- Email: [email protected]
Measuring tip–apex distance using a picture archiving and communication system (PACS)
Injury, Int. J. Care Injured (2008) 39, 786—790
www.elsevier.com/locate/injury
Measuring tip—apex distance using a picture archiving and communication system (PACS) Luke J. Johnson *, Marcus R. Cope, Shahram Shahrokhi, Peter Tamblyn Department of Orthopaedic Surgery, Flinders Medical Centre, Bedford Park, South Australia 5042, Australia Accepted 18 December 2007
KEYWORDS Tip—apex distance; Hip fracture; PACS; Digital radiography; Sliding screw prosthesis
Summary Background: The sliding hip screw is currently the most frequently used prosthesis used to fix trochanteric fractures of the hip. The ‘tip—apex distance’ (TAD) has been found to be predictive of hardware failure, with a larger TAD being associated with an increasing risk of ‘cut-out’. Previous studies have either used ‘hard-copy’ radiographs and geometrical aids or a mixture of scanned hard-copy images and extra software to measure TAD. The current study describes a new method of tip—apex distance estimation using an entirely digital picture archiving and communication system (PACS). Materials and methods: Ten radiographs were measured for TAD by four surgeons of differing experience using the described measurement protocol, at two different time points (4 weeks apart). The results were then subjected to two-tailed t-tests to determine if they differed significantly. Results: No tests attained significance (i.e. no statistical difference existed between the observers’ measurements and no difference existed over time from a single observer’s results). Conclusion: This study shows that TAD can be easily, accurately and, importantly, reproducibly measured using an entirely digitally based image capture and archiving system. The ease with which the measurements and calculations can be made will facilitate orthopaedic practitioners and trauma units in their auditing activities, and allows for quick TAD estimation in the ‘X-ray meeting’ environment. # 2007 Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: +61 408 505 765. E-mail address: [email protected] (L.J. Johnson). 0020–1383/$ — see front matter # 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2007.12.019
Measuring tip-apex distance using PACS
Introduction The sliding hip screw is currently the most frequently used prosthesis used to fix trochanteric fractures of the hip.9 This is due to its unique design features, common to both the intra- and extramedullary devices that allows for controlled impaction of the fracture into a stable position whilst maintaining a constant neck-shaft angle.1,6 The most common mode of prosthesis failure is so-called ‘cut-out’ of the lag screw from the femoral head.1—9 This is when the neck-shaft angle collapses into varus and the lag screw then extrudes superiorly through the femoral head.1,6 Baumgaertner et al.1 first introduced the concept of the ‘tip—apex distance’ (TAD) as being the strongest factor in predicting this mode of failure. The TAD as defined by Baumgaertner is the sum of the distance, in millimetres, on the anteroposterior (AP) and lateral radiographs from the tip of the lag screw to the apex of the femoral head,1,2,10 corrected for magnification. The apex of the femoral head was defined as the point of intersection between the subchondral bone and a line parallel to and in the middle of the femoral neck.1,2,10 Previous studies have found that a larger TAD is associated with an increased risk of hardware failure.1,2,10 It is accepted that a TAD value of greater than 25 mm is associated with increased rates of cut-out,1,2,6,9—11 and some studies have suggested that a value less than 20 mm should be the ideal.9 Previous studies have either used ‘hard-copy’ radiographs and geometrical aids, or more recently the radiographs have been digitised and then further software packages have been used to make the measurements.3 More recently picture archiving and communication systems (or PACS, incorporating Centricity Enterprise Web v2.1 (GE Healthcare Pty Ltd.)) have become increasingly widespread and are the standard in many centres in regards to hospital capture of radiological and other images, digital development and digital internal or intranet distribution. The standard software has many functions that allow for angle measurement, distance measurement and image enhancement. This technological advance thus enables previously laborious manual measurement protocols to be more easily and expediently executed, avoiding the use of hardcopy films by simply employing mouse clicks. This facilitates orthopaedic practitioners and units in their auditing activities and allows for quick TAD estimation in the ‘X-ray meeting’ environment. It has become routine for postoperative review in this unit. The current study aims to describe a new method of tip—apex distance estimation using new digital
787 technology and also testing the method’s accuracy and reproducibility so that the technique may be used in future studies by employing TAD.
