Abstract
Background
A laterally tracking patella is commonly seen in patients with chronic recurrent lateral patellar dislocations. Clinical appearance of the J-sign occurs when the patella is congruent with the trochlear groove in flexion and moves over the lateral border of the femoral condyle as the lower leg reaches complete extension. A Fulkerson osteotomy procedure corrects this maltracking of the patella by medially transferring the tibial tubercle. There are many radiographic patellofemoral indices that can be used describe this incongruence about the patelloformal joint. The current literature supports the use of the tibial tubercle-trochlear groove (TT- TG) index in determining the appropriate amount medialization of the extensor mechanism. However there is little agreement in how far to transfer the tibial tubercle to best achieve maximum patello- femoral congruency. It is the senior author's belief that lateral patellar edge (LPE) measure on voluntary quadriceps active hyperextension MRI scan has the strongest correlation with final operative tibial tubercle transfer distance needed to achieve maximum patellofemoral congruency.
Purpose
The purpose of this study was to show that the voluntary quadriceps active hyperextension MRI measurement of lateral patellar edge (LPE) has the strongest correlation with tibial tubercle transfer distance required to achieve maxium patellofemoral congruency intraoperatively in the terminal 30 degrees of active knee extension compared to all other patellofemoral indices measured on axial MRI scans with the knee in voluntary active knee extension to 30 degrees of flexion, passive full extension, and voluntary quadriceps active hyperextension.
Study Design
Retrospective case series via review of the electronic medical record.
Methods
Forty-three Fulkerson osteotomy patient charts were reviewed retrospectively. Three different pre-operative axial MRI views were then examined and measured for Tibial Tubercle- Troch-lear Groove (tt-tg), lateral patellar edge (LPE), bisect offset (BSO), and lateral patellar displacement (LPD). Each patient had three MRIs: one with the knee resting in extension, one in voluntary quadriceps active hyperextension, and one in voluntary quadriceps active 30 degree flexion. Statistics were then calculated using Statistical Package for the Social Sciences (SPSS) (IBM corp).
Results
Tibial tubercle transfer distances required to achieve congruency intraoperatively correlated moderately (0.500-0.300) and were statistically significant (alpha. 050) for passive extension MRI measurement of TT-TG (Pearson- 0.403, alpha 0.010) and LPD (Pearson. 362, alpha 0.022); voluntary quadriceps active hyperextension TT-TG (Pearson 0.487, alpha, 0.001); voluntary quadriceps active flexion TT-TG (Pearson. 548, alpha< 0.001), LPE (Pearson. 332, alpha 0.029), and LPD (Pearson 0.446 alpha. 003).
Conclusion
The hypothesis that voluntary quadriceps active hyperextension MRI LPE measurement best correlated with tibial tubercle transfer distance was incorrect. The data collected showed correlation and statistical significance for voluntary quadriceps active flexion LPE with required tibal tubercle transfer distance (Pearson 0.34, alpha 0.026). The MRI measurement that best correlated with tibial tubercle transfer distance was voluntary quadriceps active flexion measure of TT-TG (Pearson. 556, alpha< 0.001).
Keywords: Patellar instability, lateral patellar subluxation, Fulkerson osteotomy, lateral patellar edge (LPE), lateral patella displacement (LPD), tibial tubercle- trochlear groove (TT-TG), bisect offset (BSO), anatomy, radiographic landmarks
Introduction
The decision for operative correction of lateral patella instability hinges largely on qualitative reports of instability, the J-sign, and the apprehension test. Quantitative measures do exist, however, that utilize MRI scans of passive knee extension, voluntary active quadriceps knee extension, and voluntary quadriceps active knee flexion to 30 degrees. These add objective data, which can help to supplement clinical decision-making in determination of how far to transfer the tibial tubercle. Currently the literature supports the use of tibial tubercle to the trochlear groove (TT-TG) distance in preoperative planning of the distance the tibial tubercle needs to be transferred to achieve correction of the extensor mechanism7. The purpose of this case series is to show that the voluntary quadriceps active hyperextension MRI measurement of lateral patellar edge represents the terminal phase of the J-sign and has the strongest correlation with actual tibial tubercle transfer distance as determined intraoperatively by direct observation and palpation of the knee extensor mechanism congruency with concurrent femoral nerve stimulation.
Goals of Analysis:
1) Compare preoperative MRI measurements of tibial tubercle-trochlear groove (TT-TG), lateral patellar edge (LPE), bisect offset (BSO), and lateral patellar displacement (LPD) to actual tibial tubercle transfer distance to see which has the highest correlation.
