The Knee 21 (2014) 133–137

Contents lists available at ScienceDirect

The Knee

Triple positioning of tibial tubercle osteotomy for patellofemoral disorders

Ching-Jen Wang a,⁎,1, To Wong b,1, Jih-Yang Ko a, Ka-Kit Siu a

a Department of Orthopedic Surgery, Section of Sports Medicine, Chang Gung University College of Medicine, Kaohsiung Chang Gung Memorial Hospital, Taiwanb Department of Orthopedic Surgery, Kaohsiung Municipal United Hospital, Kaohsiung, Taiwan

⁎ Corresponding author at:Department ofOrthopedic SurKaohsiung Chang Gung Memorial Hospital, 123 Ta-Pei RdTaiwan 833. Tel.: +886 7 733 5279; fax: +886 7 733 5515

E-mail address: w281211@adm.cgmh.org.tw (C.-J. W1 Both authors contributed equally in this paper.

0968-0160/$ – see front matter © 2012 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.knee.2012.10.027

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 7 March 2012Received in revised form 26 October 2012Accepted 28 October 2012

Keywords:Triple positioningtibial tubercle osteotomyPatellofemoral disorders

Background and purpose: Patellofemoral disorders are often associated with patellofemoral malalignment.Tibial tubercle transfer is an effective method to correct the patellofemoral malalignment. This study evalu-ated the long-term results of triple positioning of tibial tubercle osteotomy for refractory patellofemoral dis-orders with 10-year follow-up.Patients and Methods: Fifty-six patients with 62 knees underwent triple positioning of tibial tubercle osteotomyfor refractory patellofemoral disorders. All patients received standard rehabilitation protocol postoperatively.The average length of follow-up was 128.5±9.8 months (range 116 to 149). The evaluations included painscore, Kujala patellofemoral score, Lysholm score and radiograph of the knee.

Results: The overall clinical results were excellent in 41.9%, good in 37%, fair in 12.9% and poor in 5% at 1 year;and 29% excellent, 41% good, 18.3% fair and 9.3% poor at 10 years. Satisfactory results were 78.9% and 70.9%, andunsatisfactory results 21.1% and 29.1% at 1 year and 10 years respectively. There was no correlation of clinicaloutcomes with age, sex, body weight and height and preoperative pain score. However, there was a positivecorrelation of clinical outcomes with the improvement of the congruence angle on postoperative X-rays ofthe knee, and a negative correlation of clinical outcome with the severity of articular cartilage damage assessedin arthroscopy. The complications included 1 non-union and 1 infection with non-union.Conclusion: Triple positioning of tibial tubercle osteotomy is effective and long lasting in patients withpatellofemoral disorders with 70.9% satisfactory results at 10-year follow-up.Level of evidence: IV (refer to instructions for detailed description on the level of evidence).

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

Patellofemoral disorder is one of the most common knee disordersand is often associated with patellofemoral malalignment [1–3]. Diag-nosis of patellofemoral disorder was made by history and physical ex-amination, and radiographic evaluation of the knee. Patellofemoraldisorder occurs predominantly in young to middle-age active individ-uals. Patients frequently present with anterior knee pain and crepitus,weakness and limping with or without overuse in daily activities. Thesymptoms are aggravated in stair climbing and kneeling or squatting,and somewhat relieved with rest in extension. Physical findings revealpatellofemoral compression pain, tenderness and crepitus, increasedQ angle and laterally tilted patella in flexion. The medial excursionof the patella is usually decreased. Radiographic examination inanteroposterior, lateral and Merchant views often revealed increasesin patellar height and congruence angle, and lateral displacement ofthe patella.

gery, Section of SportsMedicine,., Niao-Sung District, Kaohsiung,.ang).

rights reserved.

