Though multiple classification systems exist, determination of pertrochanteric fracture stability is the clinically most important determinant. Stable fractures, including AO/OTA 31-A1 and 31-A2.1 fractures, will withstand medial compressive forces after fixation [29]. Unstable fractures, including AO/OTA 31-A2.2, 31-A2.3, and 31-A3 fractures, with compromise of the posteromedial cortex, presence of subtrochanteric extension, or presence of a reversed obliquity fracture pattern, will collapse and/or displace under medial compressive forces despite axial reduction. Additionally, the importance of lateral wall integrity has been reported in the literature. Gotfried found that 100 % of the 24 patients in his study with lateral wall fracture had postoperative fixation collapse [30]. This led him to recommend inclusion of lateral wall integrity as a key component in the assessment of pertrochanteric fracture stability in addition to posteromedial comminution, subtrochanteric extension, and presence of a reversed obliquity fracture pattern. Nonetheless, fracture stability after osteosynthesis of all pertrochanteric fractures is determined by five factors: bone quality, fracture pattern, adequacy of reduction, implant type, and implant-bone positioning [31].

Non-operative treatment of pertrochanteric fractures can be considered in non-ambulatory, demented, and terminally ill patients as well as in those with recalcitrant complex medical comorbidities or who are pain-free [32]. Successful non-operative treatment requires an effective multidisciplinary team with particular attention paid to scheduled patient repositioning to avoid decubitus ulcers, nutrition and fluid balance, as well as pain control. Despite careful management, non-operative treatment often results in serious morbidity. Parker et al. noted that conservative treatment of extracapsular hip fractures is associated with femoral shortening, prolonged hospital stay, and loss of functional independence [33].

The goal of any treatment plan is to eliminate pain and to return the patient to his or her pre-injury level of function. As such, operative treatment is indicated for all patients with pertrochanteric fractures that were previously ambulatory, without dementia, and do not have significant medical comorbidities precluding operative treatment. Optimally, surgery should be initiated within 48 h of injury [34–36]. However, timing of definitive fixation in the polytraumatized patient requires special consideration.

Operative options can be categorized into three main groups: arthroplasty, extramedullary fixation, and intramedullary fixation. Arthroplasty is typically reserved as a salvage procedure for unsuccessful primary fixation of pertrochanteric fractures. However, arthroplasty as the primary treatment for pertrochanteric fractures has been shown to have satisfactory results in selected patients with osteoporosis and comminuted fracture patterns. Kim et al. found no differences in functional outcomes, hospital stay, complications, and time to weight-bearing in their comparison study of hemiarthroplasty versus intramedullary fixation for the treatment of AO/OTA 31-A2 fractures [37]. Pho et al. reported no difference in blood loss when employing a cemented short stem endoprosthetic replacement instead of an intramedullary implant for the treatment of comminuted intertrochanteric fractures in their small case series [38]. In addition, Haentjens et al. found good to excellent outcomes in 75 % of patients treated with bipolar hemiarthroplasty versus osteosynthesis in their study of 79 elderly unstable intertrochanteric fractures [39]. Furthermore, they report that the incidence of pressure sores, pneumonia, and atelectasis was lower and rehabilitation easier and quicker in the arthroplasty cohort. In another supportive study, Vahl et al. found that 77 % of patients returned to full weight-bearing after operative treatment of unstable comminuted pertrochanteric fractures with hemiarthroplasty in 22 osteoporotic patients [40]. Geiger et al. found no difference in mortality in patients treated with primary arthroplasty, dynamic hip screw, and proximal femoral nail for pertrochanteric fractures in the elderly [41]. The authors concluded that primary arthroplasty could be considered in patients with osteoporosis or pre-existing osteoarthritis.

