Complications of femoral neck fracture xation using a sliding hip screw with a side plate device in patients older than 65 years

A majority of femoral neck fractures are indicated for surgical treatment within the scope of applicable guidelines. However, the reported overall complication rate in elderly patients one year after hip fracture xation was 33.3%. The purpose of this study was to clarify the incidence of complications one year after femoral neck fracture xation using a sliding hip screw with a side plate device in elder than 65-year-old patients. We studied the results of 115 cases of the internal xation for femoral neck fractures in patients older than 65 years of age using the Twins device (HOYA Technosurgical Corporation, Tokyo, Japan). We could follow up 39 of these 115 cases at one year after xation and retrospectively investigated complications in these patients. Research outcomes were mortality, surgical site infection (SSI), loss of reduction, aseptic necrosis, late segmental collapse (LSC), nonunion, periprosthetic fracture, reoperation, and recovery of walking ability.


Abstract
Background A majority of femoral neck fractures are indicated for surgical treatment within the scope of applicable guidelines. However, the reported overall complication rate in elderly patients one year after hip fracture xation was 33.3%. The purpose of this study was to clarify the incidence of complications one year after femoral neck fracture xation using a sliding hip screw with a side plate device in elder than 65-yearold patients.

Methods
We studied the results of 115 cases of the internal xation for femoral neck fractures in patients older than 65 years of age using the Twins device (HOYA Technosurgical Corporation, Tokyo, Japan). We could follow up 39 of these 115 cases at one year after xation and retrospectively investigated complications in these patients. Research outcomes were mortality, surgical site infection (SSI), loss of reduction, aseptic necrosis, late segmental collapse (LSC), nonunion, periprosthetic fracture, reoperation, and recovery of walking ability.

Conclusions
In our study, the incidence of loss of reduction rate and periprosthetic fracture rate were relatively high. Other complications of the Twins device evaluated were equivalent to those reported previously.

Background
Recent estimates suggest that the annual incidence of hip fractures worldwide will increase to 2.6 million in 2025 and 4.5 million in 2050 [1]. A majority of these fractures are indicated for surgical treatment within the scope of applicable guidelines. However, the reported overall complication rate in elderly patients one year after hip fracture with all the currently used methods of xation was 33.3% [2].
Biomechanically, a sliding hip screw with a side plate device provides better stabilization than multiple cancellous screws for femoral neck fractures [3]. The side plate provides lateral support and prevents toggling of the cancellous screws in the femoral neck, resulting in inferior and posterior displacement of the femoral head relative to the femoral shaft [3].
The purpose of this study was to clarify the incidence of complications one year after femoral neck fracture xation using a sliding hip screw with a side plate device in patients older than 65 years of age and to evaluate the superiority of this device compared with those reported previously.

Methods
This study was conducted with the approval of the Institutional Review Board of our institution, and all participants provided written informed consent. We studied the clinical results of internal xation for femoral neck fractures in patients older than 65 years of age using Twins (HOYA Technosurgical Corporation, Tokyo, Japan), a device that has been used in Japan for the treatment of both displaced and nondisplaced femoral neck fractures ( Figure 1).
Between 2011 and 2018, we treated 115 cases of femoral neck fractures using Twins. We could follow up 39 cases of these 115 cases at one year after xation and investigated complications of the patients retrospectively. We examined the preoperative following Xp variables: Singh's index, Garden classi cation, and Pauwels angle. Additionally, we examined the femoral neck-shaft angle (NSA) on the anteroposterior (AP) view of the hip pre-and post-operatively.
The research outcomes were mortality, surgical site infection (SSI), loss of reduction, aseptic necrosis, late segmental collapse (LSC), nonunion, periprosthetic fracture, reoperation and recovery of walking ability. SSI was de ned as the ful lment of one of the following four criteria: (1) purulent drainage (2) a positive culture result from wound swab (3) diagnosis made by a surgeon, unless culture result is negative. We de ned loss of reduction as a displacement of more than 10 degrees of the post-operative NSA than the pre-operative NSA and/or severe telescoping of more than 10 mm of the sliding screw. We diagnosed aseptic necrosis of the femoral head by magnetic resonance imaging at six months after fracture and LSC as the collapse of the weight-bearing segment of the femoral head by AP hip radiography at a point during the one-year follow up, respectively. Nonunion was diagnosed by computed tomography. Using these measurements, we compared each Xp variable between the "loss of reduction group" and the "stable group". In addition, we investigated differences each Xp variables between "nonunion group" and "union group". The differences between the Xp variables were analyzed by Mann-Whitney U-test. Values of P < 0.05 were considered statistically signi cant.
All patients followed a similar postoperative protocol consisting of patient mobilization out of bed on day one with full unrestricted weight bearing. All patients were contacted at three, six, and 12 months post operation, or until death, and follow-up information was obtained.

