Original Article
Outcomes of Medial Collateral Ligament Injuries during Total Knee Arthroplasty Marcelo B. P. Siqueira, MD1 Kathryn Haller, BA1 Alison K. Klika, MS1 Wael K. Barsoum, MD1
Andrew Mulder, MD1
1 Department of Orthopaedic Surgery, Orthopaedic and
Rheumatologic Institute, Cleveland Clinic, Cleveland, Ohio
Andrew S. Goldblum, MD1
Address for correspondence Alison K. Klika, MS, Cleveland Clinic, 9500 Euclid Avenue, A41, Cleveland, OH 44195 (e-mail: [email protected]).
Abstract
Keywords
► total knee arthroplasty ► intraoperative medial collateral ligament injury ► ligament repair ► constrained implants
Intraoperative medial collateral ligament (MCL) disruption during total knee arthroplasty (TKA) is often managed with either primary repair or use of a constrained implant. A total of 23 patients with an MCL injury during TKA between 2003 and 2009 were compared with 92 matched controls. Of the 23 patients, 10 were treated with an unconstrained implant and primary MCL repair, 8 with constrained implants, 3 with constrained implants and MCL repair, and 2 with unconstrained implants and no MCL repair. After an average 5-year follow-up, patients had lower Knee Society Scores (KSS), 79 versus 87 (p ¼ 0.03), but similar Knee Function Scores (KFS), 68 versus 72 (p ¼ 0.35). The improvement between preoperative and postoperative KSS and KFS did not vary among the two groups (p ¼ 0.88 and p ¼ 0.77, respectively). Postoperative scores did not vary significantly among the four treatment modalities. Conservative treatment can provide satisfactory outcomes and avoid potential complications of increased constraint.
Medial collateral ligament (MCL) integrity is important for coronal plane stability and soft tissue balance after total knee arthroplasty (TKA).1 Iatrogenic MCL injury during TKA is a relatively infrequent but potential adverse event, which requires repair, a change in the level of prosthetic constraint, or revision surgery.2 Studies have reported intraoperative MCL injury rates between 2.2 and 2.7%.3,4 Unidentified MCL injuries can lead to adverse outcomes such as instability and accelerated implant wear, demonstrating the importance of intraoperative recognition and treatment.4,5 Surgeon-controlled risk factors for MCL injury include the use of oscillating saw blades that are wider than the femoral condyles, late removal of medial osteophytes, overstrenuous testing of varus–valgus stability, and hyperflexion of the knee.6,7 Excessive varus contracture commonly found in patients with degenerative osteoarthritis demands a more invasive medial release approach, ultimately increasing the rates of MCL tears.8 Obesity9 and severe deformities6,10,11 are patient-dependent factors,
which also increase the risk of ligament damage and attenuation. Currently, there is no consensus as to the most appropriate way to treat iatrogenic MCL injury during TKA. Various treatment modalities have been used for intraoperative injury, each with reported advantages and disadvantages. The more traditional treatment approach has been conversion to an unlinked constrained prosthesis. Concerns regarding potential loosening secondary to increased stress across the implant–cement interface prompted some surgeons to use nonconstrained prostheses in conjunction with direct primary repair (using suture or screw-and-washer reattachment),3 augmentation5 or thicker polyethylene inserts.8 Leopold et al3 reported positive outcomes in 16 patients treated with primary ligament suture or suture anchor fixation along with an unconstrained implant. Koo and Choi8 managed 15 injured MCLs with a thicker polyethylene insert in combination with unconstrained implants, and reported no instabilities or revisions. In contrast, a recent study by Lee and Lotke4 compared
received May 27, 2014 accepted after revision September 1, 2014
Copyright © by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1394166. ISSN 1538-8506.
Downloaded by: University of British Columbia. Copyrighted material.
J Knee Surg
Siqueira et al.
outcomes of patients treated with ligament repair plus a nonconstrained implant with patients managed solely with a constrained prosthesis and reported that conservative treatment of MCL injuries often led to instability requiring revision. Results showed that patients in whom an unconstrained implant was used had higher failure rates due to instability (4/7; 57%) compared with those treated with constraint (0/37; 0%), regardless of the repair technique used. With the increasing number of TKAs performed on patients at increased risk for iatrogenic MCL injury (i.e., obese patients), the potential impact of this complication requires further investigation. The mid- and long-term effects of this complication remain unclear. The primary objective of this study was to compare the postoperative functional outcomes of patients with iatrogenic MCL injury with matched controls (i.e., TKA patients without intraoperative MCL injury). A secondary objective was to compare functional outcomes for patients managed with different treatment modalities concerning implant constraint and ligament repair.
Patients and Methods After institutional review board approval for this retrospective case study, a query of the operating room information system was conducted identifying 4,183 primary TKAs (CPT code 27447) performed between January 2003 and December 2009. Patients with severe valgus deformities (> 30 degrees), prior MCL injuries, prior knee surgery, or insufficient preoperative data were excluded. Manual examination of the operative reports yielded a total of 32 (0.77%) patients who had sustained intraoperative iatrogenic MCL disruption. These injuries occurred during medial release procedure, cutting of the femur or tibia, or ranging the knee and were confirmed intraoperatively by medial compartment laxity in 30 and 90 degrees of knee flexion position. Of these 32 patients, 3 were deceased and 6 had insufficient preoperative/baseline data leaving 23 (72%) available for follow-up.
Surgeries were performed by seven fellowship trained orthopedic surgeons at a single institution, using a medial parapatellar (n ¼ 18), midvastus (n ¼ 3), or subvastus (n ¼ 2) approach. Preoperative diagnosis was osteoarthritis (n ¼ 22) and rheumatoid arthritis (n ¼ 1). Four patients underwent simultaneous bilateral TKA, none of which had MCL rupture on both the sides. Midsubstance transection occurred in 22 patients, whereas avulsion from the tibial metaphysis was the mechanism of injury in the remaining 1 patient. All MCL injuries and repair types were considered equally in the analysis. Treatment of the MCL disruption was left to the surgeon’s discretion and included an unconstrained implant with primary ligament repair (n ¼ 10), unconstrained implant without ligament repair (n ¼ 2), constrained implant with ligament repair (n ¼ 3), and finally a constrained implant without ligament repair (n ¼ 8). There are some discrepancies among authors over the definition of a “constrained implant,” and so, for that matter, we considered as “constrained” total-stabilizing (TS) implants and totalconstraint (TC) implants, whereas posterior-stabilizing (PS) and cruciate-retaining implants (CR) were considered unconstrained.3 The postoperative ambulatory protocol also varied by surgeon’s preference and ranged from immediate weight bearing with no restrictions to non–weight bearing with the knee kept in full extension. Demographic (age, gender, and body mass index [BMI]) and clinical data including MCL injury treatment, surgical approach, preoperative and postoperative Knee Society Score (KSS), and Knee Function Score (KFS)12 were collected from the electronic medical records. A total of 16 patients did not have postoperative KSS/KFS data in the electronic medical record and thus these patients were contacted by mail or by phone and the modified Knee Society Score (mKSS) and modified Knee Function Score (mKFS)13 were administered. The modified scoring system is a peer-reviewed questionnaire described by Worland et al13 and consists of eight objective questions easily conducted over phone interviews. Parameters evaluated in the mKSS included pain, range of
Table 1 Comparison of demographic and clinical variables for patients and controls Variable Female, n (%) 2
MCL injury (N ¼ 23)
Controls (N ¼ 92)
p-Value
20 (87)
80 (87)
> 0.99
BMI, kg/m , mean SD
32.7 7.8
32.8 7.5
0.94
Age, y, mean SD
66.5 9.7
69.1 9.9
0.85
OA
22 (96)
92 (100)
0.2
RA
1 (4)
0
Heart disease
5 (21.7)
9 (9.8)
0.15
Diabetes type II
4 (17.4)
13 (14.1)
0.74
Renal disease
0
5 (5.4)
0.58
Pulmonary disease
3 (13)
9 (9.8)
0.7
Liver disease
2 (8.7)
2 (2.2)
0.18
Diagnosis, n (%)
Comorbidities, n (%)
Abbreviations: BMI, body mass index; MCL, medial collateral ligament; OA, osteoarthritis; RA, rheumatoid arthritis; SD, standard deviation. The Journal of Knee Surgery
Downloaded by: University of British Columbia. Copyrighted material.
