Background and objectives: Warfarin sodium therapy is usually initiated with a loading dose to reduce the time needed to reach the target international normalized ratio (INR). The right initial dose of warfarin is controversial. This study aimed to determine the best initial dose of warfarin in terms of safety and efficacy. Patients and methods: This is a prospective hospital-based study. Fifty-nine patients who were admitted to the coronary care unit with clinical indication for warfarin were assigned to use the initial dose of warfarin 5 mg or 10 mg with the target INR of 2-3. An ethical clearance was obtained from the state ministry of health. A written consent was taken from each patient. An INR was measured at baseline and then after 72 hours. The data were then collected using a data sheet including age, gender, baseline INR, initial dose of warfarin, INR after 72 hours and whether the patient developed bleeding or not. Patients on Aspirin and other non-steroidal anti-inflammatory drugs were excluded. The data were analyzed using SPSS version 18. Results: A total number of 59 patients were enrolled in the study. In this series, 59% of patients were females. The mean age of patients was 56 years. Five mg loading dose was used in 35 of patients (59%) while 10 mg was used in 24 patients (41%). The mean INR after 72 hours was 2.8 for those with the initial dose of 5 mg. For those with the initial loading dose of 10 mg the mean INR after 72 hours was 3.2. Minor bleeding occurred in three patients, two of them used the initial loading dose of 5mg and only one patient from the group of 10mg. Conclusion: From this study, it seems that 5 mg loading dose of warfarin is reasonable and cost effective; nevertheless, 10mg initial loading dose is not associated with increased risk of bleeding.