Materials and Methods
Imaging:
In total, 43 Fulkerson osteotomy patients were retrospectively reviewed. Each patient had received three knee MRI scans using a Siemens Tesla 3. Each patient had a scan with the knee voluntarily actively flexed to 30 degrees, in passive full extension, and in voluntary active quadriceps hyperextension as part of the standard MRI workup for patients undergoing a Fulkerson osteotomy. All the patients undergoing Fulkerson osteotomy had a history of recurrent lateral subluxations and obvious patellofemoral maltracking on both clinical exam and quadriceps active knee extension MRI.
Clinical Information:
All of the patients underwent a Fulkerson osteotomy between the years of 2006-2012.
Measurements:
Each of the following patellofemoral indices were measured before surgery at voluntary active 30 degrees of knee flexion, passive extension, and voluntary active hyperextension. All measurements were performed in the axial plane. In order to normalize for the size of the patella, a bisect offset (BSO) measurement was made. First, a line was drawn tangential to the posterior aspects of both the medial and lateral femoral condyles. A perpendicular line then was drawn through the deepest point of the sulcus. The BSO is the ratio of the patella lateral to the perpendicular line compared with the total width of the patella (Figure 1A)3, 8. To measure lateral patellar displacement (LPD), a line was drawn tangential to the posterior aspect of both medial and lateral femoral condyles. A perpendicular line was then drawn through the apex of the medial femoral condyle. The distance from this line to the most medial part of the patella was then measured (Figure 1B)3,6. Lateral patellar edge (LPE) is identical to LPD, except the apex of the lateral condyle was used instead of the apex of the medial condyle (Figure 1C)1, 3.
Figure 1. Patellofemoral indices used for MRI measurements A. BSO, B. LPD, C. LPE1, 2.
To measure TT-TG, a reference line was drawn posterior to both the medial and lateral condyles then horizontal lines drawn between the deepest point in trochlear groove and apex of the anterior tibial tubercle perpendicular to the reference line. The distance was then measured between the tibial tubercle and trochlear groove (Figure 2)3, 4, 6.
Figure 2. TT-TG patella femoral measurement4.
Statistics:
Data was then analyzed using the SPSS (IBM Corp). The statistical methods used included Pearson Correlation Coefficients and Chi Square Analysis. Statistical significance was set at p=.05.
Results
Forty-three patients were included, with ages ranging from 13-51 years old with an average age of 24 (Table 1). Nineteen of the patients were male and 24 were female.
Table 1.
Patient Demographic Data
| Parameter | Mean | Standard Deviation |
|---|---|---|
| Age (years) | 24.0 | 8.64 |
| Height (cm) | 170 | 11.2 |
| Weight (kg) | 189 | 42.9 |
| BMI | 29.1 | 6.13 |
The MRI measurements for passive knee extension, active knee extension, and active knee flexion to 30 degrees are summarized on Tables 2-4 respectively. The average final operative tibial tubercle transfer distance was 21.6mm (4.53mm standard deviation, 30mm maximum, and 12mm minimum).
Table 2.
