- Email: [email protected]
Conservative Treatment for the Intraoperative Detachment of Medial Collateral Ligament from the Tibial Attachment Site during Primary Total Knee Arthroplasty
The Journal of Arthroplasty Vol. 24 No. 8 2009
Conservative Treatment for the Intraoperative Detachment of Medial Collateral Ligament from the Tibial Attachment Site during Primary Total Knee Arthroplasty Min-Hoi Koo, MD, and Choong H. Choi, MD
Abstract: For intraoperative injuries of the medial collateral ligament (MCL) during primary total knee arthroplasty (TKA), many authors recommended complex reconstructive procedures or conversion to the constrained prosthesis in previous literature. However, it is well known that medial collateral ligament has good healing potential after injuries. This retrospective study evaluated the clinical and radiological results of 15 primary TKAs complicated with intraoperative complete detachment of the MCL from tibial attachment site, which were all treated solely by nonoperative conservative treatment without any other additive procedures or braces. Compared to the MCL-intact contralateral knees, there was no significant difference in terms of clinical and radiological outcome at minimum of 2 postoperative year. There was no case showing clinical instability, either. Solely, the nonoperative conservative treatment for intraoperative injuries of the MCL during primary TKAs is expected not only to be effective with satisfactory results but also to decrease complications derived from other sophisticated reparative or reconstructive procedures. Keywords: intraoperative MCL injury, conservative treatment, total knee arthroplasty. © 2009 Elsevier Inc. All rights reserved.
In evaluating the flexion and extension gap during total knee arthroplasty (TKA), it is very important to balance the medial-lateral soft tissue tension. In most cases with degenerative osteoarthritis, varus deformity is mainly combined, increasing the necessity for the release of the contracted medial soft structures to balance mediallateral soft tissue tension. Removal of osteophytes formed around the medial femoral and tibial condyles, release of medial joint capsule containing the deep medial collateral ligament, and additional gradual release of the superficial medial collateral ligament are generally performed as the conventional techniques for medial release. However, if over-released, the superficial medial collateral ligament is detached off completely from its tibial attachment site and this intraoperative injury of the superficial medial collateral ligament is a well-known complication of TKA [1-5]. Once the tension of the medial collateral ligament is lost, the gap of medial articular compartment widens in From the Department of Orthopaedic Surgery, College of Medicine, Hanyang University, Seoul, South Korea. Submitted January 31, 2009; accepted June 6, 2009. No benefits or funds were received in support of the study. Reprint requests: Choong H. Choi, MD, Department of Orthopaedic Surgery, Hanyang University Hospital, #17 HengDang-dong, SeongDong-gu, Seoul 133-792, South Korea. © 2009 Elsevier Inc. All rights reserved. 0883-5403/09/2408-0021$36.00/0 doi:10.1016/j.arth.2009.06.007
1249
knee flexion and extension, which may cause lateral pivot movement rather than medial pivot movement. In cases of intraoperative injury of medial collateral ligament during primary TKA, many authors generally recommend a conversion to the unlinked constrained prosthesis, protection with brace after direct primary repair of the ligament with suture materials or screwand-washer reattachment, or medial collateral ligament advancement and reconstruction with bone plug recession [1,3-7]. However, the constrained prosthesis increases the stress at the implant-cement and cement-bone interfaces, which may increase the risk of polyethylene wear and resultant aseptic loosening [1-3,8,9]. It may also injure the prosthetic intercondylar post [10] and has the disadvantage of requiring substantial bone resection in the intercondylar area of the femur, which may weaken the prosthesis-bone fixation. For these reasons, most clinicians are reluctant to perform a direct conversion to an unlinked constrained prosthesis in intraoperative injury of medial collateral ligament during primary TKA [3,4,7,11-14]. Direct repair of injured medial collateral ligament also has some potential complications in that the injured ligament has the possibility of being repaired in a nonanatomical position as it is very difficult to repair the injured ligament keeping Pes anserinus in its anatomical position [15].
1250 The Journal of Arthroplasty Vol. 24 No. 8 December 2009 In this study, we examined results of previous studies showing favorable outcomes with nonoperative conservative treatment for the injury of medial collateral ligament. In particular, the medial collateral ligament was found to have excellent capacity to heal following injury [14,16-18]. In view of these reasons, we hypothesized that, in patients whose medial collateral ligament was completely detached from its tibial attachment site during primary TKA, favorable outcome could be achieved with only nonoperative conservative treatment. We evaluated the hypothesis with clinical and radiological analysis of the results.
