Introduction
Recently, it has become apparent that many patients have little knowledge regarding anterior cruciate ligament (ACL) injuries and treatment.1 One study has shown that only 30% of patients were aware of the sex-based differences associated with the risk for ACL injury, with women having a higher incidence of injury, and only 37% knew that surgery does not decrease the risk for knee osteoarthritis (OA).2 Despite this, high patient expectations have been reported in patients undergoing ACLR.3 These pre-operative expectations have been closely tied to a patient’s assessment of outcomes, which is one determinant of patient satisfaction3 as well as the success of the procedure.1,4 Because of the increasing incidence of reported ACL injuries in the literature,5,6 it is important for orthopaedic surgeons to educate their patients on a variety of postoperative considerations following ACLR, including return to sport, revision risk, risk of a contralateral ACL tear, postoperative knee OA, and eventual need for total knee arthroplasty (TKA). The purpose of this manuscript is to highlight the current literature on important postoperative considerations for patients undergoing ACLR, which surgeons can use to help educate their patients on what to expect following this procedure.
Return to Sport
Patients who sustain an ACL injury typically undergo ACLR to enable their return to sports activity and knee function.7,8 The timeframe at which patients return to sport (RTS) can be as early as 6 to 9 months postoperatively, although occasionally is restricted to 12 months postoperatively based on surgeon preferences regarding rehabilitation schedule.7,9 The rate of RTS differs between patient populations, and therefore it is important for surgeons to adequately educate patients about the factors that can influence their level of RTS timeframe to provide realistic expectations.
In a previous systematic review and meta-analysis of 69 studies comprised of 7,556 patients, Ardern et al10 found that, on average, 81% of individuals of all ages (average age, 25.8 years) return to any sport, with 65% returning to their pre-injury activity level, and 55% returning to a competitive level of sport at an average follow-up of 40 months. Although it is less common for elderly patients to undergo ACLR, a recent case series11 of 12 active patients over the age of 60 years reported that 83% of elderly patients return to any sport, with 50% resuming their pre-injury level of skiing at a mean follow-up of 49.6 months. In another systematic review and meta-analysis of 20 studies and 1,156 adolescent patients (average age, 14.3 years), Kay et al9 found that, on average, 92.0% of adolescents return to any sport, with 78.6% returning to their pre-injury activity level, and 81.0% returning to a competitive level of sport at a mean follow-up of 6.5 years. In a systematic review and meta-analysis of elite athletes by Lai et al7 of 24 studies and 1,272 patients, return to pre-injury activity level occurred at a rate of 83%. More specifically, the rate of RTS was 85% among elite soccer players (n=220), 78% among elite American football players (n=279), and 82% among elite basketball players (n=103).7 While a high proportion of adolescents and elite athletes return to their pre-injury level of sport following ACLR,7,9,10 there is a high rate of graft rupture (adolescents, 13%; elite athletes, 5.2%) and contralateral ACL injury (adolescents, 14%) among more active populations.7,9
Several factors can positively influence the level of a patient’s return to sports activity including playing an elite sport,7,10 younger age, male sex,10,12,13 a positive psychological response,14-18 symmetrical hopping performance,10 primary reconstruction,19 and use of autograft.9-11,20 It has been reported that men are approximately 1.5-times more likely than women to return to either their pre-injury level of sport (odds ratio=1.4) or competitive sport (odds ratio=1.7), though no significant difference in rate of return to any sport has been demonstrated between males and females.10 While empirical data is needed, some of the age and sex differences found in the literature may be explained by the difference in amount of time able to participate in a sport as well as social roles.10,21,22 However, because age, sex, and pre-injury sports participation level are non-modifiable factors affecting RTS, attention may be better focused on modifiable factors such as physical functioning and psychological response. Because physical functioning is a prerequisite to RTS, postoperative rehabilitation is fundamental to facilitate the knee function required to participate in sports activity.15,18 It is often seen that patients do not return to their pre-injury level of sport following ACLR despite adequate restoration of knee function,16 which is possibly attributed to the difference in psychological state among patients.14,16,17 A recent qualitative study determined that the decision to RTS after ACLR was largely based on psychosocial factors, such as hesitancy, lack of self-confidence, fear of re-injury, and changes in priorities or personal expectations, which may be independent of physical function.23 Burland et al23 suggested that many of these factors have the potential to be addressed in the rehabilitation setting. Self-confidence, optimism, and self-motivation are predictive of successful outcomes and positively influence the level of RTS and patient satisfaction.15.16,18
Subsequent Surgery
