Secondary Logo

Journal Logo


External Tibial Torsion

An Underrecognized Cause of Recurrent Patellar Dislocation

Cameron, John*; Saha, Sandipan**

Author Information
Clinical Orthopaedics and Related Research: July 1996 - Volume 328 - Issue - p 177-184
  • Free


There are many surgical procedures for patellar instability, including lateral retinacular release, lateral release combined with medial capsular plication, proximal realignment, distal realignment, proximal and distal realignment, and patellectomy. Outcome studies of these techniques consistently show poor results when the quadriceps angle has been inadequately corrected.2-4,6,19 Isolated external tibial torsion is a rare cause of patellofemoral instability. The senior author (JCC), between 1990 to 1993, operated on 214 knees with patellar instability of various causes. Eighteen were identified as cases of external tibial torsion (8%). Traditional patellar realignment procedures are unsuccessful in this group because the quadriceps angle is not corrected. The quadriceps patellar tendon tracking mechanism and the quadriceps angle for these patients can be improved only by derotating the tibia. Biomechanical studies have confirmed the relationship between external tibial torsion and patellofemoral instability.22

The etiology of tibial torsion appears to be a multifactorial combination of genetics, evolution, mechanical forces, and intrauterine position that continues in the growing child, reaching the adult level of approximately 20 ° by the age of 5 years.7,11 External tibial torsion can be screened for by noting the appearance of the squinting patella or inwardly pointing knee while the patient is standing (Fig 1). This condition has been ascribed mainly to femoral anteversion, but it can be caused by primary external tibial torsion in the absence of femoral anteversion.5,19 When these patients point their feet forward, the patella points inward, and examination of the hips demonstrates equal internal and external rotation. Computed tomography is the gold standard for quantifying the amount of torsion, but clinical examination generally yields information that is just as accurate.20 When the tibial torsion is marked, radiographs may show the femur in the anteroposterior (AP) projection and the tibia in the oblique projection (Fig 2).

Rotational tibial osteotomies have been used reluctantly in the pediatric population for idiopathic tibial torsion that causes out toeing.14 Derotational Maquet tibial osteotomy has been used successfully in treating patients for chronic knee pain.5 A group of patients with severe external tibial torsion were identified whose symptoms were unrelieved by traditional surgical techniques to realign the patella. For this series of patients, a derotational tibial osteotomy was performed when the preoperative problems were attributable to recurrent patellar dislocation.


The indication for surgery in patients with external tibial torsion was recurrent patellar dislocation. Five of these patients also had longstanding anterior knee pain and were treated operatively with a Maquet type osteotomy in addition to the derotational osteotomy. Patients in this study underwent initial conservative treatment consisting of a minimum of 12 months of physiotherapy that emphasized quadriceps strengthening programs. This involved static quadriceps exercises and straight leg raising with avoidance of resisted extension exercises. The patients continued to have recurrent patellar dislocation, and all subsequently underwent surgery. Only patients with equal internal and external hip rotation and isolated external tibial torsion were included in the study. All patients had reached skeletal maturity.

Of the 17 patients (18 knees) operated on between 1990 and 1993, 16 patients (17 knees, 9 right and 8 left) in the study were available at followup (94%). The remaining patient was unavailable for followup, and the outcome of this patient is unknown. All of the patients were female, with an average age of 27.6 years (range, 14-42 years). Prior unsuccessful surgical procedures included lateral release (5 knees), Maquet procedure (3 knees), Hauser procedure (5 knees), medialization of the patellar tendon (3 knees), and semitendinosus tenodesis (1 knee).