Materials and methods Ten consecutive patients who had sustained an intertrochanteric hip fracture and undergone fixation with a sliding screw implant (intra-medullary hip screw [IMHS], Smith & Nephew, Memphis, TN, USA) were selected. Only patients who sustained osteoporotic fractures uncomplicated by malignancy or other pathological processes were included. The mean age was 86.9 years (range 76—97 years). The male to female ratio was 1:1. The digital intra-operative fluoroscopic images were chosen for the determination of tip—apex distance. Four surgeons of varying grade and experience (one resident medical officer, one registrar, one visiting fellow and one senior consultant) made measurements on two separate occasions using the digital system and the following measurement protocol as described below. The two sets of measurements were made 4 weeks apart, with no practice of the technique between the two time points. Prior to the first measurements a brief 10-min training session was given to demonstrate the use of the measurement protocol. The standard commercially licensed Centricity Enterprise Web v2.1 software package, using all default settings was employed, with no other image viewing or measurement software used.
Measurement protocol using PACS Both the AP and lateral images underwent determination of tip—apex distance. The lag screw width is initially determined using the linear measuring tool, found in the toolbar above the image. This measurement yields a number in ‘units’, which can be compared to the actual width of the screw (13 mm with the IMHS) to give the magnification of the image. Subsequent distances determined on the same image can henceforth be corrected for magnification accordingly. This was very simply performed using a matrix table constructed in a mathematical spreadsheet programme (EXCEL 2003, Microsoft Corporation Pty Ltd.). The apex of femoral head was determined using the guidelines originally set down by Baumgaertner et al.1,2,10 Using the ‘ellipse region of interest’ tool, found in the toolbar above the image, an ellipse can be placed and stretched to exactly match the line of subchondral bone of the femoral head. Once this is in place, a line is drawn between the points where
788 the ellipse meets the neck of the femur. Using the angle measuring tool, a line perpendicular to this line arising from its mid-point is taken through the femoral head. Where this perpendicular line bisects the ellipse defines the apex of the femoral head. Once this is constructed, it is a simple matter of using the linear measuring tool to determine the number of units between the tip of the lag screw and the apex of the femoral head (see Fig. 1). Once again this number (in units) is fed into the database matrix and the corrected millimetre measurement is determined. The same methodology for determining the lag screw width, the position of the femoral head apex and the TAD are employed for the remaining digital radiographic image (see Fig. 2). The data gathered were again fed into the database matrix and the actual magnification-corrected millimetre distances determined.
Statistical analysis The original results of the senior consultant, who developed the measurement protocol based on Baumgaertner’s original methodology, were recorded and all other measurements compared with this dataset. All data were subsequently reviewed using (EXCEL 2003) for further statistical analysis. Two-tailed t-tests were employed to determine if any measurements differed significantly from the gold standard (inter-observer comparison), and the same test was used to determine any significant difference between the data from two time points from the same individual (intra-observer comparison). The null hypothesis in each case was
L.J. Johnson et al.
Figure 2 Lateral fluoroscopic view of hip and sliding screw prosthesis in situ, displaying the measurements as described in the protocol.
that a difference would exist between the measurements. A confidence interval of 95% was utilised.
Results No statistically significant difference was attained for any measurement ( p 0.05) (Table 1). Further analysis of each individual’s measurements at the two different time points also revealed no statistical differences between the two sets of measurements ( p 0.05) (Table 2).
Table 1 Comparison of all observer measurements to the original measurements Measurement comparisons of original measurements (OM) and observers (OBS)
Two-tailed t-test ( p-values)
OM—OBS1.1
0.35 0.10 0.09 0.06 0.17 0.16
OM—OBS1.2 OM—OBS2.1 OM—OBS2.2 OM—OBS3.1 OM—OBS3.2
Table 2 Intra-observer comparison between two time points
Figure 1 Anteroposterior fluoroscopic view of hip and sliding screw prosthesis in situ, displaying the measurements as described in the protocol.
Comparison of time points for each observer (intra-observer differences)
Two-tailed t-test ( p-values)
OBS1.1—OBS1.2
0.30 0.97 0.89
OBS2.1—OBS2.2 OBS3.1—OBS3.2
Measuring tip-apex distance using PACS
789
As such the null hypothesis that a difference existed between the measurements was rejected in each case. Thus there is no statistical difference between the original measurements and those measurements undertaken by other individuals. Furthermore there is no statistical difference between any individual’s TAD measurements of the same radiographs at two different time points.
images are digitally transferred from the image intensifier to the PACS system is thus essential. Secondly, the brief training session in the measurement protocol is also useful and ensures that the anatomy is correctly appreciated on radiograph. This has been found to be particularly useful for more junior members of the surgical team in the authors’ experience.
