Objective To compare the analgesic effects of ultrasound‑guided continuous costoclavicular brachial plexus block (CCB) and continuous interscalene brachial plexus block (ISB) for the treatment of upper limb fractures. Methods A total of 70 patients who were scheduled for internal fixation of upper limb fractures were selected. According to the random number table method, they were divided into two groups (n=35): a general anesthesia+continuous CCB group (group C) and a general anesthesia+continuous ISB group (group I). Patients in both groups underwent ultrasound‑guided CCB or ISB, respectively. Then, the operation was performed under general anesthesia. After operation, an electronic analgesic pump was connected. Their nerve block operation time, operation time, and the intraoperative dosages of propofol, remifentanil and dexmedetomidine were recorded. Their sensory and motor nerve blocks of the median nerve, ulnar nerve, radial nerve and musculocutaneous nerve were recorded 5, 10, 15, 20, 25, 30 min after nerve block. The Numerical Rating Scale (NRS) scores at resting and during movement 8, 12, 24 h and 48 h after surgery and the times of effective pump pressing 48 h after operation were recorded. Postoperative nausea, vomiting, lethargy and other adverse reactions were recorded. Results Puncture and catheterization were successful in both groups, and no complications related to block occurred. The operation time of nerve block in group C was shorter than that in group I (P<0.05). There was no statistical significance in the operation time and the intraoperative dosages of propofol, remifentanil and dexmedetomidine between the two groups (P>0.05). The success sensory and motor block rates of the median nerve, ulnar nerve and musculocutaneous nerve in group C were higher than those in group I, 5 min and 10 min after nerve block (P<0.05), and there was no significant difference in the success sensory and motor never block rates between the two groups at other time points (P>0.05). Compared with group I, NRS scores at resting and during movement 8, 12, 24 h and 48 h in group C were reduced (P<0.05), and the times of effective pump pressing decreased (P<0.05). There was no statistical difference in the incidence of postoperative adverse reactions between the two groups (P>0.05). Conclusions Continuous CCB can provide effective postoperative pain control for patients with upper limb fractures.