Materials and Methods After obtaining approval from our institution's investigational review board, we retrospectively analyzed 15 cases (15 patients) of primary TKA complicated with intraoperative complete detachment of medial collateral ligament from tibial attachment site. All 15 detachments happened incidentally during the procedure for medial soft tissue release using narrow osteotome and occurred at the tibial attachment site of the medial collateral ligament. Authors could hear the popping sound of detachment and confirm the detachment by medial compartment laxity in 30° and 90° of knee flexion position. As most of old Asian people do, all 15 cases had combined varus and flexion contracture deformity. Therefore, posteromedial soft tissue of their knee joints was very stiff, that was why all 15 cases needed medial release procedure which had caused the complete detachment of the medial collateral ligament from the tibial attachment site. All cases were operated by a single surgeon (CHC) at our institution between January 1998 and June 2006 and evaluated at more than 2 years after the operations on average. Of 15 cases, 11 were conducted as a part of the staged sequential bilateral TKA and medial collateral ligaments of the contralateral knees were intact during TKA. Of 15 cases, 4 were unilateral TKA and their medial collateral ligaments were injured intraoperatively as mentioned above. We also compared the results of MCL-off knees with those of MCL-intact contralateral knees in 11 patients who underwent the staged sequential bilateral TKA. The surgeon (CHC) confirmed the MCL-off injury by intraoperative valgus stress test at 0°, 30°, and 90° of knee flexion position and followed up the results postoperatively in the same way. After the intraoperative detection of the MCL-off injury, the surgeon (CHC) tried to apply thicker polyethylene insert trial to get the mediolateral soft tissue balance and stability of knee joint at 0°, 30° and 90° of knee flexion position. In most cases, a 2-mm to 4mm thicker polyethylene insert trial than initial one could stabilize the knee joint allowing some extent of lateral tightness which was permissible. Patients conducted the range of motion exercise with continuous passive motion machine for 2 weeks during the immediate postoperative period. After a patient could
flex the operated knee to 90° and straight leg raising was possible, gradual weight-bearing to tolerance was allowed and a postoperative rehabilitation program started. No other treatment such as wearing a hinged knee brace was provided. We evaluated the results with the postoperative range of motion, the American Knee Society knee score, and function score. We also checked the subjectively preferred side of knee in 11 patients who underwent bilateral TKAs with the MCL intact in one side and the MCL off in the other side. To evaluate the possibility of postoperative medial instability, physical examination and valgus stress radiography with Telos under a force of 20 lb were performed in all patients. The Telos is some kind of equipment designed to load certain extent of stress to the knee joint constantly. To eliminate the confusing effect of joint capsule and posterior cruciate ligament as the second stabilizer against valgus stress, we did valgus stress test and checked valgus stress radiography with Telos at 30° of knee flexion position. We used the nonparametric test, especially the Wilcoxon signed rank test, as statistical method because the sample size was small, and thus, the sample would not be ruled by the normal distribution.
Results The diagnosis of the patients were all degenerative osteoarthritis of the knee. Thirteen patients were women and 2 patients were men. The average age at the time of operation was 63.9 years (range, 56-73 years). Posterior cruciate ligament substituting, cemented, fixed bearing type of total knee arthroplasties were performed in all patients except 2 cases of 1 patient. For 11 patients with the MCL intact in one side and the MCL off in the other side, both knees were operated on with the same prosthesis. The types of prostheses used were Scorpio-PS (Osteonics, Allendale, NJ) in 4 patients, Genesis II-PS (Smith & Nephew, Memphis, Tenn) in 2 patients, Nexgen-LPS (Zimmer, Warsaw, Ind) in 3 patients, AGC-PS (Biomet, Warsaw, Ind) in 1 patient and Series 7000-CR (Osteonics, Allendale, NJ) in 1 patient. As for 4 patients unilaterally operated with the MCL off, Genesis II-PS (Smith & Nephew, Memphis, Tenn) were used in 2 patients and Nexgen-LPS (Zimmer) in 2 patients. Preoperative varus deformities were present in all 15 cases with the MCL-off. The average femorotibial angle was 7° ± 4° varus(range, 0°-17° varus) preoperatively and corrected to 7° ± 2° valgus (range, 5°-9° valgus) postoperatively. The average flexion contracture improved from 7° ± 7°(range, 0°-15°) to 0° ± 1°(range, 0°-5°), while the average maximal flexion angle improved from 124° ± 14°(range, 90°-140°) to 130° ± 13°(range, 105°-145°) after the operation. The average range of motion also improved from 117° (range, 100°-140°) to 130° (range, 120°-140°) postoperatively. The American knee society knee score improved from 44.42 ± 12.49 points (range, 16-63 points) to 91.00 ± 6.78
Intraoperative Detachment of Medial Collateral Ligament Koo and Choi
1251
Table 1. Results of all 15 Cases with the MCL-Off Preoperative Postoperative
TFA (°) *
FC (°) †
MFA (°) ‡
ROM (°) §
VR 7 ± 5 VL 7 ± 2
7±7 0±1
124 ± 14 130 ± 13
117 ± 9 130 ± 9
KS ‖
FS ¶
44.42 ± 12.49 91.00 ± 6.78
48.33 ± 22.60 82.50 ± 13.57
* Tibiofemoral angle. † Flexion contracture. ‡ Maximal flexion angle. § Range of motion. ‖ The American Knee Society knee score. ¶ The American Knee Society function score.
points (range, 77-100 point), whereas the American Knee Society function score improved from 48.33 ± 22.60 point (range, 0-70 point) to 82.50 ± 13.57 point (range, 50-100 point) after the operation (Table 1). On the valgus stress radiography with Telos under a force of 20 lb, only 1 of 15 cases showed a difference of more than 5° in coronal plane compared to the MCLintact contralateral knee. Some of the MCL-off knees showed more opening in medial joint space on standing anteroposterior radiography and valgus stress radiography at 30° of knee flexion position, compared to the MCL-intact knees. Statistically, the MCL-off side showed about 1 degree of more valgus angle on the valgus stress radiography and about 4° of more valgus angle on the standing anteroposterior radiography. But, none of the 15 patients with the MCL-off knee complained medial instability in daily living activities requiring full extension and mid-range flexion position of knee joint. There were no cases requiring revision surgery due to the instability. Comparing the results of the MCL-intact side with the MCL-off contralateral side in bilaterally operated 11 patients, flexion contracture improved to 0° in both sides after the operation. The average postoperative maximal flexion was 130° ± 13°(range, 105°-145°) in the MCLintact side and 131° ± 14°(range, 105°-145°) in the MCLoff side and there was no significant difference in postoperative maximal flexion (P = .563). The American knee society knee score improved to 92.20 ± 3.74 points (range, 83-97 points) in the MCL-intact side and to 91.50 ± 7.22 points (range, 77-100 points) in the MCL-off side. The American knee society function score also improved to 82.00 ± 3.59 points (range, 60-100 points) in the MCL-intact side and to 81.00 ± 4.33 points (range, 50-
100 points) in the MCL-off side. Interestingly, there was no significant difference in the American knee society knee score and function score between the MCL-intact side and the MCL-off contralateral side(P = .773, P = 1.000). On the valgus stress radiography with Telos under the force of 20lb, which was performed at the last follow up, the average valgus angle was 3° ± 1°(range, 1°-5°) in the MCL-intact side and 4° ± 1°(range, 1°-8°) in the MCL-off side. There was only the difference of 1 degree in valgus stress angle between the MCL-intact side and the MCL-off contralateral side. The MCL-off side showed almost 4° of more valgus than the MCL-intact side in tibiofemoral angle on standing anteroposterior radiography. There was no case showing clinical instability in coronal plane. Those facts meant that although there could have been some extent of medial side laxity in the MCL-off side, it might have been too minimal to cause significant clinical instability. For the response to the preferred side in 11 patients, 4 and 3 patients preferred the MCL-intact side and the MCL-off side, respectively. The other 4 patients answered that they could not feel any difference between their knees (Table 2).