However, primary arthroplasty can be technically demanding; primarily due to the need for a calcar replacing prosthesis, secondarily due to potential complications in greater trochanter fragment reattachment and any resulting hip abductor weakness. Contrary to the aforementioned findings, some studies report poorer outcomes with primary arthroplasty. Tang et al. recently found superior postoperative outcomes with intramedullary nailing of pertrochanteric fractures when compared with hemiarthroplasty in their study of 303 hip fracture patients [42]. Additionally, Chan and Gill reported their outcomes of cemented hemiarthroplasty in elderly osteoporotic patients with pertrochanteric fractures and noted that of their 54 patients, only 48 % regained their pre-injury ambulation status [43]. Furthermore, 23 % of patients in that study completely lost the ability to walk. It is our institution’s practice to not perform a primary arthroplasty for pertrochanteric fractures unless there is severe and debilitating ipsilateral hip osteoarthritis or pathologic fracture.

Of the two fixation constructs, extramedullary fixation is the oldest and includes blade plate, fixed angle nail plate, sliding nail plate, sliding hip screw, and locking plate fixation constructs. Intramedullary fixation is newer and has become more popular for the treatment of pertrochanteric fractures growing in the United States from 3 % of all cases in 1999 to 67 % in 2006 [44]. Theoretically, intramedullary devices have the advantage of more efficient load transfer due to its proximity to the medial calcar compared to extramedullary implants as well as less implant strain because of its closer positioning to the mechanical axis of the femur resulting in a shorter lever arm [45]. Several studies have noted a clinical advantage of intramedullary fixation over extramedullary fixation for the treatment of pertrochanteric fractures. Davis et al. found a higher cutout rate in sliding hips screws versus the Küntscher Y-nail in their comparison of outcomes in 230 pertrochanteric fractures [46]. Sadowski et al. noted shorter operative times, fewer blood transfusions, shorter hospital stays, and a lower incidence of implant failure or nonunion with intramedullary fixation of AO/OTA 31-A3 fractures as compared with 95° screw-plate fixation [47]. Utrilla et al. reported better postoperative walking ability in unstable trochanteric fracture patients treated with intramedullary fixation as compared with a compression hip screw fixation [48]. Additionally, Hardy et al. reported similar mortality rates in intertrochanteric fracture patients treated with an intramedullary device versus a sliding hip-screw, though patients treated with an intramedullary hip-screw demonstrated significantly better mobility at 1 and 3 months postoperatively [49]. However, the superiority of the intramedullary hip-screw was no longer present at 6 and 12 months postoperatively in that study. Platzer et al. demonstrated that cephalomedullary nailing was more successful than dynamic hip screw fixation in preventing limb length discrepancy in AO/OTA 31-A2 and 31-A3 fractures in their study of 95 patients [50]. Finally, Gill et al. showed that pertrochanteric femoral nailing had shorter operative times and a lower complication rate in their comparison study of femoral nailing versus sliding hips screw fixation [51]. In general, intramedullary fixation allows for immediate weight bearing postoperatively which is particularly important in elderly patients. Furthermore, fixation with an intramedullary device portends less biologic disruption than extramedullary fixation due to significantly less soft tissue disruption and periosteal stripping of the femoral cortex while still allowing for controlled fracture compression.

To the contrary, some studies have noted equivalent or poorer outcomes after treatment of pertrochanteric fractures with intramedullary fixation versus extramedullary fixation. Adams et al. found that treatment of intertrochanteric femoral fractures with an intramedullary nail was associated with a higher but not statistically significant risk of postoperative complications in their comparison of a short intramedullary nail versus a sliding hip screw construct in 400 patients [52]. Parker and Handoll suggested that intramedullary nailing may not be superior to extramedullary implants in the treatment of AO/OTA 31-A1 or 31-A2 pertrochanteric fractures due to a higher complication rate observed in their Cochrane Database Systemic Review [53]. Knobe et al. compared helical blade nails versus locked minimally invasive plating in AO/OTA 31-A2 pertrochanteric fractures and found no difference in the frequency of reoperation, mortality, and postoperative function between the two cohorts [54]. These findings were also supported by Yli-Kyyny et al. who compared extramedullary and intramedullary fixation for the treatment of pertrochanteric fractures in 14,915 patients but found no difference in outcomes [55]. Finally, summarizing all of these results, Huang et al. performed a meta-analysis of randomized controlled trials comparing proximal femoral nailing with dynamic hip screw fixation for trochanteric fractures and found equivalent results among the two different implants [56]. However, all of preceding studies demonstrated investigational limitations in that analyses included both stable and unstable pertrochanteric fractures.