At 6-year follow-up, it has been reported that 18.9% of ACLR patients undergo subsequent surgery on the ipsilateral leg, including cartilage procedures (13.3%), arthrofibrosis procedures (5.4%), and procedures related to hardware (2.4%).24 Similarly, in a cohort study including 14,522 primary ACLRs, the non-revision reoperation rate per 100 person-years was 1.1 for meniscus, 0.3 for cartilage, 0.4 for hardware removal, and 0.4 for arthrofibrosis.25 Risk factors for reoperations vary depending on the type of surgery evaluated. These include previous meniscal repair,26 female sex, allografts, prior surgery, older patient age (17 versus ≥26 years), and being operated on by a sports medicine fellowship-trained surgeon.25 Additionally, multiple injuries may develop due to delayed ACLR, and thus lead to subsequent injuries to meniscus and cartilage that require additional operative treatment during or following primary ACLR.27-29
Revision Risk
The rate of revision ACLR has been reported to be between 1.7% and 7.7%.24,30,31 A recent review by Kraeutler et al32 highlighted that ACL graft failure may result from a combination of technical errors, biological causes, and trauma. Although there is poor interobserver reliability among surgeons with respect to which failures are the result of technical errors,33 non-anatomic tunnel placement is a major factor that can contribute to primary ACLR graft failure (Figure 1).32,34 Additional risk factors for graft failure include younger age,32,35-38 female gender,26 higher activity level,32,35,36 use of (irradiated) allograft,32,36,39-43 lower limb malalignment,32,44 and increased tibial slope.32,45 Young female soccer players are at an extremely high risk of graft failure due to their young age and participation in a high-level, pivoting sport.12,32,46,47 This is supported by Ahldén et al46 who found that 22.0% of 15-18 year-old female soccer players reported a revision (11.8%) or contralateral ACLR (10.2%) during a 5-year period, which was significantly more than the corresponding age-matched male subgroup (revision, 5.4%; contralateral, 10.2%; p=0.02) and all patients (revision, 4.1%; contralateral, 5.0%; p<0.001).46
Graft choice for primary ACLR is a subject that is often researched and debated. Common graft choices are bone-patellar tendon-bone (BPTB) autograft, hamstring autograft, quadriceps autograft, as well as multiple allograft options.48
A recent systematic review
of overlapping meta-analyses of 16 studies and 1,396 patients by Schuette et al48 found that the current evidence is not strong enough to support a significant difference in graft failure between BPTB and hamstring autografts. Notably, this study suggested that ACLR with BPTB autograft provides superior static knee stability, while there are fewer postoperative complications following ACLR with hamstring autograft.48 Multiple studies36,39,42,43 have demonstrated that patients undergoing ACLR with autograft have superior clinical outcomes and lower rates of graft rupture compared to patients undergoing ACLR with allograft, especially among young and active patients. When allograft tissue is used during ACLR, soft tissue allografts irradiated with greater than 1.8 Mrad during chemical processing have a higher risk of graft failure, which increases with time.40 Tejwani et al41 demonstrated that BPTB allografts have a significantly higher risk of revision than soft tissue allografts, especially when irradiation is greater than 1.8 Mrad, though no differences were found in revision risk between Achilles tendon and soft tissue allografts.41
Contralateral Knee ACL Tear
In a recent population-based cohort study, Schilaty et al49 demonstrated that, between the years 1990-2000, the incidence of second ACL injuries (contralateral and ipsilateral) was 6.0%, with 63.6% of tears occurring to the contralateral ACL. In a separate study by Schilaty et al,50 the incidence of second ACL tears was 13.8% between the years 2001-2010, with 50.4% of these occurring in the contralateral knee at an average follow-up of 4.7 years. This incidence was consistent with a systematic review and meta-analysis of 19 studies by Wiggins et al,35 which reported a pooled total second ACL re-injury rate of 15% (ipsilateral,7%; contralateral, 8%).
In the same systematic review and meta-analysis mentioned above,35 the reported rate of contralateral ACL tears in patients younger than 25 years and athletes who returned to sports was 11% and 12%, respectively. A recent study reporting data from the Kaiser Permanente ACLR registry reported that for every one-year increase in age, there was a 4% lower risk of contralateral ACLR.51 Similarly, in a large cohort study by Kaeding et al,36 the odds of a contralateral ACL tear decreased by 0.04 for every one-year increase in age (p=0.04) and increased by 0.12 for every one-point increase on the preoperative Marx activity score (p<0.01). This is supported by a systematic review by Kay et al9 that evaluated the RTS rate of adolescent athletes following ACLR, which reported that 14% of adolescent athletes had sustained contralateral ACL injuries at final follow-up. In both decade-long studies mentioned above,49,50 there was high population incidence of contralateral ACL tears in younger females. Moreover, in a large cohort study of 17,682 patients from the Swedish National Knee Ligament Register by Snaebjörnsson et al,52 females had a 33.7% greater risk of contralateral ACLR surgery compared to males (hazard ratio, 1.337 [95% CI, 1.127-1.586]; p=0.001). Therefore, it is evident that younger (particularly female) patients, and those who return to a high activity level, especially in high-risk cutting/pivoting sports, have significantly increased odds of a contralateral ACL tear following primary ACLR.9,12,35,36,47,49-53