Outcomes of MCL Injuries during TKA
Outcomes of MCL Injuries during TKA
Siqueira et al.
Table 2 Results of KSS and KFS in MCL injury and controls MCL injury (n ¼ 23)
Controls (n ¼ 92)
p-Value
KSS
51.3 17.6
58.4 13.3
0.12
KFS
42.4 12.6
50.3 12.6
0.02
KSS
78.8 24.4
86.7 21
0.03
KFS
67.8 22.9
72.2 25.2
0.35
Preoperative, mean SD
Postoperative, mean SD
Difference between pre- and postoperative scores, mean SD ΔKSS
27.4 33.2
28.3 24.1
0.88
ΔKFS
24.4 24.1
21.6 25.8
0.77
motion, and knee stability, whereas mKFS were calculated by distance walked, ability to go upstairs, ability to completely straighten the knee, and whether gait aides were required for ambulation. Several other knee outcome studies have used the same modified system.14–17 We did not distinguish between the two (i.e., original or modified) in the analysis, and all are reported as KSS and KFS in results. The control group was selected from the remaining list of 4,151 patients who underwent TKA in the time frame of January 2003 to December 2009. The control cohort was matched by gender, age, and BMI. Optimal propensity score matching was done using a 1:4 of case to control patients, which was found adequate to provide the largest possible control group without significantly increasing propensity score variability within the matched groups. Continuous variables were summarized using mean, median, and interquartile range. Categorical variables were summarized using counts and percentages. Because of the small sample size, the nonparametric Wilcoxon and Kruskal– Wallis signed rank-sum tests were used for all comparisons of continuous variables. Post hoc power analysis was performed based on the means, standard deviations, and rates of enrollment in each treatment group. Considering that the Kruskal– Wallis rank-sum test was used for comparison across groups, power was determined by first computing sample sizes for an overall F-test based on analysis of variance (ANOVA). Sample sizes were subsequently inflated based on an asymptotic relative efficiency of 86% for the ANOVA compared with the Kruskal–Wallis rank-sum test. Sample sizes required to have a
power of 80 and 90% were calculated for both KSS and KFS. All analyses were done using R software and MatchIt18 package (version 3.0.2, Vienna, Austria).
Results To satisfy the 1:4 ratio, 92 controls matched by age, gender, and BMI were selected for this study. Demographic and comorbidity data of MCL injury patients and control patients are reported in ►Table 1, there was no significant difference between the two groups. Mean follow-up was 60.3 months for MCL injury patients (range, 12.9–112.6 months) and 52.0 months for control patients (range, 24–112.2 months). The MCL injury cohort had lower baseline KFS compared with the 92 matched controls, 42 versus 50 (p ¼ 0.02), but similar postoperative functional outcomes, 68 versus 72 (p ¼ 0.35). KSS scores, however, were similar to controls preoperatively, 51 versus 58 (p ¼ 0.12), but at last followup, the KSS was lower for those with MCL injury, 79 versus 87 (p ¼ 0.03). The difference between preoperative to postoperative scores did not vary among the two groups for either ΔKSS (p ¼ 0.88) or ΔKFS (p ¼ 0.77) scores (►Table 2). The type of prosthesis used for MCL injury patients and control patients varied at the surgeon’s discretion. For patients with intraoperative MCL injury, implants used included Duracon PS (n ¼ 9), Triathlon PS (n ¼ 1), Duracon TS (n ¼ 8), Triathlon TS (n ¼ 1), Duracon CR (n ¼ 1) (all manufactured by Stryker, Mahwah, NJ), Sigma TCIII (n ¼ 2), and Sigma CR (n ¼ 1) (both manufactured by DePuy, Warsaw, IN). Increased
Table 3 Treatment modality of medial collateral ligament injury Implant
Number of patients (N ¼ 23), n (%)
Unconstrained implant only
1 Duracon PS/1 Triathlon PS
2 (8.7)
Unconstrained implant with repair
8 Duracon PS/1 Duracon CR/1 Sigma PFC-CR
10 (43.5)
Constrained implant only
6 Duracon TS/2 Sigma TCIII
8 (34.8)
Constrained implant with repair
2 Duracon TS/1 Triathlon TS
3 (13.0)
Abbreviations: CR, cruciate retaining; PS, posterior stabilizing; TCIII, total constraint III; TS, total stabilizing. The Journal of Knee Surgery
Downloaded by: University of British Columbia. Copyrighted material.
Abbreviations: KFS, Knee Function Score; KSS, Knee Society Score; MCL, medial collateral ligament.
Outcomes of MCL Injuries during TKA
Siqueira et al.
Table 4 Comparisons of scores between treatment groups Unconstrained (n ¼ 2)
Unconstrained with repair (n ¼ 10)
Constrained (n ¼ 8)
Constrained with repair (n ¼ 3)
p-Value
Preoperative, mean SD KSS
56.5 4.9
58 12.3
35.9 18
67 2
0.015
KFS
45 0
45.5 5.5
33.6 17.7
51.7 7.6
0.14
Postoperative, mean SD KSS
70.5 41.8
79.2 21.6
91.1 10.5
50 36
0.21
KFS
80 28.3
78.1 22
59.3 20.3
51.7 20.8
0.20
Difference between pre- and postoperative scores, mean SD ΔKSS
14 36.8
21.2 23.4
55.3 19.5
17 34.8
0.01
ΔKFS
15 0
31.25 24
29.2 24.8
0 13.2
0.22
any of the patients and none had a diagnosis of infection. As for the 92 knees that composed the control group, 3 underwent manipulation under anesthesia due to arthrofibrosis at an average of 1.7 months and 1 had late instability and required revision surgery at 28 months. To have 80% power to detect a difference in the pre- to postoperative improvement of KSS across the four groups using a Kruskal–Wallis rank-sum test, a total of 42 patients would be required, and a 90% power would require a total of 53 patients. Similarly, to have 80% power to detect a difference in the pre- to postoperative improvement of KFS across the four groups, 61 patients would be required, and a 90% power would require having a total of 77 patients.
