Secondary Logo

Journal Logo

SECTION I: SYMPOSIUM: Papers Presented at the 2006 Meeting of the Knee Society

Total Knee Arthroplasty in Patients with Greater than 20 Degrees Flexion Contracture

Berend, Keith, R; Lombardi, Adolph, V, Jr.; Adams, Joanne, B

Section Editor(s): Laskin, Richard S MD, Guest Editor

Author Information
Clinical Orthopaedics and Related Research: November 2006 - Volume 452 - Issue - p 83-87
doi: 10.1097/01.blo.0000238801.90090.59


Because of the difficulties correcting fixed flexion contractures when performing TKA, we previously published a grading system for rating contracture.9-12 For each increasing grade we recommend a slightly different approach to soft tissue release, bony resection, and degree of prosthesis constraint.9-12 Rigid application of the algorithm would warrant use of a posterior stabilized condylar knee design in most cases of fixed preoperative flexion contracture of 20° or greater and a constrained design when ligamentous instability results from ligamentous balancing with extensive releases. Other authors have suggested correction of a fixed flexion contracture does not require cruciate substitution, additional bony resection, or increased constraint.17,20,23 The definition of a severe or surgically important preoperative flexion contracture has been debated.3,4,8-11,13,16,17,20,22,23 We initially defined 30° as severe.9,10 We have also suggested contractures greater than 15° be treated with substitution of the posterior cruciate ligament.12 Although there are sporadic reports of very severe contractures, it seems the current literature suggests a clinically important flexion contracture is present with a deformity of 20° or greater.1,17,23

We sought to examine the clinical results of primary TKA in patients with preoperative fixed flexion deformity of 20° or greater by evaluating final range of motion (ROM), Knee Society scores (KSS), residual flexion contracture, and subsequent surgery. Second, we sought to evaluate our adherence to the aforementioned algorithm in these patients. Additionally, we hypothesized no residual contracture would be present only if the algorithm was followed.


We retrospectively reviewed the electronic clinical database (Documed v2002; Documed, Inc, Ann Arbor, MI) of the senior authors (AVL, KRB), which revealed 52 knees in 40 consecutive patients with a fixed flexion contracture of 20° or greater who underwent primary TKA (2.7%) from January 2000 to December 2003. During the same period, 1943 primary TKAs were registered in the database. Inclusion criteria for study were primary TKA and preexisting fixed flexion contracture of 20° or more. No patients meeting these inclusion criteria were excluded from study. The average age at the time of TKA was 63 years (range, 42-95 years; standard deviation [SD], 13 years). There were 35 women (67%) and 28 right TKAs (54%). The minimum followup was 1.6 months (mean 38 months, range, 1.6-77 months). Two patients (three knees) died of causes unrelated to the procedure during the followup period and were included. One patient was lost to followup with last contact and examination at 10 months. Therefore all 52 patients were included for 6-week results, and 49 TKAs were reviewed for final outcome.

Twelve patients (24 knees) with bilateral contractures of 20° or more had the procedure performed simultaneously under a single anesthesia. Four patients (four knees) had bilateral simultaneous TKAs performed, but with only one knee having a fixed flexion contracture of 20° or more. Staged bilateral TKAs were performed 6 to 8 weeks apart in three knees with flexion contractures of 20° or more (three patients). The remaining 21 patients (21 knees) had isolated single-sided procedures.

At the time of surgery the authors attempted to achieve complete correction of the flexion contracture. Initial correction of the coronal deformity was accomplished by medial exposure, osteophyte removal, and posterior capsular soft tissue release. The distal femur resection was then performed using a measured resection technique. An additional 2-mm resection of distal femoral condyle is routinely done in flexion deformities of 20° or more. A trial reduction with a cruciate-retaining (CR) implant then was performed and the collateral ligaments and posterior cruciate ligament (PCL) balanced.

If full correction and full extension were not achieved, the PCL was resected and a posterior stabilized device (PS) was used. When full extension was not obtained by this point, an additional femoral resection of up to 2 to 3 mm was performed to equalize the flexion/extension gaps.

If contracture persisted after additional distal femoral bony resection, additional soft tissue release from the contracted side of the deformity was performed until full extension was achieved. If this ligamentous release and additional bony resection resulted in ligamentous instability, a posterior stabilized constrained device (PSC) was used. Infrequently, both collateral ligaments are released resulting in instability of a constrained design. In these cases, a rotating hinged component may be used.

The Maxim Complete Knee System (Biomet Inc, Warsaw, IN) was used in patients with posterior cruciate retaining (PCR), PS, or PSC implant designs. The AVL Antiluxation rotating hinge knee (Biomet Inc) was used in patients with a hinged design.

