RESULTS: Ninety-four patients were recruited with no dropouts. Mean age was 49.3 (SD = 14.2) with majority being female (74.5%). Ethnic distribution recorded 42.6% Chinese, 38.3% Malay, followed by 19.1% Indian. Mean duration of surgery was 123.8 min (SD = 56.5), and mean length of hospital stay was 4.7 days (SD = 1.8). The characteristics of patient in both groups were generally comparable except that there were more cases of total thyroidectomy in the diclofenac group (n = 31) as compared to the bupivacaine group (n = 16). Mean pain score peaked at immediate post-operative period (post-operative 0.5 h) with a score of 3.5 out of 10 and the level decreased steadily over the next 20 h starting from 4 h post-operatively. Pre-incision wound infiltration using diclofenac had better pain control as compared to bupivacaine at all time interval assessed. In the resting state, the mean post-operative pain score difference was statistically significant at 2 h [2.1 (SD = 1.5) vs. 2.8 (SD = 1.8), p = 0.04]. During neck movement, the dynamic pain score difference was statistically significant at post-operative 1 h [2.7 (SD = 1.9) vs. 3.7 (SD = 2.1), p = 0.02]; 2 h [2.7 (SD = 1.6) vs. 3.7 (SD = 2.0), p = 0.01]; 4 h [2.2 (SD = 1.5) vs. 2.9 (SD = 1.7), p = 0.04], 6 h [1.9 (SD = 1.4) vs. 2.5 (SD = 1.6), p = 0.04] and 12 h [1.5 (SD = 1.5) vs. 2.2 (SD = 1.4), p = 0.03]. Mean dose of tramadol used as rescue analgesia in 24 h duration was lower in the diclofenac group as compared to bupivacaine group [13.8 mg (SD = 24.9) vs. 36.2 mg (SD = 45.1), p = 0.01]. The total cost of analgesia used was significantly cheaper in diclofenac group as compared to bupivacaine group [RM 3.47 (SD = 1.51) vs. RM 13.43 (SD = 1.68), p