Anterior cruciate ligament (ACL) reconstruction techniques have long been studied, and 3 general modes of graft fixation have been considered: interference screws, suspension buttons, and cross pin fixation.1–4 The use of cortical buttons for femoral-sided graft fixation in ACL reconstruction is a widely accepted fixation technique, which has proven to be effective.5 Cortical button femoral fixation most commonly implements the use of soft tissue grafts such as hamstring grafts.6 There are many factors that potentially contribute to the risk of re-rupture, including the source of graft (autograft vs allograft) and the type of graft (bone–tendon–bone, hamstring, quadriceps tendon). Additional factors include patient age and activity level as well as tunnel placement and graft fixation method.2,3,7 Recent studies suggest an increased risk of revision ACL surgery with hamstring grafts in certain subsets of patients.5,8 With this potential increased risk of re-rupture, further work is necessary to evaluate and define the role that femoral cortical button position has in knee stability.
Several studies have been undertaken to evaluate the biomechanics of ACL grafts using adjustable loop devices.9–11 However, there has been no specific assessment of button positioning and its congruency with the femoral cortex. One study reported that buttons placed along the anterior, as compared with the lateral, cortex showed increased strength despite the cortex being thinner, but congruency was not specifically noted.12 An additional study discussed cortical button malposition in relation to width of the femoral tunnel but did not define placement of the button in relation to the cortex.13 To date, no study has evaluated the cortical congruency of femoral cortical buttons following ACL reconstruction, nor any classification system devised. Because the ACL is the most frequently reconstructed ligament of the knee,14 with cortical button femoral fixation being a commonly used method, further investigation is warranted to define implant position.15,16 In the current study, the authors propose a novel classification system to grade femoral cortical button placement. The benefit to developing a reproducible grading scheme is that it will provide a uniform way to describe button placement; subsequently, the effects of button cortical congruency can be studied in terms of both biomechanical strength and clinical outcomes. The purpose of this study was to develop a classification system for femoral cortical button positioning that is both accurate and reproducible and to determine the rate of cortical button malposition during ACL reconstruction. The authors hypothesized that cortical button malposition would be infrequent, with an occurrence of less than 5%.
Materials and Methods
Institutional review board approval was obtained for this study. A retrospective review was performed using a single center's ACL registry during a 5-year period from 2009 to 2014. Patients were identified using the following selection criteria: primary ACL reconstruction, graft fixation with femoral cortical button, skeletal maturity, and postoperative anteroposterior and lateral radiographs available for review. Radiographs were obtained within the first 6 weeks postoperatively. Exclusion criteria included revision ACL reconstruction, skeletal immaturity (epiphyseal buttons), multiligamentous knee injuries, and lack of appropriate postoperative imaging.
A chart review for data gathering was undertaken. Surgical variables were investigated, including femoral fixation method, tibial fixation method, graft type, meniscal involvement, articular cartilage involvement, surgical technique for drilling of the femoral tunnel, use of intraoperative imaging, and specific type of suspensory button implant. Patients meeting only the above mentioned selection criteria were included in the study.
The data from the ACL registry were sorted using the selection criteria listed above. After these criteria were met, for each patient meeting selection criteria, the distance of the cortical button from the femoral cortex was measured by 2 observers (A.C.T., A.A.D.). Postoperative button position was graded as follows: reduced and congruent (entirety of button <2 mm from cortex; Figure 1); reduced and incongruent (part of button <2 mm from cortex, part of button >2 mm from cortex; Figure 2); displaced (entirety of button >2 mm from cortex; Figure 3); intraosseous (all or part of button remains within bone; Figure 4); or ungradable (button cannot be graded based on radiographs). Radiographs were evaluated by 2 individuals (orthopedic staff surgeon and orthopedic resident physician: A.C.T., J.A.M.) at 2 time points to define inter- and intra-observer reliability. All 361 radiographs were independently graded by the same 2 individuals to determine interrater reliability. After this was completed, the data were compiled. Regrading was then performed on deidentified, randomized data after at least 2 weeks to calculate intra-rater reliability.
Anteroposterior radiograph showing a reduced and congruent button.
Anteroposterior radiograph showing a reduced and incongruent button.
Anteroposterior radiograph showing a displaced button.
Anteroposterior (A) and lateral (B) radiographs showing an intraosseous button.
The reliability of the 2 raters was determined with the use of the kappa statistic to identify what proportion of agreement existed that was not due to chance. The interrater reliability was used to determine the agreement between each observer for each patient. The intrarater reliability was used to quantify the agreement between the first and second grading of the radiographs for each rater. These were calculated with Stata version 14.2 software (Stata Corp, College Station, Texas), and Landis' guidelines for kappa interpretation were used.17 Kappa values of 0 to 0.2 indicated poor or slight agreement, 0.2 to 0.4 indicated fair agreement, 0.4 to 0.6 indicated moderate agreement, 0.6 to 0.8 indicated substantial agreement, and 0.8 to 1.0 indicated excellent or almost perfect agreement.
