- Email: [email protected]
Radiological evaluation of intertrochanteric fracture fixation by the proximal femoral nail
Injury, Int. J. Care Injured 43 (2012) 856–863
Contents lists available at SciVerse ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Radiological evaluation of intertrochanteric fracture fixation by the proximal femoral nail Amir Herman a,b,*, Yair Landau a, Gabriel Gutman a, Vladislav Ougortsin a, Aharon Chechick a, Nachshon Shazar a a b
Department of Orthopedic Surgery, Chaim Sheba Medical Center, Tel-Hashomer, Israel, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel Talpiot Medical Leadership Program, Chaim Sheba Medical Center, Tel-Hashomer, Israel, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 24 October 2011
Background: Successful treatment of intertrochanteric femoral fractures was reportedly influenced by the position of the fixation devices, by reduction quality and by fracture type. Methods: The records of 227 patients with intertrochanteric fractures treated by intramedullary hip screws were analysed retrospectively. The angle and distance from the femur head apex were transformed into Cartesian coordinates. Comparisons were performed between patients with no mechanical failure (207 patients, 90.7%), with cutouts (15 patients, 6.6%) and with secondary loss of reduction (5 patients, 2.2%). Results: The standard tip apex distance (TAD) measurement above 25 mm did not predict failure (p = 0.62). Mechanical failure rates increased from 4.8% to 34.4% when the centre of lag screw was not in the second quarter of the head–neck interface line (the so-called ‘‘safe zone’’) (p = 0.001). Lag screw insertion lower or higher than 11 mm of the head apex line were associated with failure rates of 5.5% and 18.6%, respectively (p = 0.004). Multivariate logistic regression showed that lag screw insertion not within the ‘‘safe-zone’’ was associated an Odds Ratio of 13.4 (95% CI 2.24–81) for mechanical failure (p = 0.004). Conclusions: The TAD scale focuses on length measurement and lacks the vector properties of multidirectional measurements. Vector analysis revealed that the caudal-cranial correct lag screw position is the most important factor in preventing mechanical failure. ß 2011 Elsevier Ltd. All rights reserved.
Keywords: Intertrochanteric fractures Tip apex distance Proximal femoral nail
Introduction Proximal femur fractures are amongst the most common injuries encountered by the orthopaedic surgeon. The incidence of these fractures in the US alone is expected to reach 500,000 per year in 2040.1 Intertrochanteric fractures account for about half of all proximal femur fractures.2 Fixation devices for intertrochanteric fractures include intramedullary devices (e.g., Gamma nail, intramedullary hip screws, etc.) and extramedullary devices, mainly sliding screws and plates (e.g., dynamic hip screws). The former showed an advantage in fixation of unstable intertrochanteric fractures and the latter yielded better results for fixation of stable intertrochanteric fractures.38 The correct insertion and positioning of both intra- and extramedullary devices are known to prevent implant failure
* Corresponding author at: Department of Orthopedic Surgery, Chaim Sheba Medical Center, Tel.:-Hashomer, 52621, Israel. Tel.: +972 3 5302623; fax: +972 3 5302523. E-mail address: [email protected] (A. Herman). 0020–1383/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2011.10.030
and cutout. Baumgaertner et al. (1995) had shown that small tip apex distance, i.e., less than 25 mm, is associated with a lower probability for cutout.9 Other authors have divided the femoral head into nine areas, and recommended head screw insertion at the middle-middle or lower-middle area.10 However, these works focused either on dynamic hip screws or Gamma nails, and other designs of intramedullary nails are currently available, such as is the proximal femoral nail (PFN), which has a different design by virtue of its additional antirotation hip screw proximal to the main lag screw.8 This design was shown to have biomechanical properties that are different from those of a single-head screw.11 We evaluated the application of postoperative radiographic parameters and measurements to predict success of intertrochanteric femoral fracture fixation using PFN designed nails. These parameters include reduction quality, screw placement parameters, nail length and tip apex distance amongst others. We also performed a multivariate analysis to determine which parameter is most important. Our expectation is that this analysis will prove useful for surgeons in guiding them in achieving optimal clinical results.
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
Patients and methods Institutional review board approval was obtained for this retrospective investigation. The study included 227 patients with unstable intertrochanteric fractures that were treated in our institute by PFN-like fixation devices from 2000 to 2009. Inclusion criteria were unstable intertrochanteric fracture, and a follow-up of at least six months after the initial surgery and an X-ray that demonstrated a radiological union, or patients with failed surgery. The patients with pathologic fractures were excluded. Mechanical failure was defined as either cutout (15 patients, Fig. 1A) or secondary reduction loss (5 patients Fig. 1B and C). Two hundred and seven patients did not have any mechanical failure. Clinical and radiological parameters were compared between these three groups. The patients with no mechanical failure were also compared with the twenty patients who had either cutout or secondary reduction loss. Pre and post injury mobility were grouped as follows: Group 1 = Walking without any ambulation aid or using one aid (cane, tripod etc.); Group 2 = Walking with a walking frame and Group 3 = Wheelchair use. Osteoporosis level was classified according to the Singh classification.12 Fracture classification was according to the orthopaedic trauma association (OTA) classification,13 in which scores of OTA 31.A2 and OTA 31.A3 were considered as unstable fracture patterns. Patients with subtrochanteric fractures fixed by the PFN devices were also included. All the patients in our cohort had unstable intertrochanteric fractures. One hundred and forty-seven (64.8%) fractures had been fixed by means of the Targon proximal femur (Targon PF) device (Aesculap, Tuttlingen, Germany), and eighty (35.2%) fractures were fixed with the antirotation trochanteric nailing system (ATN) device (dePuy, Warsaw, IN, USA). One hundred seventy-six (77.5%) fractures were fixed with standard length nails, and fifty-one (22.5%) with long nails (more than 300 mm). All surgeries were performed in accordance to standard surgery technique and the manufacturer’s recommendations. Postoperative rehabilitation protocol included ‘‘weight bearing as tolerated’’ for six weeks and than full weight bearing ambulation protocol was used. Radiological measurements Radiological measurements of the post-surgery fixation device were performed using postoperative radiographic images.
857
All measurements were scales by a factor of the true lag screw width divided by the radiographic lag screw width measurement. Reduction quality was defined using both anteroposterior (AP) and axial radiographs. On AP radiographs, the neck shaft angle was measured and used to assess reduction quality. On axial radiographs, reduction quality was assessed by the ‘‘reduction gap’’, defined as the displacement (in millimetres) between the medial cortexes of the distal and proximal fracture parts. We defined the neck-head transition points as the points where the head–neck contour changes from the head convex to neck concave contour. The head–neck interface line was defined as the connection of these two points (Fig. 2, Line L1). The neck centre line was defined as a line perpendicular to L1 which crosses L1 in its centre (Fig. 2, Line L2). The head apex was defined as the point at which the neck centre line crosses the femur head cortical bone. Further measurements are D1 which represents the length of L1, D2 which represents the distance from the central axis of the head lag screw to the inferior edge of L1, and D3 which represents the distance between the superior edge of L1 to the superior part of the antirotation screw (Fig. 2). On the AP radiograph, the length of the tip apex distance (TAD) vector was measured for the lag screw and antirotational screw AP ðTADAP LS and TADAR , respectively). The angle between the neck central axis line and the TAD vector for the lag screw and antirotational screw were measured (aAP and bAP, respectively). An angle cranial to the neck central axis line was taken as being positive, and an angle caudal to the neck central axis line as negative. Fig. 3A and B presents the AP radiographic TAD and angle measurements. Similar measurements were performed, only for the lag screw, on the axial radiograph. Axial measurements were Ax marked by TADAx (Fig. 3C). Axially measured angles were LS and a considered positive or negative when they were anterior or posterior to the mid-femur neck line, respectively, i.e., anterior or posterior screw placement. We used a standard TAD multiplied by sine or cosine for transformation from the Polar coordinate system (distance and angle) to the Cartesian coordinate system, in which the head–neck interface line and the neck centre line were used as bases. The centre of the axis which is the (0,0) point, is the femur head apex. For example, in order to calculate the lag screw cranial-caudal AP position on an AP radiograph, we used TADAP LS sin a . In order to
Fig. 1. (A) A cutout of an intertrochanteric fracture. (B) The postoperative position of the centre of the lag screw is inferior in this 85-year-old female. (C) The same patient as in (B) after reduction loss.
