Wael Dandachli
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Wael Dandachli has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Farhad Iranpour
Imperial College, Charing Cross Hospital
5 shared publications - Justin Peter Cobb
MSK Lab, Imperial College London, London, UK
4 shared publications - Margaret Anne Hall-Craggs
Centre for Medical Imaging University College London London United Kingdom
4 shared publications - Alister James Hart
Royal National Orthopaedic Hospital NHS Trust, Stanmore, UK
3 shared publications J. Witt
University College London Hospitals, 235 Euston Road, London NW1 2BU, UK
2 shared publicationsJustin Cobb
Department of Orthopaedic Surgery, Imperial College London, Charing Cross Hospital, London, UK
2 shared publications
Latest Publications
- 2 March 2014
- Published by SAGE Publications in HIP International
- Vol. 24 (3) , 302-305
A Pain in the Backside: A Case Report of Coxa Saltans Occurring at the Proximal Hamstring Origin
Coxa saltans occurring at the proximal hamstring origin has been rarely reported in the literature. It is better known as occurring at the iliotibial band or the iliopsoas tendon. We report a case of coxa saltans due to subluxation of the origin of the long head of biceps femoris tendon at the ischial tuberosity. This was successfully treated using a mini-open surgical technique. Six weeks postoperatively, the snapping and the associated pain were abolished and the patient was able to resume their participation in athletic activities.
- 2 March 2014
- Published by SAGE Publications in HIP International
- Vol. 24 (3) , 302-305
A Pain in the Backside: A Case Report of Coxa Saltans Occurring at the Proximal Hamstring Origin
Coxa saltans occurring at the proximal hamstring origin has been rarely reported in the literature. It is better known as occurring at the iliotibial band or the iliopsoas tendon. We report a case of coxa saltans due to subluxation of the origin of the long head of biceps femoris tendon at the ischial tuberosity. This was successfully treated using a mini-open surgical technique. Six weeks postoperatively, the snapping and the associated pain were abolished and the patient was able to resume their participation in athletic activities.
- 1 January 2013
- Published by SAGE Publications in HIP International
- Vol. 23 (1) , 87-92
The Influence of Pelvic Tilt on Acetabular Orientation and Cover: A Three-Dimensional Computerised Tomography Analysis
The orientation of the pelvis influences that of the acetabulum. In particular, pelvic tilt in the sagittal plane may lead to inaccurate interpretation of plain pelvic radiographs. We therefore quantified the relationship between this pelvic tilt and acetabular orientation in native hips, and determined whether pelvic tilt affects femoral head cover. We analysed computed tomography scans of 93 hips (36 normal, 31 dysplastic, 26 with acetabular retroversion) and measured acetabular anteversion, inclination, and femoral head cover at pelvic tilt angles ranging from −20° to 20° in relation to the anterior pelvic plane using 5° increments. Pelvic tilt changed acetabular version with a decrease in anteversion ranging from 2.5° to 5° for every 5° of forward tilt. The effect on inclination was less marked and varied. In normal and dysplastic hips pelvic tilt increased apparent femoral head cover. A greater understanding of the influence of pelvic tilt may allow improvements in the radiological diagnosis and surgical treatment of acetabular abnormalities, particularly in relation to acetabular reorientation procedures and femoroacetabular impingement.
- 1 January 2013
- Published by SAGE Publications in HIP International
- Vol. 23 (1) , 87-92
The Influence of Pelvic Tilt on Acetabular Orientation and Cover: A Three-Dimensional Computerised Tomography Analysis
The orientation of the pelvis influences that of the acetabulum. In particular, pelvic tilt in the sagittal plane may lead to inaccurate interpretation of plain pelvic radiographs. We therefore quantified the relationship between this pelvic tilt and acetabular orientation in native hips, and determined whether pelvic tilt affects femoral head cover. We analysed computed tomography scans of 93 hips (36 normal, 31 dysplastic, 26 with acetabular retroversion) and measured acetabular anteversion, inclination, and femoral head cover at pelvic tilt angles ranging from −20° to 20° in relation to the anterior pelvic plane using 5° increments. Pelvic tilt changed acetabular version with a decrease in anteversion ranging from 2.5° to 5° for every 5° of forward tilt. The effect on inclination was less marked and varied. In normal and dysplastic hips pelvic tilt increased apparent femoral head cover. A greater understanding of the influence of pelvic tilt may allow improvements in the radiological diagnosis and surgical treatment of acetabular abnormalities, particularly in relation to acetabular reorientation procedures and femoroacetabular impingement.
