We sought to compare operative times and complications for attending surgeons operating alone or with an assistant including an orthopedic resident, fellow, or physician assistant (PA) for closed reduction and percutaneous pinning (CRPP) of pediatric supracondylar humerus fractures (SCH), an archetypal procedure integral to the education of orthopedic trainees. Using a retrospective database collected following institutional review board approval, a 1-way ANOVA (non-parametric) was used to assess the effect of assistant absence or presence by type of assistant on mean operative time. We analyzed the association between the attending surgeon assistant categories and the complication rate using Fisher's Exact Test. The study was performed at Gillette Children's Specialty Healthcare, St. Paul, Minnesota, and Children's Minnesota in St. Paul and Minneapolis. These three affiliated metropolitan hospitals, the last of which is a level 1 trauma facility, all share the same group of orthopedic surgeons, trainees, and physician assistants. All patients under 14 years of age treated with CRPP for Gartland type 2, 3, 4 and flexion type closed supracondylar fractures between April 2006 and September 2016 were analyzed. Of 1053 patients identified by Current Procedure Terminology code 24358, data was available for 888 patients. Out of the 888 patients, 44.1% were operated on by a surgeon alone, 48.4% with a resident, 4.8% with a fellow and 2.6% with a physician assistant. The ANOVA revealed a statistically significant increase (p < 0.001) in operative times when a resident participated compared to the other categories. The shortest mean surgery time (34.7 minutes) occurred when an attending surgeon and PA were present. The longest mean time (44.3 minutes) occurred with a surgeon and resident. This difference, at 9.6 minutes, constituted 28% more time. There were 40 patients with a complication, a rate of 4.5%. The rate for each of the assistant categories was surgeon alone 5.6%, with a resident 3.3%, with a fellow 4.7%, with a PA 8.7%. The Fisher's Exact Test results showed no statistically significant association between the type or absence of assistant and the complication rate. Differences in operative times were observed across assistant categories although the mean time difference was statistically insignificant between the surgeon unassisted versus assisted by the fellow or a PA. The finding of statistically increased operative times with resident participation in this study for this specific procedure is consistent with results for most other reported orthopedic surgeries. Complication rates did not vary based on the participating personnel.
Read full abstract