Discussion
In general, medial collateral ligament injuries during the TKA may cause the instability of knee joint in coronal plane and thereby increase the possibility of potential complications or the necessity of revision surgery. For those cases with intraoperative medial collateral ligament injuries during primary TKA, many authors recommended various treatment options and most of them preferred the method of prosthetic conversion to unlinked constrained prosthesis [1-4,8,11].
Table 2. Results of 11 Patients with MCL-Off at One Side and MCL-Intact at the Other Side in the Same Patient MCL-off side MCL-intact side
TFA (°) *
FC (°) †
MFA (°) ‡
VA (°) §
VL 7 ± 2 VL 3 ± 2
0 0
131 ± 14 130 ± 13
4±1 3±1
KS ‖
FS ¶
P#
91.50 ± 7.22 92.20 ± 3.74
81.00 ± 4.33 82.00 ± 3.59
3 4
* Tibiofemoral angle. † Flexion contracture. ‡ Maximal flexion angle. § Valgus angle on valgus stress radiography with 30° of knee flexion. ‖ The American knee society knee score. ¶ The American knee society function score. # Number of patients’ preference. The rest of 4 patients felt no difference between the MCL-off side and the MCL-intact side.
1252 The Journal of Arthroplasty Vol. 24 No. 8 December 2009 However, unlinked constrained prosthesis proved to increase the stresses at prosthesis-cement and cementbone interfaces [2,7,8] and thereby the possibility of aseptic loosening increases with the use of unlinked constrained prosthesis [1,12]. In addition, unlinked constrained prosthesis requires the use of stem extension and much more bone cutting at the femoral intercondylar area, which has negative effect on probable revision surgery in the future and makes it difficult to use unlinked constrained prosthesis for primary TKA. Although the unlinked constrained prosthesis was used for primary TKA, recurrent instability [13] or fracture of tibial intercondylar eminence [10] has been reported. Therefore, now, most authors are reluctant to use the unlinked constrained prosthesis for the intraoperative medial collateral ligament injuries during primary TKA and try to find treatment options that are as conservative as possible [1,3,4,7,11,13]. Based on good potential healing capacity of medial collateral ligament after injuries [14,16-18], although those were not the intraoperative injuries but the injuries identified prior to primary TKA, some authors reported the result of those injuries treated by medial collateral ligament advancement with bone plug recession [19-21] instead of the use of unlinked constrained prosthesis in primary TKA. They concluded that favorable outcomes were achieved with the technique in most of the cases. There was no case of clinical instability. However, the operating time was prolonged by more than 40 minutes in elderly patients and there was a possibility of fracture in the process of medial collateral ligament advancement and fixation to distal femur or proximal tibia with bone fragment in osteoporotic patients. Although the operation ended successfully, there still remains the possibility of fracture at the fixation site. To prevent the fracture of fixation site and to promote fixation, cast immobilization was required. Thus, knee stiffness could develop after TKA. Immediate primary repair technique for intraoperative injury of medial collateral ligament during TKA [22] was reported. They used direct suture technique with nonabsorbable suture materials for midsubstance injuries. For avulsions off at the femoral or tibial insertion site, they fixed the injured medial collateral ligaments with nonabsorbable suture materials, non-absorbable suture anchors, and screw-and-washer. They did not add any other special procedure during the postoperative rehabilitation process except applying hinged knee brace without limiting the range of motion. Nor did they restrict the weight-bearing of the patients. The results were good. We also experienced the intraoperative medial collateral ligament injuries during TKAs clinically. Especially for early cases with avulsion off of the medial collateral ligament from its tibial attachment site, we finished the operation without any other additional procedure and performed the conventional postoperative rehabilitation
based on many reports that the medial collateral ligament had favorable self-healing potential after injuries, and conservative treatment for the injury of the medial collateral ligament showed good results [14,16,17]. During the follow-up period for these cases, we found that there was no significant clinical instability of knee joint in coronal plane. After that, we conservatively treated the intraoperative avulsion off of the medial collateral ligament from its tibial attachment site during TKAs without reparative or reconstructive procedures and retrospectively followed up the clinical and radiological results. In conclusion, most of those cases also showed good results and no clinical instability in coronal plane. Because most previous studies have depended on the physician's physical examination in judging whether or not the postoperative clinical instability of the knee joint existed, there could be a possibility of misjudgement derived from interobserver or intraobserver differences in the process of interpreting the examination. However, in the present study, a single surgeon (CHC) performed preoperative, intraoperative, and postoperative physical examination and provided objective result with valgus stress radiography with Telos under a force of 20 lb. By comparing the valgus stress radiography with that of MCL-intact contralateral knee, we were able to analyze the result more objectively. Moreover, most previous studies have used posterior cruciate ligament retaining prostheses. In these types of prostheses, the function of posterior cruciate ligament as the second stabilizer against valgus force around the knee joint remained to a certain extent, which could have made some positive effects on the results of previous studies about conservative treatment for medial collateral ligament injuries. In the present study, we eliminated the function of posterior cruciate ligament as the second stabilizer against valgus force with the use of posterior cruciate ligament