Mechanical Complications After Intramedullary Fixation of Extracapsular Hip Fractures

Abstract

Extracapsular hip fractures occur frequently in the elderly as a result of low-energy trauma. Achieving stable fixation in osteoporotic bone to allow early weight bearing is a key objective in the treatment of these injuries. Many of the intraoperative decisions facing surgeons are directed by first determining fracture stability. Unstable fracture patterns should be fixated with intramedullary nails, aiming to control motion around the implant and at the fracture site. Torsion control devices provide additional stability and control, although their exact indications are not precisely defined. Complications that arise as a result of the early loading can be avoided with a good surgical reduction, meticulous technique, and proper implant selection. Mechanical complications include cutout, inordinate proximal femoral shortening, delayed union, early implant fracture, nail toggle, and cut through/medial migration.

Extracapsular hip fractures in the pertrochanteric and subtrochanteric region occur frequently in the elderly as a result of low-energy trauma sustained in osteoporotic bone. In such fractures, the blood supply to the femoral head and neck remains intact postinjury. The ensuing high rate of union makes open reduction and internal fixation a viable treatment option. Therefore, failures tend to be mechanical in origin. Achieving stable fixation in osteoporotic bone to allow early weight bearing, while avoiding failure, is a key objective in the treatment of these injuries with intramedullary (IM) nails. Mechanical failure, by definition, occurs when the implant breaks in response to loading or, as more commonly seen in osteoporotic fractures, when the implant loses fixation in bone.¹ Although complications may not prevent fracture union per se, they still adversely affect the outcome potentially resulting in a need for additional procedures.

As there are few prospective studies on extracapsular hip fractures, information guiding physicians in clinical practice is limited. This issue is compounded by the fact that mechanical failures occur infrequently, and the low number of failures investigated in the literature affects the research power and validity, as demonstrated by the wide confidence intervals that often accompany the related statistical findings. This article reviews the current practices for preventing complications after IM fixation of osteoporotic extracapsular hip fractures. The importance of good surgical technique and proper implant selection will be highlighted while discussing recent developments in the literature.

Stability, Deformity, and Prevention of Mechanical Failure

Successful treatment of extracapsular hip fractures requires a comprehensive understanding of fracture stability, as stability more so than classification primarily determines the management, and prognosis of such injuries. Stability is best thought of as the capacity to resist further displacement, motion, or collapse after internal fixation.¹ Fracture geometry provides the fundamental basis for stability, in that the cortical contact of the major pieces primarily prevents displacement after reduction and fixation. Posterior medial comminution becomes destabilizing for these injuries, as the loss of cortical contact at such location results in loss of the main buttress to resist varus bending moments. Likewise, the lateral cortical wall in the proximal femur is also an important stabilizer against head/neck fragment collapse during loading. It has been reported that lateral wall fracture is associated with a 22% risk for revision surgery compared with 3% when intact (P < 0.001).²

In 2018, the OTA/AO 31A classification system for the proximal femur was revised to include lateral wall thickness as a key variable to distinguish stable from unstable injuries (31A1 from 31A2) (Figure 1).³ Hsu et al⁴ demonstrated that a lateral wall thickness of less than 20.5 mm was prognostic for iatrogenic lateral wall fractures, resulting in a significantly higher rate of failure in A2 fractures (49% versus 8.3%; P < 0.001). Under such circumstances, selecting an IM nail for treatment would allow the implant itself to substitute for the missing lateral wall and buttress collapse, which would maintain the offset. Despite this recent revision, the classification system remains imprecise, as severe and moderate collapse has still been described in 13% and 26% of stable two-part injuries, respectively, suggesting that our current understanding of such injuries' stability is incomplete.⁵

Figure 1: The OTA/AO 31A classification of pertrochanteric fractures depicted with standard radiographs and 3D CT.⁴ A1 fractures are stable injuries that include two-part fractures and three-part fractures with a lateral wall greater than 20.5 mm; A2 fractures are unstable injuries with a lateral wall thickness less than 20.5 mm; A3 injuries are reverse obliquity and transverse patterns that are also considered unstable. Depicted by the red line is the proper location for measuring lateral wall thickness at a point 3 cm below the innominate tubercle. Accurate measurement requires the extremity be internally rotated at the time of imaging. In many instances, the deformity induced by fracture geometry is more complicated than appreciated on the standard radiographs.