constraint implants (i.e., Duracon TS, Triathlon TS, and Sigma TCIII) were used in a total of 11 patients (47.8%). The remaining 12 patients were treated with unconstrained implants, 10 of who also had a primary MCL repair (►Table 3). All control patients were treated with cruciate retaining prosthesis, Triathlon CR (n ¼ 70) or Duracon CR (n ¼ 22) (Stryker, Mahwah, NJ). In all bilateral procedures (n ¼ 24), the same type of implant was used on both the sides. Comparing the treatment modalities for the intraoperative MCL injury, baseline KSS scores differed among the four groups (p ¼ 0.015). Postoperative KSS scores were not significantly different (p ¼ 0.21). KFS scores did not vary among the four treatment groups preoperatively (p ¼ 0.14) or postoperatively (p ¼ 0.2). The difference between preoperative to postoperative scores showed significant difference for KSS (p ¼ 0.01), but not for KFS (p ¼ 0.22) (►Table 4). Pairwise comparisons among the groups were performed for preoperative KSS and the difference between pre- to postoperative KSS. None of the pairwise comparisons revealed statistical differences between the groups (►Tables 5 and 6). Of our total of 23 knees, 22 (87%) with an MCL injury had a satisfactory outcome and were functioning well without evidence of instability or loosening at last follow-up. One patient treated with the TCIII implant and no ligament repair had a patellar fracture at 8 months which was managed conservatively. No further knee surgery or clinical instability was evidenced in
Iatrogenic injury of the MCL during TKA is an uncommon complication.3,4 Identification of this complication often occurs when a sudden laxity in valgus stress is found during the knee-balancing procedure. MCL disruption can be a result of an avulsion from the femoral origin, a midsubstance transection, or a tibial insertion avulsion.5 Primary repair techniques vary according to the location of injury. Avulsions from the femoral origin are generally repaired using a screw and washer,3 whereas midsubstance transections are managed with nonabsorbable sutures or tendon
Table 5 Pairwise comparisons of preoperative KSS scores between treatments
Table 6 Pairwise comparisons of ΔKSS between treatments
Unconstrained with repair
Constrained
Constrained with repair
Unconstrained
0.07
0.6
0.83
Unconstrained with repair
0.6
Constrained
0.6
Constrained with repair
Constrained Constrained with repair Abbreviation: KSS, Knee Society Score. The Journal of Knee Surgery
Discussion
0.12
Constrained
Constrained with repair
Unconstrained
0.09
0.25
> 0.99
Abbreviation: KSS, Knee Society Score
0.09
0.45 0.8
Downloaded by: University of British Columbia. Copyrighted material.
Abbreviations: KFS, Knee Function Score; KSS, Knee Society Score.
augmentation,5,19 and tibial avulsions with nonabsorbable suture anchors or ligament staple.20 In general, conversion from a standard to a constrained implant is the most common and accepted management, without the need for primary repair.21 Constrained implants, however, are associated with higher rates of wear and aseptic loosening and thus many authors tend to avoid this implant in primary TKA.22,23 Conservative treatment for iatrogenic intraoperative MCL disruption with the use of an unconstrained implant with either primary ligament repair or the use of a thicker polyethylene insert has been successfully reported.1,3,5,8 In this study, our experience with this rare complication during a 7-year period revealed that MCL disruption is a rare complication of TKA which is associated with lower postoperative KSS for pain after an average of 5 years. Although our study is potentially underpowered, our data suggest that none of the four treatment modalities analyzed in this study was statistically superior from the others. Out of 4,183 knees operated in our institution in a 7-year time frame, we had a 0.77% rate of MCL rupture. This number is below current literature, which has it between 2.2 and 2.7%.3,4 Underreporting of this complication may have played a role for this result as the method of identification of this event was a manual query through the operative reports. Our results showed that patients with intraoperative MCL disruption had a mean preoperative KSS and KFS of 51 and 42 and postoperative scores of 79 and 68, respectively. Although the postoperative KSS score was lower (p ¼ 0.03) than the average observed for the 92 knees that comprised the matched control group, KFS score did not differ significantly (p ¼ 0.35). Although the 23 patients had a lower baseline KFS score, this did not seem to influence the postoperative result. A significant increase of KSS and KFS scores after surgery was observed for both patient and control groups. This demonstrates that the MCL injury group improved a similar amount on average suggesting that rehabilitation was not hampered by MCL injury. Our results slightly differ from a similar study4 that showed that 37 MCL injured knees had both KSS and KFS significantly lower than the control group, 81 and 74 versus 91 and 87, respectively (p < 0.01). This difference might be accounted for the fact that we compared our patients to a matched control group for age, BMI, and gender, and thus, we were able to remove the influence that those variables may have exerted on the surgical outcome. Another important
Siqueira et al.
difference in our study was the lower KSS and KFS scores obtained at last follow-up when comparing to other published reports on this subject3,4,8,24(►Table 7). Two possible explanations for this is the higher average age at the time of surgery in our cohort (69 years) compared with the average age for those reports (61.9 years), and second, the longer mean follow-up period of our cohort (5 years). We had no failures for instability for any of the 23 patients at last follow-up. The analysis between the treatment modalities for the MCL injury was limited by the small number of patients. Even though preoperative KSS and the difference between preoperative to postoperative KSS was significantly different among the four treatment modalities (p ¼ 0.015 and 0.01, respectively), pairwise comparisons were not able to pin down exactly which groups differ. Considering the two groups that had most patients, unconstrained prosthesis with repair (n ¼ 10) and constrained prosthesis without repair (n ¼ 8), the former had higher postoperative scores overall (79 and 78 versus 91 and 59), but no statistical significance was found. It is important to note that a longer follow-up may show a widening of this difference because of the potential of wear and loosening in the constrained group. The other two groups (i.e., unconstrained prosthesis without repair and constrained prosthesis with repair) had an extremely small number of patients and similar or lower postoperative results. The option for using an unconstrained implant and primarily repairing the MCL instead of using a constrained implant has been the focus of many recent studies. Leopold et al3 treated 16 MCL injuries with primary repair and a cruciate-retaining implant and, after a mean follow-up of 45 months, there were no complications and a KSS score of 93 was obtained. Cruciate-retaining implants allow increased stability, as the posterior cruciate ligament is the second major contributor to maintaining medial soft tissue balance.25 Koo and Choi8 used posterior-stabilizing implants in 15 knees with MCL avulsion from the tibial attachment site. However, instead of primary repair, a thicker polyethylene insert was employed for further stabilization. They also obtained positive results with KSS and KFS scores of 91 and 82, respectively, after a 2-year follow-up, and no failures. Finally, Dragosloveanu et al24 used a combination of thicker polyethylene inserts and primary ligament repair in eight knees with intraoperative MCL rupture and, after a 1-year follow-up, mean KSS and KFS scores of 88 and 80 were
Table 7 Recent studies showing outcomes of intraoperative MCL injuries Studies
Year
N
Follow-up (mo)
Implant
Post-op KSS
Post-op KFS
Revision
Leopold et al3
2001
16
45
12CR/4PS
93
–
0
Koo and Choi8
2009
15
24
13PS/2CR
91
82
0
2011
37
54
7PS/30TCIII
81
74
4
2013
8
12
5PS/3CCK
88
80
1
2014
23
60
9TS/2TCIII/10PS/2CR
79
68
0
Lee and Lotke
4
Dragosloveanu et al
24
Siqueira et al (current study)
Abbreviations: CCK, constrained condylar knee; CR, cruciate retaining; KFS, Knee Function Score; KSS, Knee Society Score; PS, posterior stabilizing; TCIII, total constraint III; TS, total stabilizing. The Journal of Knee Surgery
Downloaded by: University of British Columbia. Copyrighted material.