substituting prostheses except only 1 case. Therefore, our study was relatively free from the confounding variable in evaluating the effectiveness of conservative treatment for intraoperative injuries of medial collateral ligament during TKAs. Despite these results, our study had several limiting factors. Firstly, we studied only the results of the avulsion-off injuries of medial collateral ligaments from tibial attachment site among various types of injuries. Secondly, this was a small study with only 15 patients and the results of short-term follow-up averaged only 2 years. Thirdly, because there is no established objective standard in evaluating the coronal instability of knee, we evaluated the existence of coronal instability in the MCL-off knees by comparing the results of physical examination and radiography of the MCL-intact contralateral knees with that of the MCL-off knees in each patient. However, in most of the patients, even the MCL-intact contralateral knees also had some degree of osteoarthritis, so there
Intraoperative Detachment of Medial Collateral Ligament Koo and Choi
might be some changes in their medial collateral ligaments. Consequently, there could have been some limitations in achieving accurate comparison. Fourthly, the dependency on the contralateral MCL-intact knees could have developed after TKAs, which might cause some confusions in evaluating the effectiveness of conservative treatment for the MCL-off TKAs. Fifthly, there is the possibility that although procedure for medial soft tissue release was done, remained medial soft tissue tension or tightness arising from combined varus deformity which was shown in most of old Asian people could have compensated medial instability caused by injury of the MCL. However, this possibility was not fully evaluated. Nevertheless, the results of our study showed that intraoperative detachment of the medial collateral ligament from tibial attachment site during TKAs could be favorably treated without any other additional procedure, and no case of coronal instability or early failure was observed clinically. There was no case requiring revision due to failure of the conservative treatment, either. In conclusion, we think that even cases complicated with intraoperative avulsion off injury of the medial collateral ligament from tibial attachment site during primary TKAs do not need conversion to the constrained prosthesis, primary repair or fixation with advancement of the avulsed medial collateral ligament, or postoperative application of hinged knee brace. By solely administering conservative treatment, these intraoperative injuries of the medial collateral ligament during TKAs can be healed favorably.
References 1. Donaldson III WF, Sculco TP, Insall JN, et al. Knee prosthesis. Long-term follow-up study. Clin Orthop 1988; 226:21. 2. Hartford JM, Goodman SB, Schurman DJ, et al. Complex primary and revision total knee arthroplasty using the condylar constrained prosthesis. An average 5-year followup. J Arthroplasty 1988;13:380. 3. Lachiewicz PF, Falatyn SP. Clinical and radiological results of the total condylar III and constrained condylar total knee arthroplasty. J Arthroplasty 1996;11:916. 4. Sculco TP. Total condylar III prosthesis in ligament instability. Orthop Clin North Am 1989;20:221. 5. Winiarsky R, Barth P, Lotke P. Total knee arthroplasty in morbidly obese patients. J Bone Joint Surg Am 1998; 80:1770. 6. Whiteside LA. Correction of ligament and bone defects in total arthroplasty of the severely valgus knee. Clin Orthop 1993;288:234.
1253
7. Lombardi Jr AV, Mallony TH, Eberie RW. Constrained knee arthroplasty. In: Scott WN, editor. The knee. St. Louis: CV Mosby; 1994; p. 1305. 8. Rosenberg AG, Verner JJ, Galante JO. Clinical results of total knee revision using the total condylar III prosthesis. Clin Orthop 1991;273:83. 9. Jacobs JJ, Berzins A, Cannestra V, et al. Surface damage in retrieved polyethylene tibial inserts. Conforming vs nonconforming geometries. Presented at the Annual Meeting of the American Academy of Orthopaedic Surgeons; March; 2000. Orlando, FL. 10. McPherson EJ, Vince KG. Breakage of a total condylar III knee prosthesis. A case report. J Arthroplasty 1993;8:561. 11. Peters CL, Hennessey R, Berden RM, et al. Revision total knee arthroplasty with a cemented posterior-stabilized or constrained condylar prosthesis. A minimum 3-year and average 5-year follow-up study. J Arthroplasty 1991; 12:896. 12. Rand JA. Revision total knee arthroplasty using the total condylar III prosthesis. J Arthroplasty 1991;6:279. 13. Vince KG. Use of constraint. “Symposium: Revision Issues,” Specialty Day, Knee Society. Presented at the Annual Meeting of the American Academy of Orthopaedic Surgeons; March; 2000. Orlando, FL. 14. Ballmer PM, Jacob RP. The non operative treatment of isolate complete tears of the medial collateral ligament of the knee. A prospective study. Arch Orthop Trauma Surg 1988;107:273. 15. Miller RH. Knee injuries. In: Canale ST, editor. Campbell's operative orthopaedics. 9th eds. St. Louis: Mosby-Year Book; 1998; p. 1113. 16. Indelicato PA, Hermansdorfer J, Huegel M. Nonoperative management of complete tears of the medial collateral ligament of the knee in intercollegiate football players. Clin Orthop 1990;107:273. 17. Sandberg R, Balkfors B, Nilsson B, et al. Operative versus nonoperative treatment of recent injuries to the ligament of the knee. A prospective randomized study. J Bone Joint Surg Am 1987;69:1120. 18. Woo SL, Chan SS, Yamaji T. Biomechanics of the knee ligament healing, repair and reconstruction. J Biomech 1997;30:431. 19. Healy WL, Iorio R, Lemos DW. Medial reconstruction during total knee arthroplasty for severe valgus deformity. Clin Orthop 1998;356:161. 20. Krackow KA. Deformity. In: Klein EA, editor. The technique of total knee arthroplasty. St. Louis: CV Mosby; 1990; p. 339. 21. Krackow KA, Jones MM, Teensy SM, et al. Primary total knee arthroplasty in patients with fixed valgus deformity. Clin Orthop 1991;273:9. 22. Leopold MSS, McStay C, Klafeta K, et al. Primary repair of intraoperative disruption of the medial collateral ligament during total knee arthroplasty. J Bone Joint Surg Am 2001; 83:86.