- 17 March 2012
- Published by Springer Nature in Skeletal Radiology
- Vol. 41 (10) , 1295-1300
Quantifying the contribution of pincer deformity to femoro-acetabular impingement using 3D computerised tomography
Objective To provide a simple, reliable method for the three-dimensional quantification of pincer-type hip deformity. Materials and methods Computerised tomography scans of 16 normal female hips and 15 female hips with clinical femoro-acetabular impingement (FAI) and radiographic signs of pincer secondary to acetabular protrusio were analysed. After orientating the pelvis in the anterior pelvic plane, the acetabular centre was determined, and the ratios of its coordinates to the corresponding pelvic dimensions were calculated. Acetabular coverage of the femoral head and centre-edge angles were also measured for the two groups. Results In hips with a pincer, the hip was medialised by 37 % (p = 0.03), more proximal by 5 % (p = 0.05) and more posterior by 9 % (p = 0.03) compared with the normal hips. Coverage of the femoral head in protrusio hips was significantly greater than normal (average 71 % vs 82 %, p = 0.0001). Both the lateral centre-edge angle and the combined anterior–posterior centre-edge angle were greater in protrusio hips than in the normal ones (48° vs 37 °, p < 0.001; and 216° vs 176°, p < 0.0001 respectively). Conclusion Displacement in acetabular protrusio occurs in all planes. This CT-based method allows for the accurate and standardised quantification of the extent of displacement, as well as 3D measurement of femoral head coverage. In the adult female population, a combined centre-edge angle of over 190° suggests an acetabulum that is too deep and a potential cause of symptoms of femoro-acetabular impingement. Conversely, an acetabulum that has a combined centre-edge angle of less than 190° may be considered to be of normal depth, and therefore not contributing a pincer to FAI should it occur.
- 17 March 2012
- Published by Springer Nature in Skeletal Radiology
- Vol. 41 (10) , 1295-1300
Quantifying the contribution of pincer deformity to femoro-acetabular impingement using 3D computerised tomography
Objective To provide a simple, reliable method for the three-dimensional quantification of pincer-type hip deformity. Materials and methods Computerised tomography scans of 16 normal female hips and 15 female hips with clinical femoro-acetabular impingement (FAI) and radiographic signs of pincer secondary to acetabular protrusio were analysed. After orientating the pelvis in the anterior pelvic plane, the acetabular centre was determined, and the ratios of its coordinates to the corresponding pelvic dimensions were calculated. Acetabular coverage of the femoral head and centre-edge angles were also measured for the two groups. Results In hips with a pincer, the hip was medialised by 37 % (p = 0.03), more proximal by 5 % (p = 0.05) and more posterior by 9 % (p = 0.03) compared with the normal hips. Coverage of the femoral head in protrusio hips was significantly greater than normal (average 71 % vs 82 %, p = 0.0001). Both the lateral centre-edge angle and the combined anterior–posterior centre-edge angle were greater in protrusio hips than in the normal ones (48° vs 37 °, p < 0.001; and 216° vs 176°, p < 0.0001 respectively). Conclusion Displacement in acetabular protrusio occurs in all planes. This CT-based method allows for the accurate and standardised quantification of the extent of displacement, as well as 3D measurement of femoral head coverage. In the adult female population, a combined centre-edge angle of over 190° suggests an acetabulum that is too deep and a potential cause of symptoms of femoro-acetabular impingement. Conversely, an acetabulum that has a combined centre-edge angle of less than 190° may be considered to be of normal depth, and therefore not contributing a pincer to FAI should it occur.