Outcomes of MCL Injuries during TKA
Siqueira et al.
obtained. Our data are in accordance with those reports in which a conservative approach to this complication can provide results comparable or even superior to those with the use of constrained implants. It is clear, nonetheless, that the wide range of treatments used in this study associated with the small number of cases prevents reaching any definitive conclusions. Our study has inevitably been limited by a small sample size. Considering the rarity of this complication, however, our sample size is in accordance with similar reports in the literature. Different subtypes of MCL injuries or repair modalities were not analyzed separately due to lack of information in the operative reports. Finally, due to the retrospective nature of the study, variables such as surgical approach, postoperative protocols, and surgical technique could not be controlled and may introduce selection bias. Recognizing intraoperative iatrogenic disruptions of the MCL is essential to guide surgical planning and decision making. Even though the most traditional treatment for this unfortunate complication has been conversion to an unlinked constrained prosthesis, a growing number of reports are showing satisfactory results with a more conservative approach that allows the use of a cruciate-retaining or a posterior-stabilized knee.
References
9 10
11
12 13
14
15
16
17
18
1 Stephens S, Politi J, Backes J, Czaplicki T. Repair of medial collateral
2
3
4
5
6
7 8
ligament injury during total knee arthoplasty. Orthopedics 2012; 35(2):e154–e159 Healy WL, Della Valle CJ, Iorio R, et al. Complications of total knee arthroplasty: standardized list and definitions of the Knee Society. Clin Orthop Relat Res 2013;471(1):215–220 Leopold SS, McStay C, Klafeta K, Jacobs JJ, Berger RA, Rosenberg AG. Primary repair of intraoperative disruption of the medical collateral ligament during total knee arthroplasty. J Bone Joint Surg Am 2001;83-A(1):86–91 Lee GC, Lotke PA. Management of intraoperative medial collateral ligament injury during TKA. Clin Orthop Relat Res 2011;469(1): 64–68 Jung KA, Lee SC, Hwang SH, Jung SH. Quadriceps tendon free graft augmentation for a midsubstance tear of the medial collateral ligament during total knee arthroplasty. Knee 2009;16(6): 479–483 Whiteside LA. Correction of ligament and bone defects in total arthroplasty of the severely valgus knee. Clin Orthop Relat Res 1993;(288):234–245 Dimitris K, Taylor BC, Steensen RN. Excursion of oscillating saw blades in total knee arthroplasty. J Arthroplasty 2010;25(1):158–160 Koo MH, Choi CH. Conservative treatment for the intraoperative detachment of medial collateral ligament from the tibial attach-
The Journal of Knee Surgery
19
20
21 22
23
24
25
ment site during primary total knee arthroplasty. J Arthroplasty 2009;24(8):1249–1253 Winiarsky R, Barth P, Lotke P. Total knee arthroplasty in morbidly obese patients. J Bone Joint Surg Am 1998;80(12):1770–1774 Healy WL, Iorio R, Lemos DW. Medial reconstruction during total knee arthroplasty for severe valgus deformity. Clin Orthop Relat Res 1998;(356):161–169 Lachiewicz PF, Soileau ES. Ten-year survival and clinical results of constrained components in primary total knee arthroplasty. J Arthroplasty 2006;21(6):803–808 Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res 1989;(248):13–14 Worland RL, Johnson GV, Alemparte J, Jessup DE, Keenan J, Norambuena N. Ten to fourteen year survival and functional analysis of the AGC total knee replacement system. Knee 2002; 9(2):133–137 Cook LE, Klika AK, Szubski CR, Rosneck J, Molloy R, Barsoum WK. Functional outcomes used to compare single radius and multiradius of curvature designs in total knee arthroplasty. J Knee Surg 2012;25(3):249–253 Yamakado K, Kitaoka K, Yamada H, Hashiba K, Nakamura R, Tomita K. Influence of stability on range of motion after cruciate-retaining TKA. Arch Orthop Trauma Surg 2003;123(1):1–4 Gaasbeek RDA, Nicolaas L, Rijnberg WJ, van Loon CJM, van Kampen A. Correction accuracy and collateral laxity in open versus closed wedge high tibial osteotomy. A one-year randomised controlled study. Int Orthop 2010;34(2):201–207 Sternheim A, Garbedian S, Backstein D. Distal femoral varus osteotomy: unloading the lateral compartment: long-term follow-up of 45 medial closing wedge osteotomies. Orthopedics 2011;34(9):e488–e490 Ho DE, Imai K, King G, Stuart EA. MatchIt: nonparametric preprocessing for parametric causal inference. J Stat Softw 2011; 42(8):1–28 Krackow KA. Management of medial collateral ligament loss. Repair and augmentation. In: Lotke PA, Garion JP, eds. Revision Total Knee Arthroplasty. Philadelphia, PA: Lippincott-Raven; 1999: 227 Mullaji AB, Padmanabhan V, Jindal G. Total knee arthroplasty for profound varus deformity: technique and radiological results in 173 knees with varus of more than 20 degrees. J Arthroplasty 2005;20(5):550–561 Sculco TP. Total condylar III prosthesis in ligament instability. Orthop Clin North Am 1989;20(2):221–226 Lombardi AV Jr, Mallony TH, Eberie RW. Constrained knee arthroplasty. In: Scott WN, ed. The Knee. St. Louis, MO: CV Mosby; 1994: 1305 Rosenberg AG, Verner JJ, Galante JO. Clinical results of total knee revision using the Total Condylar III prosthesis. Clin Orthop Relat Res 1991;(273):83–90 Dragosloveanu S, Cristea S, Stoica C, Dragosloveanu C. Outcome of iatrogenic collateral ligaments injuries during total knee arthroplasty. Eur J Orthop Surg Traumatol 2013 (Epub ahead of print); doi: 10.1007/s00590-013-1330-y Barton TM, Torg JS, Das M. Posterior cruciate ligament insufficiency. A review of the literature. Sports Med 1984;1(6):419–430
Downloaded by: University of British Columbia. Copyrighted material.