858
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
screw tip-apex distance on the neck centre axis less or greater than 11 mm, centre of lag screw position within the second distal quarter of the head–neck interface line (‘‘safe zone’’, see Fig. 4), gender, TAD less or greater than 25 mm, osteoporosis grade less or equal 2, TAD on axial view and nail system used (ATN or Targon PF). The Odds Ratio (OR) was calculated as exponent beta, and 95% confidence intervals were also calculated. Results General considerations and fixation
Fig. 2. Femur neck measurements. The head–neck interface line (L1) is a connecting line between the two curving points where the convexity of the femur head contour turns into the femur neck concavity. The centre neck line (L2) is a line perpendicular to the head–neck interface line in its mid-length. The apex is the point where the centre neck line crosses the femur head cortex. D1 = the length of the head–neck interface line. D2 = the distance to the centre of lag screw. D3 = the distance to the upper part of the antirotation screw.
determine the medial-lateral position of the lag screw, we used AP TADAP LS cos a . Beta angle measurements were used for antirotational screw transforms, and lag screw axial measurements were used for similar axial calculations. Statistical analysis Data were analysed by SPSSß 16.0 software (SPSS Inc., Chicago, IL). Categorical variables are presented as count (percent). All data was not available for all the patients. The percents are given from available data. Continuous variables are presented as mean and standard deviation (SD). The Chi-square test or Fisher’s exact test were used to test for statistical significance amongst categorical variables. The latter was used when the expected count was lower than five in at least one cell. The Wilcoxon–Mann–Whitney rank sum (Kruskal–Wallis test) test was used to calculate statistical significance amongst two groups (or more) of continuous variables. All p values are presented as two-sided. The level of significance was set at p < 0.05. A post-surgery axial ‘‘reduction gap’’ cut-point of 5 mm was found to maximize the Youden’s index (1.3). The distance of tip of the lag screw from the apex, on the neck centre axis AP ðTADAP LS cos a Þ, was dichotomized using cut-point of 11 mm which was found to maximize the Youden’s index (1.28, see Fig. 4). Multivariate logistic regression was performed using mechanical failure (either cutout or secondary reduction loss) as the dependent covariate. Independent covariates in the model included: an axial reduction gap less or greater than 5 mm, lag
The study included two hundred twenty-seven patients. The mean postoperative follow-up time was 25.7 (SD 19.9) months. Fifteen patients had a cutout of their fixation devices (Fig. 1A), and five patients had a secondary reduction loss. The mean time from surgery to cutout was 3.36 (SD 3.6) months, and the mean time from surgery to secondary reduction loss was 2.63 (SD 1.44) months (Fig. 1B and C). Demographics, and length characteristics did not differ between the three patient groups (Table 1). Fixation device types differed between patients groups. More failures were noted in ATN type nailing system (p value = 0.01). More secondary reduction losses were observed in male patients: three out of the five patients with secondary displacements were males (p = 0.047). Preinjury mobility status (aids used) was available for only 115 patients. Of 106 patients without mechanical complications, 81 patients (76.4%) used no walking aids or one aid, 25 patients (23.6%) used walking frame. Of eight patients with cutout 7 patients (87.5%) used no or one ambulation aid, and one patient (12.5%) used a wheelchair for ambulation. One patient (100% of available data) with secondary displacement used no or one walking aid. This difference was found to be statistically significant (p value = 0.003). Post injury (final visit) ambulation aids data was available in 87 patients. Of which 78 patients had no complication, 7 had a cutout and 2 had secondary displacement. Of the overall available data 19 patients (21.8%) used no walking aid or one walking aid, 62 patients (71.3%) used a walking frame and 6 patients (6.9%) were wheelchair bound. Comparing the pre and post injury mobility aids in the total patient group, 89 patients (77.4%) and 19 patients (21.8%) used none or one walking aid pre and post injury, respectively. Twentyfive patients (21.7%) and 62 patients (71.3%) had used a walking frame pre and post injury, respectively. Wheelchair was used by one patient (0.9%) and 6 patients (6.9%) pre and post injury, respectively. These differences were found to be statistically significant (p value = 0.001). The mean osteoporosis grade for the entire study group was 2.94 (SD 1.42). There was no statistically significant difference between the mechanical failure groups and patients with no failure (Table 1). Reduction quality The AP postoperative displacement, as measured by the neck shaft angle in the AP radiograph, measured a mean of 132.2 (SD 6.2) degrees. The axial ‘‘reduction gap’’ measurement were available for 122 patents. Mean axial ‘‘reduction gap’’ was 8.4 (SD 6.55 mm, p value = 0.69, see Table 2). Thirty-seven (30.3%) had an axial postoperative ‘‘reduction gap’’ less than 5 mm. Only one of these patients (2.7%) had a cutout, and none had a secondary reduction loss. Eighty-five (69.7%) had an axial ‘‘reduction gap’’ greater than 5 mm, and five (of six, 83.3%) had a cutout and 4 (100% of available 4) had a secondary displacement (p value = 0.073).
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
859
Fig. 3. (A) Transformation of the anteroposterior (AP) lag screw polar coordinates (distance and angle) to the Cartesian components of the head–neck interface axis AP AP AP ðTADAP LS sin a Þ and centre neck axis ðTADLS cos a Þ. (B) Transformation of AP antirotational polar coordinates (distance and angle) to the Cartesian components of the AP AP AP head–neck interface axis ðTADAP AR sin b Þ and centre neck axis ðTADAR cos b Þ. (C) Transformation of axial lag screw polar coordinates (distance and angle) to the Cartesian Ax Ax Ax components of the head–neck interface axis ðTADAx LS sin a Þ and centre neck axis ðTADLS cos a Þ.
Femur neck radiographic measurements The femur neck radiographic measurements revealed that the centre of lag screw positioning as a fraction of the head–neck interface line (D2/D1 in Fig. 2) had an average of 0.39 (SD 0.09). There was a statistically significant difference in the centre of lag
screw position between the three study groups (p value = 0.001). Lower values were associated with secondary displacements and higher values were associated with cutouts (Table 2). We identified the second lower quarter (D2/D1 between 0.25 and 0.5) of the head–neck interface line (L1) as being the ‘‘safe zone’’ for the centre of lag screw. The centre of the lag screw was
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
860
Fig. 4. The femur neck ‘‘safe zone’’ which is the second distal quadrant on the head–neck interface line. The 11 mm medial-lateral mark on the neck central axis is depicted.
not in the ‘‘safe zone’’ in thirty-eight patients, and twelve of them (31.6%) had either a cutout or a secondary reduction loss (Fig. 4). Of the one hundred and eighty-seven patients in whom the centre of lag screw was in the ‘‘safe zone’’, only seven (3.7%) suffered either a cutout or a secondary reduction loss (p value = 0.001).
Femur head radiographic measurements The mean TADs were calculated for the lag screw as well as for the antirotational screw and they were found to be 20.7 (SD 6.7) mm and 53.6 (SD 17.7) mm, respectively (p = NS between groups)
Table 1 Demographic and clinical data. No. failure (207 patients) Age (year) Gender Female Male Side Left Right Osteoporosis grade Nail type Targon PF ATN Nail length Standard Long Reduction Closed Open
75.3 (SD 15.74) 159 (90.9%) 48 (92.3%)
Cutout (15 patients)
Secondary reduction loss (5 patients)
p value
78.2 (SD 14.9)
82.0 (SD 8.2)
0.529
14 (8.0%) 1 (1.9%)
2 (1.1%) 3 (5.3%)
0.046
107 (92.2%) 100 (90.1%) 2.95 (SD 1.42)
8 (6.9%) 7 (6.3%) 3 (SD 1.46)
1 (0.9%) 4 (3.6%) 2.4 (1.14)
0.369 0.712
140 (95.2%) 67 (83.8%)
6 (4.1%) 9 (11.2%)
1 (0.7%) 4 (5.0%)
0.01
163 (92.6%) 44 (86.3%)
10 (5.7%) 5 (9.8%)
3 (1.7%) 2 (3.9%)
0.355
194 (90.7%) 13 (100%)
15 (7.0%) –
5 (2.3%) –
0.514
PF: proximal femur; ATN: antirotation trochanteric nailing system.