The mechanical environment of any fracture is illustrated by the deformity present on the preoperative radiographs, which is a basic orthopaedic principle observed in fractures all over the body. In cases of failure, loss of reduction often progresses as the fragments drift back into their innate position, thus resuming their original deformity (Figure 2).⁶ Currently, initial deformity dictates the reduction maneuvers performed for extracapsular hip fractures; however, little is known about its potential application to other decisions around stability and implant selection. Promising yet limited clinical evidence suggests that the degree of varus deformity on preoperative radiographs for pertrochanteric fractures is predictive of varus collapse and cutout.⁷

Figure 2: A, Mechanical failure of a pilon fracture status after internal fixation. Loss of reduction after failure resulted in the fragments resuming their original varus deformity. B, Mechanical failure of a reverse obliquity fracture after set screw fracture that resulted in displacement back to the original preoperative position with lateral trochanteric displacement and medialization of the shaft.⁴¹

One of the biggest limitations in our understanding of deformity is that our knowledge base for extracapsular hip fractures is predominately written in a two-dimensional language. The terms varus and shortened are frequently used to describe both initial deformity and mechanical failure for such fractures. However, these terms only apply to values quantified in the x and y axes. Recent technological advances have allowed for a more detailed approach using 3D reconstruction obtained from CT. CT identifies a third plane of deformity, which is often present in extracapsular hip fractures or rotational deformity of the femoral head or neck around its own axis (Figure 3). The OTA/AO classification for pertrochanteric fractures does not currently consider rotational instability, which may partially account for its imperfect description of stability. The rotational component of stability for OTA/AO 31A fractures is substantiated not only by the presence of rotational deformity due to fracture geometry but also by the rotational forces present in the mechanical environment postoperatively. In a radiostereometric analysis of intertrochanteric fractures after fixation with single-screw implants, it was noted that shortening and rotation around the anatomic axis of the femoral neck occurred concomitantly with a mean rotation of 10.6° (−28.1 to 6.1°) and mean shortening of 5.0 mm (−0.13 to 12.9) at four months.⁸

Figure 3: Radiographs and 3D reconstruction of an OTA/AO 31A1.3 fracture that is shortened with varus deformity. Rotational deformity of the femoral head/neck fragment along its axis is more apparent on the 3D reconstruction in addition to the varus and shortening.

Rotational instability has been historically described among basicervical femoral neck fractures, which are also extracapsular, where similar rotational deformities are encountered. Johnson et al⁹ noted that under unidirectional loading, rotational failures occurred frequently (27%) in basicervical fractures treated with IM implants with one point of fixation in the femoral head/neck. In addition, the higher cutout rate in basicervical femoral neck fractures treated with IM nails (45% compared with 2% to 6% reported in pertrochanteric fractures) observed by Watson et al¹⁰ was attributed to the rotational instability inherent to the fracture pattern, despite an average tip-to-apex distance (TAD) of 17.4 mm with anatomic/near anatomic reduction. Therefore, fractures at risk for rotational failure would not only be purely basicervical but also pertrochanteric fractures, with significant basicervical component where rotational deformity is encountered on the initial radiographs.

The prevention of mechanical failures begins with identifying sources of instability and tailoring implant selection and reduction goals to best meet the demands of the mechanical environment. Common mechanical complications include cutout, inordinate proximal femoral shortening, delayed union, early implant fracture, nail toggle, and cut through/medial migration.

Cutout

Cutout and loss of fixation of the implant in the femoral head occurs more commonly in osteoporotic hip fractures, arising in approximately 5% to 6% of these injuries.^11,12 This failure rate is variable and influenced by multiple factors related to host, stability, and implant. These failures occur early, most likely within the first 12 weeks.¹² Cutout occurs less frequently in a more delayed manner secondary to nonunion, osteonecrosis, or other pathologic processes. Many of the early failures are caused by several combined mechanical factors; most notably, neck-shaft malreduction (varus) in unstable patterns affects screw trajectory into the proximal fragment resulting in suboptimal TAD and failure (Figure 4)¹² Tip-to-apex distance (TAD) is the distance from the tip of the screw to the apex of the femoral head, or center position, as viewed in both the AP and lateral radiographic views. Fujii et al¹³ used a multivariable logistic regression model to establish that a TAD >20 mm was the single most important risk factor for cutout after internal fixation with an IM nail (odds ratio 12.4, P = 0.019).

