Research ethics and study population
This single center observational study was conducted at a tertiary university hospital based on a retrospective review of prospectively collected clinical and radiologic data. The study was conducted in accordance with the tenets of the Declaration of Helsinki. The study protocol was approved by the Pusan National University Hospital Institutional Review Board (H-2008-013-094). Informed consent was obtained from all subjects.
Between May 2015 and October 2019, 348 patients with intertrochanteric fracture of the femur were identified in our institution. Inclusion criteria of this study were patients with intertrochanteric fractures treated with an intramedullary nail. Exclusion criteria included patients who had 1) follow-up time of less than 6 months, 2) pathologic fracture, 3) open fracture, 4) surgery using extra-medullary device, 5) age younger than 65 years, and 6) multiple fractures in the affected limb. A total of 323 patients underwent surgical treatment, and 97 of them were excluded in this study (Figure 1).
All surgeries were performed on a radiolucent fracture table for fluoroscopic guidance with the patient in supine position. An intramedullary device (proximal femoral nail anti-rotation and trochanteric femoral nail anti-rotation, Depuy Synthes GmbH, Oberdorf, Switzerland) was used for fracture fixation. The neck shaft angle of the proximal femur on the coronal plane was restored using a traction device on the fracture table in reference to the normal side of the femur, and anteromedial cortical reduction was assessed through a trans-axial image. When posterior displacement of the proximal fragment (unacceptable reduction) was identified on the trans-axial image, indirect reduction was conducted using reduction tools to make a cortical buttress in the anteromedial part of the fracture (Figure 2). We used a helical blade in all patients and made a tip apex distance of less than 25 mm, and Cleveland index five or eight [14, 15].
We classified an intertrochanteric fracture as stable or unstable based on Evans’ classification . An unstable fracture is an intertrochanteric fracture with a large posteromedial defect. A basicervical intertrochanteric fracture refers to more than 50% of identified fracture line proximal to the intertrochanteric crest where the anterior capsule of the hip joint is attached to. However, when all fracture lines were located more than 5 mm proximal to the intertrochanteric crest, the fracture was considered to be a femoral neck fracture. This concept is a modified definition of the basicervical intertrochanteric fracture by Hu et al. , and this type of fracture can be either a partial or total an intracapsular fracture (Figure 3).
Postoperative reduction was divided into three categories, extramedullary, anatomical, and intramedullary reductions based on the displacement of the anteromedial cortex of the proximal fragment on the trans-axial image  (Figure 4). Extramedullary and anatomical reductions were considered acceptable reductions and intramedullary reduction was considered an unacceptable reduction. We assessed the status of the anterior joint capsule using 3D CT in all patients with basicervical intertrochanteric fractures. When the continuity of the anterior capsule was identified without detachment, tear, or avulsion from bony insertion, we considered it to be an intact anterior capsule. When this intact anterior capsule was located between the two fragments, we considered it to be an incarcerated anterior capsule (Figure 5). Radiologic follow-up was conducted 1, 3, 6, 9, and 12 months after surgery. We defined union as when the cortical bridging was identified at least three cortices in the anteroposterior and lateral image of the hip, with a patient that could ambulate without pain . Non-union was defined as when the proximal or lateral migration of the helical blade, cut-out or cut-through of the helical blade, varus deformation of the proximal fragment, implant failure around the fracture site, and obvious fracture line with antalgic gait at 9 months follow-up X-ray or CT were identified . All radiologic assessments were independently conducted by two orthopedic surgeons who had more than 10 years of experience and had never participated in patient treatment during this study. When they could not reach a consensus, the decision was based on a third orthopedic surgeon’s assessment.
Data were analyzed using SPSS software (ver. 17.0 for Windows; IBM, Armonk, NY). Preoperative data including age, sex, bone mass density (BMD), body mass index (BMI), and pre-injury mobility score of enrolled patients were collected based on the review of medical records. Preoperative data were used as covariates with pre- and postoperative radiologic values, including fracture patterns (unstable or basicervical type), tip apex distance, type of postoperative reduction (acceptable or unacceptable), and continuity of anterior capsule for logistic regression analysis to determine the risk factors for surgical failures. After unadjusted analysis, significant variables (p value of 0.10) were included in the model for adjusted analysis. A multivariate logistic regression analysis was performed to examine the association between possible risk factors and surgical failure. We report odds ratios (ORs) and 95% confidence intervals (CIs) for all associations. A p value of < 0.05 was considered significant. The inter-observer reliability of all radiologic values was evaluated using the interclass correlation coefficient. The results were interpreted as follows: > 0.8 = almost perfect agreement, 0.7 - 0.8 = strong agreement, 0.5- 0.6 = moderate agreement, 0.3 - 0.4 = fair agreement, and 0 - 0.2 = poor agreement. P < 0.05 was considered statistically significant. Inter-observer reliabilities for unstable fracture, basicervical fracture, postoperative reduction, and continuity of anterior capsule are shown in Table 1. A chi-square test was used to evaluate statistical correlation between continuity of anterior capsule and radiologic values in patients with a basicervical intertrochanteric fracture.