Patellofemoralmalalignmentmay be translational subluxation of thepatella or torsional patella tilt with or without patellar instability [4–7].Patellofemoral malalignment can cause anterior knee pain associatedwith chondromalacia or patellofemoral arthritis. Patellofemoral disordersecondary to arthritis shows distinction from patellar instability, andusually affects older patients with pain as the predominant symptom.The presenting symptoms include anterior knee pain and crepitus,weakness in extension and limping. The etiology of patellofemoral dis-order is multi-factorial including soft tissue imbalance and bony abnor-mality [8–10]. The initial treatment is conservative treatment thatincludes anti-inflammatory drugs, knee brace, physical therapy, musclestrengthening exercises and modification of activity levels [11]. Surgeryis indicated in knees refractory to at least 6 months of conservativetreatment [12,13]. Proximal realignment or medial patellofemoral liga-ment reconstruction is generally performed for knees with patellar in-stability [8,10,14,15], whereas distal realignment with tibial tubercletransfer for knees with patellofemoral malalignment [16–20]. For distalrealignment with tibial tubercle osteotomy, many surgical techniqueswith different magnitudes of tibial tubercle transfer are described inthe literature including Elmslie-Trillat and Fulkerson procedures, andthe results varied considerably [13,18–23]. There is no consensus ofopinions on the magnitude of tibial tubercle transfer in the treatmentof patellofemoral disorder [5,13,16–18]. Likewise, there was no

134 C.-J. Wang et al. / The Knee 21 (2014) 133–137

agreement on the best method of tibial tubercle transfer [16,17]. Basedon cadaver study, triple positioning of tibial tubercle osteotomy that in-cludes anterior, medial and proximal positioning of the tibial tuberclewas developed to optimize the patellofemoral alignment [24]. The pur-pose of this retrospective study was to analyze the long-term effects ofthis procedure in 56 patients with 62 knees with 10-year follow-up.

2. Materials and methods

The Institutional Review Board of our institution approved thisstudy protocol. Informed consent was signed prior to the participa-tion in the study.

Between July 1997 and August 2000, 66 patients with 72 knees thatmet the inclusion criteria were operated on for patellofemoral disordersby a single surgeon. Surgery is indicated when conservative treatmentshad failed. The inclusion criteria for surgery included knees with pro-gressive pain and functional disability with failure to at least 6 monthsof conservative treatment. The severity of patellofemoral pain was cat-egorized into mild with less than 3 visual analog score (VAS), moderatewith 4–6 VAS and severewith greater than 7 VAS. The exclusion criteriaincluded patients with severe radiographic arthritis of any compart-ment of the knee, infection, rheumatic disease, trochlea hypoplasia,hypermobile patella, excessive bodymass index, severe knee deformityand psychosomatic disorder. Ten patients with 10 knees were excludeddue to lost to follow-up. The remaining 56 patients with 62 knees wereincluded in the final analysis. The patient demographic characteristicsare shown in Table 1.

A complete arthroscopic examination of the knee joint wasperformed with attention to the patellofemoral articulation. The sever-ity of cartilage damage of the patellofemoral joint was assessedarthroscopically and recorded on four variables including the descrip-tion of the articular surface, the extent (depth) of involvement, the di-ameter of the lesion and the location of the lesion [25]. Arthroscopicdebridement and radiofrequency chondroplasty were performed inknees with grade II or more cartilage damage. Concomitant surgeryfor meniscus tear if any, was also performed. The tightness of the lateralretinaculum was assessed by the medial excursion of the patella withthe knee in extension. The patella was defined tight when the medialexcursion of the patella was less than half of the patellar width.

After arthroscopy, a 5-cm longitudinal incision was made on theproximal tibial. The patella tendon was dissected free on both sides. Lat-eral retinacular release was performed subcutaneously from the level of

Table 1Patient demographic characteristics.