- 22 December 2010
- Published by Springer Nature in Skeletal Radiology
- Vol. 40 (7) , 877-883
Analysis of acetabular version in the native hip: comparison between 2D axial CT and 3D CT measurements
To compare two-dimensional (2D) axial with three-dimensional (3D) computerized tomography (CT) measurements of acetabular version in native hips. CT scans of 34 hips in 17 consecutive patients being investigated for femoroacetabular impingement were analyzed. Acetabular version was measured using 2D CT at two different axial levels, one cranial (slice 2) and the other at the equator (slice 3). The measurements were repeated after correction for pelvic tilt. The results were compared to the measurements of anatomical version obtained using a 3D CT method that automatically corrects for pelvic tilt. The mean acetabular version using the 3D CT method was 15.7° (SD 6.9°). The mean version using slice 2 was 9.3° (SD 6.5°) before correction for pelvic tilt and 15.7° (SD 8.0°) after the correction. The mean version using slice 3 was 16.4° (SD 4.2°) before tilt correction and 19.0° (SD 5.0°) after the correction. In relation to the 3D method, the intraclass correlation coefficient (ICC) was 0.58 for the uncorrected and 0.93 for the corrected slice 2 method. For the uncorrected and corrected slice 3 methods, the ICC was 0.64 and 0.89, respectively. The 2D axial methods produced variable results. The results that correlated best with the 3D method were those of the cranial slice (slice 2) after correction for pelvic tilt. Interpretation of 2D axial CT measurements of acetabular version should be done with caution. The level at which the measurement is done and the presence of pelvic tilt appear to be significant factors.
- 22 December 2010
- Published by Springer Nature in Skeletal Radiology
- Vol. 40 (7) , 877-883
Analysis of acetabular version in the native hip: comparison between 2D axial CT and 3D CT measurements
To compare two-dimensional (2D) axial with three-dimensional (3D) computerized tomography (CT) measurements of acetabular version in native hips. CT scans of 34 hips in 17 consecutive patients being investigated for femoroacetabular impingement were analyzed. Acetabular version was measured using 2D CT at two different axial levels, one cranial (slice 2) and the other at the equator (slice 3). The measurements were repeated after correction for pelvic tilt. The results were compared to the measurements of anatomical version obtained using a 3D CT method that automatically corrects for pelvic tilt. The mean acetabular version using the 3D CT method was 15.7° (SD 6.9°). The mean version using slice 2 was 9.3° (SD 6.5°) before correction for pelvic tilt and 15.7° (SD 8.0°) after the correction. The mean version using slice 3 was 16.4° (SD 4.2°) before tilt correction and 19.0° (SD 5.0°) after the correction. In relation to the 3D method, the intraclass correlation coefficient (ICC) was 0.58 for the uncorrected and 0.93 for the corrected slice 2 method. For the uncorrected and corrected slice 3 methods, the ICC was 0.64 and 0.89, respectively. The 2D axial methods produced variable results. The results that correlated best with the 3D method were those of the cranial slice (slice 2) after correction for pelvic tilt. Interpretation of 2D axial CT measurements of acetabular version should be done with caution. The level at which the measurement is done and the presence of pelvic tilt appear to be significant factors.
- 1 March 2010
- Published by Wolters Kluwer Health in Clinical Orthopaedics and Related Research
- Vol. 468 (3) , 782-788
The Geometry of the Trochlear Groove
Nal geometry of the femoral trochlea and its relation to the tibiofemoral joint in terms of angles and distances. Methods Forty CT scans of femurs of healthy patients were analyzed using custom-designed imaging software. After aligning the femur using various axes, the locations and orientations of the groove and the trochlear axis were examined in relation to the conventional axes of the femur. Results The trochlear groove was circular and positioned laterally in relation to the mechanical, anatomic, and transcondylar axes of the femur; it was not aligned with any of these axes. We have defined the trochlear axis as a line joining the centers of two spheres fitted to the trochlear surfaces lateral and medial to the trochlear groove. When viewed after aligning the femur to this new axis, the trochlear groove appeared more linear than when other methods of orientation were used. Conclusions Our study shows the importance of reliable femoral orientation when reporting the shape of the trochlear groove....
- 1 March 2010
- Published by Wolters Kluwer Health in Clinical Orthopaedics and Related Research
- Vol. 468 (3) , 782-788
The Geometry of the Trochlear Groove
Nal geometry of the femoral trochlea and its relation to the tibiofemoral joint in terms of angles and distances. Methods Forty CT scans of femurs of healthy patients were analyzed using custom-designed imaging software. After aligning the femur using various axes, the locations and orientations of the groove and the trochlear axis were examined in relation to the conventional axes of the femur. Results The trochlear groove was circular and positioned laterally in relation to the mechanical, anatomic, and transcondylar axes of the femur; it was not aligned with any of these axes. We have defined the trochlear axis as a line joining the centers of two spheres fitted to the trochlear surfaces lateral and medial to the trochlear groove. When viewed after aligning the femur to this new axis, the trochlear groove appeared more linear than when other methods of orientation were used. Conclusions Our study shows the importance of reliable femoral orientation when reporting the shape of the trochlear groove....