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
861
Table 2 Radiographic reduction and femur neck measurements.
Reduction quality AP neck shaft angle (8) Axial ‘‘reduction gap’’ (mm) Axial ‘‘reduction gap’’ above 5 mm Femur neck screw positioning (see Fig. 2) Lag screw centre position (D2/D1) Lag screw centre in the ‘‘safe zone’’ (D2/D1 is 0.25-0.5) Antirotation screw position (D3/D1)
No. failure (207 patients)
Cutout (15 patients)
Secondary reduction loss (5 patients)
p value
132.3 (SD 6.1) 8.2 (SD 6.3) 76/112 (67.9%)
130.7 (SD 6.5) 12.0 (SD 10.5) 5/6 (83.3%)
132.0 (SD 7.3) 8.9 (SD 2.1) 4 (100%)
0.755 0.699 0.073
0.38 (SD 0.08) 180 (87.4%) 0.17 (SD 0.11)
(Table 3). Comparisons between the AP and axial components of the TAD for these two screws showed a statistically significant difference in the AP component alone (Table 3). One hundred and nineteen (76.8%) patients had TAD less than or equal 25 mm, of these 10 patients (8.4%) also had a cutout or secondary reduction loss. Thirty-six patients (23.2%) had TAD higher than 25, of these 4 patients (11.1%) had also a cutout or secondary reduction loss. This difference was not found to be statistically significant (p value = 0.62). The AP positions of the lag screw on the head–neck interface AP axis ðTADAP LS sin a Þ differed between the study groups. Extreme higher and lower values were significantly (p = 0.001) associated with cutouts and secondary loss of reduction, respectively (Table 3). The distance of the lag screw tip on the central neck axis AP ðTADAP LS cos a Þ was greater than 11 mm in 43 (19.1%) patients (Fig. 4). Eight of these patients (18.6% of 43 pts) had mechanical AP failure. The distance on the central neck axis ðTADAP LS cos a Þ was less than 11 mm in one hundred and eighty-two patients (80.9%), ten (5.5% of 182 pts) of whom had mechanical failure (p value = 0.004 for less than 11 mm vs. more than 11 mm). Multivariate analysis Multivariate logistic regression identified the centre of lag screw positioning in the ‘‘safe zone’’ of the head–neck interface as the most important factor in preventing mechanical failure (Table 4). Failure attributed to positioning of the centre of lag screw outside the ‘‘safe zone’’ had an OR of 13.4 (95% CI = 2.24–81, p value = 0.004). Differences in the other variables in the model were not found to be statistically significant (p value > 0.05).
0.52 (SD 0.07) 4 (28.6%) 0.08 (SD 0.04)
0.27 (SD 0.08) 3 (40%) 0.26 (SD 0.14)
0.001 0.001 0.001
Discussion The findings of this study demonstrate that the major factor contributing to treatment success is correct lag screw fixation position. Lag screw position was divided into two perpendicular axes, the head–neck interface axis and the central neck axis. Positioning of the lag screw tip within or higher than 11 mm of the head apex, on the central neck axes, were associated with failure rates of 5.5% and 19.1%, respectively. On the head–neck interface axis, we were able to define a ‘‘safe-zone’’ as being the second quarter (the lower 25–50%) of the head–neck interface line. Lag screws placed in or out of this ‘‘safe zone’’ were associated with failure rates of 3.7% and 31.6%, respectively. Multivariate analysis identified the lag screw position within the head–neck interface ‘‘safe zone’’ as the most important cofactor in preventing mechanical failure. We believe these findings are important for guiding the surgeon in correct positioning of the fixation device. It practically means that for a successful fixation the lag screw guide wire should be placed within the ‘‘safe zone’’ as defined here. We have included the vector analysis performed which led to the definition of the ‘‘safe-zone’’. However, intraoperative complicated calculations are not necessary in order to use the ‘‘safe-zone’’ concept. Our study is the first radiographic analysis of the double screw PFN fixation device. Although our calculations were done on dual screw fixation devices, our results and recommendations are not that different from those known for single screw fixation devices (e.g., DHS and Gamma nail).9,14 As such we recommend to use the ‘‘safe-zone’’ criteria also for assessing the location of single screw fixation devices.
Table 3 Radiographic femur head fixation device position measurements.
Lag screw TADLS (mm) TADAP LS (mm) TADAx LS (mm) On AP head neck interface axis AP TADAP (mm) LS sin a On AP neck centre axis AP AP TADLS cos a (mm) On axial head–neck interface axis Ax TADAx (mm) LS sin a On axial neck centre axis Ax TADAx (mm) LS cos a Antirotation screw TADAR (mm) TADAP AR (mm) TADAx AR (mm) On AP head–neck interface AP TADAP (mm) AR sin b On AP neck centre axis AP TADAP (mm) AR cos b
No. failure (207 patients)
Cutout (15 patients)
Secondary reduction loss (5 patients)
p value
20.3 (SD 6.5) 9.7 (SD 3.2) 10.0 (SD 3.9)
24.0 (SD 6.5) 11.7 (SD 2.6) 11.8 (SD 4.5)
25.5 (SD 12.9) 12.9 (SD 4) 12.2 (SD 7.2)
0.175 0.008 0.470
5.49 (SD 4.4)
0.001
10.7 (SD 4.7)
0.061
2.41 (SD 3.9) 8.6 (SD 3) 0.77 (SD 4.9) 9.3 (SD 3.6) 53.4 (SD 17) 26.9 (SD 7.3) 27.4 (SD 10.2
3.41 (SD 3.5) 10.7 (SD 3.3) 0.31 (SD 3.9)
0.67 (SD 6.4)
0.568
11.3 (SD 4.4)
11.1 (SD 6.9)
0.289
58.0 (SD 30.6) 30.2 (SD 10.3) 27.6 (SD 16.5)
– 25 (SD 11.9) –
0.84 0.68 0.94
7.6 (SD 3)
12.2 (SD 2.5)
4.5 (SD 2.7)
0.001
25.6 (SD 7.6)
27.8 (SD 11.6
24.9 (SD 12.3)
0.92
AP: anteroposterior. Mean (SD) standard deviation are presented.
862
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
Table 4 Multivariate logistic regression. Covariate
Odds ratio (95% CI)
Lag screw tip distance from the apex, on the central neck axis 1 Less than or equal to 11 mm (baseline) Greater than 11 mm 3.12 (0.25–38.9) Axial ‘‘reduction gap’’ Less than 5 mm (baseline) 1 Greater than 5 mm 9.88 (0.65–150) Centre of lag screw position 1 Within the ‘‘safe zone’’ (baseline) Outside the ‘‘safe zone’’ 13.4 (2.24–81) Gender Female (baseline) 1 Male 1.95 (0.32–11.7) Hardware system used TPF (baseline) 1 5.61 (0.81–38) ATN Tip apex distance (TAD) Less than 25 mm (baseline) 1 More than 25 mm 5.34 (0.19–150) TAD (Lag screw–axial view) 1.15 (0.83–1.61) Osteoporosis 1 Osteoporosis grade 3 (baseline) Osteoporosis grade 2 0.43 (0.07–2.55)
p value
0.375
0.099
0.004
0.464
of at least one or two years, respectively. This is, however, an acceptable follow-up period for a retrospective study of intertrochanteric fractures, with others having used similar time periods in their retrospective studies.4,9,23,24 Prospective studies tend to have longer follow-up periods, i.e., up to one year.3,6,7,17,25 Data on the mobility aids used was available for only about 50% of the patients in this cohort. Another drawback of this study is that it does not take into account several parameters that were shown to influence the outcome of proximal femoral fractures, such as the patients’ body mass index, and the American Society of Anaesthesiologist (ASA) score.15 We believe that this work may provide a guide for surgeons in optimal screw position for reducing the risk of mechanical failure when performing reduction and fixation of intertrochantric fractures.