# Patients / # knees 56/62

Gender:Females 46 (82%)Males 10 (18%)

Age (years) 45.7±11.3(Range) (22–74)

SideRight knee 34Left knee 28Bilateral knees 6

Ave. body weight (kg) 61.6±13.6(Range) (52–75)

Ave. body height (cm) 160±9.2(Range) (129–184)

Tibial tubercle transferAnterior transfer (mm) 14.3±2.7(Range) (5–20)Medial transfer (mm) 9.2±3.4(Range) (0–15)Proximal transfer (mm) 4.7±1.4(Range) (3–7)

Ave. tibial shingle (mm) 55×14×7Ave. follow-up (months) 128.5±9.8

(Range) (116–149)

tibial tubercle to the origin of vastus lateralis. A 5-cm long tibial tuberclebone block was marked and pre-drilled. An oblique osteotomy of thetibial tubercle was made 30° of the vertical line from medial to lateralaspect. The magnitudes of anterior, medial and proximal positioning ofthe tibial tubercle osteotomywere assessed and determined individuallybased on the algorithm shown in Table 2. The magnitudes of anterior,medial and proximal transfer of the tibial tubercle were affectedby the severity of patellofemoral malalignment that varied amongpatients. The severity of patellofemoral malalignment was gradedaccording to the magnitude of tibial tubercle transfer with 0–5 mm asmild, 5–10 mmasmoderate and greater than 10 mmas severe. Patientswith predominant lateral patellar tilt required more medial transferof the tibial tubercle, whereas patients with more patellofemoral com-pression pain and crepitus needed more anterior transfer of tibialtubercle. Additional proximal transfer of tibial tubercle was indicatedif persistently tight patellofemoral joint or patella baja was noted afteranteromedial transfer of the tibial tubercle. Once the location of theoptimal tibial tubercle transfer was established, the tibial tuberclebone block was rigidly fixed with two screws. The medial gap betweenbone block and tibial cortex and the gap distal to the bone block werefilled with bone grafts harvested from the lateral aspect of the proximaltibial, and the lateral defect was filled with osteoset bone substitute(Wright Medical, Arlington, Tennessee) (Fig. 1).

Postoperatively, the knee was immobilized in a knee immobilizerduring ambulation. Patients ambulated with partial weight bearingon the operated leg, and performed range of motion, quadriceps andhamstring strengthening exercise. Full weight bearing was allowedin 6 to 8 weeks when there was evidence of radiographic healing ofthe tubercle osteotomy.

The follow up examinations were scheduled at 1, 3, 6 and 12 months,and then yearly. The evaluation parameters included pain score, Kujalapatellofemoral score [26], Lysholm score [27], the Q angle, articular carti-lage damage and radiographs of the affected knees. Themaximal Lysholmscore is 100 points, and the score chart is classified as excellent with95–100 points, good with 84–94 points, fair with 65–83 points and poorwith b64 points. The severity of articular cartilage damage was docu-mented during the initial arthroscopy. Radiographic examinations includ-ing patella height, congruence angle of the patella and spur formationwere used to assess the severity of patellofemoralmalalignment andoste-oarthritis of the patellofemoral and femorotibial joints before and aftersurgery [3,28].

2.1. Statistical analysis

The data presented in the study were shown inmean±SD. The pre-and post-operative pain score, functional score, patellofemoral scoreand the Q angle were compared and analyzed statistically using pairedt test. The correlations of outcomes with the inherent factors, and thecomparison between satisfactory and unsatisfactory knees were ana-lyzed statistically using Chi-Square test. A p-value of less than 0.05was considered to be statistically significant.

3. Results

The pain score, Kujala patellofemoral score, Lysholm score and the Q angle are shownin Table 3. Significant improvements in pain score, Kujala score and Lysholm score werenoted at 1 year and 10 years postoperatively. The Q angle was 17.6±5.2 (10–30) preop-eratively and 15.3±3.0 (9–22) postoperatively. Significant improvement in Q angle wasnoticed after surgery (p=0.001). Overall, satisfactory result was 78.9% at one-year, and70.9% at 10-year follow-up. There was a trend of increase in pain score and a decreasein functional score from one-year to 10-year follow-up, however, the difference was sta-tistically not significant (p>0.05). All knees could flex 120 degrees or more, and noneshowed extension lag.

The correlations of clinical outcomes with the inherent factors are analyzed statis-tically and the results are shown in Table 4. There is no statistical correlation of clinicaloutcome with age, sex, body weight, body height and the preoperative pain score.

