Diagnosis of occult scaphoid fractures: a randomized, controlled trial comparing bone scans to radiographs for diagnosis.

BACKGROUND: Many patients with suspected scaphoid fractures but negative radiographs are immobilized for ≥ 2 weeks and are eventually found to have no fracture. Bone scans are reportedly 99% sensitive for these injuries if done ≥ 72 hours postinjury. OBJECTIVE: The purpose of this study was to determine if early bone scans would allow for shorter cast immobilization periods in patients with suspected scaphoid fractures. METHODS: Twenty-seven patients with clinically suspected scaphoid fractures and negative radiographs were randomized to early diagnosis (bone scan within 3-5 days; n  =  12) or traditional diagnosis (radiographs 10-14 days postinjury; n  =  15). The primary outcome was number of days immobilized in a cast. RESULTS: The mean number of days immobilized was 26 in the traditional group and 29 in the bone scan group. Overall, 6 patients had scaphoid fractures (2 in the traditional diagnosis group and 4 in the bone scan group; p > 0.05), and 8 had other types of fractures. These other types of fractures included four distal radius fractures, two triquetral fractures, one trapezoid fracture, and one hamate fracture. There was no significant difference in the number of other types of fractures between groups. The Kaplan-Meier survival analysis using the log-rank test revealed that there was no statistically significant difference between days immobilized between the radiograph and bone scan groups (p  =  0.38). CONCLUSIONS: The current study suggests that the use of bone scans to help diagnose occult scaphoid fractures does not reduce the number of days immobilized and that the differential diagnosis of occult scaphoid fractures should remain broad because other injuries are common.

Randomized trial of computer-assisted knee arthroplasty: impact on clinical and radiographic outcomes.

Computer-assisted surgery in knee arthroplasty is gaining popularity; however, the resulting outcome improvement is controversial. A double-blinded trial was performed with subjects randomized to undergo surgery with either computer-assisted or nonassisted instruments. Postoperatively, limb and implant alignment and rotation were assessed using both full-length radiographs and computed tomography in addition to clinical scores. One hundred twenty patients (141 knees) were randomized. No differences in Western Ontario MacMaster Osteoarthritis Score, Short Form-36, or flexion were seen. More varus limb alignment was seen in the computer-assisted group (1.9°) vs the nonassisted group (0.9°, P = .04) with no improvement in alignment precision. Rotational alignment of the components did not differ between groups. Computer-assisted surgery appeared to have minimal effect on knee implant arthroplasty with no improvement in limb alignment or early functional outcome.