In a prospective study assessing the risk of herpes zoster in more than 5000 UNC669 patients with rheumatoid arthritis treated with anti-TNF drugs, 86 episodes of herpes zoster were recorded in 82 patients; in two of these patients PHN occurred. In another study Zak-Prelich assessed serum cytokine levels and different subtypes of skin lymphocytes in patients with herpes zoster who did and did not develop PHN. Interestingly, a reduced number of cellular infiltrates was found at the lesion sites of herpes zoster patients that developed PHN compared to those who did no without association with systemic cytokine levels. The authors suggested that an impaired immune response upon zoster infection resulted in an impaired containment of infection and more damage of the dermatome as the basis for neuralgia. Here, we did not find infiltrates of T-cells and macrophages in affected skin of PHN patients compared to unaffected skin, which may be due to the different duration of disease in our study until biopsy, while in the study by Zak-Prelich et al. patients were seen at standardized time points. Since part of the pathology in PHN has been assumed to be located in the skin itself, also surgical excision of the lesioned skin has been tried. However, this approach did not lead to sustained pain relief, which indicates that the assumed pathology might be located upstream the pain pathway such as in dorsal root ganglia neurons. 
Interestingly, a recent case report described the analgesic effect of local laser therapy of the skin over a period of 13 months in a 73 year old women who had been suffering from PHN for 15 years. The author suggested that the reducing effect of laser on local cytokine expression might be the underlying mechanism of this observation, however, cytokine levels were not investigated. In another case report the analgesic effect of antiIL-2 that lasted throughout the three years of the follow-up period was described in a patient with PHN, but again cytokine levels were not measured before and after treatment. Histological assessment of skin samples also revealed denervation of affected skin compared to unaffected contralateral skin in patients with PHN. This finding is in accordance with pervious findings. Petersen et al. reported a 40% reduction in painful skin compared to painless skin. Similar results were found in the study of Buonocore et al.. In accordance with these studies we also did not find a correlation between IENFD and the presence or absence of allodynia. As outlined in Table 1, IENFD of patients with and without allodynia ranged from normal to complete denervation. Also, no correlation was found between IENFD and patients’ current pain intensity. Further studies are needed to decipher the underlying mechanisms that may range from hyperexcitable remaining nerve Butenafine hydrochloride fibers causing allodyina while IENFD is normal, to hyperactive dorsal root ganglion neurons independent of peripheral nerve fiber density. The major limitation of our study is the relatively low number of patients and controls. Although we were recruiting at two centers, only 13 patients with PHN were seen during the study period of three years. This may be due to the fact that in the majority of cases patients with PHN are not hospitalized but treated as outpatients. Our patient group was inhomogeneous with regard to disease duration, which may have influenced our results such that we missed early changes in cytokine gene expression.