A total of 2182 patients underwent ACL reconstruction between 2009 and 2014. Of these, 563 patients had cortical button fixation of the femoral side of the graft. A total of 369 of these patients had appropriate postoperative radiographs, but 8 were excluded because of revision status or skeletal immaturity. Thus, a total of 361 patients met the selection criteria. Of the 361 total buttons, 312 (86.43%) were reduced and congruent, 18 (4.99%) were reduced and incongruent, 10 (2.77%) were displaced, 13 (3.60%) were intraosseous, and 8 (2.21%) were ungradable based on the postoperative imaging.
The interrater reliability between the 2 raters showed excellent agreement, with an overall kappa value of 0.84. In addition to the overall interrater reliability calculation, separate kappa values for each grade assigned are summarized in Table 1. Intrarater reliability for rater 1 showed substantial agreement (91.84%), and intrarater reliability for rater 2 showed excellent agreement (93.88%), with kappa values of 0.77 and 0.83, respectively.
Interrater Reliability Scores Between Rater 1 and Rater 2 on the Various Cortical Button Congruency Grades
The aim of this study was to evaluate the position of cortical buttons in femoral fixation during ACL reconstruction. The results of the proposed grading system show that this classification scheme has excellent intrarater and interrater reliability. It is simple, accurate, and reproducible, as indicated by the high kappa values.
What constitutes cortical button malposition must be defined before further study. With a uniform description of button congruency, studies that examine the clinical outcomes and potential effects of button malposition can be more easily facilitated. Malpositioned buttons, as indicated in the current study, would include the following grades: reduced and incongruent, displaced, intraosseous, and ungradable. The authors' hypothesis that cortical button malposition would be an infrequent occurrence of less than 5% was found to be false, with rates of suboptimal position approximately 15%. Although this was higher than expected, there had been no prior studies either defining button malposition or describing the rate of its occurrence.
Strengths of this study included a large registry data set with many surgical variables recorded. There was a substantial number of patients meeting the clearly defined selection criteria. There are numerous reasons for a cortical button to not sit in a congruent manner on the femoral cortex, such as soft tissue interposition and deployment malfunction. The use of intraoperative imaging could be considered to aid in congruent placement, but even with fluoroscopic imaging, buttons could remain displaced and incongruent if soft tissue stays interposed and no further dissection of the soft tissues is made. Care should be taken as the button is deployed onto the femoral cortex after passing through the drilled tunnel. Tracking the button's progress with the arthroscope in a separate portal can be used to aid in this portion of the procedure.
Still the question remains, what is the clinical significance of cortical button malposition? Several studies have examined the biomechanical strength of ACL grafts using adjustable loop devices such as cortical buttons.9–11 No such study has evaluated the effect of cortical congruency of the device on the durability and strength of the construct. Theoretically, if the cortical button is not flush with the cortex, such as the case with a displaced button, the soft tissue underneath may not provide a reliable platform on which the button can function. Also, if the soft tissue under the button were to become nonviable, the button may settle and subsequently become congruent with the cortex, but there could be increased laxity introduced to the construct as a result. Further study is warranted to determine the effect that cortical button malposition has on the strength of ACL graft fixation. Another area of future research would involve prospectively collecting subjective outcome measures to determine if patients with reduced and congruent buttons have a more optimal outcome. This study provided a simple, accurate, and reproducible grading system that can form the groundwork for future studies.
This study had several limitations. It was a retrospective case series and was subject to both confirmation and measurement bias. Attempts to control for this included having the 2 raters grade the radiographs independently and then blinding them when regrading occurred to determine intrarater reliability. Additionally, the use of anteroposterior and/or lateral radiographs did result in some buttons being deemed ungradable, but this was a relatively small number. Oblique radiographs could have been used to identify and/or confirm true congruency with the femoral cortex. It is possible that some reduced and congruent buttons would have been graded differently if these views had been available. Additionally, it is conceivable that a button deemed ungradable would have been assigned a more specific grade if additional views had been obtained. Oblique radiographs were not part of the routine imaging studies obtained in this clinical setting. The greatest limitation of the study was that although it did show that the proposed grading system was reproducible, it did not prove any correlation with patient outcomes, as that was beyond its scope. It does serve as a foundation from which to build additional studies that can be designed to investigate specific variables such as revision rates and subjective outcome measures.
This study presented a novel classification system of femoral cortical button positioning in ACL reconstruction that is both reproducible and accurate, with excellent intrarater and interrater reliability. The rate of cortical button malposition in this study was higher than expected, approximately 15%. Further studies are warranted to determine the effect that cortical button malposition has on the strength of ACL graft fixation and if patients with reduced and congruent buttons have improved clinical outcomes and lower revision rates.