Figure 4: A, AP radiograph after hip fracture surgery where the tip of the greater trochanter is above the center of rotation for the femoral head indicating a varus reduction. The varus reduction forces the helical blade into a more superior position in the femoral head, which is suboptimal compared with a more central location. B, Lateral radiograph of a starting point too far posterior on the greater trochanter forcing the screw to end up anteriorly in the femoral head instead of the recommended position in the center. Ideal staring point on the lateral view is right at the junction of the anterior and middle third of the greater trochanter.

Cutout can still be observed despite good reduction and adequate TAD, which has recently led to a more detailed investigation of the underlying biomechanical forces governing failure. In a biomechanical study of unstable fractures using a multidirectional load in a cadaver hemipelvis model, it was found that the predominant motion during cutout was rotation around the anatomic axis of the femoral neck over compression/varus collapse (Video 1). Although this model lacked the iliotibial tension band to fully simulate one- and two-legged stance, multiple points of fixation in the proximal fragment via dual screws reduced failures, compared with a single unicortical screw for fixation by controlling rotation (single screw average rotation: 35.4 ± 29.3°; versus dual screw average: 5.5 ± 6.4°; P = 0.006).¹⁴ Literature support for torsion control IM nails that use dual screws proximally to improve the cutout rate is currently limited to smaller retrospective clinical studies, while randomized prospective studies are lacking.

{"href":"Single Video Player","role":"media-player-id","content-type":"play-in-place","position":"float","orientation":"portrait","label":"Video 1","caption":"Intraoperative video of failed intertrochanteric fracture. The femur was dislocated with an intact implant. Instability around the femoral head was limited to rotation around the anatomic axis of the femoral neck.","object-id":[{"pub-id-type":"doi","id":""},{"pub-id-type":"other","content-type":"media-stream-id","id":"1_8rlkfahm"},{"pub-id-type":"other","content-type":"media-source","id":"Kaltura"}]}

Technological advances have also focused on achieving torsion control and improving implant load distribution by augmenting bone quality with various injectable bone graft substitutes such as polymethyl methacrylate (PMMA) or calcium phosphate (CaPO4).¹ In a retrospective study, a lower rate of mechanical failure has been reported among PMMA augmentation compared with nonaugmented fractures (2.1% to 13.8%, P = 0.047).¹⁵ Extravasation of the PMMA into both the hip joint and the fracture site resulting in nonunion was among the intraoperative complications observed, in addition to guidewire perforation of the femoral head which, precluded augmentation. Another randomized controlled trial could not verify any significant difference between augmented and nonaugmented groups in terms of radiographic outcome (blade migration), functional outcome, or rate of mechanical failure.¹⁶ The precise indications for newer torsion control devices and their role in preventing failures currently remain undefined, and their use does not obviate the need for adequate reduction or proper technique.

The recommended treatment for cutout involves arthroplasty owing to destruction of the femoral head and erosion of the acetabulum. The incidence of complications after conversion to total hip arthroplasty (THA) was significantly higher after IM nailing than after primary THA in terms of infection (6.2% versus 2.6%), dislocation (8.1% versus 4.5%), and revision (8.4% versus 4.3%).¹⁷ In cases in which failure is detected before head violation, revision nailing with cement augmentation has been proposed as a potential alternative to arthroplasty. However, limited clinical data suggest that this technique results in a higher rate of revision surgery.¹⁸

Proximal Femoral Shortening

Proximal femoral shortening is expected by virtue of controlled collapse, which has long been established as the accepted process used to achieve union in such fractures occurring in osteoporotic bone. However, in unstable injuries, when collapse is uncontrolled, excessive shortening that develops rapidly is associated with severe pain that limits mobility. Recently, the limits of acceptable collapse were quantified by a prospective study, demonstrating that radiographic shortening over 8 mm affected the length-tension relationship of the hip abductors and correlated with several gait abnormalities (P = 0.008).¹⁹ The newer evidence regarding the effect of IM nails on proximal femoral shortening after extracapsular hip fracture fixation is summarized according to its strength in the supplemental table.