0.079
Conflict of interest
0.323 0.384
The authors declare that there is no conflict of interests in preparing this manuscript. The authors state that neither they nor any of their immediate family members has received any financial benefit from this study or any commercial company mentioned therein. No donation, due to this work, was made by any commercial company mention in the text to the institute in which the authors serve as surgeons.
0.352
CI: confidence interval; TPF: Targon proximal femur device (Aesculap, Tuttlingen, Germany); ATN: antirotation trochanteric nailing system (ATN) device (dePuy, Warsaw, IN, USA).
This is the first time the head–neck interface line has been defined. It is the line connecting the two transition points of the contour change between femur head convexity to femur neck concavity. We also defined the femur neck central axis as being a perpendicular line that crosses the head–neck interface line in its middle. We believe that the delineation of the neck central axis is more accurate and one that enables a unique definition of the head apex as being the point at which the central neck line crosses the head subchondral bone. Further studies are required in order to validate this point. We believe that these definitions will prove to be useful in other femur orthopaedic surgery fields, such as intracapsular proximal femur fractures and hip arthroplasties. In our study, the TAD as defined by Baumgaertner et al. (1995) did not emerge as being an important factor in terms of influencing either cutouts or secondary reduction loss.9 Only its AP components, when checked separately, was found to predict either cutouts or secondary reduction loss. This finding emphasizes the weakness of the TAD parameter, since it is a scalar measurement that calculates only distance and disregards direction. Our findings and definitions offer surgeons better understanding of the optimal position of lag screws in proximal femur fixation devices. The mechanical failure rate reported herein was 8.8% (twenty of two hundred and twenty-seven patients). This failure rate is comparable to other technical failure rates reported elsewhere.5,7,10,15–18 The Cochrane review by Parker and Handoll (2008)6 includes three randomized clinical trials (Pajarinen et al., 2005; Papasimos et al., 2005; Saudan et al., 2002)19–21 that compared the outcome of the PFN with that of the sliding hip screw. The overall number of patients treated by a PFN was one hundred and ninety-four, of whom five (2.57%) had cutouts of the fixation nail whilst technical failure of the fixation occurred in eleven patients (5.6%). Giraud et al. (2005) compared thirty-four patients treated by the Targon PF device with twenty-six patients treated with sliding hip screw.22 They reported three cutouts (8.8%) in the former and two (7.7%) in the latter. One weakness of our study is that it is retrospective and has a relatively short follow-up period. We included patients whose follow-up was at least six months and who had signs of radiological union. Note that of 207 patients with no mechanical failure, 159 (76.8%) and 98 (47.3%) patients had a follow-up period
Acknowledgment This research was funded by the Talpiot medical leadership programme of the Sheba medical centre, grant number 20963. No external (commercial) funding source was used. References 1. Cummings SR, Rubin SM, Black D. The future of hip fractures in the United States: numbers, costs, and potential effects of postmenopausal estrogen. Clincal Orthopedics 1990;252:163–6. 2. Koval KJ, Aharonoff GB, Rokito AS, Lyon T, Zuckerman JD. Patients with femoral neck and intertrochanteric fractures: are they the same? Clincal Orthopedics 1996;330:166–72. 3. Hardy DR, Descamps PY, Krallis P, Fabeck L, Smets P, Bertens CL, Delince PE. Use of an intramedullary hip-screw compared with a compression hip-screw with a plate for intertrochanteric femoral fractures: a prospective, randomized study of one hundred patients. JBJS (Am) 1998;80:618–30. 4. Mainds CC, Newman RJ. Implant failures in patients with proximal fractures of the femur treated with a sliding screw device. Injury 1989;20:98–109. 5. Osnes, EK, Lofthus, CM, Falch, JA, et al. More postoperative femoral fractures with the Gamma nail than the sliding screw plate in the treatment of trochanteric fractures. Acta Orthopedica Scandinavica, 2001;72:252–256. 6. Parker MJ, Handoll HH. Gamma and other cephalocondylic intramedullary nails versus extramedullary implants for extracapsular hip fractures in adults. Cochrane Database Syst Rev 2008;3:CD93. 7. Schipper IB, Steyerberg EW, Castelein RM, van der Heijden FHWM, den Hoed PT, Kerver AJH, van Vugt AB. Treatment of unstable trochanteric fractures: randomised comparison of the gamma nail and the proximal femural nail. JBJS (Br) 2004;86-B:86–94. 8. Simmermacher RK, Bosch AM, Van der Werken CH. The AO/ASIF- proximal femoral nail (PFN): a new device for the treatment of unstable proximal femur fractures. Injury 1990;30:327–32. 9. Baumgaertner MR, Curtin SL, Lindskog DM, Keggi JM. The value of the tip-apex distance in predicting failure of fixation of peritrochanteric fractures of the hip. JBJS (Am) 1995;77:1058–64. 10. Bridle SH, Patel AD, Bircher M, Calvert PT. Fixation of intertrochanteric fractures of the femur, a randomized prospective comparison of the gamma nail and the dynamic hip screw. JBJS (Br) 1991;73-B:330–4. 11. Helwig P, Faust G, Hindenlang U, Hirschmu A, Konstantinidis L, Bahrs C, Sudkamp N, Schneider R. Finite element analysis of four different implants inserted in different positions to stabilize an idealized trochanteric femoral fracture. Injury 2009;40:288–95. 12. Singh M, Nagrath AR, Maini PS. Changes in trabecular pattern of the upper end of the femur as an index of osteoporosis. JBJS (Am) 1970;52:457–67. 13. Fracture and Dislocation Classification Compendium, 2007. Orthopaedic Trauma Association Classification, Database and Outcomes Committee. Journal of Orthopedic Trauma, 2007;10, Suppl 1:31–42. 14. Davis TRC, Sher JL, Horsman A, Simpson M, Porter BB, Checketts RG. Intertrochanteric femoral fractures–mechanical failure after internal fixation. JBJS (Br) 1990;72-B:26–31.
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863 15. Forte ML, Virnig BA, Swiontkowski MF, Bhandari M, Feldman R, Eberly LE, Kane RL. Ninety-day mortality after intertrochanteric hip fracture: provider volume matter? JBJS (Am) 2010;92:799–806. 16. Lindskog DM, Baumgaertner MR. Unstable intertrochanteric hip fractures in the elderly. J Am Acad Orthop Surg 2004;12:189–90. 17. Lundy DW. Subtrochanteric femoral fractures. J Am Acad Orthop Surg 2007;15: 663–71. 18. Rogmark C, Flensburg L, Fredin H. Undisplaced femoral neck fractures—no problems? A consecutive study of 224 patients treated with internal fixation. Injury 2009;40:274–6. 19. Pajarinen J, Lindahl J, Michelsson O, Savolainen V, Hirvensalo E. Pertrochanteric femoral fractures treated with a dynamic hip screw or a proximal femoral nail; a randomised study comparing postoperative rehabilitation. Journal of Bone and Joint Surgery (Br), 2005;87. 20. Papasimos S, Koutsojannis CM, Panagopoulos A, Megas P, Lambiris E. A randomised comparison of AMBI, TGN and PFN for treatment of unstable trochanteric fractures. Archives of Orthopaedic and Trauma Surgery, 2005;125.
863
21. Saudan M, Lubbeke A, Sadowski C, Riand N, Stern R, Hoffmeyer P. Pertrochanteric fractures: is there an advantage to an intramedullary nail? A randomized, prospective study of 206 patients comparing the dynamic hip screw and proximal femoral nail. Journal of Orthopaedic Trauma 2002;16:386–93. 22. Giraud B, Dehoux E, Jovenin N, Madi K, Harisboure A, Usandizaga G, et al. Pertrochanteric fractures: a randomized prospective study comparing dynamic screw plate and intramedullary xation. Revue de Chirurgie Orthopedique et Reparatrice de l’Appareil Moteur 2005;91:732–6. 23. Laros CS, Moore JF. Complications of fixation in intertrochanteric fractures. Clincal Orthopedics 1974;11:0–119. 24. Larsson S, Friberg S, Hansson Ll. Trochanteric fractures Influence of reduction and implant position on impaction and complications. Clincal Orthopedics 1990:130–9. 25. Leung KS, So WS, Shen WY, Hui PW. Gamma nails and dynamic hip screws for peritrochanteric fractures: a randomized prospective study in elderly patients. JBJS (Br) 1992;74-B:345–51.
