Fig. 6.1
Stable femoral neck fracture on the right side in a moribund patient, axial view (a) and pelvis ap (b). Patient was bedridden due to spinal stenosis and multiple cardiovascular diseases. (c) and (d) showing the same patient after 3 months and (e) and (f) after 6 months. The fracture is healed, the patient has no pain. Arrows indicate the fracture
Fig. 6.2
CT-scan of the right hip from the same patient as in Fig. 6.1, (a) and (b) showing the frontal, (c) the sagittal, and (d) the axial plain at the time of injury. Arrows indicate the fracture
Another interesting approach to this field was published by Buord et al. in 2009 [8]. They treated 57 Garden I fractures in patients age 65 and older (the mean was 82), with a standardized “early functional training” with full weight bearing and frequent radiographic follow-up on post-injury days 2, 7, 21, and 45, and months 3, 6, and 12 to evaluate the predictive factors of displacement and the results of the functional training. If displacement occurred, then arthroplasty was performed. One-third of the patients had a displacement at a mean of 10 days; in fact, they reported comparable results in the functional successful vs. the arthroplasty group with Parker Score (6.9 vs. 7) and Harris Hip Score (HHS) (82 vs. 85), but one has to admit that the arthroplasty “control” group was the failed functional training group. Unfortunately, they were unable to identify predictive parameters for displacement, such as age, gender, side, fracture type, inclination angle, degree of outward displacement, sagittal displacement, and general status. In view of the missing predictive values and the disadvantages of secondary arthroplasty after primary osteosynthesis, [9] the approach with this trial-and-error management could be an option for borderline patients, as early mobilization has multiple advantages. However, as long as there is no clear evidence for this kind of treatment, conservative treatment should be an individual decision, especially for moribund patients, and the standard should be the osteosynthesis.
Type of Implant for Osteosynthesis in Stable Femoral Neck Fractures
Choosing the type of implant leads directly to the next step. Usually, there are two implant types that are feasible: two or three parallel cannulated screws , and fixed-angle devices such as the sliding hip screw (SHS) or cephalomedullary nails [6, 10–13]. Most authors prefer cannulated screws, as they are a fast, inexpensive method.
In 112 consecutive patients, Krastman and colleagues reported positive results for stable and undisplaced femoral neck fractures if treated with two cannulated screws , but the patient collective was very heterogeneous due to age and fracture type [6]. In addition, Manohara concluded that the cancellous screw fixation for undisplaced femoral neck fractures in the elderly is associated with relatively few complications and revision rates [11]. However, he found longer hospital stays and a higher mortality rate in the >75 years patient collective.
Unfortunately, there is no significant evidence-level study comparing the outcome of the two internal fixation (IF) methods , especially not in the elderly. In 2008, Liporace et al. tried to compare the fixed-angle devices with cannulated screws in 76 displaced high vertical femoral neck fractures (Pauwels Type 3, Orthopaedic Trauma Association (OTA) type 31 B2.3) and found a non-union rate of 19% in the cannulated screw treated group vs. 8% in the fixed-angle device group, although this was not significant [14]. Siavashi et al. demonstrated significantly better results for the DHS compared to the cannulated screws after 1 year in the young with no fixation failure in the DHS group vs. an 18% failure rate in the cannulated screw group (p < 0.001) [15]. It seems that the fixed-angle devices provide better stability, but further research is necessary, since any comparative study is in displaced and/or young patients. Figure 6.3a–d shows an example of DHS fixation .
Fig. 6.3
Garden II femoral neck fracture in an active patient on the left side with a posterior tilt of 30° (see Fig. 6.5), pelvis ap (a) and axial view (b). Same patient after treatment with DHS, pelvis ap (c) and axial view (d). Arrows indicate the fracture
However, there are some interesting considerations regarding the biomechanics, which allows for cautious recommendations. First of all, the question remains whether two screws are enough, or whether a third one is necessary. Maurer et al. tested, in a cadaveric model, the anterior loading, incremental axial loading, and cycling loading of two and three screws in two matched pairs of human cadaveric femurs with femoral neck fractures and found that three screws yield better results, with greater resistance to anterior loading, less inferior femoral head displacement, and less superior gapping at the osteotomy site [16]. Furthermore, Yang and colleagues evaluated the influence of the relative screw position of three screws in young patients with femoral neck fractures and found a significantly better union rate for the “inverted triangle configuration ” (91 vs. 77%, p = 0.018) (Fig. 6.4a–d) which means one screw placed distally near the calcar femoris, and two screws parallel above it anteriorly and posteriorly to form an inverted triangle [17].