Discussion
Conclusion
As far as we the authors are aware this is the first paper to describe and test for accuracy and reproducibility a method of TAD estimation using an entirely digital system of image capture and measurement. All previous studies have used hard-copy film, or a combination of hard-copy film and secondary digital scanning of the former. As digital picture archiving and communication systems become more prevalent in the hospital setting and surgical practices, the traditional means by which research and evaluation of surgical interventions is undertaken using radiological images will also have to evolve. This paper offers a technique for TAD estimation using soft images only, thus enabling the technique to be used in future research. The paper demonstrates that surgeons of differing experience and levels of training can perform these measurements accurately after a brief training session. The data also shows that once the technique has been learned the ability to accurately reproduce the measurements is retained, even after 4 weeks without practice of the technique. As such, the measurement protocol described in this paper represents an accurate and reproducible method of TAD estimation that is employable by any member of the surgical team, regardless of their level of training and experience. The level of accuracy attained is likely to be higher than previously described methods as many errors of parallax have been eliminated through digitally accurate measurements are attainable through enhancement and magnification of the images. During this study it became clear that several factors are involved in accurate measurement of TAD using this technique. First, image quality is paramount. Low quality images (particularly from operative fluoroscopy) can be very pixelated and can significantly alter where one believes the subchondral bone of the femoral head lies, thus creating potential errors in apex estimation. Thus the authors suggest a relatively high-resolution capture/upload of images to minimise the occurrence of errors of parallax. Ensuring that the captured
The measurement protocol described in this paper establishes a new method of TAD estimation for future use as a research and audit tool in those institutions using PACS. The ease with which the measurements and calculations can be made will facilitate orthopaedic practitioners and units in their auditing activities and allows for quick TAD estimation in the ‘X-ray meeting’ environment. It is shown to allow TAD to be easily, accurately and reproducibly measured using an entirely digital image capture and archiving system. When using this protocol, the authors suggest that high-resolution image capture is used to ensure the most accurate results.
References 1. Baumgaertner MR, Curtin SL, Lindskog DM, Keggi JM. The value of the tip—apex distance in predicting failure of fixation of perithrochanteric fractures of the hip. J Bone Joint Surg Am 1995;77:1058—64. 2. Baumgaertner MR, Solberg BD. Awareness of tip—apex distance reduces failure of fixation of trochanteric fractures of the hip. J Bone Joint Surg Am 1997;79(6 (November)): 969—71. 3. Kumar AJS, Parmar VN, Bankart J, et al. Comparison of accuracy of lag screw placement in cephalocondylic nails and sliding hip screw plate fixation for extracapsular fractures of the neck of femur. Int Orthop 2006;30: 320—4. 4. Kumar AJS, Parmar VN, Kolpattil S, et al. Significance of hip rotation on measurement of ‘tip apex distance’ during fixation of extracapsular proximal femoral fractures. Injury 2007;38:792—6. 5. Lindskog DM, Baumgaertner MR. Unstable intertrochanteric hip fractures in the elderly. J Am Acad Orthop Surg 2004;12: 179—90. 6. Lorich DG, Geller DS, Nielson JH. Osteoporotic pertrochanteric hip fractures. Management and current controversies. J Bone Joint Surg Am 2004;86:398—410. 7. Parker MJ. Cutting-out of the dynamic hip screw related to its position. J Bone Joint Surg Br 1992;74(4):625. 8. Parmar V, Kumar S, Aster A, Harper W. Review of methods to quantify lag screw placement in hip fracture fixation. Acta Orthop Belg 2005;71(3):260—3. Jun. 9. Pervez H, Parker MJ, Vowler S. Prediction of fixation failure after sliding hip screw fixation. Injury 2004;35: 994—8.
790 10. Stapley SA, Kumar BA, Parker MJ. The prediction of fixation failure of intertrochanteric fractures of the femur using the sliding hip screw: which is the best method? J Bone Joint Surg Br 2000;82-B(Suppl. 1):56.