As previously noted, pertrochanteric fractures are extracapsular hip fractures spanning of the region between the femoral neck and femoral shaft. This region of transition is supported by an intraosseous framework of dense trabecular bone oriented to withstand extreme compressive and tensile forces. Originally described by Ward, this framework includes primary compressive trabeculae oriented from the medial femoral calcar to the superior dome of the femoral head, primary tensile trabeculae oriented from the foveal region of the femoral head to the lateral border of the femur just distal to the greater trochanter, as well as secondary compressive trabeculae oriented from the lesser trochanter to the greater trochanter, and secondary tensile trabeculae oriented from the mid basicervical region to the lateral border of the femur distal to the greater trochanter (Fig. 16.4) [67, 68]. The greater trochanter has also been described to have its own unique intraosseous trabecular framework. As bone is weaker in tension than in compression, the stress configuration of the proximal femur is important in that fractures occur along the path of least resistance [69, 70]. This is responsible for the commonly observed fracture patterns seen clinically.

Another important consideration when assessing a pertrochanteric fracture is the proximal femoral muscle attachments and deforming muscle forces. The iliopsoas muscle inserts onto the lesser trochanter and is responsible for a flexion and external rotation deforming force. Hip abductors muscles, including the gluteus medius and minimus muscles, are lateral deforming forces. The gluteus medius and gluteus minimus muscles insert onto the posterior and anterior portions of the greater trochanter, respectively. The adductor longus, brevis, and magnus muscles as well as the gracilis muscle comprise the hip adductor muscles. Originating proximally at the pelvis and inserting distal to the fracture on the medial surface of the femur, these muscles act as varus and adducting deforming force. External rotator muscles (including the piriformis, superior and inferior gemelli, obturator internus and externus, and quadratus femoris muscles) insert in the pertrochanteric region and cause external rotation deformity of proximal fracture fragments. The hamstring muscles and the gluteus maximus muscle comprise the hip extensors muscles and primarily shorten proximal hip fractures due to their proximal pelvic origin and distal insertion [71]. However, all deforming muscular forces play some role in shortening the femur after pertrochanteric fracture. Understanding the deforming muscle forces can help with the indirect and direct reduction maneuvers mentioned later in this section.

Perfusion to the pertrochanteric region of the hip is supplied by branches of the medial femoral circumflex artery (MFCA) and the lateral femoral circumflex artery (LFCA). Both arise from the profunda femoris artery with the MFCA coursing posteriorly and the LFCA coursing anteriorly. Perfusion is often maintained in the setting of a pertrochanteric fracture due to the rich and redundant blood supply of the proximal femoral metaphysis including anastomotic contributions from the inferior gluteal artery [72, 73]. Though intramedullary fixation may pose a risk of iatrogenic neurovascular or musculotendinous injury, a precise intramedullary nail starting point and skillful operative technique may reduce this risk. In fact, Ansari et al. determined that there was no musculotendinous, neurovascular, or capsular injury with antegrade femoral nailing via a trochanteric starting point [74]. Anatomic dissections have been performed to determine if a trochanteric start point without tendinous attachments exists to further reduce iatrogenic damage to the gluteal musculotendinous unit. Gardner et al. described an ellipsoid 21 mm trochanteric ‘bald spot’ free of all tendinous insertions indicating a safe starting point for intramedullary trochanteric nailing (Fig. 16.5) [75]. Ozsoy et al. found that hip flexion and adduction during antegrade femoral nailing can decrease the risk of damage to the superior gluteal nerve and gluteus medius muscle [76]. However, iatrogenic injury can still occur if an inappropriately positioned medial starting point within the piriformis fossa is obtained. Dora et al. observed a higher incidence of injury to the gluteal musculotendinous and neurovascular structures with a piriformis starting point compared with a trochanteric starting point for antegrade femoral nailing [77]. When intramedullary nailing of pertrochanteric fractures is performed, a trochanteric entry point minimizing postoperative pain and abductor weakness has to be chosen. The design of the implant must be adapted to this entry portal.