- Open Access
- 1 October 2009
- Published by SAGE Publications in HIP International
- Vol. 19 (4) , 323-329
Large Ball Metal on Metal Hips Obscure Cup Angle Measurement on Plain Radiographs
CT when compared to plain radiograph is known to be a more valid measure of acetabular component orientation. The validity of plain radiographs may be further compromised by large diameter metal femoral heads because of obscuration of the acetabular rim. We quantified this effect by measuring acetabular cup angles (inclination and version) of 49 metal on metal (MOM) hip resurfacings using plain radiographs and 3D CT based measurement. Bland-Altman plots revealed poor agreement between plain radiographic and CT based measurement with 2 standard deviation limits of agreements of: +7 to –15 degrees for cup inclination angle; and +16 to –31 degrees for cup version angle. The large differences between plain radiographic and CT measurement of cup positions are probably due to the large diameter metal femoral head that can obscure the cup margin. We have used a metal artefact reduction CT protocol with a 3D imaging software package to overcome this problem and measure cup position relative to the Anterior Pelvic Plane.
- 1 October 2009
- Published by SAGE Publications in HIP International
- Vol. 19 (4) , 323-329
Open AccessLarge Ball Metal on Metal Hips Obscure Cup Angle Measurement on Plain Radiographs
CT when compared to plain radiograph is known to be a more valid measure of acetabular component orientation. The validity of plain radiographs may be further compromised by large diameter metal femoral heads because of obscuration of the acetabular rim. We quantified this effect by measuring acetabular cup angles (inclination and version) of 49 metal on metal (MOM) hip resurfacings using plain radiographs and 3D CT based measurement. Bland-Altman plots revealed poor agreement between plain radiographic and CT based measurement with 2 standard deviation limits of agreements of: +7 to –15 degrees for cup inclination angle; and +16 to –31 degrees for cup version angle. The large differences between plain radiographic and CT measurement of cup positions are probably due to the large diameter metal femoral head that can obscure the cup margin. We have used a metal artefact reduction CT protocol with a 3D imaging software package to overcome this problem and measure cup position relative to the Anterior Pelvic Plane.
- 1 August 2009
- Published by British Editorial Society of Bone & Joint Surgery in The Journal of Bone and Joint Surgery. British volume
- Vol. 91-B (8) , 1031-1036
Three-dimensional CT analysis to determine acetabular retroversion and the implications for the management of femoro-acetabular impingement
This study examined the relationship between the cross-over sign and the true three-dimensional anatomical version of the acetabulum. We also investigated whether in true retroversion there is excessive femoral head cover anteriorly. Radiographs of 64 hips in patients being investigated for symptoms of femoro-acetabular impingement were analysed and the presence of a cross-over sign was documented. CT scans of the same hips were analysed to determine anatomical version and femoral head cover in relation to the anterior pelvic plane after correcting for pelvic tilt. The sensitivity and specificity of the cross-over sign were 92% and 55%, respectively for identifying true acetabular retroversion. There was no significant difference in total cover between normal and retroverted cases. Anterior and posterior cover were, however, significantly different (p < 0.001 and 0.002). The cross-over sign was found to be sensitive but not specific. The results for femoral head cover suggest that retroversion is characterised by posterior deficiency but increased cover anteriorly.
- 1 August 2009
- Published by British Editorial Society of Bone & Joint Surgery in The Journal of Bone and Joint Surgery. British volume
- Vol. 91-B (8) , 1031-1036
Three-dimensional CT analysis to determine acetabular retroversion and the implications for the management of femoro-acetabular impingement
This study examined the relationship between the cross-over sign and the true three-dimensional anatomical version of the acetabulum. We also investigated whether in true retroversion there is excessive femoral head cover anteriorly. Radiographs of 64 hips in patients being investigated for symptoms of femoro-acetabular impingement were analysed and the presence of a cross-over sign was documented. CT scans of the same hips were analysed to determine anatomical version and femoral head cover in relation to the anterior pelvic plane after correcting for pelvic tilt. The sensitivity and specificity of the cross-over sign were 92% and 55%, respectively for identifying true acetabular retroversion. There was no significant difference in total cover between normal and retroverted cases. Anterior and posterior cover were, however, significantly different (p < 0.001 and 0.002). The cross-over sign was found to be sensitive but not specific. The results for femoral head cover suggest that retroversion is characterised by posterior deficiency but increased cover anteriorly.