L.J. Johnson et al. 11. Walton NP, Wynn-Jones H, et al. Femoral neck-shaft angle in extra-capsular proximal femoral fracture fixation; does it make a TAD of difference? Injury 2005;36: 1362—4.
www.elsevier.com/locate/injury
Measuring tip—apex distance using a picture archiving and communication system (PACS) Luke J. Johnson *, Marcus R. Cope, Shahram Shahrokhi, Peter Tamblyn Department of Orthopaedic Surgery, Flinders Medical Centre, Bedford Park, South Australia 5042, Australia Accepted 18 December 2007
KEYWORDS Tip—apex distance; Hip fracture; PACS; Digital radiography; Sliding screw prosthesis
Summary Background: The sliding hip screw is currently the most frequently used prosthesis used to fix trochanteric fractures of the hip. The ‘tip—apex distance’ (TAD) has been found to be predictive of hardware failure, with a larger TAD being associated with an increasing risk of ‘cut-out’. Previous studies have either used ‘hard-copy’ radiographs and geometrical aids or a mixture of scanned hard-copy images and extra software to measure TAD. The current study describes a new method of tip—apex distance estimation using an entirely digital picture archiving and communication system (PACS). Materials and methods: Ten radiographs were measured for TAD by four surgeons of differing experience using the described measurement protocol, at two different time points (4 weeks apart). The results were then subjected to two-tailed t-tests to determine if they differed significantly. Results: No tests attained significance (i.e. no statistical difference existed between the observers’ measurements and no difference existed over time from a single observer’s results). Conclusion: This study shows that TAD can be easily, accurately and, importantly, reproducibly measured using an entirely digitally based image capture and archiving system. The ease with which the measurements and calculations can be made will facilitate orthopaedic practitioners and trauma units in their auditing activities, and allows for quick TAD estimation in the ‘X-ray meeting’ environment. # 2007 Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: +61 408 505 765. E-mail address: [email protected] (L.J. Johnson). 0020–1383/$ — see front matter # 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2007.12.019
Measuring tip-apex distance using PACS
Introduction The sliding hip screw is currently the most frequently used prosthesis used to fix trochanteric fractures of the hip.9 This is due to its unique design features, common to both the intra- and extramedullary devices that allows for controlled impaction of the fracture into a stable position whilst maintaining a constant neck-shaft angle.1,6 The most common mode of prosthesis failure is so-called ‘cut-out’ of the lag screw from the femoral head.1—9 This is when the neck-shaft angle collapses into varus and the lag screw then extrudes superiorly through the femoral head.1,6 Baumgaertner et al.1 first introduced the concept of the ‘tip—apex distance’ (TAD) as being the strongest factor in predicting this mode of failure. The TAD as defined by Baumgaertner is the sum of the distance, in millimetres, on the anteroposterior (AP) and lateral radiographs from the tip of the lag screw to the apex of the femoral head,1,2,10 corrected for magnification. The apex of the femoral head was defined as the point of intersection between the subchondral bone and a line parallel to and in the middle of the femoral neck.1,2,10 Previous studies have found that a larger TAD is associated with an increased risk of hardware failure.1,2,10 It is accepted that a TAD value of greater than 25 mm is associated with increased rates of cut-out,1,2,6,9—11 and some studies have suggested that a value less than 20 mm should be the ideal.9 Previous studies have either used ‘hard-copy’ radiographs and geometrical aids, or more recently the radiographs have been digitised and then further software packages have been used to make the measurements.3 More recently picture archiving and communication systems (or PACS, incorporating Centricity Enterprise Web v2.1 (GE Healthcare Pty Ltd.)) have become increasingly widespread and are the standard in many centres in regards to hospital capture of radiological and other images, digital development and digital internal or intranet distribution. The standard software has many functions that allow for angle measurement, distance measurement and image enhancement. This technological advance thus enables previously laborious manual measurement protocols to be more easily and expediently executed, avoiding the use of hardcopy films by simply employing mouse clicks. This facilitates orthopaedic practitioners and units in their auditing activities and allows for quick TAD estimation in the ‘X-ray meeting’ environment. It has become routine for postoperative review in this unit. The current study aims to describe a new method of tip—apex distance estimation using new digital
787 technology and also testing the method’s accuracy and reproducibility so that the technique may be used in future studies by employing TAD.
