Compressive buttress compared with off-axial screw fixation for vertical femoral neck fractures in young adults: a prospective, randomized controlled trial

Hip fractures are a commonly encountered traumatic injury in orthopedic practice that are associated with substantial morbidity, mortality, and costs [1‐5]. Femoral neck fracture (FNF) is a typical representative of hip fracture, and its treatment often varies according to the age of the patient [6]. Although occurring at a lower incidence than in the elderly, FNF in the young population presents a unique clinical challenge for the orthopedic traumatologist due to higher postoperative functional demands for work or recreational activities. Additionally, FNF in young adult patients is typically associated with high-energy trauma and displaced fracture patterns, resulting in a biomechanically disadvantageous environment for fracture healing [7‐10]. The high proportion of fixation failures resulting in reoperation has generated controversy about the most appropriate surgical treatment for FNF [7]. Regardless of the fixation method used, pooled incidences of complications after FNF fixation in young patients are common, including reoperation (18.0%), avascular necrosis of the femoral head (ANFH) (14.3%), and nonunion (9.3%) [7]. As such, treatment of FNF in young patients presents a clinical challenge when choosing the optimum fixation approach and is associated with significant complications [8,10‐12].

The available internal fixation techniques for FNF are classified based on mechanism and type of fixation: traditional sliding compression fixation, angular stabilization fixation, length-stable implant fixation, medial buttress plate fixation, and off-axial screw fixation, etc. [13]. Except the traditional sliding compression fixation, including multiple parallel partial-threaded cancellous cannulated lag screw (PTCS) and dynamic/sliding hip screw (DHS&SHS) fixation [14], other fixations are not recommended as common treatments for FNF [8]. Although biomechanical and clinical studies have evaluated various options, the optimal fixation construct to allow for healing and prevention of complications after FNF is still unknown [10]. The PTCS fixation is a widely accepted technique with reported advantages of less tissue invasiveness and blood loss in addition to shorter hospital stay and operation time [15,16]. However, recent studies have shown high complication rates in vertical FNFs fixed with PTCS, attributed to poor biomechanical performance [12]. Therefore, off-axis screw configuration in which a horizontal or transverse screw implanted orthogonal to the vertical fracture line besides the PTCS, has been proposed as a surgical option for the treatment of vertical FNFs. Although clinical data are limited [17‐19], studies have demonstrated superior biomechanical stability with off-axis screw configurations when compared with traditional PTCS fixation [19‐24]. In particular, the use of OPTCS (three parallel screws plus with an off-axis trochanteric transverse screw) for the treatment of vertical FNFs has demonstrated promising clinical results in recent studies [19,25,26]. At this time, there is no definite agreement on the effectiveness of this new screw configuration for FNF treatment.

The fully threaded headless cannulated screw (FTHCS) was originally introduced as length-stable fixation implant to prevent the femoral neck shortening for the FNF treatment with the characteristics of normal head, cylindrical profile. Meanwhile, another kind of fully threaded screw, FTHCS possessing the specific features of tapered profile and variable pitch, fully threaded headless cannulated screw have been introduced [27]. It was shown by our previous biomechanical experiments and clinical study that comparing with conventional PTCS fixation, the new fixation configuration using FTHCS could not only supply better biomechanical stability but also reduce the complication rate significantly for vertical FNF [28‐31]. In response to a recent clinical study demonstrating a distinctive failure model of medial migration and superior cut-out of the proximal screw in three FTHCSs fixation for vertical FNF in young patients [31], a novel screw configuration named compression buttress screw (CBS) fixation has been proposed [13]. In CBS fixation, two distal FTHCSs are combined with one proximal partial threaded cannulated screw (PTCS) in a regular triangular configuration. To the best of our knowledge, no prospective study has compared the clinical outcomes of these two configurations (CBS vs. OPTCS) on young patients with vertical FNFs. Thus, the objective of this prospective randomized, controlled trial was to evaluate the efficacy of CBS fixation with the OPTCS fixation on complication rates, hip function and life quality in young patients with vertical FNF. The null hypothesis was that there is no difference in complication rates, pain, or functional outcomes between groups at 24 months.

This prospective, randomized controlled trial was approved by the institutional review board of our center and registered at www.​chictr.​org.​cn (ChiCTR1900026283). From January 2016 to July 2017, all adult patients younger than 55 years presenting to our level-I trauma center with FNFs were considered eligible for enrollment.