The correlations of the clinical outcome and the severity of articular cartilage damageassessedduring the initial arthroscopy are shown in Table 5. There is a negative correlation

Table 2The algorithm of triple positioning of tibial tubercle osteotomy for patellofemoral disorders.

Additional 0 to 5mm proximal displacement

Tight patellofemoral

joint or patella baja

Arthroscopic examination and debridement

Lateral retinacular release

20 mm anterior

0 mm medial

0 to 5mm proximal

15 mm anterior

5 to 10 mm medial

0 to 5mm proximal

10 mm anterior

10 to 15 mm medial

0 to 5mm proximal

Mild malalignment Severe malalignmentModerate malalignment

135C.-J. Wang et al. / The Knee 21 (2014) 133–137

of the clinical outcome with the severity of articular cartilage damage. The majority ofknees with unsatisfactory results showed advanced articular cartilage damage.

The correlations of clinical outcome and radiographic evaluation are shown inTable 6. There is no difference in patella height between satisfactory and unsatisfactoryknees before and after operation. The difference in congruence angle between satisfac-tory and unsatisfactory knees was statistically not significant preoperatively (p>0.05),however, the difference became statistically significant postoperatively (pb0.05). Thisindicated that improvement in the correction of the congruence angle is associatedwith satisfactory clinical results after surgery.

3.1. Complications

One knee with non-union of tubercle osteotomy was successfully treated withbone grafting. Infection and non-union occurred in one knee that was satisfactorily

Fig. 1. Intra-operative photograph showed triple positioning (medial, anterior andproximal)of tibial tubercle osteotomy to optimize the correction of patellofemoral malalignment.

treated with surgical debridement and bone grafting and antibiotic therapy. Twelveknees (19%) had screw removal due to local pain and irritation.

4. Discussion

The results of the current study showed that this new surgical tech-nique produced satisfactory outcome in 78.9% at one-year and 70.9% at10-year follow-up. Eighteen knees (21%) showed unsatisfactory resultsat 10 years due to reasons including the natural progression of osteoar-thritis with age, the articular cartilage damage, inadequate surgicaltechnique and socioeconomic factors. In this technique, arthroscopyprovides prudent information in the assessment of articular cartilagedamage and the treatment strategy. Lateral retinacular release relievesthe lateral strain due to tight lateral retinaculum. Triple positioning oftibial tubercle osteotomy is patient specific, and themagnitudes of tibialtubercle transfer in three directions must be individualized. Therefore,patient selection and precision in surgical technique with adequate

Table 3Pain score, Lysholm score, Kujala score and Q angle.

Preoperative Postoperative(1 year)

Postoperative(10 years)

p-values

⁎Pain score p-1b0.001Mean±SD 3.9±4.1 21.8±4.1 21.3±4.1 p-2b0.001

(Range)⁎ (0–15) (10–25) (10–25) p-3=0.057Lysholm score p-1b0.001Mean±SD 71.1±15.7 91.3±9.3 91.0±7.8 p-2b0.001

(Range) (21–95) (70–100) (70–100) p-3=0.597Kujala score p-1b0.001Mean±SD 34.7±11.6 88.3±9.9 87.4±8.6 p-2b0.001

(Range) (14–54) (65–100) (69–100) p-3=0.213The Q angle ( ) p-2=0.001Mean±SD 17.6±5.2 15.3±3.0(Range) (10–30) (9–22)

p-1: Pre-op vs Post-op (1 year); p-2: Pre-op. vs Post-op. (10 years); p-3: Post-op(1 year) vs Post-op. (10 years). p-values: Comparison of pre- and postoperative dataat 1 and 10 years using paired t test.⁎ 0—severe pain; 5—moderated and continuous pain; 10—moderated and occasional

pain; 15—mild pain on walking and stairs; 20—mild pain on stairs only; 25—no pain.

Table 4The correlation of clinical outcome with the inherent factors.

