Current Trends in the Evaluation and Management of Nondisplaced Femoral Neck Fractures in the Elderly

Kanu Okike, MD, MPH; Ian G. Hasegawa, MD


J Am Acad Orthop Surg. 2021;29(4):e154-e164. 

In This Article

Fracture Fixation

Fracture fixation has long been considered the standard of care for nondisplaced FN fractures in the elderly. In the past, fixation was accomplished with devices such as hook pins or Knowles pins (Figure 3). These historical devices have since been replaced with more modern implants, most commonly cannulated cancellous screws and sliding hip screw devices, as described below.

Figure 3.

Radiographs demonstrating an example of a femoral neck fracture fixed with a historical implant (Knowles pins) on the AP (A) and lateral (B) views.

In the 2015 American Academy of Orthopaedic Surgeons Clinical Practice Guideline on the Management of Hip Fractures in the Elderly,[19] a moderate recommendation was made for the surgical fixation of stable (nondisplaced) FN fractures. However, several important issues regarding fracture fixation remain, including the possible role of fracture reduction before fixation, the type of implant to be used including novel devices, and the predictors of fixation failure.

Reduction Before Fixation

As noted below, recent publications have suggested that malunion of the FN may be associated with deficiencies in hip function.[9,22] As a result, some surgeons have questioned whether any apparent displacement of the fracture should be corrected before fixation. For example, Noda et al[23] recently described a technique for disimpacting Garden I fractures in an attempt to maintain hip offset. However, this strategy has not been evaluated in any type of clinical trial and, as such, the outcomes and potential complications (such as nonunion and/or osteonecrosis) associated with disimpaction before fixation are unknown.

Regarding the sagittal plane, one study suggested that correction of posterior tilt (also termed apex anterior angulation or retroversion deformity) before fixation might reduce secondary revision surgery rates in nondisplaced fractures. In particular, Yamamoto et al[24] reported on 40 nondisplaced FN fractures with posterior angulation that were treated with closed reduction, followed by Hansson Pinloc fixation (Stryker). A sufficient closed reduction, defined as ≤5° of residual posterior angulation, was achieved in 85% of patients. In this sufficiently reduced group, the rate of revision surgery was 14.7% as compared to 50% in the insufficiently reduced group (odds ratio = 5.8, no P-value reported).

From a practical standpoint, fracture reduction can have both positive and negative consequences. On the plus side, fracture reduction can restore anatomy and also improve implant positioning within the femoral head. (When a retroversion deformity is present, implants placed in the center of the FN will often end in an anterior position within the femoral head.) However, fracture reduction also unlocks FN impaction, which can in turn destabilize the fracture and lead to greater demand on the fixation implant.

Cannulated Screws

In the treatment of nondisplaced FN fractures in the elderly, cannulated screws currently represent one of the most popular fixation choices. Typically, three parallel screws are placed percutaneously in an inverted triangle configuration. Proper technique is important when placing these screws to minimize the risk of fixation failure and other complications. The start point for the inferior screw should not be below the lesser trochanter because this creates a stress riser which can predispose to iatrogenic subtrochanteric fracture. To best resist fracture displacement, the inferior screw should abut the medial calcar and the posterior screw should abut the posterior cortex (Figure 4).

Figure 4.

Radiographs demonstrating an example of a Garden II femoral neck fracture fixed with cannulated screws. On the AP fluoroscopic image (A), note the position of the inferior screw, which starts from a point not below the lesser trochanter and abuts the medial cortex. On the lateral fluoroscopic image (B), note the position of the posterior screw, which abuts the posterior cortex.

The advantages of cannulated screw fixation include lower implant cost, percutaneous insertion, shorter surgical time,[25] and lower surgical blood loss.[26] In addition, the multiple points of fixation spanning the FN provide rotational stability across the fracture site. However, a limitation of the cannulated screw constructs is that they do not provide fixed-angle stability.

Notable complication rates have been described after cannulated screw fixation of nondisplaced FN fractures in the elderly. Fixation failure rates have been found to be as high as 19%.[14] The rates of nonunion and femoral head osteonecrosis have each been as high as 11% in previous reports.[27,28] Overall, the rates of revision surgery after cannulated screw fixation have ranged from 8% to 27%.[4–6,29] Several factors exist that have been found to predict revision surgery after fixation of nondisplaced FN fractures, which are discussed in detail below.