Postoperatively, patients were allowed full weightbearing with an assistive device. They started ROM exercises the day of surgery. A splint or immobilizer was worn at night for several weeks to prevent recurrence. Patients were examined at 2 to 3 weeks and 6 weeks postoperatively and yearly thereafter.

Final ROM was measured by clinical evaluation at 6 weeks, 6 months, and annually thereafter. Measurements were obtained by the operative surgeon (KRB, AVL) and were not blinded to examiner. Outcome measures examined were preoperative and postoperative ROM, KSS, and amount and incidence of residual flexion contracture. Flexion contracture was categorized as none, less than 10°, 11° to 15°, 16° to 20°, and greater than 20° for outcome interpretation. The type of TKA device, level of constraint, and subsequent procedures are reported.

Statistical analyses were performed using StatsDirect (Stats Direct, Ltd; Cheshire, UK). All analyses were performed using 95% confidence intervals (CI). Multivariate analysis and oneway analysis of variance (ANOVA) were used to determine if preoperative arc of motion or combined severity of sagittal and coronal deformity predicted the type of device implanted. The type of device necessary after intraoperative correction was examined to determine if there was any influence on the outcome measures of arc of motion, residual contracture, and final KSS using ANOVA (Table 1). Significance was defined as p < 0.05.

Multivariate Analysis of Final Implant Type Utilized


The average preoperative arc of motion of 90° (range, 46°-99°; SD, 14°) improved (p < 0.05) to 112° (range, 95°-125°; SD, 9°) postoperatively. The average preoperative KSS of 32 points improved (p < 0.05) to 90 points at final followup. Forty two of the 52 knees (80%) had a good or excellent KSS at most recent followup.

At the time of patient discharge 14 knees (28%) had full extension and 30 had a residual contracture of less than 10° (60%). Six weeks postoperatively, 26 knees (53%) had full extension and 21 knees (42%) had 5° to 10° contracture. At final followup, of the 49 knees with greater than 6 months followup, 33 (67%) had complete full extension and 13 (27%) had a residual contracture of less than 10°, yielding a 94% rate of knees achieving less than 10° contracture at final followup (Table 2).

Final Postoperative Flexion Contracture

In knees with no residual contracture at the time of discharge (14 TKAs), all maintained less than 10° contracture with eight knees having full extension at final followup. In knees with a flexion contracture of less than 10° at hospital discharge, 22 of 30 (73%) improved to no contracture at final followup, six had a contracture of less than 10°, and one knee worsened to a greater than 20° contracture. Of the five knees with 11° to 15° contracture at hospital discharge, three improved to full extension, one improved to a final contracture of less than 10°, and one knee had a final contracture of 11° to 15°. The one knee with greater than 20° flexion contracture at hospital discharge improved to 11° to 15° contracture.

Five of the 52 knees (10%) underwent subsequent procedures. Three manipulations were performed when motion was unacceptable to the patient and more than 90° was not achieved by 6 to 10 weeks followup. No knees underwent implant revision for residual flexion contracture or stiffness. One knee was revised for subsequent instability in the PS subset of TKA. One deep infection underwent two-stage treatment. No loosening or wear-related failures occurred. No nerve injuries were identified in any patient regardless of preoperative sagittal or coronal deformity.

A PCR device was used in 31 TKAs (60%) after the trial reduction. A PS device then was implanted in 14 patients (14 knees; 27%). A PSC design was used in five patients (five knees; 10%). A rotating hinge was used in two patients (two knees; 4%) with a gross disparity between the flexion/extension gap and substantial ligamentous instability after correction of the fixed flexion contracture.


We previously published a grading system and algorithm for addressing fixed flexion contractures in primary TKA.9-11 As noted in our previous descriptions, flexion contractures of greater than 30° were considered severe.9-11 Recent literature suggests fixed flexion contractures of 20° or more should be considered severe.1,17,23 The current study evaluates the clinical outcomes, ROM, residual flexion contracture, and rate of subsequent procedures in a group of primary TKA with preoperative flexion contracture of 20° or greater. We further examined the use and outcomes of cruciate retaining (CR), PS, PSC, and hinged prostheses for these cases. Conflicting reports suggest a PS or more highly constrained design may or may not be required to accomplish full extension in contractures greater than 20°.12,17,23 In the current series, an attempt was made to fully correct all knees to full extension intraoperatively and increasing constraint was used to achieve stability. Before hospital discharge, 84.5% of knees had full extension or less than 10° contracture. This worsened slightly at early followup, but 94.2% of knees maintained full extension or less than 10° of flexion contracture at final followup.