Outcomes of MCL Injuries during TKA
Outcomes of Medial Collateral Ligament Injuries during Total Knee Arthroplasty Marcelo B. P. Siqueira, MD1 Kathryn Haller, BA1 Alison K. Klika, MS1 Wael K. Barsoum, MD1
Andrew Mulder, MD1
1 Department of Orthopaedic Surgery, Orthopaedic and
Rheumatologic Institute, Cleveland Clinic, Cleveland, Ohio
Andrew S. Goldblum, MD1
Address for correspondence Alison K. Klika, MS, Cleveland Clinic, 9500 Euclid Avenue, A41, Cleveland, OH 44195 (e-mail: [email protected]).
Abstract
Keywords
► total knee arthroplasty ► intraoperative medial collateral ligament injury ► ligament repair ► constrained implants
Intraoperative medial collateral ligament (MCL) disruption during total knee arthroplasty (TKA) is often managed with either primary repair or use of a constrained implant. A total of 23 patients with an MCL injury during TKA between 2003 and 2009 were compared with 92 matched controls. Of the 23 patients, 10 were treated with an unconstrained implant and primary MCL repair, 8 with constrained implants, 3 with constrained implants and MCL repair, and 2 with unconstrained implants and no MCL repair. After an average 5-year follow-up, patients had lower Knee Society Scores (KSS), 79 versus 87 (p ¼ 0.03), but similar Knee Function Scores (KFS), 68 versus 72 (p ¼ 0.35). The improvement between preoperative and postoperative KSS and KFS did not vary among the two groups (p ¼ 0.88 and p ¼ 0.77, respectively). Postoperative scores did not vary significantly among the four treatment modalities. Conservative treatment can provide satisfactory outcomes and avoid potential complications of increased constraint.
Medial collateral ligament (MCL) integrity is important for coronal plane stability and soft tissue balance after total knee arthroplasty (TKA).1 Iatrogenic MCL injury during TKA is a relatively infrequent but potential adverse event, which requires repair, a change in the level of prosthetic constraint, or revision surgery.2 Studies have reported intraoperative MCL injury rates between 2.2 and 2.7%.3,4 Unidentified MCL injuries can lead to adverse outcomes such as instability and accelerated implant wear, demonstrating the importance of intraoperative recognition and treatment.4,5 Surgeon-controlled risk factors for MCL injury include the use of oscillating saw blades that are wider than the femoral condyles, late removal of medial osteophytes, overstrenuous testing of varus–valgus stability, and hyperflexion of the knee.6,7 Excessive varus contracture commonly found in patients with degenerative osteoarthritis demands a more invasive medial release approach, ultimately increasing the rates of MCL tears.8 Obesity9 and severe deformities6,10,11 are patient-dependent factors,
which also increase the risk of ligament damage and attenuation. Currently, there is no consensus as to the most appropriate way to treat iatrogenic MCL injury during TKA. Various treatment modalities have been used for intraoperative injury, each with reported advantages and disadvantages. The more traditional treatment approach has been conversion to an unlinked constrained prosthesis. Concerns regarding potential loosening secondary to increased stress across the implant–cement interface prompted some surgeons to use nonconstrained prostheses in conjunction with direct primary repair (using suture or screw-and-washer reattachment),3 augmentation5 or thicker polyethylene inserts.8 Leopold et al3 reported positive outcomes in 16 patients treated with primary ligament suture or suture anchor fixation along with an unconstrained implant. Koo and Choi8 managed 15 injured MCLs with a thicker polyethylene insert in combination with unconstrained implants, and reported no instabilities or revisions. In contrast, a recent study by Lee and Lotke4 compared
received May 27, 2014 accepted after revision September 1, 2014
Copyright © by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1394166. ISSN 1538-8506.
Downloaded by: University of British Columbia. Copyrighted material.
J Knee Surg
Siqueira et al.
outcomes of patients treated with ligament repair plus a nonconstrained implant with patients managed solely with a constrained prosthesis and reported that conservative treatment of MCL injuries often led to instability requiring revision. Results showed that patients in whom an unconstrained implant was used had higher failure rates due to instability (4/7; 57%) compared with those treated with constraint (0/37; 0%), regardless of the repair technique used. With the increasing number of TKAs performed on patients at increased risk for iatrogenic MCL injury (i.e., obese patients), the potential impact of this complication requires further investigation. The mid- and long-term effects of this complication remain unclear. The primary objective of this study was to compare the postoperative functional outcomes of patients with iatrogenic MCL injury with matched controls (i.e., TKA patients without intraoperative MCL injury). A secondary objective was to compare functional outcomes for patients managed with different treatment modalities concerning implant constraint and ligament repair.
Patients and Methods After institutional review board approval for this retrospective case study, a query of the operating room information system was conducted identifying 4,183 primary TKAs (CPT code 27447) performed between January 2003 and December 2009. Patients with severe valgus deformities (> 30 degrees), prior MCL injuries, prior knee surgery, or insufficient preoperative data were excluded. Manual examination of the operative reports yielded a total of 32 (0.77%) patients who had sustained intraoperative iatrogenic MCL disruption. These injuries occurred during medial release procedure, cutting of the femur or tibia, or ranging the knee and were confirmed intraoperatively by medial compartment laxity in 30 and 90 degrees of knee flexion position. Of these 32 patients, 3 were deceased and 6 had insufficient preoperative/baseline data leaving 23 (72%) available for follow-up.
Surgeries were performed by seven fellowship trained orthopedic surgeons at a single institution, using a medial parapatellar (n ¼ 18), midvastus (n ¼ 3), or subvastus (n ¼ 2) approach. Preoperative diagnosis was osteoarthritis (n ¼ 22) and rheumatoid arthritis (n ¼ 1). Four patients underwent simultaneous bilateral TKA, none of which had MCL rupture on both the sides. Midsubstance transection occurred in 22 patients, whereas avulsion from the tibial metaphysis was the mechanism of injury in the remaining 1 patient. All MCL injuries and repair types were considered equally in the analysis. Treatment of the MCL disruption was left to the surgeon’s discretion and included an unconstrained implant with primary ligament repair (n ¼ 10), unconstrained implant without ligament repair (n ¼ 2), constrained implant with ligament repair (n ¼ 3), and finally a constrained implant without ligament repair (n ¼ 8). There are some discrepancies among authors over the definition of a “constrained implant,” and so, for that matter, we considered as “constrained” total-stabilizing (TS) implants and totalconstraint (TC) implants, whereas posterior-stabilizing (PS) and cruciate-retaining implants (CR) were considered unconstrained.3 The postoperative ambulatory protocol also varied by surgeon’s preference and ranged from immediate weight bearing with no restrictions to non–weight bearing with the knee kept in full extension. Demographic (age, gender, and body mass index [BMI]) and clinical data including MCL injury treatment, surgical approach, preoperative and postoperative Knee Society Score (KSS), and Knee Function Score (KFS)12 were collected from the electronic medical records. A total of 16 patients did not have postoperative KSS/KFS data in the electronic medical record and thus these patients were contacted by mail or by phone and the modified Knee Society Score (mKSS) and modified Knee Function Score (mKFS)13 were administered. The modified scoring system is a peer-reviewed questionnaire described by Worland et al13 and consists of eight objective questions easily conducted over phone interviews. Parameters evaluated in the mKSS included pain, range of
Table 1 Comparison of demographic and clinical variables for patients and controls Variable Female, n (%) 2
MCL injury (N ¼ 23)
Controls (N ¼ 92)
p-Value
20 (87)
80 (87)
> 0.99
BMI, kg/m , mean SD
32.7 7.8
32.8 7.5
0.94
Age, y, mean SD
66.5 9.7
69.1 9.9
0.85
OA
22 (96)
92 (100)
0.2
RA
1 (4)
0
Heart disease
5 (21.7)
9 (9.8)
0.15
Diabetes type II
4 (17.4)
13 (14.1)
0.74
Renal disease
0
5 (5.4)
0.58
Pulmonary disease
3 (13)
9 (9.8)
0.7
Liver disease
2 (8.7)
2 (2.2)
0.18
Diagnosis, n (%)
Comorbidities, n (%)
Abbreviations: BMI, body mass index; MCL, medial collateral ligament; OA, osteoarthritis; RA, rheumatoid arthritis; SD, standard deviation. The Journal of Knee Surgery
Downloaded by: University of British Columbia. Copyrighted material.