- Ahmad CS, Gardner TR, Groh M, Arnouk J, Levine WN. Mechanical properties of soft tissue femoral fixation devices for anterior cruciate ligament reconstruction. Am J Sports Med. 2004;32(3):635–640. doi:10.1177/0363546503261714 [CrossRef]
- Beynnon BD, Johnson RJ, Abate JA, Fleming BC, Nichols CE. Treatment of anterior cruciate ligament injuries, part 1. Am J Sports Med. 2005;33(10):1579–1602. doi:10.1177/0363546505279913 [CrossRef]
- Beynnon BD, Johnson RJ, Abate JA, Fleming BC, Nichols CE. Treatment of anterior cruciate ligament injuries, part 2. Am J Sports Med. 2005;33(11):1751–1767. doi:10.1177/0363546505279922 [CrossRef]
- Colvin A, Sharma C, Parides M, Glashow J. What is the best femoral fixation of hamstring autografts in anterior cruciate ligament reconstruction? A meta-analysis. Clin Orthop Relat Res. 2011;469(4):1075–1081. doi:10.1007/s11999-010-1662-4 [CrossRef]
- Rahr-Wagner L, Thillemann TM, Pedersen AB, Lind M. Comparison of hamstring tendon and patellar tendon grafts in anterior cruciate ligament reconstruction in a nationwide population-based cohort study: results from the Danish registry of knee ligament reconstruction. Am J Sports Med. 2014;42(2):278–284. doi:10.1177/0363546513509220 [CrossRef]
- Samuelsson K, Andersson D, Ahldén M, Fu FH, Musahl V, Karlsson J. Trends in surgeon preferences on anterior cruciate ligament reconstructive techniques. Clin Sports Med. 2013;32(1):111–126. doi:10.1016/j.csm.2012.08.011 [CrossRef]
- Shelton WR, Fagan BC. Autografts commonly used in anterior cruciate ligament reconstruction. J Am Acad Orthop Surg. 2011;19(5):259–264. doi:10.5435/00124635-201105000-00003 [CrossRef]
- Persson A, Fjeldsgaard K, Gjertsen JE, et al. Increased risk of revision with hamstring tendon grafts compared with patellar tendon grafts after anterior cruciate ligament reconstruction: a study of 12,643 patients from the Norwegian Cruciate Ligament Registry, 2004–2012. Am J Sports Med. 2014;42(2):285–291. doi:10.1177/0363546513511419 [CrossRef]
- Barrow AE, Pilia M, Guda T, Kadrmas WR, Burns TC. Femoral suspension devices for anterior cruciate ligament reconstruction: do adjustable loops lengthen?Am J Sports Med.2014;42(2):343–349. doi:10.1177/0363546513507769 [CrossRef]
- Petre BM, Smith SD, Jansson KS, et al. Femoral cortical suspension devices for soft tissue anterior cruciate ligament reconstruction: a comparative biomechanical study. Am J Sports Med. 2013;41(2):416–422. doi:10.1177/0363546512469875 [CrossRef]
- Kamelger FS, Onder U, Schmoelz W, Tecklenburg K, Arora R, Fink C. Suspensory fixation of grafts in anterior cruciate ligament reconstruction: a biomechanical comparison of 3 implants. Arthroscopy. 2009;25(7):767–776. doi:10.1016/j.arthro.2009.01.021 [CrossRef]
- Conner CS, Perez BA, Morris RP, Buckner JW, Buford WL Jr, Ivey FM. Three femoral fixation devices for anterior cruciate ligament reconstruction: comparison of fixation on the lateral cortex versus the anterior cortex. Arthroscopy. 2010;26(6):796–807. doi:10.1016/j.arthro.2009.10.015 [CrossRef]
- Okazaki K, Matsubara H, Osaki K, et al. Femoral tunnel apertures on the lateral cortex in anterior cruciate ligament reconstruction: an analysis of cortical button fixation. Arthroscopy. 2014;30(7):841–848. doi:10.1016/j.arthro.2014.03.004 [CrossRef]
- Magnussen RA, Lawrence JTR, West RL, Toth AP, Taylor DC, Garrett WE. Graft size and patient age are predictors of early revision after anterior cruciate ligament reconstruction with hamstring autograft. Arthroscopy. 2012;28(4):526–531. doi:10.1016/j.arthro.2011.11.024 [CrossRef]
- Ahldén M, Samuelsson K, Sernert N, Forssblad M, Karlsson J, Kartus J. The Swedish National Anterior Cruciate Ligament Register: a report on baseline variables and outcomes of surgery for almost 18,000 patients. Am J Sports Med. 2012;40(10):2230–2235. doi:10.1177/0363546512457348 [CrossRef]
- Andernord D, Björnsson H, Petzold M, et al. Surgical predictors of early revision surgery after anterior cruciate ligament reconstruction: results from the Swedish National Knee Ligament Register on 13,102 patients. Am J Sports Med. 2014;42(7):1574–1582. doi:10.1177/0363546514531396 [CrossRef]
- Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159–174. doi:10.2307/2529310 [CrossRef]
Interrater Reliability Scores Between Rater 1 and Rater 2 on the Various Cortical Button Congruency Grades