From January 2016 through July 2017, a total of 178 patients who suffered from FNF scheduled for fixation were screened for eligibility. Before operation, 32 patients were excluded because they had not met the inclusion criteria or had met exclusion criteria. Intraoperatively, 12 patients were excluded due to irreducibility and inevitable open reduction. A total of 134 patients were randomly assigned to received CBS fixation or OPTCS fixation. After operation, 10 patients in CBS fixation were lost to follow-up, and 7 patients in OPTCS lost. A total of 57 patients in the CBS fixation group and 60 in the OPTCS fixation group were included in the pre-protocol population. The follow-up time was 31.3 ± 3.9 months in the CBS group and 31.4 ± 3 months in the OPTCS group (p = 0.967). Enrollment, allocation, and follow-up are summarized in Fig. 2. The baseline characteristics of patients were comparable between the two groups (all p > 0.05, Table 1).

Patients who underwent CBS treatment had less surgery time [48 (40–55)] vs. 50 (45–60), p = 0.017] and less blood loss [50 (45–60) vs. 70 (57.5–90), p < 0.001]. There was no significant difference in bone healing time between the two groups [15 (13–16) vs. 14 (13–16), p = 0.340] (Table 2).

Patients allocated to CBS fixation were less likely to develop complications than those allocated to OPTCS fixation at 24-month follow-up, including fixation failure (10.5% vs. 25.0%, p = 0.041), fracture nonunion (1.8% vs. 18.3%, p = 0.003). However, there was no statistical difference in the incidence of the ANFH (7.0% vs. 11.7%, p = 0.389) between the two groups (Table 3). In CBS fixation group, one patient who experienced fixation failure and nonunion simultaneously, underwent total hip arthroplasty (THA) at 12 months (Fig. 3). Another 1 patient who encountered the ANFH received the revision surgery with free vascularized fibular grafting (FVFG) and new internal fixation. In the OPTCS, 10 patients underwent various revision surgeries due to nonunion or/and ANFH.

In both groups the most common mechanism of fixation loosening was screw lateral withdrawal; however, the OPTCS fixation group was more likely to develop screw lateral withdrawal than those receiving CBS fixation (19.3% vs. 58.3%, p =  < 0.001) (Fig. 4). Only one patient in the CBS fixation group suffered from the medial migration, superior cut-out of the proximal screw (Fig. 3), compared with none in the OPTCS group. The patients managed with CBS fixation showed significantly less severe femoral neck shortening and varus collapse (10.5% vs. 25.0%, p = 0.007). 70.2% and 19.3% patients in CBS fixation experienced none/mild and moderate shortening and carus collapse, whereas, the incidences in OPTCS group were 41.7% and 33.3% (p = 0.007).

With regard to the function, the CBS fixation group demonstrated significantly higher overall HHS score [93 (82.5–95) vs. 83 (73–93), p = 0.001] and HHS grade (p = 0.008) compared with the OPTCS group, although these did not meet the minimal clinically important difference between groups [51]. There were also significantly higher EQ-5D-5L [0.814 (0.609–1.000) vs. 0.581 (0.542–0.814), p < 0.001] and EQ-VAS [85 (80–90) vs. 80 (65–90), p = 0.002] scores in the CBS fixation group when compared with OPTCS (Table 3).

The variables with p values < 0.05 in univariate analysis including surgical approach (CBS vs. OPTCS), age, smoking status, hypertension, diabetes, BMI, and cause of injury were included in the multivariate logistic regression analysis to determine the risk factors of fixation loosening in vertical FNFs (Table 4). Multivariate logistic regression analysis demonstrated surgical approach (p < 0.001), smoking status (p = 0.004), diabetes (p < 0.001), and cause of injury (p = 0.042) to be significant risk factors for fixation loosening. In order to predict the risk of fixation loosening in vertical FNF, a nomogram based on the multivariable logistic regression results was constructed (Fig. 5A). Using the bootstrap method, a calibration plot was constructed to compare the predicted outcome of fixation loosening with the actual outcome. A close fit between the predictive curve and the ideal curve indicates good predictive ability (Fig. 5B). For internal verification, the ROC showed that the resulting model had a fairly good discriminatory ability with an AUC of 0.868 (0.800–0.936) (Fig. 5C).

This prospective, randomized trial demonstrates that the occurrence of severe complications at 24 months is significantly less with the use of CBS fixation for the treatment of vertical FNFs in young patients when compared with OPTCS fixation. Additionally, functional scores and quality of life outcomes at 24 months are significantly improved in the CBS fixation group. We were able to identify surgical approach (CBS vs. OPTCS), smoking status, diabetes, and cause of injury (traffic accident) as predictive factors in risk of fixation loosening, which is an important reason for the complications of fixation failure, nonunion, etc., and the poor function. To our knowledge, this is the first randomized controlled trial to compare CBS to OPTCS fixation for the operative treatment of vertical FNFs in the young population.