Conservative Treatment for the Intraoperative Detachment of Medial Collateral Ligament from the Tibial Attachment Site during Primary Total Knee Arthroplasty Min-Hoi Koo, MD, and Choong H. Choi, MD
Abstract: For intraoperative injuries of the medial collateral ligament (MCL) during primary total knee arthroplasty (TKA), many authors recommended complex reconstructive procedures or conversion to the constrained prosthesis in previous literature. However, it is well known that medial collateral ligament has good healing potential after injuries. This retrospective study evaluated the clinical and radiological results of 15 primary TKAs complicated with intraoperative complete detachment of the MCL from tibial attachment site, which were all treated solely by nonoperative conservative treatment without any other additive procedures or braces. Compared to the MCL-intact contralateral knees, there was no significant difference in terms of clinical and radiological outcome at minimum of 2 postoperative year. There was no case showing clinical instability, either. Solely, the nonoperative conservative treatment for intraoperative injuries of the MCL during primary TKAs is expected not only to be effective with satisfactory results but also to decrease complications derived from other sophisticated reparative or reconstructive procedures. Keywords: intraoperative MCL injury, conservative treatment, total knee arthroplasty. © 2009 Elsevier Inc. All rights reserved.
In evaluating the flexion and extension gap during total knee arthroplasty (TKA), it is very important to balance the medial-lateral soft tissue tension. In most cases with degenerative osteoarthritis, varus deformity is mainly combined, increasing the necessity for the release of the contracted medial soft structures to balance mediallateral soft tissue tension. Removal of osteophytes formed around the medial femoral and tibial condyles, release of medial joint capsule containing the deep medial collateral ligament, and additional gradual release of the superficial medial collateral ligament are generally performed as the conventional techniques for medial release. However, if over-released, the superficial medial collateral ligament is detached off completely from its tibial attachment site and this intraoperative injury of the superficial medial collateral ligament is a well-known complication of TKA [1-5]. Once the tension of the medial collateral ligament is lost, the gap of medial articular compartment widens in From the Department of Orthopaedic Surgery, College of Medicine, Hanyang University, Seoul, South Korea. Submitted January 31, 2009; accepted June 6, 2009. No benefits or funds were received in support of the study. Reprint requests: Choong H. Choi, MD, Department of Orthopaedic Surgery, Hanyang University Hospital, #17 HengDang-dong, SeongDong-gu, Seoul 133-792, South Korea. © 2009 Elsevier Inc. All rights reserved. 0883-5403/09/2408-0021$36.00/0 doi:10.1016/j.arth.2009.06.007
1249
knee flexion and extension, which may cause lateral pivot movement rather than medial pivot movement. In cases of intraoperative injury of medial collateral ligament during primary TKA, many authors generally recommend a conversion to the unlinked constrained prosthesis, protection with brace after direct primary repair of the ligament with suture materials or screwand-washer reattachment, or medial collateral ligament advancement and reconstruction with bone plug recession [1,3-7]. However, the constrained prosthesis increases the stress at the implant-cement and cement-bone interfaces, which may increase the risk of polyethylene wear and resultant aseptic loosening [1-3,8,9]. It may also injure the prosthetic intercondylar post [10] and has the disadvantage of requiring substantial bone resection in the intercondylar area of the femur, which may weaken the prosthesis-bone fixation. For these reasons, most clinicians are reluctant to perform a direct conversion to an unlinked constrained prosthesis in intraoperative injury of medial collateral ligament during primary TKA [3,4,7,11-14]. Direct repair of injured medial collateral ligament also has some potential complications in that the injured ligament has the possibility of being repaired in a nonanatomical position as it is very difficult to repair the injured ligament keeping Pes anserinus in its anatomical position [15].
1250 The Journal of Arthroplasty Vol. 24 No. 8 December 2009 In this study, we examined results of previous studies showing favorable outcomes with nonoperative conservative treatment for the injury of medial collateral ligament. In particular, the medial collateral ligament was found to have excellent capacity to heal following injury [14,16-18]. In view of these reasons, we hypothesized that, in patients whose medial collateral ligament was completely detached from its tibial attachment site during primary TKA, favorable outcome could be achieved with only nonoperative conservative treatment. We evaluated the hypothesis with clinical and radiological analysis of the results.
Materials and Methods After obtaining approval from our institution's investigational review board, we retrospectively analyzed 15 cases (15 patients) of primary TKA complicated with intraoperative complete detachment of medial collateral ligament from tibial attachment site. All 15 detachments happened incidentally during the procedure for medial soft tissue release using narrow osteotome and occurred at the tibial attachment site of the medial collateral ligament. Authors could hear the popping sound of detachment and confirm the detachment by medial compartment laxity in 30° and 90° of knee flexion position. As most of old Asian people do, all 15 cases had combined varus and flexion contracture deformity. Therefore, posteromedial soft tissue of their knee joints was very stiff, that was why all 15 cases needed medial release procedure which had caused the complete detachment of the medial collateral ligament from the tibial attachment site. All cases were operated by a single surgeon (CHC) at our institution between January 1998 and June 2006 and evaluated at more than 2 years after the operations on average. Of 15 cases, 11 were conducted as a part of the staged sequential bilateral TKA and medial collateral ligaments of the contralateral knees were intact during TKA. Of 15 cases, 4 were unilateral TKA and their medial collateral ligaments were injured intraoperatively as mentioned above. We also compared the results of MCL-off knees with those of MCL-intact contralateral knees in 11 patients who underwent the staged sequential bilateral TKA. The surgeon (CHC) confirmed the MCL-off injury by intraoperative valgus stress test at 0°, 30°, and 90° of knee flexion position and followed up the results postoperatively in the same way. After the intraoperative detection of the MCL-off injury, the surgeon (CHC) tried to apply thicker polyethylene insert trial to get the mediolateral soft tissue balance and stability of knee joint at 0°, 30° and 90° of knee flexion position. In most cases, a 2-mm to 4mm thicker polyethylene insert trial than initial one could stabilize the knee joint allowing some extent of lateral tightness which was permissible. Patients conducted the range of motion exercise with continuous passive motion machine for 2 weeks during the immediate postoperative period. After a patient could
flex the operated knee to 90° and straight leg raising was possible, gradual weight-bearing to tolerance was allowed and a postoperative rehabilitation program started. No other treatment such as wearing a hinged knee brace was provided. We evaluated the results with the postoperative range of motion, the American Knee Society knee score, and function score. We also checked the subjectively preferred side of knee in 11 patients who underwent bilateral TKAs with the MCL intact in one side and the MCL off in the other side. To evaluate the possibility of postoperative medial instability, physical examination and valgus stress radiography with Telos under a force of 20 lb were performed in all patients. The Telos is some kind of equipment designed to load certain extent of stress to the knee joint constantly. To eliminate the confusing effect of joint capsule and posterior cruciate ligament as the second stabilizer against valgus stress, we did valgus stress test and checked valgus stress radiography with Telos at 30° of knee flexion position. We used the nonparametric test, especially the Wilcoxon signed rank test, as statistical method because the sample size was small, and thus, the sample would not be ruled by the normal distribution.