When fixation failure, nonunion, and/or osteonecrosis occur, the salvage procedure is often conversion arthroplasty. In the FAITH trial,[4] the overall revision surgery rate for nondisplaced fractures treated with cannulated screws was 16.3% (60 of 369), with conversion arthroplasty representing 55% of the secondary procedures performed. Gjertsen et al[3] reported an overall revision surgery rate of 9.8% in a cohort of 4,468 Garden I and II fractures, with 85% of those requiring revision surgery undergoing conversion arthroplasty. A systematic review of 11 studies found that the mean time to conversion arthroplasty after screw fixation varied from 2 to 23 months, with nonunion resulting in an earlier conversion as compared to osteonecrosis.[29]

Sliding Hip Screw

Another popular choice in the treatment of nondisplaced FN fractures in the elderly is the sliding hip screw, a device consisting of a single large diameter, partially threaded lag screw affixed to the proximal femur via a short side plate and barrel (Figure 5). The advantages of the sliding hip screw include controlled fracture collapse and the fixed-angle stability provided by the side plate.

Figure 5.

Radiographs demonstrating an example of a Garden I femoral neck fracture fixed with a sliding hip screw device plus antirotational screw, as depicted here on the AP (A) and lateral (B) views.

One concern with sliding hip screw insertion is rotational control of the head fragment. This can be addressed via the addition of an accessory "antirotational" screw placed superiorly, before insertion of the partially threaded lag screw (Figure 5). Interestingly, the addition of an antirotational screw has not been shown to influence clinical outcomes. In particular, Makki et al[30] retrospectively reviewed 65 Garden I and II FN fractures fixed in situ using a two-hole sliding hip screw. Thirty-four patients received a permanent 6.5-mm partially threaded screw placed in the superior femoral head and neck before sliding hip screw insertion, whereas 31 patients did not. The authors found that adding the antirotational screw resulted in longer surgical times, greater radiation exposure, and higher cost but did not yield any difference in radiographic union, osteonecrosis, or revision surgery at the two-year follow up.

Cannulated Screws Versus Sliding Hip Screw

A number of studies have recently compared the outcomes after sliding hip screw and cannulated screw fixation. Sliding hip screw fixation has been associated with slightly longer surgical times (on the order of 5 minutes)[25] and slightly greater intraoperative blood loss (on the order of 100 mL).[26] However, the clinical significance of these differences is questionable because they have not appeared to translate into any increased risk of perioperative complications.[31]

Sliding hip screw devices and cannulated screws have also been compared on the basis of fracture-related complications and revision surgery rates. In a subanalysis of nondisplaced FN fractures conducted as part of the FAITH trial, investigators found that implant failure, fracture shortening, nonunion, and osteonecrosis did not differ between cannulated screws and sliding hip screw fixation.[4] Overall, the revision surgery rate was similar between the two types of fixation (16.3% versus 16.7%, respectively, P = 0.99). When the specific types of revision surgery were compared, however, notable differences were observed as conversion total hip arthroplasty was found to be more common after sliding hip screw fixation (11.1% versus 4.9%, P = 0.004), whereas implant removal was more common after cannulated screw fixation (6.8% versus 1.7%, P = 0.001). Watson et al[31] also compared cannulated screws with the sliding hip screw in a randomized clinical trial and found no differences in the overall rates of revision surgery, although the type of secondary procedure was not specifically analyzed. Taken together, these findings suggest that neither cannulated screws nor the sliding hip screw may be considered superior in the management of nondisplaced FN fractures in the elderly, although revision surgery after sliding hip screw fixation may involve a more extensive surgery.

Predictors of Revision Surgery After Fixation

In recent years, several factors have been found to predict revision surgery after fixation of nondisplaced FN fractures in the elderly (Figure 6). One of the most consistent has been the degree of posterior tilt (ie, retroversion deformity) on the preoperative lateral radiograph.[6,14,15,32] In a study of 322 Garden I and II FN fractures treated with two parallel screws, fixation failure was found to occur at a rate of 19% in patients with a preoperative posterior tilt angle ≥20° as compared to 11% in patients with a preoperative posterior tilt angle between 10° and 20° and 6% in patients with preoperative posterior tilt <10° (P = 0.03).[14] Similarly, Palm et al[6] observed a 56% revision surgery rate among fractures with a preoperative posterior tilt angle ≥20° as compared to 14% among fractures with a posterior tilt angle below this value (P < 0.001). Most recently, Okike et al[15] performed a secondary analysis of nondisplaced fractures in the FAITH trial and also found posterior tilt ≥20° to be associated with an increased risk of revision surgery after fixation. As noted previously, these findings emphasize the importance of carefully evaluating the preoperative lateral radiograph in patients with nondisplaced FN fractures.

Figure 6.

Radiographs demonstrating a case example of an 87-year-old woman who presented with a Garden I femoral neck fracture (A) and underwent fixation with a sliding hip screw plus antirotational screw (B). At 9 months postoperatively, the patient presented with worsening pain and was found to have osteonecrosis and a nonunion (C). She was treated with hardware removal and conversion arthroplasty.