Fig. 6.4
Stable femoral neck fracture on the right side treated with three cannulated screws in the inverted triangle, hip ap (a) and axial view (b). One screw lies in a steep angle near the calcar femoris, two screws are parallel above it, forming an inverted triangle highlighted in (b). In (c) and (d) there is a stable femoral neck fracture on the left side treated with three cannulated screws not lying in the preferred triangle configuration
As it is known that for trochanteric fractures, the so-called “tip-apex” distance of the sliding hip screw should be less than 25 mm [18], one could assume similar results for the femoral neck fracture, but there is as yet little data on the exact screw position in relation to the head for the undisplaced femoral neck fractures [19]. In 2009, Palm et al. described the posterior tilt for the undisplaced femoral neck fractures [13]. They treated 113 patients ≥60 years old with Garden I and II fractures with two cannulated screws and found that the posterior tilt (Fig. 6.5a, b) was a predictor for reoperation with a rate of 14/25 to 12/88 if the posterior tilt was ≥ 20° (p < 0.001). Clement and colleagues proved that the posterior tilt ≥20° was an independent predictor of internal fixation failure in 162 elderly patients; moreover, they found the ASA (American Society of Anesthesiologists) grade to be an independent predictor for failure [12]. The reason for this finding is not clear; one possible explanation could be the higher rate of osteoporotic bone. In a biomechanical in vitro study, Paech et al. found better results for polymer-augmented sliding hip screws in osteoporotic bone with a decrease of failure in terms of cut-out [20]. However, there are still no studies regarding cement-augmented cannulated screws for femoral neck fractures.
Fig. 6.5
Stable femoral neck fracture on the right side, axial view (a). A posterior tilt of 32° is shown. In (b) there is a stable femoral neck fracture on the right side, axial view, a posterior tilt of 5° is shown. The posterior tilt is the angle between the mid-collum line (MCL) and the radius collum line (RCL) defined at the line from the head center to the crossing of the MCL with the radius of the head [13]
Overall, the implant situation is not clear, and in some ways disappointing. There are some new implants, such as the angular stable multiple screw fixation (Targon FN) that has yielded good results, [21] but other studies have shown no difference for this implant against cannulated screws [22]. In our opinion, the treatment of the stable femoral neck fracture in the elderly should be pragmatic: If there is no posterior tilt and good bone stock, then three cannulated screws in an inverted triangle should be used. If there is poor bone stock and/or posterior tilt ≥20°, then a sliding hip screw with additional anti-rotational screw should be used, augmented with bone cement eventually, depending on the bone quality.
Unstable Fractures
Internal Fixation vs. Arthroplasty
Although there have been multiple discussions on the best implant for femoral neck fractures in the young patient, it is beyond dispute that a primary reduction and fixation should be obtained in a timely, acceptable manner [23]. In the elderly population, the discussion on the best strategy for displaced femoral neck fractures has been going on for ages. The general considerations are about early mobilization, failure rate, functional outcome, mortality, and socioeconomic costs. Since many patients in this group have numerous diseases, and since internal fixation is mostly performed with a sliding hip screw or 2–3 cancellous parallel screws, it is understandable that surgeons perform such small, fast operation [24, 25]. Many studies have shown that internal fixation takes less operating time, there is less blood loss and fewer needs for transfusion [25, 26]. However, the problems occur post-operatively (Fig. 6.6a–h): high failure rates with the necessity for reoperation in 30–43% [9, 24, 25, 27], compared with 6–11% for hemiarthroplasty (HA) [9, 28, 29], and there is a high rate of avascular necrosis of the femoral head in the internal fixation group [9]. Despite these data, internal fixation is still a common treatment for patients with severe comorbidities—and especially dementia. One reason could be the expected higher mortality of hemiarthroplasty (HA) vs. internal fixation (IF), but many studies could not confirm this in active elderly patients [9, 25, 27]. A small exclusion has to be made: Parker et al. found a tendency towards improved survival after IF in patients aged 90 or above, and in those with a low mobility score, although this was not statistically significant [26]. Figure 6.7a–d shows the treatment of a displaced femoral neck fracture with HA.
