Objectives: At present, the dose conversion ratio for a continuous intravenous infusion of fentanyl (CIV) and fentanyl transdermal patches (TP), which are widely used in Japan, is not based on the results of clinical studies in Japanese patients. Studies comparing serum fentanyl concentrations in patients with cancer pain treated by TP showed large differences between Japanese patients and those in other countries. We therefore studied the dose conversion ratio in Japanese patients. Methods: From October 2003 through October 2008, we extracted information on all patients with gastrointestinal cancer who underwent rotation from CIV to TP in the gastrointestinal ward of Kitasato University East Hospital. We selected patients in whom the daily dose of CIV or TP (i.e., the basic dose) was unchanged for 10 days after rotation and the difference in the number of rescue doses (per day) as compared with immediately before rotation was 1 or less on at least 3 consecutive days. All TP preparations used in this study were reservoir-type. Regression lines were plotted on the basis of the relation of “the basic released dose of TP” to “the basic prescribed dose of CIV,” and the dose conversion ratio was calculated. Results: 47 patients underwent opioid rotation, and 11 of them satisfied the eligibility criteria. Eleven patients were studied. The following regression equation was obtained: Y=1.0227X+1.0103, r²=0.9188, indicating a strong correlation. The dose conversion ratio of CIV to TP (released dose) derived by regression analysis was 1:1. Conclusions: Our results obtained in Japanese patients will allow dose conversion at the time of opioid rotation from CIV to TP to be more appropriately performed.