Tibial rotation was assessed goniometrically (Fig 3). The amount of surgical correction was calculated such that the rotation in the involved knee would match that of the uninvolved knee, which was presumed to be in correct anatomic alignment, as evidenced by lack of symptomatology. In addition to the rotational abnormalities, 9 patients also had an associated patella alta, as determined radiographically and clinically using the method of Insall and Salvati.13

Surgical Technique

The operations were performed through a straight anterior incision and a lateral parapatellar arthrotomy. The patellar tendon is identified medially and laterally to its insertion. The inner aspect of the fibular head is resected, leaving the lateral cortex and the attachment of the lateral ligament to protect the common peroneal nerve. Two Kirschner wires (K wires) are placed offset from each other by the degree of correction, with 1 above and 1 below the osteotomy site. A transverse osteotomy of the tibia is made 2.5 cm distal to the articular surface of the tibia, just proximal to the patellar tendon insertion. The distal fragment is internally rotated to bring the K wires parallel to achieve the desired degree of correction (Fig 4). Selected patients with well established anterior knee pain also undergo a Maquet type osteotomy with anterior translation of the distal fragment 1.5 cm to relieve the patellofemoral joint reaction force. Staples are used to fix the osteotomy and are verified radiographically (Fig 5). A distal transfer of the patellar tendon insertion is used for correction in patients with patella alta. This is held with 2 cancellous screws and washers. The postoperative course includes progressive range of motion (ROM) with the patient avoiding weightbearing for 5 weeks, after which union of the osteotomy site is verified radiographically and the patient is put on a combined quadriceps and hamstring rehabilitation program.

Postoperative followup periods ranged from 1 to 3.2 years, with a mean followup of 25 months (only 4 patients had a followup of less than 24 months). The charts for all 16 patients (17 knees) were reviewed for details of the history, preoperative physical findings, and documented postoperative problems. Postoperative assessment included an objective clinical examination that evaluated ROM, crepitus, thigh girth, quadriceps strength, effusion, and pain to palpation. A subjective functional assessment also was performed using a modified Lysholm Functional Knee Score questionnaire21(Table 1); the assessment yielded preoperative and postoperative scores from 0 to 100. Clinical and functional subjective scores were compared with preoperative and postoperative findings to determine which factors affected outcome. A Student's t-test and chi-square analysis were used to identify findings associated with significantly more favorable, or with compromised, results.

Outcome assessment was divided into excellent, good, or poor using the following criteria: excellent, postoperative subjective (Lysholm) score of 80-100 with no episodes of patellar dislocation; good, a postoperative improvement of 20 or greater in the subjective score with an absolute postoperative score of less than 80 and no episodes of patellar dislocation; and poor, a postoperative improvement of less than 20 with episodes of patellar dislocation.


The preoperative external tibial torsion averaged 40 ° (range, 37 °-65 °) with an average rotational correction of 24 ° (range, 15 °-40 °). The preoperative quadriceps angle averaged 23 ° (±5 °), with an average postoperative quadriceps angle of 14 ° (±3 °). The average preoperative functional knee score was 40 (±22), and the postoperative score was 70 (±28). Individual patient knee data are listed in Table 2.

Outcome assessment on the 17 knees showed 8 excellent, 5 good, and 4 poor results. Overall, 13 of the 17 (76%) knees were rated good to excellent. Of the 5 patients with concurrent anterior knee pain, 4 obtained good to excellent results when treated with the combined derotational Maquet osteotomy. Patients with less painful symptoms preoperatively had significantly better outcomes (p < 0.01). Knees that had undergone multiple unsuccessful surgical procedures had significantly poorer outcomes (p < 0.01). Table 3 lists those variables that had significant impact on outcome (p < 0.05).

All patients reported that activities undergone preoperatively were done so with hesitation for fear of dislocation and pain; postoperatively, 12 patients reported these same activities were able to be performed without such fears. Nine of these patients also were able to increase their activity level, and 2 patients returned to work after being on disability preoperatively. Most patients were able to discontinue the use of antiinflammatory medications and pain killers postoperatively.

The only complication noted was staple pain at the osteotomy site in 2 patients. This was eliminated with subsequent removal of the staple. Ligamentous laxity after derotation was not detected clinically in any patient postoperatively.