Contents lists available at SciVerse ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Radiological evaluation of intertrochanteric fracture fixation by the proximal femoral nail Amir Herman a,b,*, Yair Landau a, Gabriel Gutman a, Vladislav Ougortsin a, Aharon Chechick a, Nachshon Shazar a a b
Department of Orthopedic Surgery, Chaim Sheba Medical Center, Tel-Hashomer, Israel, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel Talpiot Medical Leadership Program, Chaim Sheba Medical Center, Tel-Hashomer, Israel, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 24 October 2011
Background: Successful treatment of intertrochanteric femoral fractures was reportedly influenced by the position of the fixation devices, by reduction quality and by fracture type. Methods: The records of 227 patients with intertrochanteric fractures treated by intramedullary hip screws were analysed retrospectively. The angle and distance from the femur head apex were transformed into Cartesian coordinates. Comparisons were performed between patients with no mechanical failure (207 patients, 90.7%), with cutouts (15 patients, 6.6%) and with secondary loss of reduction (5 patients, 2.2%). Results: The standard tip apex distance (TAD) measurement above 25 mm did not predict failure (p = 0.62). Mechanical failure rates increased from 4.8% to 34.4% when the centre of lag screw was not in the second quarter of the head–neck interface line (the so-called ‘‘safe zone’’) (p = 0.001). Lag screw insertion lower or higher than 11 mm of the head apex line were associated with failure rates of 5.5% and 18.6%, respectively (p = 0.004). Multivariate logistic regression showed that lag screw insertion not within the ‘‘safe-zone’’ was associated an Odds Ratio of 13.4 (95% CI 2.24–81) for mechanical failure (p = 0.004). Conclusions: The TAD scale focuses on length measurement and lacks the vector properties of multidirectional measurements. Vector analysis revealed that the caudal-cranial correct lag screw position is the most important factor in preventing mechanical failure. ß 2011 Elsevier Ltd. All rights reserved.
Keywords: Intertrochanteric fractures Tip apex distance Proximal femoral nail
Introduction Proximal femur fractures are amongst the most common injuries encountered by the orthopaedic surgeon. The incidence of these fractures in the US alone is expected to reach 500,000 per year in 2040.1 Intertrochanteric fractures account for about half of all proximal femur fractures.2 Fixation devices for intertrochanteric fractures include intramedullary devices (e.g., Gamma nail, intramedullary hip screws, etc.) and extramedullary devices, mainly sliding screws and plates (e.g., dynamic hip screws). The former showed an advantage in fixation of unstable intertrochanteric fractures and the latter yielded better results for fixation of stable intertrochanteric fractures.38 The correct insertion and positioning of both intra- and extramedullary devices are known to prevent implant failure
* Corresponding author at: Department of Orthopedic Surgery, Chaim Sheba Medical Center, Tel.:-Hashomer, 52621, Israel. Tel.: +972 3 5302623; fax: +972 3 5302523. E-mail address: [email protected] (A. Herman). 0020–1383/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2011.10.030
and cutout. Baumgaertner et al. (1995) had shown that small tip apex distance, i.e., less than 25 mm, is associated with a lower probability for cutout.9 Other authors have divided the femoral head into nine areas, and recommended head screw insertion at the middle-middle or lower-middle area.10 However, these works focused either on dynamic hip screws or Gamma nails, and other designs of intramedullary nails are currently available, such as is the proximal femoral nail (PFN), which has a different design by virtue of its additional antirotation hip screw proximal to the main lag screw.8 This design was shown to have biomechanical properties that are different from those of a single-head screw.11 We evaluated the application of postoperative radiographic parameters and measurements to predict success of intertrochanteric femoral fracture fixation using PFN designed nails. These parameters include reduction quality, screw placement parameters, nail length and tip apex distance amongst others. We also performed a multivariate analysis to determine which parameter is most important. Our expectation is that this analysis will prove useful for surgeons in guiding them in achieving optimal clinical results.
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
Patients and methods Institutional review board approval was obtained for this retrospective investigation. The study included 227 patients with unstable intertrochanteric fractures that were treated in our institute by PFN-like fixation devices from 2000 to 2009. Inclusion criteria were unstable intertrochanteric fracture, and a follow-up of at least six months after the initial surgery and an X-ray that demonstrated a radiological union, or patients with failed surgery. The patients with pathologic fractures were excluded. Mechanical failure was defined as either cutout (15 patients, Fig. 1A) or secondary reduction loss (5 patients Fig. 1B and C). Two hundred and seven patients did not have any mechanical failure. Clinical and radiological parameters were compared between these three groups. The patients with no mechanical failure were also compared with the twenty patients who had either cutout or secondary reduction loss. Pre and post injury mobility were grouped as follows: Group 1 = Walking without any ambulation aid or using one aid (cane, tripod etc.); Group 2 = Walking with a walking frame and Group 3 = Wheelchair use. Osteoporosis level was classified according to the Singh classification.12 Fracture classification was according to the orthopaedic trauma association (OTA) classification,13 in which scores of OTA 31.A2 and OTA 31.A3 were considered as unstable fracture patterns. Patients with subtrochanteric fractures fixed by the PFN devices were also included. All the patients in our cohort had unstable intertrochanteric fractures. One hundred and forty-seven (64.8%) fractures had been fixed by means of the Targon proximal femur (Targon PF) device (Aesculap, Tuttlingen, Germany), and eighty (35.2%) fractures were fixed with the antirotation trochanteric nailing system (ATN) device (dePuy, Warsaw, IN, USA). One hundred seventy-six (77.5%) fractures were fixed with standard length nails, and fifty-one (22.5%) with long nails (more than 300 mm). All surgeries were performed in accordance to standard surgery technique and the manufacturer’s recommendations. Postoperative rehabilitation protocol included ‘‘weight bearing as tolerated’’ for six weeks and than full weight bearing ambulation protocol was used. Radiological measurements Radiological measurements of the post-surgery fixation device were performed using postoperative radiographic images.
857
All measurements were scales by a factor of the true lag screw width divided by the radiographic lag screw width measurement. Reduction quality was defined using both anteroposterior (AP) and axial radiographs. On AP radiographs, the neck shaft angle was measured and used to assess reduction quality. On axial radiographs, reduction quality was assessed by the ‘‘reduction gap’’, defined as the displacement (in millimetres) between the medial cortexes of the distal and proximal fracture parts. We defined the neck-head transition points as the points where the head–neck contour changes from the head convex to neck concave contour. The head–neck interface line was defined as the connection of these two points (Fig. 2, Line L1). The neck centre line was defined as a line perpendicular to L1 which crosses L1 in its centre (Fig. 2, Line L2). The head apex was defined as the point at which the neck centre line crosses the femur head cortical bone. Further measurements are D1 which represents the length of L1, D2 which represents the distance from the central axis of the head lag screw to the inferior edge of L1, and D3 which represents the distance between the superior edge of L1 to the superior part of the antirotation screw (Fig. 2). On the AP radiograph, the length of the tip apex distance (TAD) vector was measured for the lag screw and antirotational screw AP ðTADAP LS and TADAR , respectively). The angle between the neck central axis line and the TAD vector for the lag screw and antirotational screw were measured (aAP and bAP, respectively). An angle cranial to the neck central axis line was taken as being positive, and an angle caudal to the neck central axis line as negative. Fig. 3A and B presents the AP radiographic TAD and angle measurements. Similar measurements were performed, only for the lag screw, on the axial radiograph. Axial measurements were Ax marked by TADAx (Fig. 3C). Axially measured angles were LS and a considered positive or negative when they were anterior or posterior to the mid-femur neck line, respectively, i.e., anterior or posterior screw placement. We used a standard TAD multiplied by sine or cosine for transformation from the Polar coordinate system (distance and angle) to the Cartesian coordinate system, in which the head–neck interface line and the neck centre line were used as bases. The centre of the axis which is the (0,0) point, is the femur head apex. For example, in order to calculate the lag screw cranial-caudal AP position on an AP radiograph, we used TADAP LS sin a . In order to
Fig. 1. (A) A cutout of an intertrochanteric fracture. (B) The postoperative position of the centre of the lag screw is inferior in this 85-year-old female. (C) The same patient as in (B) after reduction loss.
