Opioid-switching (OS) is usually performed with conversion methods based on the equianalgesic dose table. However, the conversion ratios might lead to significant differences in clinical practice. No clear guideline exists for safe, effective switching from transdermal fentanyl (TF) to oxycodone injection (OXJ). We retrospectively investigated the adequate analgesia dose of OXJ in OS from TF by comparing with the equianalgesic calculated dose based on the conversion ratio of 1.0 : 41.7 between TF and OXJ. Patients with a pain scale score of 0 were assigned to the “NRS/VRS=0” group (n=4), and the remaining patients were assigned to the “NRS/VRS>0” group (n=27). During a 4-year period, 31 of 49 patients with cancer-related pain who underwent OS from TF to OXJ were investigated. All patients in the NRS/VRS=0 group (4/4, 100%) and most in the NRS/VRS>0 group (23/27, 85%) achieved adequate analgesia. Among the 27 patients with adequate analgesia, the median effective OXJ dose was 28% (interquartile range, 21-47) of the equianalgesic calculated dose in the well-controlled group and 103% (interquartile range, 71-164) in the poor analgesia group. Strong drowsiness developed in two patients the day after OS, and the OXJ dose was decreased. Our findings suggest that in patients without pain, it might be necessary to reduce the OXJ dose to approximately 30% of the equianalgesic calculated dose for safe OS from TF. Pain intensity and drowsiness due to an opioid overdose should be carefully monitored and may require dose adjustment.