Outcomes of MCL Injuries during TKA
Outcomes of MCL Injuries during TKA
Siqueira et al.
Table 2 Results of KSS and KFS in MCL injury and controls MCL injury (n ¼ 23)
Controls (n ¼ 92)
p-Value
KSS
51.3 17.6
58.4 13.3
0.12
KFS
42.4 12.6
50.3 12.6
0.02
KSS
78.8 24.4
86.7 21
0.03
KFS
67.8 22.9
72.2 25.2
0.35
Preoperative, mean SD
Postoperative, mean SD
Difference between pre- and postoperative scores, mean SD ΔKSS
27.4 33.2
28.3 24.1
0.88
ΔKFS
24.4 24.1
21.6 25.8
0.77
motion, and knee stability, whereas mKFS were calculated by distance walked, ability to go upstairs, ability to completely straighten the knee, and whether gait aides were required for ambulation. Several other knee outcome studies have used the same modified system.14–17 We did not distinguish between the two (i.e., original or modified) in the analysis, and all are reported as KSS and KFS in results. The control group was selected from the remaining list of 4,151 patients who underwent TKA in the time frame of January 2003 to December 2009. The control cohort was matched by gender, age, and BMI. Optimal propensity score matching was done using a 1:4 of case to control patients, which was found adequate to provide the largest possible control group without significantly increasing propensity score variability within the matched groups. Continuous variables were summarized using mean, median, and interquartile range. Categorical variables were summarized using counts and percentages. Because of the small sample size, the nonparametric Wilcoxon and Kruskal– Wallis signed rank-sum tests were used for all comparisons of continuous variables. Post hoc power analysis was performed based on the means, standard deviations, and rates of enrollment in each treatment group. Considering that the Kruskal– Wallis rank-sum test was used for comparison across groups, power was determined by first computing sample sizes for an overall F-test based on analysis of variance (ANOVA). Sample sizes were subsequently inflated based on an asymptotic relative efficiency of 86% for the ANOVA compared with the Kruskal–Wallis rank-sum test. Sample sizes required to have a
power of 80 and 90% were calculated for both KSS and KFS. All analyses were done using R software and MatchIt18 package (version 3.0.2, Vienna, Austria).
Results To satisfy the 1:4 ratio, 92 controls matched by age, gender, and BMI were selected for this study. Demographic and comorbidity data of MCL injury patients and control patients are reported in ►Table 1, there was no significant difference between the two groups. Mean follow-up was 60.3 months for MCL injury patients (range, 12.9–112.6 months) and 52.0 months for control patients (range, 24–112.2 months). The MCL injury cohort had lower baseline KFS compared with the 92 matched controls, 42 versus 50 (p ¼ 0.02), but similar postoperative functional outcomes, 68 versus 72 (p ¼ 0.35). KSS scores, however, were similar to controls preoperatively, 51 versus 58 (p ¼ 0.12), but at last followup, the KSS was lower for those with MCL injury, 79 versus 87 (p ¼ 0.03). The difference between preoperative to postoperative scores did not vary among the two groups for either ΔKSS (p ¼ 0.88) or ΔKFS (p ¼ 0.77) scores (►Table 2). The type of prosthesis used for MCL injury patients and control patients varied at the surgeon’s discretion. For patients with intraoperative MCL injury, implants used included Duracon PS (n ¼ 9), Triathlon PS (n ¼ 1), Duracon TS (n ¼ 8), Triathlon TS (n ¼ 1), Duracon CR (n ¼ 1) (all manufactured by Stryker, Mahwah, NJ), Sigma TCIII (n ¼ 2), and Sigma CR (n ¼ 1) (both manufactured by DePuy, Warsaw, IN). Increased
Table 3 Treatment modality of medial collateral ligament injury Implant
Number of patients (N ¼ 23), n (%)
Unconstrained implant only
1 Duracon PS/1 Triathlon PS
2 (8.7)
Unconstrained implant with repair
8 Duracon PS/1 Duracon CR/1 Sigma PFC-CR
10 (43.5)
Constrained implant only
6 Duracon TS/2 Sigma TCIII
8 (34.8)
Constrained implant with repair
2 Duracon TS/1 Triathlon TS
3 (13.0)
Abbreviations: CR, cruciate retaining; PS, posterior stabilizing; TCIII, total constraint III; TS, total stabilizing. The Journal of Knee Surgery
Downloaded by: University of British Columbia. Copyrighted material.
Abbreviations: KFS, Knee Function Score; KSS, Knee Society Score; MCL, medial collateral ligament.
Outcomes of MCL Injuries during TKA
Siqueira et al.
Table 4 Comparisons of scores between treatment groups Unconstrained (n ¼ 2)
Unconstrained with repair (n ¼ 10)
Constrained (n ¼ 8)
Constrained with repair (n ¼ 3)
p-Value
Preoperative, mean SD KSS
56.5 4.9
58 12.3
35.9 18
67 2
0.015
KFS
45 0
45.5 5.5
33.6 17.7
51.7 7.6
0.14
Postoperative, mean SD KSS
70.5 41.8
79.2 21.6
91.1 10.5
50 36
0.21
KFS
80 28.3
78.1 22
59.3 20.3
51.7 20.8
0.20
Difference between pre- and postoperative scores, mean SD ΔKSS
14 36.8
21.2 23.4
55.3 19.5
17 34.8
0.01
ΔKFS
15 0
31.25 24
29.2 24.8
0 13.2
0.22
any of the patients and none had a diagnosis of infection. As for the 92 knees that composed the control group, 3 underwent manipulation under anesthesia due to arthrofibrosis at an average of 1.7 months and 1 had late instability and required revision surgery at 28 months. To have 80% power to detect a difference in the pre- to postoperative improvement of KSS across the four groups using a Kruskal–Wallis rank-sum test, a total of 42 patients would be required, and a 90% power would require a total of 53 patients. Similarly, to have 80% power to detect a difference in the pre- to postoperative improvement of KFS across the four groups, 61 patients would be required, and a 90% power would require having a total of 77 patients.