Materials and methods Ten consecutive patients who had sustained an intertrochanteric hip fracture and undergone fixation with a sliding screw implant (intra-medullary hip screw [IMHS], Smith & Nephew, Memphis, TN, USA) were selected. Only patients who sustained osteoporotic fractures uncomplicated by malignancy or other pathological processes were included. The mean age was 86.9 years (range 76—97 years). The male to female ratio was 1:1. The digital intra-operative fluoroscopic images were chosen for the determination of tip—apex distance. Four surgeons of varying grade and experience (one resident medical officer, one registrar, one visiting fellow and one senior consultant) made measurements on two separate occasions using the digital system and the following measurement protocol as described below. The two sets of measurements were made 4 weeks apart, with no practice of the technique between the two time points. Prior to the first measurements a brief 10-min training session was given to demonstrate the use of the measurement protocol. The standard commercially licensed Centricity Enterprise Web v2.1 software package, using all default settings was employed, with no other image viewing or measurement software used.
Measurement protocol using PACS Both the AP and lateral images underwent determination of tip—apex distance. The lag screw width is initially determined using the linear measuring tool, found in the toolbar above the image. This measurement yields a number in ‘units’, which can be compared to the actual width of the screw (13 mm with the IMHS) to give the magnification of the image. Subsequent distances determined on the same image can henceforth be corrected for magnification accordingly. This was very simply performed using a matrix table constructed in a mathematical spreadsheet programme (EXCEL 2003, Microsoft Corporation Pty Ltd.). The apex of femoral head was determined using the guidelines originally set down by Baumgaertner et al.1,2,10 Using the ‘ellipse region of interest’ tool, found in the toolbar above the image, an ellipse can be placed and stretched to exactly match the line of subchondral bone of the femoral head. Once this is in place, a line is drawn between the points where
788 the ellipse meets the neck of the femur. Using the angle measuring tool, a line perpendicular to this line arising from its mid-point is taken through the femoral head. Where this perpendicular line bisects the ellipse defines the apex of the femoral head. Once this is constructed, it is a simple matter of using the linear measuring tool to determine the number of units between the tip of the lag screw and the apex of the femoral head (see Fig. 1). Once again this number (in units) is fed into the database matrix and the corrected millimetre measurement is determined. The same methodology for determining the lag screw width, the position of the femoral head apex and the TAD are employed for the remaining digital radiographic image (see Fig. 2). The data gathered were again fed into the database matrix and the actual magnification-corrected millimetre distances determined.
Statistical analysis The original results of the senior consultant, who developed the measurement protocol based on Baumgaertner’s original methodology, were recorded and all other measurements compared with this dataset. All data were subsequently reviewed using (EXCEL 2003) for further statistical analysis. Two-tailed t-tests were employed to determine if any measurements differed significantly from the gold standard (inter-observer comparison), and the same test was used to determine any significant difference between the data from two time points from the same individual (intra-observer comparison). The null hypothesis in each case was
L.J. Johnson et al.
Figure 2 Lateral fluoroscopic view of hip and sliding screw prosthesis in situ, displaying the measurements as described in the protocol.
that a difference would exist between the measurements. A confidence interval of 95% was utilised.
Results No statistically significant difference was attained for any measurement ( p 0.05) (Table 1). Further analysis of each individual’s measurements at the two different time points also revealed no statistical differences between the two sets of measurements ( p 0.05) (Table 2).
Table 1 Comparison of all observer measurements to the original measurements Measurement comparisons of original measurements (OM) and observers (OBS)
Two-tailed t-test ( p-values)
OM—OBS1.1
0.35 0.10 0.09 0.06 0.17 0.16
OM—OBS1.2 OM—OBS2.1 OM—OBS2.2 OM—OBS3.1 OM—OBS3.2
Table 2 Intra-observer comparison between two time points
Figure 1 Anteroposterior fluoroscopic view of hip and sliding screw prosthesis in situ, displaying the measurements as described in the protocol.
Comparison of time points for each observer (intra-observer differences)
Two-tailed t-test ( p-values)
OBS1.1—OBS1.2
0.30 0.97 0.89
OBS2.1—OBS2.2 OBS3.1—OBS3.2
Measuring tip-apex distance using PACS
789
As such the null hypothesis that a difference existed between the measurements was rejected in each case. Thus there is no statistical difference between the original measurements and those measurements undertaken by other individuals. Furthermore there is no statistical difference between any individual’s TAD measurements of the same radiographs at two different time points.
images are digitally transferred from the image intensifier to the PACS system is thus essential. Secondly, the brief training session in the measurement protocol is also useful and ensures that the anatomy is correctly appreciated on radiograph. This has been found to be particularly useful for more junior members of the surgical team in the authors’ experience.