858
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
screw tip-apex distance on the neck centre axis less or greater than 11 mm, centre of lag screw position within the second distal quarter of the head–neck interface line (‘‘safe zone’’, see Fig. 4), gender, TAD less or greater than 25 mm, osteoporosis grade less or equal 2, TAD on axial view and nail system used (ATN or Targon PF). The Odds Ratio (OR) was calculated as exponent beta, and 95% confidence intervals were also calculated. Results General considerations and fixation
Fig. 2. Femur neck measurements. The head–neck interface line (L1) is a connecting line between the two curving points where the convexity of the femur head contour turns into the femur neck concavity. The centre neck line (L2) is a line perpendicular to the head–neck interface line in its mid-length. The apex is the point where the centre neck line crosses the femur head cortex. D1 = the length of the head–neck interface line. D2 = the distance to the centre of lag screw. D3 = the distance to the upper part of the antirotation screw.
determine the medial-lateral position of the lag screw, we used AP TADAP LS cos a . Beta angle measurements were used for antirotational screw transforms, and lag screw axial measurements were used for similar axial calculations. Statistical analysis Data were analysed by SPSSß 16.0 software (SPSS Inc., Chicago, IL). Categorical variables are presented as count (percent). All data was not available for all the patients. The percents are given from available data. Continuous variables are presented as mean and standard deviation (SD). The Chi-square test or Fisher’s exact test were used to test for statistical significance amongst categorical variables. The latter was used when the expected count was lower than five in at least one cell. The Wilcoxon–Mann–Whitney rank sum (Kruskal–Wallis test) test was used to calculate statistical significance amongst two groups (or more) of continuous variables. All p values are presented as two-sided. The level of significance was set at p < 0.05. A post-surgery axial ‘‘reduction gap’’ cut-point of 5 mm was found to maximize the Youden’s index (1.3). The distance of tip of the lag screw from the apex, on the neck centre axis AP ðTADAP LS cos a Þ, was dichotomized using cut-point of 11 mm which was found to maximize the Youden’s index (1.28, see Fig. 4). Multivariate logistic regression was performed using mechanical failure (either cutout or secondary reduction loss) as the dependent covariate. Independent covariates in the model included: an axial reduction gap less or greater than 5 mm, lag
The study included two hundred twenty-seven patients. The mean postoperative follow-up time was 25.7 (SD 19.9) months. Fifteen patients had a cutout of their fixation devices (Fig. 1A), and five patients had a secondary reduction loss. The mean time from surgery to cutout was 3.36 (SD 3.6) months, and the mean time from surgery to secondary reduction loss was 2.63 (SD 1.44) months (Fig. 1B and C). Demographics, and length characteristics did not differ between the three patient groups (Table 1). Fixation device types differed between patients groups. More failures were noted in ATN type nailing system (p value = 0.01). More secondary reduction losses were observed in male patients: three out of the five patients with secondary displacements were males (p = 0.047). Preinjury mobility status (aids used) was available for only 115 patients. Of 106 patients without mechanical complications, 81 patients (76.4%) used no walking aids or one aid, 25 patients (23.6%) used walking frame. Of eight patients with cutout 7 patients (87.5%) used no or one ambulation aid, and one patient (12.5%) used a wheelchair for ambulation. One patient (100% of available data) with secondary displacement used no or one walking aid. This difference was found to be statistically significant (p value = 0.003). Post injury (final visit) ambulation aids data was available in 87 patients. Of which 78 patients had no complication, 7 had a cutout and 2 had secondary displacement. Of the overall available data 19 patients (21.8%) used no walking aid or one walking aid, 62 patients (71.3%) used a walking frame and 6 patients (6.9%) were wheelchair bound. Comparing the pre and post injury mobility aids in the total patient group, 89 patients (77.4%) and 19 patients (21.8%) used none or one walking aid pre and post injury, respectively. Twentyfive patients (21.7%) and 62 patients (71.3%) had used a walking frame pre and post injury, respectively. Wheelchair was used by one patient (0.9%) and 6 patients (6.9%) pre and post injury, respectively. These differences were found to be statistically significant (p value = 0.001). The mean osteoporosis grade for the entire study group was 2.94 (SD 1.42). There was no statistically significant difference between the mechanical failure groups and patients with no failure (Table 1). Reduction quality The AP postoperative displacement, as measured by the neck shaft angle in the AP radiograph, measured a mean of 132.2 (SD 6.2) degrees. The axial ‘‘reduction gap’’ measurement were available for 122 patents. Mean axial ‘‘reduction gap’’ was 8.4 (SD 6.55 mm, p value = 0.69, see Table 2). Thirty-seven (30.3%) had an axial postoperative ‘‘reduction gap’’ less than 5 mm. Only one of these patients (2.7%) had a cutout, and none had a secondary reduction loss. Eighty-five (69.7%) had an axial ‘‘reduction gap’’ greater than 5 mm, and five (of six, 83.3%) had a cutout and 4 (100% of available 4) had a secondary displacement (p value = 0.073).
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
859
Fig. 3. (A) Transformation of the anteroposterior (AP) lag screw polar coordinates (distance and angle) to the Cartesian components of the head–neck interface axis AP AP AP ðTADAP LS sin a Þ and centre neck axis ðTADLS cos a Þ. (B) Transformation of AP antirotational polar coordinates (distance and angle) to the Cartesian components of the AP AP AP head–neck interface axis ðTADAP AR sin b Þ and centre neck axis ðTADAR cos b Þ. (C) Transformation of axial lag screw polar coordinates (distance and angle) to the Cartesian Ax Ax Ax components of the head–neck interface axis ðTADAx LS sin a Þ and centre neck axis ðTADLS cos a Þ.
Femur neck radiographic measurements The femur neck radiographic measurements revealed that the centre of lag screw positioning as a fraction of the head–neck interface line (D2/D1 in Fig. 2) had an average of 0.39 (SD 0.09). There was a statistically significant difference in the centre of lag
screw position between the three study groups (p value = 0.001). Lower values were associated with secondary displacements and higher values were associated with cutouts (Table 2). We identified the second lower quarter (D2/D1 between 0.25 and 0.5) of the head–neck interface line (L1) as being the ‘‘safe zone’’ for the centre of lag screw. The centre of the lag screw was
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
860
Fig. 4. The femur neck ‘‘safe zone’’ which is the second distal quadrant on the head–neck interface line. The 11 mm medial-lateral mark on the neck central axis is depicted.
not in the ‘‘safe zone’’ in thirty-eight patients, and twelve of them (31.6%) had either a cutout or a secondary reduction loss (Fig. 4). Of the one hundred and eighty-seven patients in whom the centre of lag screw was in the ‘‘safe zone’’, only seven (3.7%) suffered either a cutout or a secondary reduction loss (p value = 0.001).
Femur head radiographic measurements The mean TADs were calculated for the lag screw as well as for the antirotational screw and they were found to be 20.7 (SD 6.7) mm and 53.6 (SD 17.7) mm, respectively (p = NS between groups)
Table 1 Demographic and clinical data. No. failure (207 patients) Age (year) Gender Female Male Side Left Right Osteoporosis grade Nail type Targon PF ATN Nail length Standard Long Reduction Closed Open
75.3 (SD 15.74) 159 (90.9%) 48 (92.3%)
Cutout (15 patients)
Secondary reduction loss (5 patients)
p value
78.2 (SD 14.9)
82.0 (SD 8.2)
0.529
14 (8.0%) 1 (1.9%)
2 (1.1%) 3 (5.3%)
0.046
107 (92.2%) 100 (90.1%) 2.95 (SD 1.42)
8 (6.9%) 7 (6.3%) 3 (SD 1.46)
1 (0.9%) 4 (3.6%) 2.4 (1.14)
0.369 0.712
140 (95.2%) 67 (83.8%)
6 (4.1%) 9 (11.2%)
1 (0.7%) 4 (5.0%)
0.01
163 (92.6%) 44 (86.3%)
10 (5.7%) 5 (9.8%)
3 (1.7%) 2 (3.9%)
0.355
194 (90.7%) 13 (100%)
15 (7.0%) –
5 (2.3%) –
0.514
PF: proximal femur; ATN: antirotation trochanteric nailing system.