Fig. 6.6
Unstable and displaced femoral neck fracture on the left side, hip ap (a) and axial view (b), arrows indicate the fracture. Treatment was with three cannulated screws, (c) showing the post-operative hip ap view (d) the axial view. Although the treatment was done in an adequate technique, there was internal fixation failure after 6 weeks with shortening and a bad functional result (e, f). Arrows indicate the screw movement in relation to the washer. After 6 months (g, h), the fracture is still visible and the head collapses, resulting in a revision with hemiarthroplasty
Fig. 6.7
Unstable and displaced Garden III femoral neck fracture in active patients with no radiological osteoarthritis on the left side (a) before and (b) after surgery with hemiarthroplasty with cemented stem and bipolar head. (c) and (d) showing the same for a Garden IV femoral neck fracture. Arrows indicate the fractures
As nearly one-third of the patients with femoral neck fracture suffer from dementia or other mental deficiencies, in following the post-
operative treatment protocols, the question emerges as to whether this group would benefit. However, Olofsson et al. found no differences in mortality between patients with or without dementia treated with IF or arthroplasty, and all patients with arthroplasty had better functional results after 4 months and 1 year [5]. In 2014, Johansson published long-term results of 146 fractures and included 38% mentally impaired patients in his study [30]. The failure rate for the IF was very high, with 55% for the lucid and only 16% for the mentally impaired, whereas failure was defined as early redisplacement, non-union, symptomatic segmental collapse, or severe infection. For arthroplasty, the failure rates were 5% and 16% respectively, whereas failure was defined as two dislocations or more, implant loosening, severe infection, or a periprosthetic fracture. Most of the complications occurred within the first 2 years. Unfortunately, this study has a high bias due to patient loss. After 2 years, only 50% of the mentally impaired patients were still alive, and after 5 years 13% (n = 7); this makes interpretation of the data on the mentally impaired very difficult. Further research is needed to highlight this large, increasing patient population, although the trend is toward arthroplasty even in this cohort.
With regard to functional outcome, many studies show either equal [24] or better results for arthroplasty [9, 25, 26, 31, 32]. Furthermore, a secondary arthroplasty after failed IF seems to have worse results for hip function [33]. Blomfeldt et al. compared the outcomes of 43 patients with a primary hip due to femoral neck fractures with 41 patients with secondary arthroplasty due to IF failure; both the hip function and the health-related quality of life were significantly better after hip arthroplasty [33]. In contrast, Parker et al. found in one of the largest long-term studies (with 455 patients), no difference in the outcomes of IF vs. HA after 11 years [26]. Due to the study design, uncemented stems were used, and no THA control group was evaluated. Also Ravikumar and Marsh showed—in 290 patients after 13 years—equally poor outcomes with regard to function for IF and HA, but they had good functional results for the total hip arthroplasty (THA) group, which will be discussed in the HA vs. THA section [27].
Some studies dealt with the socioeconomic outcome and performed a cost effectiveness analysis of IF vs. HA or THA. Obviously, the sheer costs of the implants and the shorter operating time for IF speak for themselves, but if one keeps the higher failure rate and necessity for reoperations in mind, this recommendation may change. Bjørnelv et al. made a cost effectiveness” analysis alongside a randomized, controlled trial in Norway, and found that besides a better health-related quality of life for HA, there were higher overall costs for IF in comparison with HA [34], whereas Johansson and colleagues found no difference in the costs for 143 patients receiving either THA or IF in addition to better results for the THA group for the Harris Hip Score [35].
To summarize, regarding the actual evidence on unstable femoral neck fractures in view of treatment with internal fixation or arthroplasty, one has to conclude that the internal fixation with the high failure rate and the worse functional and health-related results cannot be recommended as the implant of first choice for the active elderly. In our opinion, an HA or THA should be performed for the elderly and active, whereas the internal fixation can be considered to be a fast and gentle tool for moribund and bedridden patients. For the active mentally impaired elderly, there is no clear evidence as yet, but the tendency is towards hemiarthroplasty.
Uni- vs. Bipolar Head
As noted, the HA still remains the “work horse” in the treatment of displaced femoral neck fractures in the elderly. However, depending on the region and economic considerations, the decision to use a uni- or bipolar head varies. The basic idea to use a bipolar head is the reduction of the acetabular erosion, as the movement mainly takes place in the bipolar head rather than in the joint, and thereby reduces pain levels and increases the clinical outcome [36, 37]. Unfortunately, the evidence in the existing literature varies as much as the real treatment. In a randomized control trial study of 120 patients, Hedbeck and colleagues found that at a mean age of 86 years, there was an almost identical clinical outcome after 1 year, but a significantly higher incidence of acetabular erosion in the unipolar HA group. As they found trends towards worse Harris Hip Scores (HHS) in patients with acetabular erosion, they concluded that a bipolar HA should be the preferred treatment [37]. In contrast, Calder et al. reported no difference in clinical outcomes and complication rates between both groups [36]. They evaluated 250 patients with a median age of 85, and found significantly better results for the unipolar group in view of the return to pre-injury state, in comparison to the bipolar group. In summary, current evidence is not able to provide a conclusive recommendation for or against unipolar or bipolar hemiarthroplasties in octogenarians.