Satisfactory Unsatisfactory p-value

Clinical outcome 44 (71%) 18 (29%)Age (years) 49±12 53±11 0.635Sex (female/male) 38/6 14/4 0.945Body height (cm) 157±10 161±15 0.765Body weight (kg) 65±9 68±12 0.625Pre-op. pain score 7.5±4.1 6.9±4.2 0.703

p-value: Comparison of satisfactory and unsatisfactory knees in different factors usingpaired t test.

Table 6The correlation of clinical outcome with radiographic evaluation.

Patella height⁎ Satisfactory Unsatisfactory p-value

Preoperative 1.09±0.18 1.1±0.18 0.765Postoperative 0.86±0.17 0.89±0.20 0.310

Congruence angle# Normal Abnormal Normal Abnormal p-value

Preoperative 19(43%) 25(57%) 9(50%) 9(69%) 0.973Postoperative 35(87%) 9(20%) 10(56%) 8 (44%) 0.021

⁎ The Insall–Salvati method was used in the measurement of patellar height.# The congruence angle is measured between the apex of patella to bisected femoral

trochlea on Merchant view. The congruence angle is positive if lateral to the congru-ence line, and negative if medial to the congruence line. The normal congruenceangle is defined as b−16 degrees.

136 C.-J. Wang et al. / The Knee 21 (2014) 133–137

improvement in the correction of the congruence angle on radiographand a reduction in patellofemoral pressure appear to be critical to thesuccess of this operation in long-term.

Many factors may contribute to the etiology of patellofemoral disor-ders including joint geometry, limb alignment and soft tissue imbalance[3,8,10,21,29–32]. Several surgical techniques are designed to improvethe patellofemoral malalignment including lateral retinacular release,proximal realignment, medial patellofemoral ligament reconstructionand distal realignment with tibial tubercle transfer [14–18,20,29]. For pa-tients with patellofemoral malalignment, tibial tubercle transfer is an ef-fective surgical method with more consistent clinical results [5,16–18].However, there is a disparity of opinions on the magnitudes of tibialtubercle transfer. The Elmslie-Trillat procedure was designed to reducethe Q angle and improve the patellofemoral malalignment by medialtransfer of the tibial tubercle [13,23]. Fulkerson procedure consisted of afixed 14.8 mm anterior and 4.8 mmmedial transfers of tibial tubercle toreduce the patellofemoral contact pressure and balance of the facet pres-sure [16]. However, the severity of patellofemoral malalignment and thesize and shape of the knee showed considerable individual variations[24,25,31,32]. Therefore, surgical correction using fixed magnitudes oftibial tubercle transfer is unable to satisfy all patients. One operationthat fits all approach will predictably lead to uncertain outcomes. Withfixed tibial tubercle transfer, satisfactory results were reported rangingfrom 38% to 93% [23]. Therefore, a more versatile surgical technique thatincludes anterior, medial and proximal positioning of the tibial tubercleosteotomy to optimize the patellofemoral malalignment appears to bemore desirable for patients with patellofemoral disorders.

Previous cadaver study revealed that combination of anterior, me-dial and proximal transfer of the tibial tubercle can reduce thepatellofemoral contact pressure and alter the patellofemoral contactarea in a ‘see-saw’ phenomenon [6,7,24,31,32]. Triple positioning oftibial tubercle osteotomy was then established in the treatment ofpatellofemoral disorders. We believed that 3-dimensional translationof tibial tubercle provides better reduction in patellofemoral contactpressure and alter the patellofemoral contact area. Additional lateralreticular release further reduces the lateral retinacular strain due totight lateral retinaculum.

Patellofemoral disorder with osteoarthritic changes differs frompatellofemoral malalignment without OA changes. The current method

Table 5The correlation of clinical outcome with the severity of cartilage damage.