MRI Measurements passive knee extension
| Passive Quadriceps Knee Extension MRI | |||||
|---|---|---|---|---|---|
| Patient | Actual transfer distance (mm) | TT-TG (mm) | LPE (mm) | LPD (mm) | BSO (mm) |
| 1 | 19 | 20 | 4.7 | 16.2 | 1.70 |
| 2 | 20 | 23.5 | 13.2 | 24.4 | 8.95 |
| 3 | 22 | 12.7 | 7 | 13.7 | 8.07 |
| 4 | 20 | 22 | 9.8 | 19.2 | 8.72 |
| 5 | 21 | 20.2 | 4.2 | 16.9 | 1.86 |
| 6 | 15 | 18.6 | 15.2 | 24.8 | ∼ |
| 7 | 22 | 23.2 | 15.1 | 27.6 | ∼ |
| 8 | 24 | 26.3 | 14.1 | 22.6 | ∼ |
| 9 | 22 | 25.9 | 9.9 | 26.1 | ∼ |
| 10 | 23 | 20.3 | 14.5 | 31.7 | ∼ |
| 11 | 20 | 20.6 | 6.2 | 13 | 3.93 |
| 12 | 24 | 23 | 15 | 25.8 | ∼ |
| 13 | 15 | 25.8 | 8 | 13.6 | 4.67 |
| 14 | 22 | 29.1 | 0.09 | 16.4 | 2.54 |
| 15 | 20 | 7.9 | 6.3 | 4.08 | |
| 16 | 20 | 17.8 | 9.8 | 18.8 | 19.85 |
| 17 | 30 | 31.1 | 15.6 | 38.2 | ∼ |
| 18 | 28 | 23.5 | -2.1 | 12.6 | 1.79 |
| 19 | 20 | 20.2 | 7.5 | 17.4 | 12.27 |
| 20 | 25 | ||||
| 21 | 28 | 22.8 | -5.7 | 12 | 1.64 |
| 22 | 20 | 19.7 | 4.2 | 21.1 | 8.98 |
| 23 | 30 | 27.7 | 17.4 | 28.1 | ∼ |
| 24 | 25 | 24.9 | 13.3 | 15.9 | 6.62 |
| 25 | 26 | 20.8 | 5 | 15.6 | 3.16 |
| 26 | 30 | 26.8 | 14.5 | 27.2 | ∼ |
| 27 | 15 | 24.4 | 1.9 | 5.2 | 2.76 |
| 28 | 25 | 25.4 | 19.3 | 35 | ∼ |
| 29 | 20 | 19.1 | 6.8 | 14.7 | 2.97 |
| 30 | 18 | 22.1 | 15.9 | 18.3 | ∼ |
| 31 | 25 | 25.5 | -4.2 | 7.6 | 1.56 |
| 32 | 25 | 22.5 | 3.8 | 14.1 | 1.73 |
| 33 | 17 | 25.9 | 3.3 | 8 | 2.77 |
| 34 | 13 | 22.1 | 13.2 | 25.4 | ∼ |
| 35 | 12 | 17.9 | 2.9 | 9.9 | 3.17 |
| 36 | 15 | 19.7 | -3 | 2.7 | 1.21 |
| 37 | 23 | 29.7 | 12 | 32.1 | ∼ |
| 38 | 26 | ||||
| 39 | 20 | ||||
| 40 | 27 | 18.5 | 13.6 | 13.2 | 8.33 |
| 41 | 21 | 14.7 | 1.4 | 6.6 | 1.96 |
| 42 | 14 | 18.6 | 10 | 18.6 | 3.81 |
| 43 | 24 | 30.8 | 14.9 | 25.3 | ∼ |
Table 4.
MRI measurements active knee extension
| Quadriceps Active Knee Flexion 30 degrees MRI | |||||
|---|---|---|---|---|---|
| Patient | Actual transfer distance (mm) | TT-TG (mm) | LPE (mm) | LPD (mm) | BSO (mm) |
| 1 | 19 | 19.4 | 0 | 20 | 5.37 |
| 2 | 20 | 16.9 | 3.1 | 7.5 | 3.41 |
| 3 | 22 | 13.3 | 11.3 | 15.3 | 3.44 |
| 4 | 20 | 15.1 | 4.4 | 11.9 | 3.14 |
| 5 | 21 | 11.1 | 3.3 | 11.5 | 2.13 |
| 6 | 15 | 9.4 | 2.5 | 7.5 | 1.90 |
| 7 | 22 | 24.4 | 3.1 | 5.6 | 1.55 |
| 8 | 24 | 21.9 | 15.1 | 33.1 | ∼ |
| 9 | 22 | 15.6 | 11.3 | 20.6 | ∼ |
| 10 | 23fl | 18.8 | 21.9 | 24.2 | 6.21 |
| 11 | 20 | 12.5 | 3.8 | 5.7 | 1.54 |
| 12 | 24 | 19.9 | 11.4 | 20.3 | ∼ |
| 13 | 15 | 9.5 | 14.4 | 17.6 | 11 |
| 14 | 22 | 15.7 | 16.9 | 23.8 | ∼ |
| 15 | 20 | 13.1 | 7.1 | 5.3 | 2.54 |
| 16 | 20 | 14.4 | 8.8 | 10 | 2.79 |
| 17 | 30 | 28.5 | 14 | 33.9 | ∼ |
| 18 | 28 | 20.4 | 3.4 | 13.6 | 1.97 |
| 19 | 20 | 23.8 | 13.2 | 10 | 3.40 |
| 20 | 25 | 23.2 | 5.6 | 3.1 | 1.04 |
| 21 | 28 | 9.2 | 2.3 | 9.3 | 2.19 |
| 22 | 20 | 17.1 | 3.8 | 7.6 | 2.40 |
| 23 | 30 | 22.4 | 14.1 | 27.9 | ∼ |
| 24 | 25 | 25.8 | 10.2 | 9.9 | 4.99 |