But the question remains of whether the “young” old patients would benefit from a bipolar head. It was again Calder et al. evaluating a questionnaire from a randomized control trial of a group of 110 patients age 65–79 who were treated with either hip screw, unipolar, or bipolar arthroplasty [38]. The bipolar arthroplasty group showed a trend towards better results in almost all questionnaire values in this patient collective. But in 2001, Davison et al. showed in a prospective, randomized control trial of the treatment of 280 displaced fractures of the femoral neck (187 patients had had arthroplasty, and 93 sliding hip screws) in patients aged 65–79, that there was no advantage to the bipolar over the unipolar head [24]. So again, there is no clear evidence for or against one of the two options. Theoretically, younger patients would have a higher acetabular erosion rate due to longer survival, although the existing literature does not support the use of the four times higher costs of bipolar head in the 65 to 79-year-old age group [39]. Moreover, the actual trend goes toward total hip arthroplasty in this group, under certain circumstances.
Cemented vs. Uncemented Stem
Today, there are many studies that deal with this subject, and there is convincing evidence in favor of cemented or uncemented stems. The reason for choosing cement, or an uncemented implant, depends mostly on the surgeons’ experience, education, and personal preference. Surgeons who use uncemented implants are afraid of the revision surgery, and the rare, but severe cardiopulmonary effects of cement in this patient cohort; the other group (who use cement) may worry about early loosening with pain and worsening function [40]. Figved and colleagues found no difference in HHS scores after 3 and 12 months in 220 patients (112/108 cemented/uncemented stems, 83.4/83.0 years) [41]. Also, DeAngelis et al. found no differences in mortality and various activities of daily living post-operatively and after 12 months in a prospective control trial with 125 fractures [42]. However, there are many more studies that show better results for the cement group in this cohort. It seems that cemented stems have better function and mobility results in the short-term [43], that the patients have statistically less pain, lower rates of complication [44], and lower periprosthetic fracture rates [45]. Furthermore, the results regarding walking ability, the use of walking aids, and activities of daily living were statistically better [40]. In 2010, a Cochrane Systematic Review dealing with this topic was published [39]. Many studies were included, but most had weaknesses in form and content. Nevertheless, the Cochrane Review corroborated the better results for cemented stems.
In the last few years, new models of the hydroxyapatite-coated stems have come on the market, and it remains to be seen whether they are associated with similar results. In 2014, Bell and colleagues showed, in a case-control study of nearly 180 patients, better results for the hydroxyapatite-coated Corail stem (DePuy Ortho- paedics Inc., Warsaw, Indiana) in view of further surgery, less operating time, and lower peri-prosthetic fracture rates in comparison with a cemented Exeter stem (Stryker Howmedica, Newbury, United Kingdom) [46]. Whether this is a really new way, or just a trend, cannot be assessed yet. Hence, the recommendation today should be a cemented stem for treating femoral neck fractures in the elderly.
Hemiarthroplasty vs. Total Hip Arthroplasty
As highlighted above, arthroplasty is the recommended treatment for active patients with unstable femoral neck fractures. However, the question increasingly emerges about whether active patients should receive a THA (Fig. 6.8a–d) rather than an HA. Van den Bekerom et al. found in 252 patients (>70 years) who had either cemented HA (n = 137) or THA (n = 115), that there were no differences in the modified HHS, revision rate of the prosthesis, local and general complications, or mortality [47]. Furthermore, they reported lower intra-operative blood loss for HA (7% > 500 ml) than for THA (26% > 500 ml), shorter surgery for HA (12% > 1.5 h vs. 28% > 1.5 h). and no dislocations of any HA, but 8 dislocations of the THA during their 1- and 5-year follow-ups. Because of the dislocation rate, they concluded that they would not recommend THA for these patients in the absence of advanced radiological osteoarthritis or rheumatoid arthritis of the hip [47]. But is the higher dislocation rate a factor for recommendation, or should we rather ask the patients whether they are satisfied? In 2013, Leonardson et al. showed—in a national survey of 4467 patients—better results for those below and above 70 years of age who were treated with THA; they had less pain and more satisfaction compared with those treated with IF or HA [48]. This shows the conflict in the debate about the best strategy for or against THA pretty well. Finally, it is a question of outcome parameter definition.