- Open Access
- 1 November 2008
- Published by British Editorial Society of Bone & Joint Surgery in The Journal of Bone and Joint Surgery. British volume
- Vol. 90-B (11) , 1428-1434
Analysis of cover of the femoral head in normal and dysplastic hips
We present a new CT-based method which measures cover of the femoral head in both normal and dysplastic hips and allows assessment of acetabular inclination and anteversion. A clear topographical image of the head with its covered area is generated. We studied 36 normal and 39 dysplastic hips. In the normal hips the mean cover was 73% (66% to 81%), whereas in the dysplastic group it was 51% (38% to 64%). The significant advantage of this technique is that it allows the measurements to be standardised with reference to a specific anatomical plane. When this is applied to assessing cover in surgery for dysplasia of the hip it gives a clearer understanding of where the corrected hip stands in relation to normal and allows accurate assessment of inclination and anteversion.
- 1 November 2008
- Published by British Editorial Society of Bone & Joint Surgery in The Journal of Bone and Joint Surgery. British volume
- Vol. 90-B (11) , 1428-1434
Open AccessAnalysis of cover of the femoral head in normal and dysplastic hips
We present a new CT-based method which measures cover of the femoral head in both normal and dysplastic hips and allows assessment of acetabular inclination and anteversion. A clear topographical image of the head with its covered area is generated. We studied 36 normal and 39 dysplastic hips. In the normal hips the mean cover was 73% (66% to 81%), whereas in the dysplastic group it was 51% (38% to 64%). The significant advantage of this technique is that it allows the measurements to be standardised with reference to a specific anatomical plane. When this is applied to assessing cover in surgery for dysplasia of the hip it gives a clearer understanding of where the corrected hip stands in relation to normal and allows accurate assessment of inclination and anteversion.
- 1 August 2008
- Published by Wolters Kluwer Health in Journal of Bone and Joint Surgery
- Vol. 90 (Supplement) , 57-64
Learning How to Resurface Cam-Type Femoral Heads with Acceptable Accuracy and Precision: The Role of Computed Tomography-Based Navigation
Background: Resurfacing arthroplasty for cam-type deformities, which are a common cause of early osteoarthritis, is a technically demanding operation. Like any other arthroplasty, it requires both accuracy and precision. On the basis of the results of series reported by expert surgeons, we considered it desirable that this operation should be performed within ±10° of the desired angular orientation and ±6 mm of entry-point translation in 95% of hips. Technological aids are now available to help surgeons achieve that level of accuracy. Three models of cam-type hips of increasing severity were used to assess the efficacy of three systems of instrumentation at delivering the required level of accuracy and precision. Methods: Thirty-two students of surgical technology were instructed in hip resurfacing and shown detailed plans of the desired operative outcome for the three hips with cam-type deformity. They then used conventional instruments, imageless navigation, and computed tomography-based navigation to perform the operation as accurately as possible. Results: Conventional instrumentation produced an unacceptably wide range of entry-point errors. Imageless navigation was able to deliver adequate accuracy and precision in varus-valgus angulation and superoinferior translation, but was less satisfactory in version and anteroposterior translation. Computed tomography-based navigation enabled novice surgeons to navigate hips that had difficult cam-type deformity with acceptable precision in all four degrees of freedom measured. Conclusions: Only computed tomography-based navigation appears to be appropriate for delivering both the accuracy and the precision needed by surgeons on the steep part of their learning curve. Neither conventional neck-based instrumentation nor imageless navigation provided enough help for novice surgeons learning to perform this technically challenging operation. Clinical Relevance: Training with this computed tomography-based navigation system may shorten the learning curve for inexperienced surgeons, leading to a reduction in the prevalence of poor results and revision surgery.
