Is a Novel Fluoroscopic Intraoperative Reference System Superior to Conventional Management for Distal Radius Fracture Reduction? A Propensity-matched Comparative Study

Abstract

Abstract Background Preoperative planning is generally performed to simulate the process of reduction as well as to determine the size and placement of implants in patients undergoing distal radius fracture surgery. We previously described a three-dimensional (3D) digital preoperative planning system for the osteosynthesis of distal radius fractures, and we have developed a novel intraoperative referencing system that superimposes preoperative planning (such as plate position and length) onto fluoroscopic images during surgery; however, its efficacy has not been evaluated compared with conventional planning and surgery. Questions/purposes Does use of a novel intraoperative referencing system result in (1) better Mayo wrist scores at 3 and 6 months after surgery and (2) less loss of reduction in terms of ulnar variance, palmar tilt, and radial inclination on plain radiographs taken 1 week, 3 months, and 6 months after surgery compared with conventional preoperative planning? Methods Between April 2014 and October 2021, we treated 294 patients with open reduction and volar plate fixation for distal radius fractures. Of 294 patients, 65% (191) underwent surgery using either conventional preoperative planning or a novel intraoperative referencing system. The remaining patients were excluded because they were younger than 18 years, they had some missing medical records related to the clinical outcomes, or they had a previous history of upper extremity injuries. During that time, we generally treated fractures with volar plates when there was: more than 2 mm of stepoff/gap in the articular surface, a dorsal tilt more than 15°, radial inclination less than 15°, or radial shortening more than 5 mm. Generally, we used a flexor carpi radialis approach. In some patients who had dorsal fragments, we added a dorsal approach. At that time, we were developing the new intraoperative referencing system, so it was not used consistently. To arrive at a fair assessment, we opted to perform propensity matching based on age, gender, and AO fracture type. During the period in question, 36% (69 of 191) of patients with distal radius fractures who received a volar plate were treated using our novel intraoperative referencing system, and 64% (122 of 191) had surgery using conventional preoperative planning (control group). Of those, 91% (63 of 69) of patients who were treated with the intraoperative referencing system and 89% (108 of 122) of those in the control group were available for follow-up with all imaging and Mayo wrist scores at least 6 months after surgery. After propensity matching, that left us with two groups of 39 patients, who were well matched in terms of age and fracture type; these were the study groups. We also tried to match them according to gender, but there were fewer patients in the intraoperative referencing group, and the percentage of women for each group differed: 70% (44 of 63) in the intraoperative referencing group and 76% (82 of 108) in the control group. Also, there were fewer men with C3 fractures in the control group. Therefore, 64% (25 of 39) of patients in the intraoperative referencing group were women and 77% (30 of 39) of patients in the control group were women. In the intraoperative referencing group, our novel intraoperative referencing system was used in combination with the 3D digital preoperative planning system for preoperative planning. In the control group, preoperative planning was performed manually in a conventional manner using tracing paper and implant templates or using a digital template. We compared the groups in terms of operative duration, the radiation dose used in surgery, and Mayo wrist scores at 3 and 6 months after surgery. We also compared the groups in terms of loss of reduction on ulnar variance, palmar tilt, and radial inclination on plain radiographs taken 3 months and 6 months after surgery. We considered the plain radiograph taken 1 week after surgery as a baseline. Each item was compared between the image fusion and control groups using a Welch t-test. Results Mayo wrist scores were no different between the intraoperative referencing system and the control group at 3 months (71 ± 7 versus 72 ± 11, mean difference 1 [95% CI -3.7 to 5.7]; p = 0.07) or at 6 months after surgery (76 ± 6 versus 79 ± 11, mean difference 3 [95% CI -3.5 to 7.9]; p = 0.12). There were no differences in surgical duration or radiation doses between the intraoperative referencing and control groups. We found only a small advantage in favor of the intraoperative referencing system in terms of loss of reduction on ulnar variance (3 months after surgery: 0.2 ± 0.4 mm versus 0.6 ± 0.7 mm, mean difference 0.4 mm [95% CI 0.15 to 0.69]; p = 0.003, 6 months after surgery: 0.4 ± 0.6 mm versus 0.8 ± 0.8 mm, mean difference 0.4 mm [95% CI 0.05 to 0.73]; p = 0.02 for the intraoperative referencing system and the control group, respectively). This difference in radial shortening was so small that it was not likely to have been clinically important. Conclusion We found no clinically important advantages from the use of our novel intraoperative referencing system except a slight improvement in ulnar variance. Therefore, we recommend against its use in everyday practice at this time. However, future improvements may lead to better clinical outcomes, so we plan further investigations. Level of Evidence Level III, therapeutic study.