| 25 | 26 | 20 | 13.8 | 13.1 | 18.56 |
| 26 | 30 | 25.6 | 15.7 | 31.3 | ∼ |
| 27 | 15 | 15.7 | 6.4 | 9.4 | 1.95 |
| 28 | 25 | 20.8 | 24 | 41.6 | ∼ |
| 29 | 20 | 15.8 | -5.8 | 2.6 | 0.45 |
| 30 | 18 | 14.4 | 7.5 | 8.1 | 2.23 |
| 31 | 25 | 20.1 | 5.7 | 9.4 | 2.60 |
| 32 | 25 | 17.2 | 8.1 | 10.7 | 1.84 |
| 33 | 17 | 14.5 | 1.3 | 4.4 | 1.58 |
| 34 | 13 | 12.6 | 10 | 16.9 | 7.20 |
| 35 | 12 | 15.1 | 7 | 6.5 | 2.94 |
| 36 | 15 | 10 | 0.5 | 5.1 | 1.05 |
| 37 | 23 | 34 | 19.3 | 42.9 | ∼ |
| 38 | 26 | 19 | 14.6 | 29.5 | ∼ |
| 39 | 20 | 17.6 | 5.7 | 13.8 | 4.6 |
| 40 | 27 | 22.8 | 12.7 | 14 | 6.77 |
| 41 | 21 | 11.9 | 0 | 5 | 2.00 |
| 42 | 14 | 20.5 | 12 | 13.5 | 3.25 |
| 43 | 24 | 26.4 | 7.5 | 15.1 | 4.43 |
Table 3.
MRI measurements active knee extension
| Active Quadriceps Knee Hyperextension MRI | |||||
|---|---|---|---|---|---|
| Patient | Actual transfer distance (mm) | TT-TG (mm) | LPE (mm) | LPD (mm) | BSO (mm) |
| 1 | 19 | 22.3 | 18.5 | 28.1 | ∼ |
| 2 | 20 | 27.6 | 24 | 37.2 | ∼ |
| 3 | 22 | 14 | 21.5 | 28.5 | ∼ |
| 4 | 20 | 22.7 | 14 | 22 | 39.0 |
| 5 | 21 | 22.5 | 16.3 | 21.3 | ∼ |
| 6 | 15 | 20.2 | 17.6 | 25.3 | ∼ |
| 7 | 22 | 25.5 | 18.9 | 33.8 | ∼ |
| 8 | 24 | 26 | 19.9 | 38.7 | ∼ |
| 9 | 22 | 26.8 | 18.4 | 36.3 | ∼ |
| 10 | 23 | 22.5 | 28.2 | 37.6 | ∼ |
| 11 | 20 | 22 | 18 | 22.6 | ∼ |
| 12 | 24 | 19.4 | 18.2 | 29.5 | ∼ |
| 13 | 15 | 15.8 | 15.2 | 20.9 | ∼ |
| 14 | 22 | 30.2 | 21.5 | 36 | ∼ |
| 15 | 20 | 16 | 21.1 | 18 | ∼ |
| 16 | 20 | 24.4 | 12.6 | 28.8 | ∼ |
| 17 | 30 | 31.3 | 16.9 | 32.6 | ∼ |
| 18 | 28 | 24.8 | 9.2 | 25.8 | ∼ |
| 19 | 20 | 21.3 | 13.8 | 20.2 | ∼ |
| 20 | 25 | 25.5 | 28 | 23.4 | ∼ |
| 21 | 28 | 23.5 | 1.8 | 17.1 | 3.18 |
| 22 | 20 | 22.5 | 15 | 25 | 8.93 |
| 23 | 30 | 30.1 | 18.2 | 30.7 | ∼ |
| 24 | 25 | 17.5 | 4.9 | 17.5 | 6.55 |
| 25 | 26 | 23.1 | 7.6 | 15.1 | 10.50 |
| 26 | 30 | 28.2 | 20 | 31.3 | ∼ |
| 27 | 15 | 24 | 12.7 | 13.3 | 4.52 |
| 28 | 25 | 27.3 | 20 | 33.7 | ∼ |
| 29 | 20 | 19.1 | 16 | 30.8 | ∼ |
| 30 | 18 | 19.4 | 20.7 | 27.1 | ∼ |
| 31 | 25 | 21.9 | 11.3 | 26.3 | ∼ |
| 32 | 25 | 23.4 | 13.4 | 24 | ∼ |
| 33 | 17 | 20.8 | 7.6 | 14.6 | 4.27 |
| 34 | 13 | 25.7 | 18.2 | 30.7 | ∼ |
| 35 | 12 | 19.4 | 11.6 | 23.9 | 11.21 |
| 36 | 15 | 19.4 | 5.1 | 9.2 | 2.69 |
| 37 | 23 | 31 | 17 | 41.1 | ∼ |
| 38 | 26 | 20.2 | 17.1 | 27.2 | ∼ |
| 39 | 20 | 22.4 | 2.6 | 5.8 | 1.6 |
| 40 | 27 | 31.9 | 11.9 | 22.5 | ∼ |
| 41 | 21 | 22.1 | 7.6 | 16.5 | 19.89 |
| 42 | 14 | 19.6 | 15.5 | 23.5 | 16.43 |
| 43 | 24 | 17.2 | 35.8 | ∼ | |
MRI scan of passive quadriceps knee extension had patellofemoral indices measures of TT-TG (22.6mm average, 4.18mm standard deviation, 12.7mm minimum, 31.1mm maximum), LPE (8.40mm average, 6.44mm standard deviation, -5.7mm minimum, 19.3mm maximum), LPD (18.5mm average, 8.53mm standard deviation, 2.70mm minimum, 38.2mm maximum), and BSO (4.97mm average, 4.29mm standard deviation, 1.21mm minimum, 19.8mm maximum).