- 1 August 2008
- Published by Wolters Kluwer Health in Journal of Bone and Joint Surgery
- Vol. 90 (Supplement) , 57-64
Learning How to Resurface Cam-Type Femoral Heads with Acceptable Accuracy and Precision: The Role of Computed Tomography-Based Navigation
Background: Resurfacing arthroplasty for cam-type deformities, which are a common cause of early osteoarthritis, is a technically demanding operation. Like any other arthroplasty, it requires both accuracy and precision. On the basis of the results of series reported by expert surgeons, we considered it desirable that this operation should be performed within ±10° of the desired angular orientation and ±6 mm of entry-point translation in 95% of hips. Technological aids are now available to help surgeons achieve that level of accuracy. Three models of cam-type hips of increasing severity were used to assess the efficacy of three systems of instrumentation at delivering the required level of accuracy and precision. Methods: Thirty-two students of surgical technology were instructed in hip resurfacing and shown detailed plans of the desired operative outcome for the three hips with cam-type deformity. They then used conventional instruments, imageless navigation, and computed tomography-based navigation to perform the operation as accurately as possible. Results: Conventional instrumentation produced an unacceptably wide range of entry-point errors. Imageless navigation was able to deliver adequate accuracy and precision in varus-valgus angulation and superoinferior translation, but was less satisfactory in version and anteroposterior translation. Computed tomography-based navigation enabled novice surgeons to navigate hips that had difficult cam-type deformity with acceptable precision in all four degrees of freedom measured. Conclusions: Only computed tomography-based navigation appears to be appropriate for delivering both the accuracy and the precision needed by surgeons on the steep part of their learning curve. Neither conventional neck-based instrumentation nor imageless navigation provided enough help for novice surgeons learning to perform this technically challenging operation. Clinical Relevance: Training with this computed tomography-based navigation system may shorten the learning curve for inexperienced surgeons, leading to a reduction in the prevalence of poor results and revision surgery.
- Open Access
- 1 August 2008
- Published by British Editorial Society of Bone & Joint Surgery in The Journal of Bone and Joint Surgery. British volume
- Vol. 90-B (8) , 1032-1038
The anatomical tibial axis
The rotational alignment of the tibia is an unresolved issue in knee replacement. A poor functional outcome may be due to malrotation of the tibial component. Our aim was to find a reliable method for positioning the tibial component in knee replacement. CT scans of 19 knees were reconstructed in three dimensions and orientated vertically. An axial plane was identified 20 mm below the tibial spines. The centre of each tibial condyle was calculated from ten points taken round the condylar cortex. The tibial tubercle centre was also generated as the centre of the circle which best fitted eight points on the outside of the tubercle in an axial plane at the level of its most prominent point. The derived points were identified by three observers with errors of 0.6 mm to 1 mm. The medial and lateral tibial centres were constant features (radius 24 mm (sd 3), and 22 mm (sd 3), respectively). An anatomical axis was created perpendicular to the line joining these two points. The tubercle centre was found to be 20 mm (sd 7) lateral to the centre of the medial tibial condyle. Compared with this axis, an axis perpendicular to the posterior condylar axis was internally rotated by 6° (sd 3). An axis based on the tibial tubercle and the tibial spines was also internally rotated by 5° (sd 10). Alignment of the knee when based on this anatomical axis was more reliable than either the posterior surfaces or any axis involving the tubercle which was the least reliable landmark in the region.
- 1 August 2008
- Published by British Editorial Society of Bone & Joint Surgery in The Journal of Bone and Joint Surgery. British volume
- Vol. 90-B (8) , 1032-1038
Open AccessThe anatomical tibial axis
The rotational alignment of the tibia is an unresolved issue in knee replacement. A poor functional outcome may be due to malrotation of the tibial component. Our aim was to find a reliable method for positioning the tibial component in knee replacement. CT scans of 19 knees were reconstructed in three dimensions and orientated vertically. An axial plane was identified 20 mm below the tibial spines. The centre of each tibial condyle was calculated from ten points taken round the condylar cortex. The tibial tubercle centre was also generated as the centre of the circle which best fitted eight points on the outside of the tubercle in an axial plane at the level of its most prominent point. The derived points were identified by three observers with errors of 0.6 mm to 1 mm. The medial and lateral tibial centres were constant features (radius 24 mm (sd 3), and 22 mm (sd 3), respectively). An anatomical axis was created perpendicular to the line joining these two points. The tubercle centre was found to be 20 mm (sd 7) lateral to the centre of the medial tibial condyle. Compared with this axis, an axis perpendicular to the posterior condylar axis was internally rotated by 6° (sd 3). An axis based on the tibial tubercle and the tibial spines was also internally rotated by 5° (sd 10). Alignment of the knee when based on this anatomical axis was more reliable than either the posterior surfaces or any axis involving the tubercle which was the least reliable landmark in the region.