Satisfactory Unsatisfactory p-value-1

Clinical outcome 44 (71%) 18 (29%)Articular cartilage lesion

b5% involvement 21 (47.7%) 0 b0.0015–10% involvement 18 (40.9%) 5 (27.8%) 0.00710–15% involvement 5 (11.4%) 4 (22.2%) 0.73915–20% involvement 0 3 (16.7%) b0.001>20% involvement 0 6 (33.3%) b0.001p-value-2 b0.001

p-value-1: Comparisons of satisfactory and unsatisfactory knees in different categoryusing paired t test. p-value-2: Overall comparison of satisfactory and unsatisfactoryknees using Chi Square test.

appears to be equally effective in patients with patellofemoral disordersassociated with mild to moderate osteoarthritis of the patellofemoraljoint.

In conclusion, triple positioning of tibial tubercle osteotomy for re-fractory patellofemoral disorder yields satisfactory long-term resultswith 10-year follow-up.

5. Conflict of interest statement

No external funding from a commercial party was received for thisproject. The authors declared that they did not receive any honorariaor consultancy fee for the writing of this paper. The authors furtherdeclaimed no conflict of interest in term of stock or option owner-ships from any company.

Acknowledgment

No fund was received or will be received from a commercial partydirectly or indirectly related to the subject in this study.

DisclosureThe authors declare that they did not receive any honoraria or

consultancy fee in the writing of this article. The authors further dis-close no conflict of interest.

References

[1] Colvin Alexis Chiang, West Robin V. Patellar instability. J Bone Joint Surg Am2008;90:2751-62.

[2] Collado H, Fredericson M. Patellofemoral pain syndrome. Clin Sports Med2010;29(3):379-98.

[3] Dejour H, Walch G, Nove-Josserand L, Guier C. Factors of patella instability: an an-atomical radiographic study. Knee Surg Sports Traumatol Arthrosc 1994;2:19-26.

[4] Goudakos IG, Konig C, Schottle PB, Taylor WR, Hoffmann JE, Popplau BM, et al.Regulation of the patellofemoral contact area: an essential mechanism inpatellofemoral joint mechanics? J Biomech 2010;43(16):3237-9.

[5] Mani S, Kirkpatrick MS, Saranathan A, Smith LG, Cosgarea AJ, Elias JJ. Tibial tuber-osity osteotomy for patellofemoral realignment alters tibiofemoral kinematics.Am J Sports Med 2011;39(5):1024-31.

[6] Koshino T. Changes in patellofemoral compression force after anterior oranteromedial displacement of tibial tuberosity for chondromalacia patellae. ClinOrthop Relat Res 1991;266:133-8.

[7] Van Eijden TM, Kouwenhoven E, Weijs WA. The influence of anterior displace-ment of the tibial tuberosity on patellofemoral biomechanics. Int Orthop1987;11(3):215-21.

[8] Pal S, Draper CE, Fredericson M, Gold GE, Delp SL, Beaupre GS, et al. Patellarmaltracking correlates with vastus medialis activation delay in patellofemoralpain patients. Am J Sports Med 2011;39(3):590-8.

[9] Servien E, Verdonk PC, Neyret P. Tibial tuberosity transfer for episodic patellardislocation. Sports Med Arthrosc 2007;15(2):61-7.

[10] Oberlander MA, Baker CL, Morgan BE. Patellofemoral arthrosis: the treatmentoptions. Am J Orthop (Belle Mead NJ) 1998;27(4):263-70.

[11] Crossley KM, Marino GP, Macilquham MD, Schache AG, Hinman RS. Can patellarbrace reduce the patellar malalignment and pain associated with patellofemoralosteoarthritis? Arthritis Rheum 2009;61(12):1719-25.

[12] Wetzels T, Bellemans J. Patellofemoral osteoarthritis treated by partial lateralfacetectomy: results at long-term follow up. Knee 2012;19(4):411–5.

[13] Barber FA, McGarry JE. Elmslie-Trillat procedure for the treatment of recurrentpatellar instability. Arthroscopy 2008;24(1):77-81.