We note several limitations, including the retrospective nature of the study design. A prospective randomized trial of various constrained designs may provide a clearer understanding of the degree of soft tissue release and constraint necessary to achieve full extension. The followup period averaged approximately 3 years and longer followup is necessary to draw long-term conclusions about wear, loosening, and late instability. The current study includes only clinical outcome measures. Radiographic outcomes are not reported. Lastly, ROM data were measured and recorded or supervised by the operative surgeon and may represent bias.

One reported complication of treating major degrees of fixed flexion contracture is nerve injury, particularly the peroneal nerve with surgical correction of large combined deformities.5-7 No peroneal nerve palsies occurred despite large flexion contractures and 14% of knees having a pre-operative fixed valgus deformity greater than 10°. It seems the step-wise approach outlined above is a safe method of addressing these deformities.

The ability to achieve full extension has been related to preoperative severity.2,15,19,21 Tanzer and Miller reported on 35 knees in 33 patients with preoperative flexion contractures of ≤ 30°. They reported an average flexion contracture of 2.9° at final followup.22 Their study included only five knees with ≥ 20° flexion contracture.22 Of these five knees, one failed to improve postoperatively and was considered a failure.22 In the current series, only patients with flexion contractures of 20° or greater were evaluated. If a residual contracture of less than 10° is used as an indicator of success, using the methods described, a successful outcome was achieved in 94% of cases, not dissimilar from Tanzer and Miller.22

Firestone et al4 reported on results from 40 knees with greater than 20° of preoperative fixed flexion deformity. The overall residual deformity was 3.1° after surgery, which increased to 10.1° at 3 months postoperatively and decreased to 7° at 2 years postoperatively. The authors resected up to 5 mm of additional distal femur to obtain intraoperative correction, and did not observe instability, extensor lag, or decreased quadriceps function.4 Tanzer and Miller22 urged against over resection of the distal femur, postulating this can result in abnormal kinematics of the prosthetic knee. We did not specifically examine the need for distal femoral resection greater than 5 mm in the current series, but routinely resect an additional 2 mm in all cases of substantial flexion contracture. We revised one patient with instability in our PS design group (the group in which additional bony resection of up to 5 mm may have been required).

We attempted to achieve full correction of flexion deformity intraoperatively in all cases. McPherson et al16 reported improved ROM in patients treated with TKA for residual flexion contractures. In their series of 29 knees, the average postoperative flexion contracture of 10.5° decreased to 1° at 24 months followup.16 In our series, all knees with no contracture at hospital discharge were considered successful with 57% maintaining full extension and 43% having a less than 10° contracture at final followup. We noted improvement in 26 of 35 knees in the groups with residual contracture at discharge. This includes three knees of five with a residual contracture of 11° to 15° at hospital discharge and full extension at followup. Of the three knees considered clinical failures with greater than 10° contracture at final followup, one had less than 10° contracture, one had 11° to 15°, and one had greater than 20° contracture at hospital discharge.

One of the critical aspects of successfully treating fixed flexion deformity is addressing the PCL.18 Laskin et al8 observed an average postoperative residual flexion contracture of 11° in PCR designs used in patients with fixed flexion deformities greater than 10° to 15°. The current results would refute these concerns as we were able to achieve a clinically successful result in 97% of CR knees when the deformity can be fully corrected intraoperatively without creating sagittal or coronal instability.8 Posterior-stabilized designs are warranted if the deformity is not eliminated after the initial steps of correction.12 Successful results were achieved in 93% of PS TKA.

If more than 2 mm of additional distal femoral resection is performed and/or PCL resection is required to obtain full extension, we recommend using a PS or more highly constrained design. In contrast, in a series of 37 TKAs for severe flexion contracture (≥ 60°) (average, 78°), Lu et al13 reported an average final flexion deformity of 7°, with no patients having a final deformity of greater than 15° postoperatively. These good results were observed with the use of a CR device and severe contracture with large amounts of distal femoral bony resection; however PS devices were not available at the time.13 These authors now use PS devices when additional distal femoral resection is required.13

When treating fixed flexion contracture of 20° or more in primary TKA, we use the surgical algorithm outlined above (Table 3). In contrast to our previously published reports, we now base our implant selection on intraoperative findings of sagittal and coronal stability instead of preoperative severity of deformity.9-12 We continue to limit the amount of distal femoral resection to no more than 2 mm greater than the thickness of the femoral component if a PCR device is considered.14 A PS device should be used if greater than 2 mm of distal femoral resection or release of the PCL is required to obtain full extension after release of the posterior capsular tissues. Constrained TKA is used if obtaining full extension requires releasing the collateral ligaments to the point of instability. A rotating hinge is used when full extension is achieved at the expense of an unstable flexion gap and flexion instability resulting in dislocation of the constrained device. Using this updated algorithm resulted in 94% of patients achieving within 10° of full extension and 80% good to excellent results at followup. This severity of contracture is difficult to fully correct, but good results can be achieved.