Results
The patients consisted of 13 men and 26 women (mean age 80.5 years, range 65-104 years). The demographic preoperative Xp variables are presented in Table 1. Mortality was observed in four cases (14.2%), SSI in one case (2.5%), loss of reduction in six cases (15.4%), aseptic necrosis in seven cases (17.9%), LSC in three cases (7.6%), nonunion in eight cases (20.5%), periprosthetic fracture in three cases (7.6%), and reoperation was performed in seven cases (17.9%), respectively. The number of patients who could walk independently before injury was 28; this decreased to 19 after injury (67.9%).

Discussion
Previously reported studies showed that complication rates one year after osteosynthesis for femoral neck fracture in elderly patients were as follows: mortality (12.7%) [4], SSI (1%) [5], loss of reduction (4.3%) [5], aseptic necrosis (4-39%) [6,7], LSC (12.5%) [8], nonunion (8.9%) [9], periprosthetic fracture (1-2%) [5]. In our study, the incidence of loss of reduction rate and periprosthetic fracture rate were relatively high. Our de nition of "loss of reduction" included telescoping of more than 10 mm because Slobogean et. al. showed that more than 10 mm telescoping of the hip screw resulted in disfunction of gluteus muscle power, which can lead to the inability to walk [10]. In our study, although three of the six "loss of reduction" cases could walk before fracture, two of them were unable to walk for a year. The de nition of "loss of reduction" differs across the reported studies. Therefore, it is di cult to compare the loss of reduction rate to those reported in other studies.
In contrast, other complication rates were almost equivalent to those reported previously [4,6,7,8,9]. Four cases of mortality were due to deteriorating pre-existing disease. From our results, none of the Xp variables (Singh index, Garden classi cation, and Pauwels angle) were responsible for the incidence of loss of reduction or bone union. These complications may have instead been caused by reoperation. Details of the cases of reoperation are shown in Table 2.
Six of the eight patients with nonunion were reoperated by bipolar head replacement or total hip arthroplasty. The other two patients, who were wheelchair-bound, recovered to the pre xation physical status conservatively within six months after primary xation. All the three periprosthetic fracture cases were of subtrochanteric fracture. One of these cases (Case 7 in Table 1) resulted in intramedullary nail conversion. However, the other two cases, which were of nondisplaced fracture, were not reoperated, and after eight weeks of non-weightbearing, had fully recovered to normal physical activity. One case of SSI had super cial infection and was treated only by wound irrigation, which resulted in full recovery of normal physical activity.
Several studies have attempted to identify determinants of the increased complication rate after hip fracture in the elderly; however, most evaluated populations were heterogeneous with respect to prefracture health status, living situation, and mental health [11]. Other prefracture factors frequently have acted as uncontrolled confounding variables. In addition, some complications may have been caused by technical failures.

Conclusions
The complications arising from the use of the Twins device described in this study were equivalent to the complications reported previously regarding the xation of femoral neck fractures for elderly patients. Further studies with larger patient numbers and longer follow-up are required to determine the roles of various devices in the treatment of both nondisplaced and displaced femoral neck fractures.   Figure 1 Femoral neck fracture xation using Twins device. The Twins device is designed to combine twin 7-mm cancellous lag screws that telescope within a barrel, which locks to a side plate, and sliding hip screw xation for stabilization of femoral neck fractures. The twin lag screws are oriented in an inverted parallel con guration at a 130-or 135-degree angle to the side plate. Each of the screws has 19 mm of available slide within the barrel when fully extended.