137C.-J. Wang et al. / The Knee 21 (2014) 133–137

[14] Fisher B, Nyland J, Brand E, Curtin B. Medial patellofemoral ligament reconstruc-tion for recurrent patellar dislocation: a systematic review including rehabilita-tion and return-to-sports efficacy. Arthroscopy 2010;26(10):1384-94.

[15] Ostermeier S, Stubenberg-Colsman C, Hurschler C, Wirth CJ. In vitro investigationof the effect of medial patellofemoral ligament reconstruction and medial tibialtuberosity transfer on lateral patellar stability. Arthroscopy 2006;22(3):308-19.

[16] Fulkerson JP. Anteromedialization of the tibial tuberosity for patellofemoralmalalignment. Clin Orthop Relat Res 1983;177:176-81.

[17] Karlsson J, Lansinger O, Sward L. Anterior advancement of the tibial tuberosity inthe treatment of patellofemoral pain syndrome. Arch Orthop Trauma Surg1985;103(6):392-5.

[18] Morshuis WJ, Pavlov PW, deRooy KP. Anteriomedialization of the tibial tuberosityin the treatment of paellofemoral pain and malalignment. Clin Orthop Relat Res1990;255:242-50.

[19] Pritsch T, Haim A, Arbel R, Snir N, Shasha N, Dekel S. Tailored tibial tubercle trans-fer for patellofemoral malalignment: analysis of clinical outcomes. Knee SurgSports Traumatol Arthrosc 2007;15(8):994–1002.

[20] Koskinen SK, Hurme M, Kujala UM. Restoration of patellofemoral congruity bycombined lateral release and tibial tuberosity transposition as assessed by MRIanalysis. Int Orthop 1991;15(4):363-6.

[21] Wang CJ, Chan YS, Chen HH, Wu ST. Factors affecting the outcome of distalrealignment for patellofemoral disorder of the knee. Knee 2005;12:195-200.

[22] Akgun U, Nuran R, Karahan M. Modified Fulkerson osteotomy in recurrentpatellofemoral dislocations. Acta Orthop Traumatol Turc 2010;44(1):27-35.

[23] Dannawi Z, Khanduja V, Palmer CR, El-Zebdeh M. Evaluation of the modifiedElmslie-Trillat procedure for patellofemoral dysfunction.Orthopedics 2010;33(1):13.

[24] Wang CJ, Chen LM, Chen YC, Chan YS. The influence of tibial tubercle transfer onpatellofemoral contact: A cadaver study. J Orthop Surg 1998;6(1):31-9.

[25] Noyes FR, Stabler CL. A system for grading articular cartilage lesions at arthrosco-py. Am J Sports Med 1989;17(4):505-13.

[26] Kujala UM, Jaakkola LH, Koskinen SK, Taimela S, Hurme M, Nelimarkka O. Scoringof patellofemoral disorders. Arthroscopy 1993;9(2):159-63.

[27] Lysholm J, Gillquist J. Evaluation of knee ligament surgery results with specialemphasis on use of a scrong scale. Am J Sports Med 1982;10:150-4.

[28] Kijowskin R, Blankenbaker D, Stanton P, Fine J, De Smet A. Correlation betweenradiographic findings of osteoarthritis and arthroscopic findings of articular carti-lage degeneration within the patellofemoral joint. Skeletal Radiol 2006;35(12):895-902.

[29] Wu CC. Combined lateral retinacular release with drilling chondroplasty for treat-ment of patellofemoral osteoarthritis associated with patellar malalignment inelderly patients. Knee 2011;18(1):24-9.

[30] Weaver JK,WiedrD, DerkashRS. Patellofemoral arthritis resulting frommalalignment.A long-term evaluation of treatment options. Orthop Rev 1991;20(12):1075-81.

[31] Quintelier J, Skalski K, Verdonk P, De Baets P, Almqvist F. Patellofemoral contactpressure. Acta Bioeng Biomech 2008;10(2):23-8.

[32] Kuroda R, Kambic H, Valdevit A, Andrish JT. Articular cartilage contact pressureafter tibial tuberosity transfer. A cadaver study. Am J Sports Med 2001;29(4):403-9.