Discussion
Conclusion
As far as we the authors are aware this is the first paper to describe and test for accuracy and reproducibility a method of TAD estimation using an entirely digital system of image capture and measurement. All previous studies have used hard-copy film, or a combination of hard-copy film and secondary digital scanning of the former. As digital picture archiving and communication systems become more prevalent in the hospital setting and surgical practices, the traditional means by which research and evaluation of surgical interventions is undertaken using radiological images will also have to evolve. This paper offers a technique for TAD estimation using soft images only, thus enabling the technique to be used in future research. The paper demonstrates that surgeons of differing experience and levels of training can perform these measurements accurately after a brief training session. The data also shows that once the technique has been learned the ability to accurately reproduce the measurements is retained, even after 4 weeks without practice of the technique. As such, the measurement protocol described in this paper represents an accurate and reproducible method of TAD estimation that is employable by any member of the surgical team, regardless of their level of training and experience. The level of accuracy attained is likely to be higher than previously described methods as many errors of parallax have been eliminated through digitally accurate measurements are attainable through enhancement and magnification of the images. During this study it became clear that several factors are involved in accurate measurement of TAD using this technique. First, image quality is paramount. Low quality images (particularly from operative fluoroscopy) can be very pixelated and can significantly alter where one believes the subchondral bone of the femoral head lies, thus creating potential errors in apex estimation. Thus the authors suggest a relatively high-resolution capture/upload of images to minimise the occurrence of errors of parallax. Ensuring that the captured
The measurement protocol described in this paper establishes a new method of TAD estimation for future use as a research and audit tool in those institutions using PACS. The ease with which the measurements and calculations can be made will facilitate orthopaedic practitioners and units in their auditing activities and allows for quick TAD estimation in the ‘X-ray meeting’ environment. It is shown to allow TAD to be easily, accurately and reproducibly measured using an entirely digital image capture and archiving system. When using this protocol, the authors suggest that high-resolution image capture is used to ensure the most accurate results.
References 1. Baumgaertner MR, Curtin SL, Lindskog DM, Keggi JM. The value of the tip—apex distance in predicting failure of fixation of perithrochanteric fractures of the hip. J Bone Joint Surg Am 1995;77:1058—64. 2. Baumgaertner MR, Solberg BD. Awareness of tip—apex distance reduces failure of fixation of trochanteric fractures of the hip. J Bone Joint Surg Am 1997;79(6 (November)): 969—71. 3. Kumar AJS, Parmar VN, Bankart J, et al. Comparison of accuracy of lag screw placement in cephalocondylic nails and sliding hip screw plate fixation for extracapsular fractures of the neck of femur. Int Orthop 2006;30: 320—4. 4. Kumar AJS, Parmar VN, Kolpattil S, et al. Significance of hip rotation on measurement of ‘tip apex distance’ during fixation of extracapsular proximal femoral fractures. Injury 2007;38:792—6. 5. Lindskog DM, Baumgaertner MR. Unstable intertrochanteric hip fractures in the elderly. J Am Acad Orthop Surg 2004;12: 179—90. 6. Lorich DG, Geller DS, Nielson JH. Osteoporotic pertrochanteric hip fractures. Management and current controversies. J Bone Joint Surg Am 2004;86:398—410. 7. Parker MJ. Cutting-out of the dynamic hip screw related to its position. J Bone Joint Surg Br 1992;74(4):625. 8. Parmar V, Kumar S, Aster A, Harper W. Review of methods to quantify lag screw placement in hip fracture fixation. Acta Orthop Belg 2005;71(3):260—3. Jun. 9. Pervez H, Parker MJ, Vowler S. Prediction of fixation failure after sliding hip screw fixation. Injury 2004;35: 994—8.
790 10. Stapley SA, Kumar BA, Parker MJ. The prediction of fixation failure of intertrochanteric fractures of the femur using the sliding hip screw: which is the best method? J Bone Joint Surg Br 2000;82-B(Suppl. 1):56.
L.J. Johnson et al. 11. Walton NP, Wynn-Jones H, et al. Femoral neck-shaft angle in extra-capsular proximal femoral fracture fixation; does it make a TAD of difference? Injury 2005;36: 1362—4.