The first step in controlling collapse is to maximize contact of cortical bone at the fracture site with a good reduction. The need for proximal dynamization, or the slide, should be determined after evaluating fracture stability and the quality of the reduction. Any gaps or distraction of the fracture can be compensated for by unlocking the proximal set screw a quarter turn to allow some collapse of the construct. Alternatively, when gaps are eliminated by intraoperative compression, fully locking the set screw to minimize collapse has also been shown to achieve union without making constructs overly rigid and increasing complications.^20,21 Although fully locking the set screw is recommended in unstable fracture patterns, failure to dynamize or allow controlled collapse in the presence of a persistent fracture gap may lead to nonunion. It is essential for the surgeon to determine the set point of control required for each injury by balancing reduction quality/gapping with the need for proximal versus distal construct dynamization to achieve union and minimize shortening.

After reduction, implant selection is an important consideration because in several randomized prospective studies, IM nails have demonstrated less radiographic shortening compared with sliding hip screws (SHS).^22-25 These results have only correlated with enhanced functional outcome scores and have been validated by prospective randomized studies that performed a subgroup analysis of patients with high baseline function. A randomized prospective study reported that patients who independently ambulated more than 150 feet before injury treated with SHS had more proximal femoral shortening and poorer functional outcome scores than those treated with torsion control nails.²³ Another study confirmed that improvements in mobility were significantly superior for patients over the age of 80 with a preinjury mobility score ≥7 (good) treated with a nail compared with SHS, at all follow-up intervals after 8 weeks.²⁴

In several retrospective studies, torsion control IM nails have also been shown to reduce radiographic shortening compared with standard IM nails.^5,20,26,27 A study of over 400 subjects showed significantly less shortening with a torsion control IM nail over a standard IM nail (5.10 versus 2.36 mm at 12 months; P < 0.001) in unstable patterns with good reduction.²⁰ Ricci et al⁵ reported that even in stable two-part injuries, there was less collapse with torsion control IM nails compared with standard IM nails and SHS (P < 0.001). Surprisingly, among this cohort with a standard IM nail, 26% still collapsed >1 cm, and 5% >2 cm. What is considered acceptable in terms of these fractures' shortening should be reconciled with newer data using modern implants, particularly in active patients with higher levels of baseline function.

Nonunion/Delayed Union

Delayed union and nonunion are generally defined as lack of healing beyond four and six months, respectively, for extracapsular hip fractures, occurring more frequently in subtrochanteric than pertrochanteric patterns (7% to 20% compared with 1 to 2%).^28,29 In addition to metabolic causes such as diabetes mellitus, smoking history, or infection, a frequent source of nonunion among extracapsular hip fractures is poor initial reduction. It has been observed that an average varus malreduction of 9° (4 to 14° range) was a feature in almost every subtrochanteric nonunion and associated implant failure.²⁹ Krappinger et al³⁰ state that varus malreduction >5° and lack of medial calcar support from either malreduction or comminution were predictive of nonunion in subtrochanteric fractures after IM nailing. A difference in neck-shaft angle (NSA) of >4.4° in varus angulation when compared with the contralateral hip (sensitivity 60%; specificity 90.91%; P = 0.0035), and sagittal angulation >5.5° (sensitivity 60%; specificity 93.94%; P = 0.0001) were both predictive of nonunion and failure in atypical subtrochanteric fractures.³¹ In the current study, 31% were maligned; and alignment was shown to be a greater risk factor for nonunion compared with opening the fracture site and diphosphonate use. It is recommended that the neck-shaft angle be restored to within 5° of the contralateral side for optimal results.

Varus malreduction in extracapsular hip fractures is iatrogenic in several scenarios. In a cadaver study of subtrochanteric femur fractures, an entry point lateral to the tip of the trochanter was created significant varus angulation regardless of proximal bend design (P < 0.0001) than an entry point just medial to the tip of the greater trochanter.³² Eccentric reaming can cause the starting point to drift laterally, and good technique during reaming involves stopping the reamer before and after passing through the entry point to prevent eccentric reaming. In pertrochanteric fractures, a starting point in the fracture itself can create a varus wedge deformity that can obstruct collapse and prevent union (Figure 5).³³

Figure 5: Wedge deformity of a pertrochanteric fracture resulting in varus malreduction. Having a starting point in the fracture can result in further displacement of the fragments during reaming and failure to ream a proximal hole for the implant. The implant itself then acts as a wedge during insertion creating varus and can even prevent controlled collapse resulting in nonunion.