Results The diagnosis of the patients were all degenerative osteoarthritis of the knee. Thirteen patients were women and 2 patients were men. The average age at the time of operation was 63.9 years (range, 56-73 years). Posterior cruciate ligament substituting, cemented, fixed bearing type of total knee arthroplasties were performed in all patients except 2 cases of 1 patient. For 11 patients with the MCL intact in one side and the MCL off in the other side, both knees were operated on with the same prosthesis. The types of prostheses used were Scorpio-PS (Osteonics, Allendale, NJ) in 4 patients, Genesis II-PS (Smith & Nephew, Memphis, Tenn) in 2 patients, Nexgen-LPS (Zimmer, Warsaw, Ind) in 3 patients, AGC-PS (Biomet, Warsaw, Ind) in 1 patient and Series 7000-CR (Osteonics, Allendale, NJ) in 1 patient. As for 4 patients unilaterally operated with the MCL off, Genesis II-PS (Smith & Nephew, Memphis, Tenn) were used in 2 patients and Nexgen-LPS (Zimmer) in 2 patients. Preoperative varus deformities were present in all 15 cases with the MCL-off. The average femorotibial angle was 7° ± 4° varus(range, 0°-17° varus) preoperatively and corrected to 7° ± 2° valgus (range, 5°-9° valgus) postoperatively. The average flexion contracture improved from 7° ± 7°(range, 0°-15°) to 0° ± 1°(range, 0°-5°), while the average maximal flexion angle improved from 124° ± 14°(range, 90°-140°) to 130° ± 13°(range, 105°-145°) after the operation. The average range of motion also improved from 117° (range, 100°-140°) to 130° (range, 120°-140°) postoperatively. The American knee society knee score improved from 44.42 ± 12.49 points (range, 16-63 points) to 91.00 ± 6.78
Intraoperative Detachment of Medial Collateral Ligament Koo and Choi
1251
Table 1. Results of all 15 Cases with the MCL-Off Preoperative Postoperative
TFA (°) *
FC (°) †
MFA (°) ‡
ROM (°) §
VR 7 ± 5 VL 7 ± 2
7±7 0±1
124 ± 14 130 ± 13
117 ± 9 130 ± 9
KS ‖
FS ¶
44.42 ± 12.49 91.00 ± 6.78
48.33 ± 22.60 82.50 ± 13.57
* Tibiofemoral angle. † Flexion contracture. ‡ Maximal flexion angle. § Range of motion. ‖ The American Knee Society knee score. ¶ The American Knee Society function score.
points (range, 77-100 point), whereas the American Knee Society function score improved from 48.33 ± 22.60 point (range, 0-70 point) to 82.50 ± 13.57 point (range, 50-100 point) after the operation (Table 1). On the valgus stress radiography with Telos under a force of 20 lb, only 1 of 15 cases showed a difference of more than 5° in coronal plane compared to the MCLintact contralateral knee. Some of the MCL-off knees showed more opening in medial joint space on standing anteroposterior radiography and valgus stress radiography at 30° of knee flexion position, compared to the MCL-intact knees. Statistically, the MCL-off side showed about 1 degree of more valgus angle on the valgus stress radiography and about 4° of more valgus angle on the standing anteroposterior radiography. But, none of the 15 patients with the MCL-off knee complained medial instability in daily living activities requiring full extension and mid-range flexion position of knee joint. There were no cases requiring revision surgery due to the instability. Comparing the results of the MCL-intact side with the MCL-off contralateral side in bilaterally operated 11 patients, flexion contracture improved to 0° in both sides after the operation. The average postoperative maximal flexion was 130° ± 13°(range, 105°-145°) in the MCLintact side and 131° ± 14°(range, 105°-145°) in the MCLoff side and there was no significant difference in postoperative maximal flexion (P = .563). The American knee society knee score improved to 92.20 ± 3.74 points (range, 83-97 points) in the MCL-intact side and to 91.50 ± 7.22 points (range, 77-100 points) in the MCL-off side. The American knee society function score also improved to 82.00 ± 3.59 points (range, 60-100 points) in the MCL-intact side and to 81.00 ± 4.33 points (range, 50-
100 points) in the MCL-off side. Interestingly, there was no significant difference in the American knee society knee score and function score between the MCL-intact side and the MCL-off contralateral side(P = .773, P = 1.000). On the valgus stress radiography with Telos under the force of 20lb, which was performed at the last follow up, the average valgus angle was 3° ± 1°(range, 1°-5°) in the MCL-intact side and 4° ± 1°(range, 1°-8°) in the MCL-off side. There was only the difference of 1 degree in valgus stress angle between the MCL-intact side and the MCL-off contralateral side. The MCL-off side showed almost 4° of more valgus than the MCL-intact side in tibiofemoral angle on standing anteroposterior radiography. There was no case showing clinical instability in coronal plane. Those facts meant that although there could have been some extent of medial side laxity in the MCL-off side, it might have been too minimal to cause significant clinical instability. For the response to the preferred side in 11 patients, 4 and 3 patients preferred the MCL-intact side and the MCL-off side, respectively. The other 4 patients answered that they could not feel any difference between their knees (Table 2).