Controversy exists with regard to the optimal screw choice and fixation configuration for the treatment of vertical FNF. Okcu and colleagues conducted a prospective, randomized study comparing the use of FTHCS with PTCS for treatment of 22 patients with FNFs encompassing all Pauwels classifications [52]. In each group, three or four screw configurations were used with the authors demonstrating that PTCS provided a shorter union time and less complication rate while providing equivalent functional results when compared with FTHCS. A second study on geriatric patients with nondisplaced Pauwels I-III FNF demonstrated similar complication rates between groups treated with three parallel FTHCSs and PTCSs [53]. However, recent studies demonstrates superior biomechanical stability and significantly reduced complication rate with the use of FTHCS for fixation for vertical FNFs when compared with PTCS [28‐31]. Interestingly, a distinctive “medial screw migration” fixation loosening mechanism of three FTHCSs fixation of FNF was detected in a recent clinical study [31], highlighting the need for further study on screw configuration.

Studies have demonstrated superior biomechanical stability and lower complication rates with off-axis screw configuration when compared with traditional PTCS fixation [19,24‐26]. The transverse off-axis screw fixation for FNFs can be subdivided into two distinct configurations based on the level of screw implantation: the trochanteric transverse screw (TTS) and the calcar screw (CTS). The TTS is placed close to the proximal cortex, in a trajectory from lateral to inferior-medial quadrant of the femoral head (uni-cortical fixation of OPTCS, TTS-OPTCS) [17,20‐22,26]. The CTS is inserted proximal to the level of lesser trochanter and fixed to the calcar cortex (bi-cortical fixation of OPTCS, CTS-OPTCS) [18,23,54,55] (Fig. 6). Although there are no clinical studies comparing these two screw configurations, a recent biomechanical study on synthetic femurs demonstrated statistically significant increases in axial stiffness and ultimate failure load for CTS when compared with TTS [56]. However, it should be noted that CTS-OPTCS fixation is not suitable for all Pauwels type III fractures, including the subcapital type and the transcervical type with medial cortex comminution. As such, TTS-OPTCS fixation was chosen for use in the present study.

A retrospective study comparing three PTCS fixation (97 patients) to those augmented with a TTS (60 patients) for vertical FNFs, demonstrated a significantly lower rate of femoral neck shortening in the TTS-OPTCS fixation cohort; however, the rates of nonunion, ANFH, and reoperation were lower in TTS-OPTCS but not statistically significant [25]. Another retrospective cohort study comparing PTCS (107 patients), TTS-OPTCS (65 patients), and dynamic hip screw (32 patients) for the treatment of vertical FNF found significantly lower fixation failure rates in the TTS-OPTCS group [19]. Taking all these into account, it is not difficult to find out that for vertical FNFs the capacity of TTS-OPTCS fixation to reduce the complication rates is limited, which is perhaps why the potency of CBS fixation for vertical FNFs was assessed and compared to the TTS-OPTCS. The results of the study proved our hypotheses.

CBS fixation, a novel fixation configuration combining FTHCS and PTCS for vertical FNF has been proposed based on clinical practice and verified by biomechanical testing [13,28‐30]. Furthermore, pilot data from our group suggest that FTHCSs fixation in this new CBS configuration improves clinical outcomes in young patients with vertical FNF [29,31]. At least two inferior FTHCS in the regular triangular configuration can resist axial displacement as firmly as the non-comminuted femoral neck cortex in vertical FNF and even rebuild the lost posteromedial buttressing effect in the comminuted cases. It might be defined as “intraosseous medial buttressing effect” which is clearly distinct from the medial buttress plate fixation for treatment of vertical FNF ^13,57.

The primary strengths of this study include the clinical relevance, strict inclusion and exclusion criteria, the large number of patients recruited, and the high rate of follow-up. Other strengths include use of a predictive nomogram based on the multivariable logistic regression analysis. As with all studies, there are limitations. First, the comparison was based on a single-center study, potentially limiting generalizability. Secondly, because the relative distribution of subjects suffered from complications, such as fixation failure and nonunion, was incomparable to that in the patients without complications in both groups, the regression analysis and nomogram for these complications could not been executed. Finally, the average follow-up time is insufficient to detect the complication of ANFH in a long-term.

In conclusion, CBS fixation for the treatment of vertical FNFs in young patients is an effective surgical technique for reducing complications, decreasing fixation loosening and femoral neck shortening/varus collapse, and improving patient’s function and quality of life when compared with OPTCS fixation. Future clinical research with larger populations should clarify if CBS fixation is superior to the traditional PTCS fixation and OPTCS, particularly bearing the biomechanical stability in mind.