constraint implants (i.e., Duracon TS, Triathlon TS, and Sigma TCIII) were used in a total of 11 patients (47.8%). The remaining 12 patients were treated with unconstrained implants, 10 of who also had a primary MCL repair (►Table 3). All control patients were treated with cruciate retaining prosthesis, Triathlon CR (n ¼ 70) or Duracon CR (n ¼ 22) (Stryker, Mahwah, NJ). In all bilateral procedures (n ¼ 24), the same type of implant was used on both the sides. Comparing the treatment modalities for the intraoperative MCL injury, baseline KSS scores differed among the four groups (p ¼ 0.015). Postoperative KSS scores were not significantly different (p ¼ 0.21). KFS scores did not vary among the four treatment groups preoperatively (p ¼ 0.14) or postoperatively (p ¼ 0.2). The difference between preoperative to postoperative scores showed significant difference for KSS (p ¼ 0.01), but not for KFS (p ¼ 0.22) (►Table 4). Pairwise comparisons among the groups were performed for preoperative KSS and the difference between pre- to postoperative KSS. None of the pairwise comparisons revealed statistical differences between the groups (►Tables 5 and 6). Of our total of 23 knees, 22 (87%) with an MCL injury had a satisfactory outcome and were functioning well without evidence of instability or loosening at last follow-up. One patient treated with the TCIII implant and no ligament repair had a patellar fracture at 8 months which was managed conservatively. No further knee surgery or clinical instability was evidenced in
Iatrogenic injury of the MCL during TKA is an uncommon complication.3,4 Identification of this complication often occurs when a sudden laxity in valgus stress is found during the knee-balancing procedure. MCL disruption can be a result of an avulsion from the femoral origin, a midsubstance transection, or a tibial insertion avulsion.5 Primary repair techniques vary according to the location of injury. Avulsions from the femoral origin are generally repaired using a screw and washer,3 whereas midsubstance transections are managed with nonabsorbable sutures or tendon
Table 5 Pairwise comparisons of preoperative KSS scores between treatments
Table 6 Pairwise comparisons of ΔKSS between treatments
Unconstrained with repair
Constrained
Constrained with repair
Unconstrained
0.07
0.6
0.83
Unconstrained with repair
0.6
Constrained
0.6
Constrained with repair
Constrained Constrained with repair Abbreviation: KSS, Knee Society Score. The Journal of Knee Surgery
Discussion
0.12
Constrained
Constrained with repair
Unconstrained
0.09
0.25
> 0.99
Abbreviation: KSS, Knee Society Score
0.09
0.45 0.8
Downloaded by: University of British Columbia. Copyrighted material.
Abbreviations: KFS, Knee Function Score; KSS, Knee Society Score.
augmentation,5,19 and tibial avulsions with nonabsorbable suture anchors or ligament staple.20 In general, conversion from a standard to a constrained implant is the most common and accepted management, without the need for primary repair.21 Constrained implants, however, are associated with higher rates of wear and aseptic loosening and thus many authors tend to avoid this implant in primary TKA.22,23 Conservative treatment for iatrogenic intraoperative MCL disruption with the use of an unconstrained implant with either primary ligament repair or the use of a thicker polyethylene insert has been successfully reported.1,3,5,8 In this study, our experience with this rare complication during a 7-year period revealed that MCL disruption is a rare complication of TKA which is associated with lower postoperative KSS for pain after an average of 5 years. Although our study is potentially underpowered, our data suggest that none of the four treatment modalities analyzed in this study was statistically superior from the others. Out of 4,183 knees operated in our institution in a 7-year time frame, we had a 0.77% rate of MCL rupture. This number is below current literature, which has it between 2.2 and 2.7%.3,4 Underreporting of this complication may have played a role for this result as the method of identification of this event was a manual query through the operative reports. Our results showed that patients with intraoperative MCL disruption had a mean preoperative KSS and KFS of 51 and 42 and postoperative scores of 79 and 68, respectively. Although the postoperative KSS score was lower (p ¼ 0.03) than the average observed for the 92 knees that comprised the matched control group, KFS score did not differ significantly (p ¼ 0.35). Although the 23 patients had a lower baseline KFS score, this did not seem to influence the postoperative result. A significant increase of KSS and KFS scores after surgery was observed for both patient and control groups. This demonstrates that the MCL injury group improved a similar amount on average suggesting that rehabilitation was not hampered by MCL injury. Our results slightly differ from a similar study4 that showed that 37 MCL injured knees had both KSS and KFS significantly lower than the control group, 81 and 74 versus 91 and 87, respectively (p < 0.01). This difference might be accounted for the fact that we compared our patients to a matched control group for age, BMI, and gender, and thus, we were able to remove the influence that those variables may have exerted on the surgical outcome. Another important
Siqueira et al.
difference in our study was the lower KSS and KFS scores obtained at last follow-up when comparing to other published reports on this subject3,4,8,24(►Table 7). Two possible explanations for this is the higher average age at the time of surgery in our cohort (69 years) compared with the average age for those reports (61.9 years), and second, the longer mean follow-up period of our cohort (5 years). We had no failures for instability for any of the 23 patients at last follow-up. The analysis between the treatment modalities for the MCL injury was limited by the small number of patients. Even though preoperative KSS and the difference between preoperative to postoperative KSS was significantly different among the four treatment modalities (p ¼ 0.015 and 0.01, respectively), pairwise comparisons were not able to pin down exactly which groups differ. Considering the two groups that had most patients, unconstrained prosthesis with repair (n ¼ 10) and constrained prosthesis without repair (n ¼ 8), the former had higher postoperative scores overall (79 and 78 versus 91 and 59), but no statistical significance was found. It is important to note that a longer follow-up may show a widening of this difference because of the potential of wear and loosening in the constrained group. The other two groups (i.e., unconstrained prosthesis without repair and constrained prosthesis with repair) had an extremely small number of patients and similar or lower postoperative results. The option for using an unconstrained implant and primarily repairing the MCL instead of using a constrained implant has been the focus of many recent studies. Leopold et al3 treated 16 MCL injuries with primary repair and a cruciate-retaining implant and, after a mean follow-up of 45 months, there were no complications and a KSS score of 93 was obtained. Cruciate-retaining implants allow increased stability, as the posterior cruciate ligament is the second major contributor to maintaining medial soft tissue balance.25 Koo and Choi8 used posterior-stabilizing implants in 15 knees with MCL avulsion from the tibial attachment site. However, instead of primary repair, a thicker polyethylene insert was employed for further stabilization. They also obtained positive results with KSS and KFS scores of 91 and 82, respectively, after a 2-year follow-up, and no failures. Finally, Dragosloveanu et al24 used a combination of thicker polyethylene inserts and primary ligament repair in eight knees with intraoperative MCL rupture and, after a 1-year follow-up, mean KSS and KFS scores of 88 and 80 were
Table 7 Recent studies showing outcomes of intraoperative MCL injuries Studies
Year
N
Follow-up (mo)
Implant
Post-op KSS
Post-op KFS
Revision
Leopold et al3
2001
16
45
12CR/4PS
93
–
0
Koo and Choi8
2009
15
24
13PS/2CR
91
82
0
2011
37
54
7PS/30TCIII
81
74
4
2013
8
12
5PS/3CCK
88
80
1
2014
23
60
9TS/2TCIII/10PS/2CR
79
68
0
Lee and Lotke
4
Dragosloveanu et al
24
Siqueira et al (current study)
Abbreviations: CCK, constrained condylar knee; CR, cruciate retaining; KFS, Knee Function Score; KSS, Knee Society Score; PS, posterior stabilizing; TCIII, total constraint III; TS, total stabilizing. The Journal of Knee Surgery
Downloaded by: University of British Columbia. Copyrighted material.