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
861
Table 2 Radiographic reduction and femur neck measurements.
Reduction quality AP neck shaft angle (8) Axial ‘‘reduction gap’’ (mm) Axial ‘‘reduction gap’’ above 5 mm Femur neck screw positioning (see Fig. 2) Lag screw centre position (D2/D1) Lag screw centre in the ‘‘safe zone’’ (D2/D1 is 0.25-0.5) Antirotation screw position (D3/D1)
No. failure (207 patients)
Cutout (15 patients)
Secondary reduction loss (5 patients)
p value
132.3 (SD 6.1) 8.2 (SD 6.3) 76/112 (67.9%)
130.7 (SD 6.5) 12.0 (SD 10.5) 5/6 (83.3%)
132.0 (SD 7.3) 8.9 (SD 2.1) 4 (100%)
0.755 0.699 0.073
0.38 (SD 0.08) 180 (87.4%) 0.17 (SD 0.11)
(Table 3). Comparisons between the AP and axial components of the TAD for these two screws showed a statistically significant difference in the AP component alone (Table 3). One hundred and nineteen (76.8%) patients had TAD less than or equal 25 mm, of these 10 patients (8.4%) also had a cutout or secondary reduction loss. Thirty-six patients (23.2%) had TAD higher than 25, of these 4 patients (11.1%) had also a cutout or secondary reduction loss. This difference was not found to be statistically significant (p value = 0.62). The AP positions of the lag screw on the head–neck interface AP axis ðTADAP LS sin a Þ differed between the study groups. Extreme higher and lower values were significantly (p = 0.001) associated with cutouts and secondary loss of reduction, respectively (Table 3). The distance of the lag screw tip on the central neck axis AP ðTADAP LS cos a Þ was greater than 11 mm in 43 (19.1%) patients (Fig. 4). Eight of these patients (18.6% of 43 pts) had mechanical AP failure. The distance on the central neck axis ðTADAP LS cos a Þ was less than 11 mm in one hundred and eighty-two patients (80.9%), ten (5.5% of 182 pts) of whom had mechanical failure (p value = 0.004 for less than 11 mm vs. more than 11 mm). Multivariate analysis Multivariate logistic regression identified the centre of lag screw positioning in the ‘‘safe zone’’ of the head–neck interface as the most important factor in preventing mechanical failure (Table 4). Failure attributed to positioning of the centre of lag screw outside the ‘‘safe zone’’ had an OR of 13.4 (95% CI = 2.24–81, p value = 0.004). Differences in the other variables in the model were not found to be statistically significant (p value > 0.05).
0.52 (SD 0.07) 4 (28.6%) 0.08 (SD 0.04)
0.27 (SD 0.08) 3 (40%) 0.26 (SD 0.14)
0.001 0.001 0.001
Discussion The findings of this study demonstrate that the major factor contributing to treatment success is correct lag screw fixation position. Lag screw position was divided into two perpendicular axes, the head–neck interface axis and the central neck axis. Positioning of the lag screw tip within or higher than 11 mm of the head apex, on the central neck axes, were associated with failure rates of 5.5% and 19.1%, respectively. On the head–neck interface axis, we were able to define a ‘‘safe-zone’’ as being the second quarter (the lower 25–50%) of the head–neck interface line. Lag screws placed in or out of this ‘‘safe zone’’ were associated with failure rates of 3.7% and 31.6%, respectively. Multivariate analysis identified the lag screw position within the head–neck interface ‘‘safe zone’’ as the most important cofactor in preventing mechanical failure. We believe these findings are important for guiding the surgeon in correct positioning of the fixation device. It practically means that for a successful fixation the lag screw guide wire should be placed within the ‘‘safe zone’’ as defined here. We have included the vector analysis performed which led to the definition of the ‘‘safe-zone’’. However, intraoperative complicated calculations are not necessary in order to use the ‘‘safe-zone’’ concept. Our study is the first radiographic analysis of the double screw PFN fixation device. Although our calculations were done on dual screw fixation devices, our results and recommendations are not that different from those known for single screw fixation devices (e.g., DHS and Gamma nail).9,14 As such we recommend to use the ‘‘safe-zone’’ criteria also for assessing the location of single screw fixation devices.
Table 3 Radiographic femur head fixation device position measurements.
Lag screw TADLS (mm) TADAP LS (mm) TADAx LS (mm) On AP head neck interface axis AP TADAP (mm) LS sin a On AP neck centre axis AP AP TADLS cos a (mm) On axial head–neck interface axis Ax TADAx (mm) LS sin a On axial neck centre axis Ax TADAx (mm) LS cos a Antirotation screw TADAR (mm) TADAP AR (mm) TADAx AR (mm) On AP head–neck interface AP TADAP (mm) AR sin b On AP neck centre axis AP TADAP (mm) AR cos b
No. failure (207 patients)
Cutout (15 patients)
Secondary reduction loss (5 patients)
p value
20.3 (SD 6.5) 9.7 (SD 3.2) 10.0 (SD 3.9)
24.0 (SD 6.5) 11.7 (SD 2.6) 11.8 (SD 4.5)
25.5 (SD 12.9) 12.9 (SD 4) 12.2 (SD 7.2)
0.175 0.008 0.470
5.49 (SD 4.4)
0.001
10.7 (SD 4.7)
0.061
2.41 (SD 3.9) 8.6 (SD 3) 0.77 (SD 4.9) 9.3 (SD 3.6) 53.4 (SD 17) 26.9 (SD 7.3) 27.4 (SD 10.2
3.41 (SD 3.5) 10.7 (SD 3.3) 0.31 (SD 3.9)
0.67 (SD 6.4)
0.568
11.3 (SD 4.4)
11.1 (SD 6.9)
0.289
58.0 (SD 30.6) 30.2 (SD 10.3) 27.6 (SD 16.5)
– 25 (SD 11.9) –
0.84 0.68 0.94
7.6 (SD 3)
12.2 (SD 2.5)
4.5 (SD 2.7)
0.001
25.6 (SD 7.6)
27.8 (SD 11.6
24.9 (SD 12.3)
0.92
AP: anteroposterior. Mean (SD) standard deviation are presented.
862
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863
Table 4 Multivariate logistic regression. Covariate
Odds ratio (95% CI)
Lag screw tip distance from the apex, on the central neck axis 1 Less than or equal to 11 mm (baseline) Greater than 11 mm 3.12 (0.25–38.9) Axial ‘‘reduction gap’’ Less than 5 mm (baseline) 1 Greater than 5 mm 9.88 (0.65–150) Centre of lag screw position 1 Within the ‘‘safe zone’’ (baseline) Outside the ‘‘safe zone’’ 13.4 (2.24–81) Gender Female (baseline) 1 Male 1.95 (0.32–11.7) Hardware system used TPF (baseline) 1 5.61 (0.81–38) ATN Tip apex distance (TAD) Less than 25 mm (baseline) 1 More than 25 mm 5.34 (0.19–150) TAD (Lag screw–axial view) 1.15 (0.83–1.61) Osteoporosis 1 Osteoporosis grade 3 (baseline) Osteoporosis grade 2 0.43 (0.07–2.55)
p value
0.375
0.099
0.004
0.464
of at least one or two years, respectively. This is, however, an acceptable follow-up period for a retrospective study of intertrochanteric fractures, with others having used similar time periods in their retrospective studies.4,9,23,24 Prospective studies tend to have longer follow-up periods, i.e., up to one year.3,6,7,17,25 Data on the mobility aids used was available for only about 50% of the patients in this cohort. Another drawback of this study is that it does not take into account several parameters that were shown to influence the outcome of proximal femoral fractures, such as the patients’ body mass index, and the American Society of Anaesthesiologist (ASA) score.15 We believe that this work may provide a guide for surgeons in optimal screw position for reducing the risk of mechanical failure when performing reduction and fixation of intertrochantric fractures.
0.079
Conflict of interest
0.323 0.384
The authors declare that there is no conflict of interests in preparing this manuscript. The authors state that neither they nor any of their immediate family members has received any financial benefit from this study or any commercial company mentioned therein. No donation, due to this work, was made by any commercial company mention in the text to the institute in which the authors serve as surgeons.
