Patients
The present study enrolled a consecutive series of patients with femoral intertrochanteric fractures who received osteosynthesis at our institute between January and December in 2015. The inclusion criteria of this study were 1) skeletally matured patients aged over 55 years old, 2) acute closed femoral intertrochanteric fractures as the single fracture of the body, 3) received osteosynthesis either with extramedullary or intramedullary implants, and 4) were followed up for at least 6 months postoperatively. The clinical management for the patients proceeded prospectively and the data were recorded retrospectively. This study protocol was approved by the Institutional Review Board of Chang Gung Memorial Hospital (IRB No. 201601598B0). All methods were performed in accordance with the relevant guidelines and regulations of the institution.
Surgical procedure
Once the patient was admitted to the hospital, the osteosynthesis was arranged as soon as possible after their medical condition had been optimized. All the osteosynthesis was conducted on a radiolucent traction table under real-time fluoroscopic examination throughout the surgical procedure. We used the Arbeitsgemeinschaft für Osteosynthesefragen (AO) classification as the reference to determine a “stable” or “unstable” femoral intertrochanteric fracture9,18. According to the AO classification, type 31-A1.1, 1.2, 1.3, and 2.1 were classified as stable fracture patterns while the rest types of 31-A were classified as unstable ones. We usually used an extramedullary plate (dynamic hip compression screw/plate [DHS], Depuy Synthes, Raynham, MA) for stable fracture patterns and intramedullary nails (Gamma 3 nail, Stryker, Kalamazoo, MI) and Proximal femoral nail anti-rotation II (PFNA II, Depuy Synthes, Raynham, MA) for unstable ones. However, intramedullary nails might occasionally be applied to the stable femoral intertrochanteric fracture for those with potentially unstable patterns.
Radiographic assessments
Standard X-ray assessments consisted of anteroposterior (AP) and lateral views before and after the osteosynthesis were applied. Patients were assessed at intervals of 1, 3, 6, 12 months after discharge. Thereafter, annual follow-ups were conducted annually. Routine radiographic follow-ups were arranged immediately after the surgery and on each clinic visit. Several parameters were applied during radiographic assessments, including:
- TAD: Sum of the distances between the tip of the lag screw to the apex of the femoral head on the AP and lateral views (Figure 1a & 1c) [10]. The true implant diameters were used for the correction of image magnification.
- CalTAD: Sum of the distance between the tip of the lag screw to the intersection of the femoral head contour and the tangential line of the medial femoral neck cortex on the AP view (Figure 1b), and the distance between the lag screw tip to the apex of the femoral head on the lateral view (Figure 1c) [13].
- The quality of the reduction was categorized into three grades based on a modification of the method developed by Baumgaertner et al [10]. The first criterion used was a normal or slightly valgus neck-shaft angle (130° to 150°) on the AP view and an angulation < 20° on the lateral view. The second criterion used was the presence of a < 4 mm displacement of any fragments in the AP and lateral views. A reduction was categorized as good if both criteria were met, acceptable if only one criterion was met, and poor if neither criteria were met.
- The lag screw tip position in the femoral head was classified by the Cleveland index (Figure 2a) [22].
- Femoral neck shortening: The femoral neck shortening distance was measured between the last follow-up and the immediate postoperative radiographs with the correction of the true implant diameters.
- Fixation failure was defined as lag screw cutout or progressive varus collapse of the femoral head that needs revision surgery.
Two surgeons (Y.-C. Y. and C.-H. L.) interpreted the postoperative AP and lateral plain films at each radiographic follow-up to evaluate the applied parameters. Regarding continuous data during assessments, such as TAD, CalTAD, and femoral neck shortening, the average value from the measurements of the 2 surgeons was recorded. For categorical data, the data was directly recorded when both surgeons had a similar opinion. When the result was different between the 2 main evaluators, a senior surgeon (Y.-H. H.) was consulted for the final decision.
Patients with a commonly placed lag screw or blade positions (center-center vs. inferior-center) were compared for radiographic statistics and incidences of fixation failure. The modes of fixation failure, which include screw cutout and progressive varus collapse, were recorded specifically. Subgroup comparisons in patients who received either extramedullary or intramedullary implants were also made for different lag screw positions, and TAD and CalTAD comparisons (TAD ≤ 25 mm vs. TAD > 25 mm and CalTAD ≤ 25 mm vs. CalTAD > 25 mm).
Statistical analysis
Statistical analysis was conducted using SPSS version 18.0 (SPSS Inc., Chicago, IL, USA). Continuous variables are presented as the mean and standard deviation. Categorical variables are reported as frequency and percentage. Parametric Student t-test and nonparametric Mann-Whitney U-test was used on continuous dependent variables. Pearson’s chi-squared test was used on categorical variables. A two-tailed p-value of 0.05 was considered significant.