The derotational osteotomy presented in this study realigns the extensor mechanism in cases of recurrent patellar dislocation secondary to external tibial torsion. Patellofemoral instability and symptomatic chondromalacia patellae have been documented in patients with external tibial torsion values of greater than 24 °.23 Medialization of the extensor mechanism to a more anatomically correct alignment in patients with such large external tibial torsion values can be obtained only via derotation of the tibia because of the technical restrictions of present tibial tubercle medialization procedures6,8,16,17 to a maximal rotational correction of 20 °. Patients with concurrent anterior knee pain and treated with the combined derotational Maquet osteotomy had favorable outcomes, with 80% obtaining good to excellent results. This illustrates the need for anteriorization in addition to medialization for patients with instability and anterior knee pain.

Knees that had been operated on multiple times were associated with a poorer result (p < 0.01), as has been observed in other realignment procedures.1 These prior operations were uniplanar techniques that attempted to solve a rotational problem and as such were ineffective in correcting the malalignment. As a result, the damage caused by persistent patellar maltracking was too advanced to benefit from proper realignment of the extensor mechanism. Knees that subjectively and functionally demonstrated less painful symptoms preoperatively were associated with excellent results (p < 0.01), probably reflecting the lack of anterior knee pain symptomatology. Knees with a poor result have not undergone additional surgery. All patients had some improvement with the surgery, but the patients with poor outcome continued to have anterior knee pain. Patellectomy may be considered in the future if pain becomes significantly disabling. The amount of documented preoperative chondromalacia patellae did not affect outcome, a finding that has been observed in other studies.10,12,15,18 Other factors, including a history of knee injury, amount of preoperative tibial torsion, amount of rotational correction, patient age at the time of operation, evidence of a hypoplastic trochlea, preoperative patella alta, and length of followup did not affect outcome.

Cox6 has recommended derotation osteotomies in cases of external tibial torsion. Fulkerson and Schutzer9 report that some patients may require unique surgical approaches when excessive femoral anteversion or extreme knee valgus significantly contributes to patellofemoral malalignment; the authors recommend that external tibial torsion be included in this list, with derotational osteotomy of the tibia being the procedure of choice. Only by correcting the biomechanical axis via derotational osteotomy in patients with large quadriceps angles secondary to isolated external tibial torsion can such patients be effectively treated.


The results of this study suggest that patients with patellofemoral disorders be screened for isolated external tibial torsion. In patients with external tibial torsion that requires rotational correction of greater than 20 °, the derotational tibial osteotomy should be used to achieve medialization in patients with recurrent patellar dislocation. In patients with concurrent established anterior knee pain, anteromedialization of the extensor mechanism can be attained with the use of a Maquet type osteotomy in addition to the derotational osteotomy.


The authors thank Dr. I. J. Harrington for reviewing this article. They also thank Mrs. E. Wood for her assistance with the study.

Fig 1
Fig 1:
. The inwardly pointing right knee of a 28-year-old man with 53 ° of external tibial torsion and no femoral anteversion.
Fig 2A-B
Fig 2A-B:
. Radiographs of a patient with marked right tibial torsion. This patient had a prior Hauser procedure. (A) AP radiograph. Note that the femur appears in the normal AP projection, but the tibia appears in the oblique projection, as evidenced by the prominence of the fibula. (B) Lateral radiograph illustrating the oblique projection of the tibia and fibula.
Fig 3
Fig 3:
. Measurement of tibial rotation using a goniometer. The subject is positioned prone, the knee flexed to 90 °, and the ankle positioned in neutral dorsiflexion/plantar flexion. Line A corresponds to the transcondylar axis, and line B to the transmalleolar axis. The amount of tibial rotation equals angle AB.20 (Reproduced with permission.)
Fig 4A-B
Fig 4A-B:
. Intraoperative photos of left knee undergoing derotational osteotomy procedure. (A) The 2 K wires are shown, 1 above and 1 below the osteotomy site, offset from each other by the amount of desired correction (30 ° in this case). (B) The distal fragment has been internally rotated with the K wires now parallel. A lateral staple is used to fix the positioning.
Fig 5A-B
Fig 5A-B:
. Postoperative radiographs, right knee. (A) AP radiograph showing staple placement. (B) Lateral radiograph illustrating the combined Maquet type osteotomy with the derotational osteotomy. The tibia is advanced anteriorly 1.5 cm.