Reverse obliquity fractures represent a unique mechanical environment in which the geometry of the pieces creates shear forces that predispose implants to nonunion and failure. Haidukewych et al³⁴ reported that fixed-angle constructs (blade plates) outperformed sliding constructs (SHS) for such fracture patterns, as prevention of sliding minimized the shear forces. The use of IM nails for these fractures has improved outcomes. In a retrospective study looking at 193 patients with OTA/AO 31A3 fractures treated with long IM nails, an overall nonunion rate of 6% and higher incidence of failure among malreduced fractures (11.1% versus 2.6%; P = 0.021) have been observed.³⁵ There was a higher nonunion rate in this cohort with the use of long stem recon-type IM nail (4.5% versus 23.5%; P = 0.002), which may reflect the lack of fixed-angle proximal locking options in those specific implants. Cephalomedullary nails are not truly fixed angle devices, as demonstrated by the continued sliding observed after fully locking the set screw reported in several studies. However, when treating a reverse obliquity fracture with an IM implant, fully locking the proximal construct is recommended to minimize sliding.

One of the most common deformities encountered after IM nailing of the proximal femur supine on a fracture table is an internal rotation deformity.³⁶ With reverse obliquity fractures, excessive internal rotation of the shaft causes asymmetric cortical engagement at the oblique fracture line, resulting in a large posterior and medial calcar gap (Figure 6). In such circumstances, traction is released and rotation adjusted before distal locking. Distal dynamization would further help close the gaps and allow compression in the fracture plane, particularly with fracture lines that extend below the proximal blade/screw. In cases in which the blade/lag screw prevents axial compression, lateral cortical notching is required to dynamize distally and improve compression across the fracture.³⁷

Figure 6: A, Sawbones model with a revere obliquity fracture is anatomically reduced. B, Excessive internal rotation of the distal fragment as demonstrated by the offset line causes asymmetric cortical engagement of the obliquity resulting in a large calcar gap posterior and medial. In this circumstance, it is recommended to release traction and adjust rotation before distal locking. Any persistent gaps below the level of the blade/lag screw are best addressed with distal dynamization.

The algorithm for surgical management of extracapsular nonunion is based on the condition of the femoral head, location of the broken implant, and presence of fixed deformity to include shortening. Arthroplasty is recommended in cases in which there is destruction or concern regarding the viability of the femoral head. Valgus osteotomy with revision internal fixation is considered in the presence of a fixed deformity/shortening to restore the proper neck-shaft angle. Implant selection for the revision procedure is geared toward achieving secure fixation in the proximal fragment and should target the area where bone stock remains after implant removal. After restoration of the neck-shaft angle, revision internal fixation is typically successful regardless of the type of implant used.³⁸

Early Implant Fracture

Eventual implant fracture is the expected consequence of repetitive loading in the absence of healing or nonunion. Early implant fracture, defined as failure before four months, is concerning because it occurs before the effects of delayed union and may be related to the implant, including its damage. Implant notching is the result of iatrogenic damage to the titanium implant that occurs during insertion; it is the largest contributor to early implant fracture, as observed by Klima³⁹ in a large retrospective study of 342 reported IM nail failures. Although notching can occur at any point along the implant with an aperture for drilling, implant fractures occur most commonly through the smallest cross-sectional area located at the proximal screw aperture (Figure 7)

Figure 7: A, Looking at the proximal lateral aspect of an intramedullary implant with a large notch at the proximal screw aperture. IM implants fracture most commonly at this location through the smallest cross-sectional area. B, Notching of the distal interlocking oblong hole directly at the point of failure of a short stem implant. Notching of the implant occurs at any point along the implant that has an aperture for drilling and can result in implant fracture at that specific location. C, Multiple points of contact and notches at the distal aspect of an IM nail after attempted interlocking. IM, intramedullary.

A key aspect to consider to extend implant survival is preventing implant damage during its insertion through meticulous technique. An eccentrically placed Kirschner wire in the proximal screw aperture will be closer to the implant wall, resulting in wall contact with the reamer and notching. Excessive bending forces applied to the drill/targeting device during insertion, and the continued use of bent Kirschner wires also contributes to insertional notching. Not all notches are attributed to the stepped reamer, as a lag screw/blade inserted over an eccentrically placed guidewire can still cause minor notches and abrasions in titanium that compromise the fatigue strength. Guidewires used for rotational control of the proximal implant should be strategically placed to avoid making contact with the proximal aspect of the implant.