Discussion
In general, medial collateral ligament injuries during the TKA may cause the instability of knee joint in coronal plane and thereby increase the possibility of potential complications or the necessity of revision surgery. For those cases with intraoperative medial collateral ligament injuries during primary TKA, many authors recommended various treatment options and most of them preferred the method of prosthetic conversion to unlinked constrained prosthesis [1-4,8,11].
Table 2. Results of 11 Patients with MCL-Off at One Side and MCL-Intact at the Other Side in the Same Patient MCL-off side MCL-intact side
TFA (°) *
FC (°) †
MFA (°) ‡
VA (°) §
VL 7 ± 2 VL 3 ± 2
0 0
131 ± 14 130 ± 13
4±1 3±1
KS ‖
FS ¶
P#
91.50 ± 7.22 92.20 ± 3.74
81.00 ± 4.33 82.00 ± 3.59
3 4
* Tibiofemoral angle. † Flexion contracture. ‡ Maximal flexion angle. § Valgus angle on valgus stress radiography with 30° of knee flexion. ‖ The American knee society knee score. ¶ The American knee society function score. # Number of patients’ preference. The rest of 4 patients felt no difference between the MCL-off side and the MCL-intact side.
1252 The Journal of Arthroplasty Vol. 24 No. 8 December 2009 However, unlinked constrained prosthesis proved to increase the stresses at prosthesis-cement and cementbone interfaces [2,7,8] and thereby the possibility of aseptic loosening increases with the use of unlinked constrained prosthesis [1,12]. In addition, unlinked constrained prosthesis requires the use of stem extension and much more bone cutting at the femoral intercondylar area, which has negative effect on probable revision surgery in the future and makes it difficult to use unlinked constrained prosthesis for primary TKA. Although the unlinked constrained prosthesis was used for primary TKA, recurrent instability [13] or fracture of tibial intercondylar eminence [10] has been reported. Therefore, now, most authors are reluctant to use the unlinked constrained prosthesis for the intraoperative medial collateral ligament injuries during primary TKA and try to find treatment options that are as conservative as possible [1,3,4,7,11,13]. Based on good potential healing capacity of medial collateral ligament after injuries [14,16-18], although those were not the intraoperative injuries but the injuries identified prior to primary TKA, some authors reported the result of those injuries treated by medial collateral ligament advancement with bone plug recession [19-21] instead of the use of unlinked constrained prosthesis in primary TKA. They concluded that favorable outcomes were achieved with the technique in most of the cases. There was no case of clinical instability. However, the operating time was prolonged by more than 40 minutes in elderly patients and there was a possibility of fracture in the process of medial collateral ligament advancement and fixation to distal femur or proximal tibia with bone fragment in osteoporotic patients. Although the operation ended successfully, there still remains the possibility of fracture at the fixation site. To prevent the fracture of fixation site and to promote fixation, cast immobilization was required. Thus, knee stiffness could develop after TKA. Immediate primary repair technique for intraoperative injury of medial collateral ligament during TKA [22] was reported. They used direct suture technique with nonabsorbable suture materials for midsubstance injuries. For avulsions off at the femoral or tibial insertion site, they fixed the injured medial collateral ligaments with nonabsorbable suture materials, non-absorbable suture anchors, and screw-and-washer. They did not add any other special procedure during the postoperative rehabilitation process except applying hinged knee brace without limiting the range of motion. Nor did they restrict the weight-bearing of the patients. The results were good. We also experienced the intraoperative medial collateral ligament injuries during TKAs clinically. Especially for early cases with avulsion off of the medial collateral ligament from its tibial attachment site, we finished the operation without any other additional procedure and performed the conventional postoperative rehabilitation
based on many reports that the medial collateral ligament had favorable self-healing potential after injuries, and conservative treatment for the injury of the medial collateral ligament showed good results [14,16,17]. During the follow-up period for these cases, we found that there was no significant clinical instability of knee joint in coronal plane. After that, we conservatively treated the intraoperative avulsion off of the medial collateral ligament from its tibial attachment site during TKAs without reparative or reconstructive procedures and retrospectively followed up the clinical and radiological results. In conclusion, most of those cases also showed good results and no clinical instability in coronal plane. Because most previous studies have depended on the physician's physical examination in judging whether or not the postoperative clinical instability of the knee joint existed, there could be a possibility of misjudgement derived from interobserver or intraobserver differences in the process of interpreting the examination. However, in the present study, a single surgeon (CHC) performed preoperative, intraoperative, and postoperative physical examination and provided objective result with valgus stress radiography with Telos under a force of 20 lb. By comparing the valgus stress radiography with that of MCL-intact contralateral knee, we were able to analyze the result more objectively. Moreover, most previous studies have used posterior cruciate ligament retaining prostheses. In these types of prostheses, the function of posterior cruciate ligament as the second stabilizer against valgus force around the knee joint remained to a certain extent, which could have made some positive effects on the results of previous studies about conservative treatment for medial collateral ligament injuries. In the present study, we eliminated the function of posterior cruciate ligament as the second stabilizer against valgus force with the use of posterior cruciate ligament