Operative Algorithm Used During Study Period


1. Ansari S, Ackroyd CE, Newman JH. Kinematic posterior cruciate ligament-retaining total knee replacements: a 10-year survivorship study of 445 arthroplasties. Am J Knee Surg. 1998;11:9-14.
2. Dorr LD. Total knee replacement: from exposure to soft tissue balance. Orthopedics Today. 1993;13.
3. Faris PM, Herbst SA, Ritter MA, Keating EM. The effect of pre-operative knee deformity on the initial results of cruciate-retaining total knee arthroplasty. J Arthroplasty. 1992;7:527-530.
4. Firestone TP, Krackow KA, Davis JD, Teeny SW, Hungerford DS. The management of fixed flexion contractures during total knee arthroplasty. Clin Orthop Relat Res. 1992;284:221-227.
5. Insall JN. Technique of total knee replacement. Instr Course Lect. 1981;30:324-341.
6. Insall JN. Choices and compromises in total knee arthroplasty. Clin Ortho Relat Res. 1988;226:43-48.
7. Insall JN, Easley ME. Surgical Techniques and Instrumentation in Total Knee Arthroplasty. In: Insall JN, Scott WN, ed. Surgery of the Knee, 3rd Ed. Philadelphia: Churchill Livingstone; 2001:1553-1620.
8. Laskin RS, Rieger M, Schob C, Turen C. The posterior-stabilized total knee prosthesis in the knee with severe fixed deformity. Am J Knee Surg. 1988;1:199-203.
9. Lombardi AV Jr. Soft tissue balancing the knee-flexion. In: Callaghan JJ, Rosenberg AG, Rubash HE, Simonian PT, Wickiewicz TL, eds. The Adult Knee, Vol 2. Philadelphia: Lippincott Williams & Wilkins; 2003:1223-1232.
10. Lombardi AV Jr, Mallory TH. Dealing with flexion contractures in total knee arthroplasty. Bone resection versus soft tissue releases. In: Insall JN, Scott WN, Scuderi GR, eds. Current Concepts in Primary and Revision Total Knee Arthroplasty. Philadelphia: Lippincott-Raven Publishers; 1996:191-202.
11. Lombardi AV Jr, Mallory TH, Adams JB, Herrington SM. A stepwise algorithmic approach to flexion contractures in total knee arthroplasty. Arch Am Acad Orthop Surg. 1997;1:1-8. (Selected scientific exhibits)
12. Lombardi AV Jr, Mallory TH, Fada RA, Hartman JF, Capps SG, Kefauver CA, Adams JB. An algorithm for the posterior cruciate ligament in total knee arthroplasty. Clin Orthop Relat Res. 2001; 392:75-87.
13. Lu H, Mow CS, Lin J. Total knee arthroplasty in the presence of severe flexion contracture. J Arthroplasty. 1999;14:775-780.
14. Martin JW, Whiteside LA. The influence of joint line position on knee stability after condylar knee arthroplasty. Clin Orthop Relat Res. 1990;259:146-156.
15. McAuley JP, Harrer MF, Ammeen D, Engh GA. Outcome of knee arthroplasty in patients with poor preoperative range of motion. Clin Orthop Relat Res. 2002;404:203-207.
16. McPherson EJ, Cushner FD, Schiff CF, Friedman RJ. Natural history of uncorrected flexion contractures following total knee arthroplasty. J Arthroplasty. 1994;9:499-502.
17. Mihalko WM, Whiteside LA. Bone resection and ligament treatment for flexion contracture in knee arthroplasty. Clin Orthop Relat Res. 2003;405:141-147.
18. Ritter MA, Faris PM, Keating EM. Posterior cruciate ligament balancing during total knee arthroplasty. J Arthroplasty. 1988;3: 323-326.
19. Ritter MA, Harty LD, Davis KE, Meding JB, Berend ME. Predicting range of motion after total knee arthroplasty. J Bone Joint Surg Am. 2003;85:1278-1285.
20. Rodriguez-Merchan EC. Correction of fixed contractures during total knee arthroplasty in haemophiliacs. Haemophilia. 1999;5: 33-38.
21. Schurman DJ, Parker JN, Ornstein D. Total condylar knee replacement: a study of factors influencing range of motion as late as two years after arthroplasty. J Bone Joint Surg Am. 1985;67:1006-1014.
22. Tanzer M, Miller J. The natural history of flexion contracture in total knee arthroplasty: a prospective study. Clin Orthop Relat Res. 1989;248:129-134.
23. Whiteside LA, Mihalko WM. Surgical procedure for flexion contracture and recurvatum in total knee arthroplasty. Clin Orthop Relat Res. 2002;404:189-195.
© 2006 Lippincott Williams & Wilkins, Inc.