Fractured implants pose several challenges during their removal. Some titanium alloys fracture via a stepped propagation pathway, resulting in irregular pieces with sharp edges, as observed in fatigue failures of these implants.⁴⁰ Secondary fracture lines, also described among these alloys, can contribute to fragmentation during removal if increased force is required for extraction (Figure 8). Additional preoperative planning is recommended to ensure that the surgeon is familiar with the implant, removal procedure, and any broken components before extraction to ensure safe removal. In cases of early failure, a thorough workup to rule out infection is also recommended.

Figure 8: Titanium alloy fracture via a stepped propagation pathway resulting in irregular implants pieces with sharp edges as observed in fatigue failures of these implants. The secondary fracture lines of this alloy indicated by the red arrows can contribute to fragmentation during removal if increased force is required for extraction

Early implant breakage has also been described among various implant components.⁴¹ Auto dynamization is a process of early distal screw breakage that occurs around 3 months in subtrochanteric fracture cases and predates proximal implant breakage by 2 months.²⁸ It is an indicator of mechanical instability. Early revision to a stable construct is recommended to avoid the consequences of prolonged immobility.

Nail Toggle

A randomized prospective study of 220 patients showed that short IM nails maintained reduction with up to 3 cm of subtrochanteric extension with distal cross-locking.⁴² However, this study did not evaluate the effect of canal fill on loss of reduction, as virtually all fractures were treated with the same size implant. In the setting of a large or capacious proximal femoral canal, concerns regarding the use of short IM implants remain. A study revealed that varus deformation/toggle under compressive loading occurred, to some degree, in all short IM specimens loaded; however, it was greatest (17.9 ± 2.6; P < 0.05) in the unlocked specimens in wide canals.⁴³ In this study, distal interlocking reduced the amount of toggle in the femoral canal but did not eliminate toggle entirely. In another biomechanical study, it was observed that peri-implant fracture of short IM implants was preceded by almost immediate toggle of the implant in a large femoral canal that occurred even before loading.⁴⁴

As there are no formal criteria for deciding when the width of the femoral canal requires the use of a long nail, it is at the surgeon's discretion to evaluate when a longer implant that contacts the femoral isthmus is required to prevent displacement (Figure 9). In circumstances where the tip of a short implant does not make direct contact with the lateral femoral cortex, the oval shape of the distal interlocking aperture permits accessory motion in the coronal plane or toggle and varus deformation.⁴⁵ Implant toggle may be a precursor to predicting peri-implant fractures in short IM implants.

Figure 9: Radiographs of implant toggle or early displacement into varus. When the tip of a short implant does not make direct contact with the lateral femoral cortex, the oval shape of the distal interlocking aperture permits accessary motion in the coronal plane resulting in loss of reduction.

Medial Migration

Medial migration, also referred to as axial migration or cut through, occurs when blade/lag screw cuts outs from the femoral head medially instead of superiorly, as seen in typical cutout. In contrast to other mechanical failures, the blade/lag screw fails to slide in the proximal screw aperture and, paradoxically, continues to migrate against gravity into the pelvis after penetrating the acetabulum, thus placing intra-abdominal structures at risk. In some cases, medial migration is preceded by a period of uncontrolled collapse that progresses to the design limits of the proximal set screw. Further collapse or sliding would then be prevented by the implant itself, resulting in medially directed penetration of the femoral head and acetabulum.⁴⁶

A meta-analysis confirmed that medial migration or cut through (OR = 5.33; 95% CI, 2.09 to 13.56; P < 0.01) was more common with helical blades than with lag screws.⁴⁷ This suggests that the in vivo behavior of the helical blade differs compared with the large sliding lag screw used in cephalomedullary nails. This may also be attributed to surgical technique as described by Flores et al,⁴⁸ who found increased odds for combined axial cutout and medial migration with a TAD <20 mm when implanting a helical blade (OR = 1.15, P = 0.01). It is proposed that maintaining a slightly longer TAD with helical blade use will help prevent medial cut through.