Outcomes of MCL Injuries during TKA
Siqueira et al.
obtained. Our data are in accordance with those reports in which a conservative approach to this complication can provide results comparable or even superior to those with the use of constrained implants. It is clear, nonetheless, that the wide range of treatments used in this study associated with the small number of cases prevents reaching any definitive conclusions. Our study has inevitably been limited by a small sample size. Considering the rarity of this complication, however, our sample size is in accordance with similar reports in the literature. Different subtypes of MCL injuries or repair modalities were not analyzed separately due to lack of information in the operative reports. Finally, due to the retrospective nature of the study, variables such as surgical approach, postoperative protocols, and surgical technique could not be controlled and may introduce selection bias. Recognizing intraoperative iatrogenic disruptions of the MCL is essential to guide surgical planning and decision making. Even though the most traditional treatment for this unfortunate complication has been conversion to an unlinked constrained prosthesis, a growing number of reports are showing satisfactory results with a more conservative approach that allows the use of a cruciate-retaining or a posterior-stabilized knee.
References
9 10
11
12 13
14
15
16
17
18
1 Stephens S, Politi J, Backes J, Czaplicki T. Repair of medial collateral
2
3
4
5
6
7 8
ligament injury during total knee arthoplasty. Orthopedics 2012; 35(2):e154–e159 Healy WL, Della Valle CJ, Iorio R, et al. Complications of total knee arthroplasty: standardized list and definitions of the Knee Society. Clin Orthop Relat Res 2013;471(1):215–220 Leopold SS, McStay C, Klafeta K, Jacobs JJ, Berger RA, Rosenberg AG. Primary repair of intraoperative disruption of the medical collateral ligament during total knee arthroplasty. J Bone Joint Surg Am 2001;83-A(1):86–91 Lee GC, Lotke PA. Management of intraoperative medial collateral ligament injury during TKA. Clin Orthop Relat Res 2011;469(1): 64–68 Jung KA, Lee SC, Hwang SH, Jung SH. Quadriceps tendon free graft augmentation for a midsubstance tear of the medial collateral ligament during total knee arthroplasty. Knee 2009;16(6): 479–483 Whiteside LA. Correction of ligament and bone defects in total arthroplasty of the severely valgus knee. Clin Orthop Relat Res 1993;(288):234–245 Dimitris K, Taylor BC, Steensen RN. Excursion of oscillating saw blades in total knee arthroplasty. J Arthroplasty 2010;25(1):158–160 Koo MH, Choi CH. Conservative treatment for the intraoperative detachment of medial collateral ligament from the tibial attach-
The Journal of Knee Surgery
19
20
21 22
23
24
25
ment site during primary total knee arthroplasty. J Arthroplasty 2009;24(8):1249–1253 Winiarsky R, Barth P, Lotke P. Total knee arthroplasty in morbidly obese patients. J Bone Joint Surg Am 1998;80(12):1770–1774 Healy WL, Iorio R, Lemos DW. Medial reconstruction during total knee arthroplasty for severe valgus deformity. Clin Orthop Relat Res 1998;(356):161–169 Lachiewicz PF, Soileau ES. Ten-year survival and clinical results of constrained components in primary total knee arthroplasty. J Arthroplasty 2006;21(6):803–808 Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res 1989;(248):13–14 Worland RL, Johnson GV, Alemparte J, Jessup DE, Keenan J, Norambuena N. Ten to fourteen year survival and functional analysis of the AGC total knee replacement system. Knee 2002; 9(2):133–137 Cook LE, Klika AK, Szubski CR, Rosneck J, Molloy R, Barsoum WK. Functional outcomes used to compare single radius and multiradius of curvature designs in total knee arthroplasty. J Knee Surg 2012;25(3):249–253 Yamakado K, Kitaoka K, Yamada H, Hashiba K, Nakamura R, Tomita K. Influence of stability on range of motion after cruciate-retaining TKA. Arch Orthop Trauma Surg 2003;123(1):1–4 Gaasbeek RDA, Nicolaas L, Rijnberg WJ, van Loon CJM, van Kampen A. Correction accuracy and collateral laxity in open versus closed wedge high tibial osteotomy. A one-year randomised controlled study. Int Orthop 2010;34(2):201–207 Sternheim A, Garbedian S, Backstein D. Distal femoral varus osteotomy: unloading the lateral compartment: long-term follow-up of 45 medial closing wedge osteotomies. Orthopedics 2011;34(9):e488–e490 Ho DE, Imai K, King G, Stuart EA. MatchIt: nonparametric preprocessing for parametric causal inference. J Stat Softw 2011; 42(8):1–28 Krackow KA. Management of medial collateral ligament loss. Repair and augmentation. In: Lotke PA, Garion JP, eds. Revision Total Knee Arthroplasty. Philadelphia, PA: Lippincott-Raven; 1999: 227 Mullaji AB, Padmanabhan V, Jindal G. Total knee arthroplasty for profound varus deformity: technique and radiological results in 173 knees with varus of more than 20 degrees. J Arthroplasty 2005;20(5):550–561 Sculco TP. Total condylar III prosthesis in ligament instability. Orthop Clin North Am 1989;20(2):221–226 Lombardi AV Jr, Mallony TH, Eberie RW. Constrained knee arthroplasty. In: Scott WN, ed. The Knee. St. Louis, MO: CV Mosby; 1994: 1305 Rosenberg AG, Verner JJ, Galante JO. Clinical results of total knee revision using the Total Condylar III prosthesis. Clin Orthop Relat Res 1991;(273):83–90 Dragosloveanu S, Cristea S, Stoica C, Dragosloveanu C. Outcome of iatrogenic collateral ligaments injuries during total knee arthroplasty. Eur J Orthop Surg Traumatol 2013 (Epub ahead of print); doi: 10.1007/s00590-013-1330-y Barton TM, Torg JS, Das M. Posterior cruciate ligament insufficiency. A review of the literature. Sports Med 1984;1(6):419–430
Downloaded by: University of British Columbia. Copyrighted material.
Outcomes of MCL Injuries during TKA