1. Aglietti P, Buzzi R, De Biase P, Giron F: Surgical treatment of recurrent dislocation of the patella. Clin Orthop 308:8-17, 1994.
2. Bessette GC, Hunter RE: The Maquet procedure: A retrospective review. Clin Orthop 232:159-167, 1988.
3. Brief LP: Lateral patellar instability: Treatment with a combined open-arthroscopic approach. Arthroscopy 9:617-623, 1993.
4. Brown DE, Alexander AH, Lichtman DM: The Elmslie-Trillat procedure: Evaluation in patellar dislocation and subluxation. Am J Sports Med 12:104-109, 1984.
5. Cooke TDV, Price N, Fisher B, Hedden D: The inwardly pointing knee: An unrecognized problem of external rotational malalignment. Clin Orthop 260:56-60, 1990.
6. Cox JS: Evaluation of the Roux-Elsmlie-Trillat procedure for knee extensor realignment. Am J Sports Med 10:303-310, 1982.
7. Eckhoff DG: Effect of limb malrotation on malalignment and osteoarthritis. Orthop Clin North Am 25:405-414, 1994.
8. Fulkerson JP, Becker GJ, Meaney JA, Miranda M, Folcik MA: Anteromedial tubercle transfer without bone graft. Am J Sports Med 18:490-497, 1990.
9. Fulkerson JP, Schutzer SF: After failure of conservative treatment for painful patellofemoral malalignment: Lateral release or realignment? Orthop Clin North Am 17:283-287, 1986.
10. Gecha SR, Torg JS: Clinical prognosticators for the efficacy of retinacular release surgery to treat patellofemoral pain. Clin Orthop 253:203-208, 1990.
11. Guidera KJ, Ganey TM, Keneally CR, Ogden JA: The embryology of lower extremity torsion. Clin Orthop 302:17-21, 1994.
12. Huid I, Anderson LI, Schmidt H: Chondromalacia patellae: The relationship to abnormal patellofemoral joint mechanics. Acta Orthop Scand 52:661-666, 1981.
13. Insall J, Salvati E: Patella position in the normal knee joint. Radiology 101:101-104, 1971.
14. Krengel WF, Staheli LT: Tibial rotational osteotomy for idiopathic torsion: A comparison of the proximal and distal osteotomy levels. Clin Orthop 283:285-289, 1992.
15. Metcalf RW: An arthroscopic method for lateral release of the subluxating or dislocating patella. Clin Orthop 167:9-18, 1982.
16. Miller BJ, Larochelle PJ: The treatment of patellofemoral pain by combined rotation and elevation of the tibial tubercle. J Bone Joint Surg 68A:419-423, 1986.
17. Noll BJ, Ben-Itzhak I, Rossouw P: Modified technique for tibial tubercle elevation with realignment for patellofemoral pain. Clin Orthop 234:178-182, 1988.
18. Royle SG, Noble J, Davies DR, Kay PR: The significance of chondromalacic changes on the patella. Arthroscopy 7:158-160, 1992.
19. Scuderi GR: Surgical treatment for patellar instability. Orthop Clin North Am 23:619-630, 1992.
20. Stuberg W, Temme J, Kaplan P, Clarke A, Fuchs R: Measurement of tibial torsion and thigh-foot angle using goniometry and computed tomography. Clin Orthop 272:208-212, 1991.
21. Tegner Y, Lysholm J: Rating systems in the evaluation of knee ligament injuries. Clin Orthop 198:43-49, 1985.
22. Turner MS, Smillie IS: The effect of tibial torsion on the pathology of the knee. J Bone Joint Surg 63B:396-398, 1981.

Section Description



© Lippincott-Raven Publishers.