In case of new-onset postoperative pain, medial migration detected radiographically is progressive. Hence, limited weight bearing or revision procedures before perforation are recommended. Ideally, arthroplasty is advised, as revision IM nailing and blade exchange with and without augmentation results in a higher failure rate than that with THA alone.¹⁸

Summary

It is essential that surgeons understand how surgical reduction, proper implant selection, and meticulous technique reduce mechanical complications after IM fixation of extracapsular hip fractures in osteoporotic bone. Unstable fracture patterns should be fixated with IM nails with the goal of controlling motion around the implant and at the fracture site. When indicated, torsion control devices provide additional stability and control. Although new technology can be useful, new medical devices alone cannot obviate the need for good reduction or surgical technique.

Table 1 - Literature Review of Level I, II, and III Studies Within Last Five Years Pertaining to IM Nails and Proximal Femoral Shortening

Author	Study Type	Number of Fractures	Implants	Outcomes	Salient Findings
Ong 24 , 2019	Subgroup analysis from Parker, 2017	T: 1000 170 stable 806 unstable 23 basicervical	SHS Taragon IMN	Functional: mobility score	Significant improvement in the mobility score with IMN over SHS if independent before fracture, age <80 yrs, and cognitively intact
Parker 25 , 2017	Prospective randomized, level I	T: 1000 170 stable 806 unstable 23 basicervical	SHS Taragon IMN	Complications Functional: mobility score and pain score	No difference between two implants (LOS, need for revision surgery, blood loss, mortality, and medical/surgical complications). Trend toward more mobility with IM nails was not statistically significant
Sanders 23 2017	Prospective randomized, level I	T: 249 43 stable 206 unstable	Intertan SHS	Radiographic: PFS and alignment Functional: FIM, TUG, LEM, and 2MWT	IMN had fewer patients with limb shortening greater than 1 cm (17%) compared with SHS (42%). Significantly higher TUG/FIM scores with IMN over SHS among subgroup of patients able to ambulate more than 150 feet prior to injury
Berger-Groch 49 , 2016	Prospective randomized, level I	T: 104 28 stable 76 unstable	Intertan Gamma3	Radiographic: CCD and PFS Functional: HHS and SF-36	No differences in outcomes at 5 yrs, only 33 patients available at 5‐yr follow‐up. Average PFS in Intertan was 2mm versus 4mm for Gamma 3, although the difference was not found to be significant
Reindl 22 , 2015	Prospective randomized, level I	T: 167 167 unstable	SHS, TFN, Intertan, and Gamma3	Radiographic: PFS Functional: LEM, FIM, and TUG	Less radiographic shortening with IMN, no difference in functional outcome scores, and mean difference 8-9 mm
Gausden 19 , 2018	Prospective, prognostic, level II	T: 72 30 stable 42 unstable	TFN	Radiographic: PFS Functional: HHS and SF-36	Mean PFS of 4.7mm, shortening more than 8mm correlated with decreaed cadence, decreased step length
Ciufo 27 , 2021	Retrospective, level III	T: 290 290 stable	SHS, Gamma3, and TFN	Radiographic: NSA and PFS	More PFS in SHS compared with IMN. No difference in implant failure rates
Parry 26 , 2020	Retrospective, level III	T: 158 49 stable 62 unstable	Gamma3 TFN	Radiographic: PFS and NSA	Reduction quality and stability of fracture patterns associated with more shortening; calcar gap >4 mm on average 3 mm more of PFS
Ricci 5 , 2019	Retrospective, level III	T: 144 144 stable	SHS Intertan Gamma3	Radiographic: PFS and NSA	26% treated with SHS and SSF collapsed greater than 10 mm and 17% more than 20 mm. More PFS in SHS compared with DSF, P = 0.021, more PFS in SSF compared with DSF P = 0.001
Serrano 20 , 2017	Retrospective, level III	T: 413 234 stable 179 unstable	Intertan Gamma3	Radiographic: PFS and NSA	Standard IMN 2.5 more varus collapse and 2 times more PFS compared with torsion control IMN regardless of stability of fracture

2MWT = two-minute walk time; CCD = caput-collum-diaphyseal; DSF = dual screw fixation; FIM = Functional Independence Measure; HHS = Harris Hip Score; LEM = lower extremity measure; NSA = neck-shaft angle; PFS = proximal femoral shortening; SSF = single-screw fixation; standard IMN = TFN, Gamma3; Torsion Control IMN: Intertan, Taragon IMN; TUG = Timed Up and Go Test.