substituting prostheses except only 1 case. Therefore, our study was relatively free from the confounding variable in evaluating the effectiveness of conservative treatment for intraoperative injuries of medial collateral ligament during TKAs. Despite these results, our study had several limiting factors. Firstly, we studied only the results of the avulsion-off injuries of medial collateral ligaments from tibial attachment site among various types of injuries. Secondly, this was a small study with only 15 patients and the results of short-term follow-up averaged only 2 years. Thirdly, because there is no established objective standard in evaluating the coronal instability of knee, we evaluated the existence of coronal instability in the MCL-off knees by comparing the results of physical examination and radiography of the MCL-intact contralateral knees with that of the MCL-off knees in each patient. However, in most of the patients, even the MCL-intact contralateral knees also had some degree of osteoarthritis, so there
Intraoperative Detachment of Medial Collateral Ligament Koo and Choi
might be some changes in their medial collateral ligaments. Consequently, there could have been some limitations in achieving accurate comparison. Fourthly, the dependency on the contralateral MCL-intact knees could have developed after TKAs, which might cause some confusions in evaluating the effectiveness of conservative treatment for the MCL-off TKAs. Fifthly, there is the possibility that although procedure for medial soft tissue release was done, remained medial soft tissue tension or tightness arising from combined varus deformity which was shown in most of old Asian people could have compensated medial instability caused by injury of the MCL. However, this possibility was not fully evaluated. Nevertheless, the results of our study showed that intraoperative detachment of the medial collateral ligament from tibial attachment site during TKAs could be favorably treated without any other additional procedure, and no case of coronal instability or early failure was observed clinically. There was no case requiring revision due to failure of the conservative treatment, either. In conclusion, we think that even cases complicated with intraoperative avulsion off injury of the medial collateral ligament from tibial attachment site during primary TKAs do not need conversion to the constrained prosthesis, primary repair or fixation with advancement of the avulsed medial collateral ligament, or postoperative application of hinged knee brace. By solely administering conservative treatment, these intraoperative injuries of the medial collateral ligament during TKAs can be healed favorably.
References 1. Donaldson III WF, Sculco TP, Insall JN, et al. Knee prosthesis. Long-term follow-up study. Clin Orthop 1988; 226:21. 2. Hartford JM, Goodman SB, Schurman DJ, et al. Complex primary and revision total knee arthroplasty using the condylar constrained prosthesis. An average 5-year followup. J Arthroplasty 1988;13:380. 3. Lachiewicz PF, Falatyn SP. Clinical and radiological results of the total condylar III and constrained condylar total knee arthroplasty. J Arthroplasty 1996;11:916. 4. Sculco TP. Total condylar III prosthesis in ligament instability. Orthop Clin North Am 1989;20:221. 5. Winiarsky R, Barth P, Lotke P. Total knee arthroplasty in morbidly obese patients. J Bone Joint Surg Am 1998; 80:1770. 6. Whiteside LA. Correction of ligament and bone defects in total arthroplasty of the severely valgus knee. Clin Orthop 1993;288:234.
1253
7. Lombardi Jr AV, Mallony TH, Eberie RW. Constrained knee arthroplasty. In: Scott WN, editor. The knee. St. Louis: CV Mosby; 1994; p. 1305. 8. Rosenberg AG, Verner JJ, Galante JO. Clinical results of total knee revision using the total condylar III prosthesis. Clin Orthop 1991;273:83. 9. Jacobs JJ, Berzins A, Cannestra V, et al. Surface damage in retrieved polyethylene tibial inserts. Conforming vs nonconforming geometries. Presented at the Annual Meeting of the American Academy of Orthopaedic Surgeons; March; 2000. Orlando, FL. 10. McPherson EJ, Vince KG. Breakage of a total condylar III knee prosthesis. A case report. J Arthroplasty 1993;8:561. 11. Peters CL, Hennessey R, Berden RM, et al. Revision total knee arthroplasty with a cemented posterior-stabilized or constrained condylar prosthesis. A minimum 3-year and average 5-year follow-up study. J Arthroplasty 1991; 12:896. 12. Rand JA. Revision total knee arthroplasty using the total condylar III prosthesis. J Arthroplasty 1991;6:279. 13. Vince KG. Use of constraint. “Symposium: Revision Issues,” Specialty Day, Knee Society. Presented at the Annual Meeting of the American Academy of Orthopaedic Surgeons; March; 2000. Orlando, FL. 14. Ballmer PM, Jacob RP. The non operative treatment of isolate complete tears of the medial collateral ligament of the knee. A prospective study. Arch Orthop Trauma Surg 1988;107:273. 15. Miller RH. Knee injuries. In: Canale ST, editor. Campbell's operative orthopaedics. 9th eds. St. Louis: Mosby-Year Book; 1998; p. 1113. 16. Indelicato PA, Hermansdorfer J, Huegel M. Nonoperative management of complete tears of the medial collateral ligament of the knee in intercollegiate football players. Clin Orthop 1990;107:273. 17. Sandberg R, Balkfors B, Nilsson B, et al. Operative versus nonoperative treatment of recent injuries to the ligament of the knee. A prospective randomized study. J Bone Joint Surg Am 1987;69:1120. 18. Woo SL, Chan SS, Yamaji T. Biomechanics of the knee ligament healing, repair and reconstruction. J Biomech 1997;30:431. 19. Healy WL, Iorio R, Lemos DW. Medial reconstruction during total knee arthroplasty for severe valgus deformity. Clin Orthop 1998;356:161. 20. Krackow KA. Deformity. In: Klein EA, editor. The technique of total knee arthroplasty. St. Louis: CV Mosby; 1990; p. 339. 21. Krackow KA, Jones MM, Teensy SM, et al. Primary total knee arthroplasty in patients with fixed valgus deformity. Clin Orthop 1991;273:9. 22. Leopold MSS, McStay C, Klafeta K, et al. Primary repair of intraoperative disruption of the medial collateral ligament during total knee arthroplasty. J Bone Joint Surg Am 2001; 83:86.