68-2

37/110

inclusions of both early and late visitors22-24).However, our study clearly demonstrated that pain intensities in the late phase, rather than early phase, of acute HZ could be a much more reliable predictor of PHN. Because pain spontaneously resolves with healing of skin rashes within 2-3 weeks in many patients not developing PHN while pain persist beyond such periods in most patients developing PHN lasting for months or more1, 13, 14), it seemed quite reasonable that pain intensities in the late phase of acute HZ, rather than those in the early phase, could be more significantly associated with the development of PHN.The present study showed that pain scores at Day 7 and at Week 2-3 were significantly associ-ated with the development of PHN, and that AMNV reduced pain at Day 7 and Week 2-3 more effec-tively than FCV before and after PSM in the total cohort. Therefore, it could be expected that the use of AMNV would lead to a reduced incidence of PHN. In the present study, however, AMNV did not more effectively reduce the incidence of PHN, compared with FCV. Such discrepancies might be explained by the fact that intensities of acute pain account only for a part of the multifactorial etiology of PHN1, 12-15), as was suggested also by the rela-tively low coefficient of determination (r2) values provided by univariate logistic regression analyses (0.088 and 0.180 for NRS pain scores at Day 7 and Week 2-3, respectively). As mentioned above, however, AMNV could reduce pain at Day 7 and Week 2-3 more effectively than FCV not in early visitors but only in late visitors both before and after PSM. Further, in PSM pairs of late visitors, AMNV tended to be associated with a lower inci-dence of PHN, compared with FCV (p = 0.0633). Therefore, the possibility could not be completely excluded that compared with FCV, AMNV would reduce pain more effectively in late visitors, thereby reducing the risk of PHN in patients receiving the therapy later than 3 days of HZ onset.Clearly, our study had limitations resulting from the retrospective, uncontrolled fashion, such as some inaccuracy and/or incompleteness of data, and further studies are required to confirm the pain-re-lieving effect of AMNV, and to examine whether its use can help to reduce the incidence of PHN.AMNV could reduce acute HZ pain more effec-tively than FCV in late visitors receiving the therapy later than 3 days after onset, but not in early visitors receiving the therapy within 3 days. Intensities of acute HZ pain in the late phase of acute HZ was more predictive of PHN than those in early phase. Further, compared with FCV, AMNV might more effectively reduce the incidence of PHN in late visitors, although this possibility should be confirmed in a further prospective study.The authors wish to acknowledge Dr. Kazuhisa Takahashi, Hospital Director of Juntendo Univer-sity Hospital, for permitting us to access patients’ data in the medical record system. We thank all the staff in outpatient clinics at Departments of Dermatology; Anesthesiology and Pain Medicine; Otorhinolaryngology, Head and Neck Surgery; Ophthalmology; General Medicine; Rheumatology; Hematology; Diabetes, Endocrinology and Metabo-lism; Cardiovascular Medicine; Neurology; Plastic and Reconstructive Surgery; Gastroenterology; Respiratory Medicine; Nephrology and Hyperten-sion; Emergency Medicine; Obstetrics and Gyne-cology; Pediatrics and Adolescent Medicine; and Radiology and Radiation Oncology; for treating patients with acute herpes zoster, of whom valu-able data were utilized in the present study.No funding was received.All authors meet every criteria recommended at the International Committee of Medical Journal Editors (ICMJE). YK collected data, analyzed them using a simple statistical analysis software, and wrote the first draft of the manuscript. EI contrib-uted to conceptualization of this study, reviewed the manuscripts, and helped revise the first draft. KY reviewed the manuscript and advised the authors about additional data collection in improving the manuscript. MH was a major contributor in writing and editing the manuscript and analyzing all of the data. All authors read and approved the final manuscript.129ConclusionAcknowledgementFundingAuthor contributions