0.352
CI: confidence interval; TPF: Targon proximal femur device (Aesculap, Tuttlingen, Germany); ATN: antirotation trochanteric nailing system (ATN) device (dePuy, Warsaw, IN, USA).
This is the first time the head–neck interface line has been defined. It is the line connecting the two transition points of the contour change between femur head convexity to femur neck concavity. We also defined the femur neck central axis as being a perpendicular line that crosses the head–neck interface line in its middle. We believe that the delineation of the neck central axis is more accurate and one that enables a unique definition of the head apex as being the point at which the central neck line crosses the head subchondral bone. Further studies are required in order to validate this point. We believe that these definitions will prove to be useful in other femur orthopaedic surgery fields, such as intracapsular proximal femur fractures and hip arthroplasties. In our study, the TAD as defined by Baumgaertner et al. (1995) did not emerge as being an important factor in terms of influencing either cutouts or secondary reduction loss.9 Only its AP components, when checked separately, was found to predict either cutouts or secondary reduction loss. This finding emphasizes the weakness of the TAD parameter, since it is a scalar measurement that calculates only distance and disregards direction. Our findings and definitions offer surgeons better understanding of the optimal position of lag screws in proximal femur fixation devices. The mechanical failure rate reported herein was 8.8% (twenty of two hundred and twenty-seven patients). This failure rate is comparable to other technical failure rates reported elsewhere.5,7,10,15–18 The Cochrane review by Parker and Handoll (2008)6 includes three randomized clinical trials (Pajarinen et al., 2005; Papasimos et al., 2005; Saudan et al., 2002)19–21 that compared the outcome of the PFN with that of the sliding hip screw. The overall number of patients treated by a PFN was one hundred and ninety-four, of whom five (2.57%) had cutouts of the fixation nail whilst technical failure of the fixation occurred in eleven patients (5.6%). Giraud et al. (2005) compared thirty-four patients treated by the Targon PF device with twenty-six patients treated with sliding hip screw.22 They reported three cutouts (8.8%) in the former and two (7.7%) in the latter. One weakness of our study is that it is retrospective and has a relatively short follow-up period. We included patients whose follow-up was at least six months and who had signs of radiological union. Note that of 207 patients with no mechanical failure, 159 (76.8%) and 98 (47.3%) patients had a follow-up period
Acknowledgment This research was funded by the Talpiot medical leadership programme of the Sheba medical centre, grant number 20963. No external (commercial) funding source was used. References 1. Cummings SR, Rubin SM, Black D. The future of hip fractures in the United States: numbers, costs, and potential effects of postmenopausal estrogen. Clincal Orthopedics 1990;252:163–6. 2. Koval KJ, Aharonoff GB, Rokito AS, Lyon T, Zuckerman JD. Patients with femoral neck and intertrochanteric fractures: are they the same? Clincal Orthopedics 1996;330:166–72. 3. Hardy DR, Descamps PY, Krallis P, Fabeck L, Smets P, Bertens CL, Delince PE. Use of an intramedullary hip-screw compared with a compression hip-screw with a plate for intertrochanteric femoral fractures: a prospective, randomized study of one hundred patients. JBJS (Am) 1998;80:618–30. 4. Mainds CC, Newman RJ. Implant failures in patients with proximal fractures of the femur treated with a sliding screw device. Injury 1989;20:98–109. 5. Osnes, EK, Lofthus, CM, Falch, JA, et al. More postoperative femoral fractures with the Gamma nail than the sliding screw plate in the treatment of trochanteric fractures. Acta Orthopedica Scandinavica, 2001;72:252–256. 6. Parker MJ, Handoll HH. Gamma and other cephalocondylic intramedullary nails versus extramedullary implants for extracapsular hip fractures in adults. Cochrane Database Syst Rev 2008;3:CD93. 7. Schipper IB, Steyerberg EW, Castelein RM, van der Heijden FHWM, den Hoed PT, Kerver AJH, van Vugt AB. Treatment of unstable trochanteric fractures: randomised comparison of the gamma nail and the proximal femural nail. JBJS (Br) 2004;86-B:86–94. 8. Simmermacher RK, Bosch AM, Van der Werken CH. The AO/ASIF- proximal femoral nail (PFN): a new device for the treatment of unstable proximal femur fractures. Injury 1990;30:327–32. 9. Baumgaertner MR, Curtin SL, Lindskog DM, Keggi JM. The value of the tip-apex distance in predicting failure of fixation of peritrochanteric fractures of the hip. JBJS (Am) 1995;77:1058–64. 10. Bridle SH, Patel AD, Bircher M, Calvert PT. Fixation of intertrochanteric fractures of the femur, a randomized prospective comparison of the gamma nail and the dynamic hip screw. JBJS (Br) 1991;73-B:330–4. 11. Helwig P, Faust G, Hindenlang U, Hirschmu A, Konstantinidis L, Bahrs C, Sudkamp N, Schneider R. Finite element analysis of four different implants inserted in different positions to stabilize an idealized trochanteric femoral fracture. Injury 2009;40:288–95. 12. Singh M, Nagrath AR, Maini PS. Changes in trabecular pattern of the upper end of the femur as an index of osteoporosis. JBJS (Am) 1970;52:457–67. 13. Fracture and Dislocation Classification Compendium, 2007. Orthopaedic Trauma Association Classification, Database and Outcomes Committee. Journal of Orthopedic Trauma, 2007;10, Suppl 1:31–42. 14. Davis TRC, Sher JL, Horsman A, Simpson M, Porter BB, Checketts RG. Intertrochanteric femoral fractures–mechanical failure after internal fixation. JBJS (Br) 1990;72-B:26–31.
A. Herman et al. / Injury, Int. J. Care Injured 43 (2012) 856–863 15. Forte ML, Virnig BA, Swiontkowski MF, Bhandari M, Feldman R, Eberly LE, Kane RL. Ninety-day mortality after intertrochanteric hip fracture: provider volume matter? JBJS (Am) 2010;92:799–806. 16. Lindskog DM, Baumgaertner MR. Unstable intertrochanteric hip fractures in the elderly. J Am Acad Orthop Surg 2004;12:189–90. 17. Lundy DW. Subtrochanteric femoral fractures. J Am Acad Orthop Surg 2007;15: 663–71. 18. Rogmark C, Flensburg L, Fredin H. Undisplaced femoral neck fractures—no problems? A consecutive study of 224 patients treated with internal fixation. Injury 2009;40:274–6. 19. Pajarinen J, Lindahl J, Michelsson O, Savolainen V, Hirvensalo E. Pertrochanteric femoral fractures treated with a dynamic hip screw or a proximal femoral nail; a randomised study comparing postoperative rehabilitation. Journal of Bone and Joint Surgery (Br), 2005;87. 20. Papasimos S, Koutsojannis CM, Panagopoulos A, Megas P, Lambiris E. A randomised comparison of AMBI, TGN and PFN for treatment of unstable trochanteric fractures. Archives of Orthopaedic and Trauma Surgery, 2005;125.
863
21. Saudan M, Lubbeke A, Sadowski C, Riand N, Stern R, Hoffmeyer P. Pertrochanteric fractures: is there an advantage to an intramedullary nail? A randomized, prospective study of 206 patients comparing the dynamic hip screw and proximal femoral nail. Journal of Orthopaedic Trauma 2002;16:386–93. 22. Giraud B, Dehoux E, Jovenin N, Madi K, Harisboure A, Usandizaga G, et al. Pertrochanteric fractures: a randomized prospective study comparing dynamic screw plate and intramedullary xation. Revue de Chirurgie Orthopedique et Reparatrice de l’Appareil Moteur 2005;91:732–6. 23. Laros CS, Moore JF. Complications of fixation in intertrochanteric fractures. Clincal Orthopedics 1974;11:0–119. 24. Larsson S, Friberg S, Hansson Ll. Trochanteric fractures Influence of reduction and implant position on impaction and complications. Clincal Orthopedics 1990:130–9. 25. Leung KS, So WS, Shen WY, Hui PW. Gamma nails and dynamic hip screws for peritrochanteric fractures: a randomized prospective study in elderly patients. JBJS (Br) 1992;74-B:345–51.