MRI scan of active quadriceps knee hyperextension had patellofemoral indices measures of TT-TG (23.2mm average, 4.27mm standard deviation, 14.0mm minimum, 31.9mm maximum), LPE (15.5mm average, 6.03mm standard deviation, 1.8mm minimum, 28.2mm maximum), LPD (25.8mm average, 8.15mm standard deviation, 5.80mm minimum, 41.1mm maximum), and BSO (10.7mm average, 10.5mm standard deviation, 1.6mm minimum, 39mm maximum).
MRI scan of quadriceps active knee flexion to 30 degrees had patellofemoral indices measures of TT-TG (18.0mm average, 5.58mm standard deviation, 9.20mm minimum, 34.0mm maximum), LPE (8.63mm average, 6.35mm standard deviation, -5.80mm minimum, 24.0mm maximum), LPD (15.1mm average, 10.3mm standard deviation, 2.60mm minimum, 42.9mm maximum), and BSO (3.71mm average, 3.42mm standard deviation, 0.450mm minimum, 18.6mm maximum).
In the analysis of actual tibial tubercle transfer distances with the measures made on the three different preoperative MRI scans actual tibial tubercle transfer distances correlated moderately (0.5-0.3) and were statistically significant (alpha. 05) for passive extension MRI measurement of TT-TG (Pearson 0.403, alpha 0.011), LPD (Pearson 0.362, alpha 0.022), voluntary quad active hyperextension TT-TG (Pearson 0.487, alpha 0. 001), voluntary quad active flexion TT-TG (Pearson. 548, alpha <0.001), LPE (Pearson. 332, alpha 0.029), and LPD (Pearson 0.446 alpha. 003). MRI measurements compared with actual transfer distance are summarized on Table 5.
Table 5.
Actual transfer distance with MRI measurement correlations
| Passive extension MRI correlated with actual transfer distance | ||
| Measurement | Pearson Correlation Coefficient | Significance |
| TT-TG | 0.400 | 0.012 |
| LPE | 0.155 | 0.338 |
| LPD | 0.363 | 0.021 |
| BSO | -0.149 | 0.359 |
| Active extension MRI Correlated with actual transfer distance | ||
| Measurement | Pearson Correlation Coefficient | Significance |
| TT-TG | 0.485 | 0.001 |
| LPE | 0.034 | 0.828 |
| LPD | 0.272 | 0.078 |
| BSO | -0.220 | 0.157 |
| Flexion MRI correlated with actual transfer distance | ||
| Measurement | Pearson Correlation Coefficient | Significance |
| TT-TG | 0.556 | < 0.001 |
| LPE | 0.34 | 0.026 |
| LPD | 0.444 | 0.003 |
| BSO | 0.117 | 0.517 |
Discussion
The purpose of this study was to show that LPE measurement on a voluntary quadriceps active MRI scan has the highest correlation with final operative tibial tubercle transfer distance. Our hypothesis that voluntary quadriceps active hyperextension MRI LPE measurement best correlated with actual tibial tubercle transfer distance (determined by intra-operative end point of visualization and palpation of the extensor mechanism showing no overlap of the patella on the lateral femoral condyle with femoral nerve stimulation) was incorrect (Pearson 0.155, alpha 0.338).
