testing (QST) battery (German Research Network on Neuropathic Pain) was applied on untreated ("control") and topical capsaicin-hypersensitized ("test") skin. Z-transformed QST-parameter values obtained at the test site were compared with corresponding values published from 1236 patients with neuropathic pain using Bayesian statistics. Subjects were clustered for the resemblance of their QST pattern to neuropathic pain. Although QST parameter values from the untreated site agreed with reference values, several QST parameters acquired at the test site treated with topical capsaicin deviated from normal. These deviations resembled in 0 to 7 parameters of the QST pattern observed in patients with neuropathic pain. Higher degrees (50%-60%) of resemblance to neuropathic QST pattern were obtained in 18% of the subjects. Inclusion in the respective clusters was predictable at a cross-validated accuracy of 86.9% by a classification and regression tree comprising 3 QST parameters (mechanical pain sensitivity, wind-up ratio, and z-transformed thermal sensory limen) from the control sites. Thus, we found that topical capsaicin partly induced the desired clinical pattern of neuropathic pain, in a preselectable subgroup of healthy subjects to a degree that fuels expectations that experimental pain models can be optimized toward mimicking clinical pain. The subjects, therefore, qualify for enrollment in analgesic drug studies that use highly selected cohorts to enhance predictivity for clinical analgesia.