Once in the operating room, the patient is typically positioned supine on a fracture table or other radiolucent table. We prefer a fracture table because of its ability to perform hands-free manipulation including longitudinal traction and axial rotation of the operative extremity. The injured extremity is appropriately padded and secured to a foot post. The contralateral uninjured extremity is either placed in an abducted and flexed position using a well leg holder or scissored posteriorly. Of note, while we prefer to use a well leg holder for the uninjured extremity, there have been reports of compartment syndrome [78]. Therefore, great care and appropriate padding must be employed. Finally, a well-padded perineal post is positioned and attached to the operating room table. This post should not impinge on the patient’s labia or scrotum to prevent iatrogenic damage to these structures. Finally, the ipsilateral arm is typically placed in an elevated sling or secured across the patient’s body to increase operative exposure. Of note, the upper torso, if possible, should be positioned on the opposite side of the table to provide more unimpeded access to the femoral medullary canal.

The C-arm image intensifier should be positioned prior to prepping and draping the operative extremity so as to obtain optimal orientation for intraoperative use. The image intensifier is best positioned approaching the injured limb from the contralateral side of the table. If one fluoroscopy machine is used for obtaining both anteroposterior and lateral images, the uninjured leg must be in the appropriate position so as to not obstruct the movement of the image intensifier. At our institution, we prefer a dual image intensifier arrangement with one machine oriented in the anteroposterior plane and another in the lateral plane to obtain immediate biplanar images (Fig. 16.6).

Whether performed directly or indirectly, reduction of the pertrochanteric fracture is recommended prior to placement of the intramedullary nail. Indirect reduction is primarily achieved with manipulation of the injured lower extremity via the foot holder. In our experience, successful closed reduction is most often achieved with longitudinal traction and slight internal rotation. Adduction versus abduction or hip flexion versus extension is dependent on the fracture pattern and resulting deformity (Fig. 16.7a–g). If all attempts at an acceptable closed reduction are unsuccessful, then open reduction maneuvers must be employed. Open reduction instruments may be helpful depending on the type of deformity and the number of fracture fragments that requires reduction. Commonly used implements include 5.0 mm Schanz pins, ball spikes, crutches, Cobb elevators, bone hooks and additional threaded guidewires to provisionally secure the reduction or prevent the proximal segment from rotating during insertion of the sliding screw or helical blade. With open reduction, pointed reduction clamps positioned through a small incision can also be used to further reduce the fracture with minimal risk of fracture hematoma evacuation or significant soft tissue disruption [58]. Flexion of the proximal segment is best controlled with a ball spike. Safe insertion of the ball spike is performed by first palpating and marking the course of the femoral artery on the thigh. An approximately 3 cm skin and fascial incision is made over the inferior femoral neck. Blunt dissection is performed until the proximal segment can be palpated and a ball spike is inserted with application of downward pressure to perform the reduction. To counteract posterior sag of the distal segment, a well-positioned crutch, Cobb elevator from the proximal or lateral incision, or a 5.0 mm Schanz pin allows for manipulation of the distal shaft segment (Fig. 16.8). We prefer to use a ball spike to control proximal fragment flexion and a laterally inserted 5.0 mm Schanz pin into the distal fragment (Fig. 16.9a–f). Additionally, depending on the brand of implant used, the nail and its insertion instruments can aid in reduction. For example, the guide sleeve of the Trochanteric Fixation Nail (TFN; Synthes, West Chester, PA, USA) can be used to buttress the lateral cortex of the proximal fragment in a reversed obliquity fracture pattern to aid in reduction (Fig. 16.10a–d). Regardless of the techniques used, it should be noted that all reduction maneuvers are designed to counteract the deforming muscular forces and anatomically reduce the fracture. Once reduction is obtained, the intramedullary nailing process can be initiated.