The measurement with the highest correlation of all measures with actual tibial tubercle transfer distance was voluntary quadriceps active 30 degrees of knee flexion measure of TT-TG (Pearson 0.548, alpha < 0.001). This is in contrast to the universally accepted TT-TG measurement which is done in passive full knee extension as described Dejour et al.2 While we did include these measures in our calculations, quadriceps active knee flexion had the highest correlation in achieving maximum congruency at the patellofemaoral joint as determined intraoperatively.
Among all of the MRI scans when the leg was held in mid-air with the quadriceps contracted (quad active flexion) and the knee in 30 degrees of voluntary flexion, LPE had the highest correlation with final tibial tubercle transfer distance of any LPE measurement (Pearson 0. 332, alpha 0.029). However, all LPE measurements correlated less than TT-TG measurement in quad active flexion with final tibial tubercle transfer distance.
These findings where the measurement of TT-TG is more reliable when the patella in the trochlear grove in flexion is consistent with finding of McDermott et al. that demonstrated that TT-TG should be used in at least a 1:1 transfer distance ratio when the goal for preoperative planning is to achieve maximum congruency at the patellofemoral joint.
Conclusions
In conclusion our hypothesis was incorrect that quadriceps active measure of LPE is the best predictor of how far the tibial tubercle needs to be transferred to in order to eliminate the J-sign and achieve maximum patellofemoral congruency of the extensor mechanism with femoral nerve stimulation at the time of surgery. The findings of this study are consistent with the recommendation in the current literature of the use of TT-TG in at least a 1:1 ratio for preoperative tubercle transfer distance when the goal is to restore maximum congruency at the patellofemoral joint7.
References
- 1.Brossmann J, Muhle C, Büll CC, Zieplies J, Melchert UH, Brinkmann G, Schröder C, Heller M. Cine MR imaging before and after realignment surgery for patellar maltracking - comparison with axial radiographs. Skeletal Radiol. 1995;24(3):191–6. doi: 10.1007/BF00228921. Apr. [DOI] [PubMed] [Google Scholar]
- 2.Dejour D, Le Coultre B. Osteotomies in patello- femoral instabilities. Sports Med Arthrosc. 2007;15(1):39–46. doi: 10.1097/JSA.0b013e31803035ae. Mar Review. [DOI] [PubMed] [Google Scholar]
- 3.Duchman K, Mellecker c, El-Hattab AY, Albright JP. Quantitative MRI of tibial tubercle transfer during active quadriceps contraction. Clinical Orthopaedic Related Research. 2011;469:294–299. doi: 10.1007/s11999-010-1598-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Duchman K, Mellecker C, Thedens DR, Albright JP. Quantifying the effects of extensor mechanism medializatlon procedures using MRI: a cadaver-based study. Iowa Orthop J. 2011;31:90–8. [PMC free article] [PubMed] [Google Scholar]
- 5.Laurin CA, Dussault R, Levesque HP. The tangential x-ray investigation of the patellofemoral joint: x-ray technique, diagnostic criteria and their interpretation. Clin Orthop Relat Res. 1979;144:16–26. Oct. [PubMed] [Google Scholar]
- 6.Lustig S, Servien E, Aït Si Selmi T, Neyret P. Factors affecting reliability of TT-TG measurements before and after medialization: a CT-scan study. Rev Chir Orthop Reparatrice Appar Mot. 2006;92(5):429–36. doi: 10.1016/s0035-1040(06)75829-9. Sep. [DOI] [PubMed] [Google Scholar]
- 7.McDermott SE, Anthony CA, McCunniff P, Duchman K, Albright JP. A clinical and radiographic approach for establishing proper tibial tubercle transfer when using quad active femoral nerve stimulation. Iowa Orthop J. 2013;33:84–9. [PMC free article] [PubMed] [Google Scholar]
- 8.Stanford W, Phelan J, Kathol MH, Roohol-amini SA, el-Khoury GY, Palutsis GR, Albright JP. Patellofemoral joint motion: evaluation by ultrafast computed tomography. Skeletal Radiol. 1988;17(7):487–92. doi: 10.1007/BF00364042. [DOI] [PubMed] [Google Scholar]