Other factors have been identified as potential predictors of revision surgery after fixation of nondisplaced FN fractures, including disruption of the medial cortex,[33] age older than 80 years,[15] poor bone quality, and malnutrition.[32] However, these findings have been less consistent in the literature to date.

Femoral Neck Shortening

An increasing appreciation exists that even when fractures unite without complications such as nonunion or osteonecrosis, the patient's functional outcome may not be optimal. Both cannulated screws and sliding hip screw devices allow fracture compression with weight-bearing, which is thought to promote healing. However, this compression may also result in considerable FN shortening (Figure 7), which may, in turn, be associated with decreased hip function.[9,22] The incidence of FN shortening greater than 5 mm after in situ fixation of Garden I and II fractures has been reported to range from 24% to 66%.[9,22] In a recent study, Felton et al[9] observed a significant association between FN shortening and hip function (P < 0.01), with hip function worsening as the degree of FN shortening increased (P < 0.01). It has been theorized that the decrease in hip function observed with FN shortening may stem from a loss of hip abductor strength. For example, Noda et al[10] demonstrated that with 15° of valgus impaction, the abductor moment arm decreased by 10%. In this same study, all subjects demonstrated significant loss of hip abduction strength compared with the nonsurgical hip (P < 0.05), even though the mean valgus angle was only 8.9°.

Figure 7.

Radiograph demonstrating a case example of an 81-year-old man who underwent cannulated screw fixation of a Garden II FN fracture. Radiographs performed at 4 months postoperatively demonstrate a fracture which has healed, but with notable shortening of the FN.

Novel Fixation Implants

Recently, a new generation of devices for FN fracture fixation have been introduced by several companies. Incorporating features of both cannulated screws and the sliding hip screw, these novel fixation implants were designed to reduce the high rates of complications and revision surgery associated with current FN fracture fixation devices. Examples of these novel fixation implants include the Targon FN (B. Braun Melsungen AG), the Femoral Neck System (DePuy Synthes), the Conquest FN (Smith & Nephew), and the Telescoping Hip Plate (Zimmer Biomet). Currently, the Femoral Neck System and Conquest FN are available for use in the United States, whereas the Targon FN is available in Europe. As of now, the Telescoping Hip Plate is no longer available after a class II device recall by the United States Food and Drug Administration in January 2020. It should be noted that all of these novel fixation implants require a formal incision (similar to the sliding hip screw), as opposed to cannulated screws which can be placed percutaneously.

Published data on these novel fixation implants in the treatment of nondisplaced FN fractures in the elderly are currently limited to the Targon FN. In three retrospective studies, the implant was associated with complication rates on the order of 0 to 2.7% for nonunion, 4.5% to 5.5% for osteonecrosis, and 9.9% to 16.4% for revision surgery.[34–36] These results suggest that when compared with cannulated screws and sliding hip screw devices, the Targon FN may be associated with lower rates of nonunion and osteonecrosis, but similar rates of revision surgery. To date, only one study has compared the Targon FN to cannulated screw fixation in the treatment of nondisplaced FN fractures in the elderly. In particular, Warschawski et al[37] retrospectively compared the Targon FN with cannulated screws in the treatment of 115 Garden I and II FN fractures. No differences in complications (5.9% versus 8.6%, respectively, P = 0.72) or revision surgery (5.9% versus 6.2%, respectively, P = 1.0) were noted between the two modes of fixation. However, the authors' conclusions favored cannulated screw fixation given the lower implant cost. At our institution, the price of one of these novel fixation implants is approximately twice that of a sliding hip screw and three times that of three partially threaded 6.5-mm cannulated screws. Additional research is needed to further evaluate these novel devices, including randomized clinical trials with greater patient numbers and longer follow-up.

Cement Augmentation

Recently, there has been renewed interest in the use of resorbable (eg, calcium phosphate) and nonresorbable (eg, polymethyl methacrylate) bone cement to augment the fixation of proximal femur fractures. Regarding nondisplaced FN fractures specifically, however, data in support of this practice are lacking. Hofmann-Fliri et al[38] conducted a biomechanical evaluation of nondisplaced FN fractures treated with multiple cannulated screws and found no advantage to polymethyl methacrylate augmentation regarding secondary displacement. In fact, the study found that cement augmentation actually resulted in greater varus displacement with cyclic loading. The authors suggest that this could have occurred if the cement increased screw fixation in the head, thereby shifting the weak point of the construct laterally to the screw insertion points. Conceptually, a fixed angled device such as a sliding hip screw could counteract this shift in stress, although this has not been studied to date.