Epidemiological studies49,50 have demonstrated that for patients aged 17-25 years49,50 and 26-45 years,49 use of allograft significantly increases the risk of second ACL injuries, including both graft failure and contralateral ACL tears, compared to bone-patellar tendon-bone autograft (p<0.0001), whereas in the age groups <16 years (p=0.0093) and 17-25 years (0.0067), use of allograft significantly increases the risk of second ACL injuries compared to hamstring autograft.50 Additionally, among female patients undergoing reconstruction using a hamstring autograft harvested from the contralateral leg, a significantly increased risk of contralateral ACL tear has been reported (relative risk, 3.4 [95% CI, 1.4-7.9]; p=0.005).53 Harvesting autograft from a contralateral healthy knee might alter and presumably impair knee kinematics, which in turn may increase the risk of injury.53 In addition to graft type and risk factors that predispose these patients to initial injury,54 altered motor patterns that protect the reconstructed knee and increase stress on the contralateral limb may play a role in the increased incidence of contralateral ACL tears.35,55,56 This is demonstrated by the increased risk of contralateral tears with non-operative treatment of the ACL compared to ACLR,50 along with the higher future rate of contralateral ACLR (16%) in soccer athletes who underwent ACLR on their non-dominant limb compared to those who underwent ACLR on their dominant limb (3.5%) (p=0.03).12
Postoperative Knee Osteoarthritis
The incidence of radiographic knee OA has been estimated to appear in more than 50% of patients 10 to 20 years following ACLR (Figure 2).57 According to a systematic review of 27 studies, the current literature does not support the prophylactic benefit of ACLR in reducing the incidence of knee OA after ACL injury.58 Although ACLR improves knee function and stability, the initial impaction force that results in the ACL tear often causes injury to the articular cartilage.59 As a result, chondrocytes within the knee joint produce increased levels of apoptotic, inflammatory and catabolic factors that are associated with time from injury contributing to cartilage degeneration and eventually OA development.60
By using T1 magnetic resonance imaging (MRI), previous studies61,62 have observed lower proteoglycan density in the articular cartilage of the femur and tibia at one year following ACLR. More recently, studies have revealed that this reduced proteoglycan density of the articular cartilage in ACL-injured knees was associated with significantly worse patient-reported outcomes including pain, knee-related quality of life, and activity level at one year following ACLR.63,64 Furthermore, a prospective case-control study by Ware et al65 revealed lower preoperative Knee Injury and Osteoarthritis Outcome Scores and 36-Item Short-Form Health Survey scores in patients who developed imaging evidence of symptomatic knee OA 7 years following ACLR. Because the initial stages of OA include proteoglycan loss, increased water content, and disorganization of the collagen network,66,67 the lower articular cartilage proteoglycan density seen on MRI following an ACL injury and subsequent ACLR may be used as an early biomarker of the initial stages of OA.
Until recently, the association between graft type and long-term OA progression was unclear. The results of a recent systematic review59 of 8 randomized controlled trials suggested that, at a mean follow-up of 11.5 years, no significant difference exists in clinical outcomes (graft failure rate, radiographic signs of knee OA, or patient-reported outcomes) between patients undergoing ACLR with BPTB autograft versus hamstring autograft. Previous randomized controlled trials68-74 have similarly failed to demonstrate significant differences in clinical outcomes between single-bundle (SB) versus double-bundle (DB) grafts for ACLR. While it remains unclear if there is a difference between SB and DB grafts for ACLR in terms of progression of knee OA, several studies68-71,73 have found that DB reconstruction is not superior to SB reconstruction in terms of knee stability70 and OA progression at 4-year,71 5-year,70,73 6-year,68 and 10-year follow-up thus far.69 Additionally, Järvelä et al69 reported that the most severe OA changes were found in patients who had the longest delay from the primary injury to ACLR and in patients who underwent partial meniscectomy at the time of ACLR, underscoring the relative importance of meniscal and cartilage injury in ultimate degenerative progression.
In a nested cohort of 358 young, active patients from t
he Multicenter Orthopa
edic Outcomes Network (MOON) group, the lateral joint space width was 0.11 mm narrower on the ACL-reconstructed knee compared to the contralateral healthy knee (p<0.01) at two-year follow-up, and decreases in joint space can be expected in patients who had a lateral meniscectomy (p<0.001) and a lower baseline Marx activity score (p<0.001).75 This is consistent with previous findings that meniscectomy at the time of ACLR is considered a significant risk factor for future knee OA.76-78 It is therefore recommended that ACLR should be performed within 6 months after the initial ACL injury to minimize further meniscal or chondral lesions.79
Total Knee Arthroplasty
Because injury to the ACL increases the risk for knee OA, it is also associated with an increased risk for TKA. In a study by Leroux et al,80 the cumulative incidence of TKA at 15 years following ACLR (1.4%) was significantly higher than that of a case-matched control cohort (0.2%) (p<0.001). This study demonstrated that several factors increased the risk of a TKA including age of 50 years or more at the time of ACLR, female sex, higher comorbidity score (≥5 points) using the Collapsed Aggregated Diagnosis Groups method,81,82 low surgeon annual volume of ACLR, and ACLR performed in a university-affiliated hospital.80
In a matched case-control study of all TKAs performed between 1990 and 2011 and recorded in the Clinical Practice Research Datalink, Khan et al83 investigated whether ACL injury or meniscal injury increases the risk of end-stage knee OA resulting in TKA (n=49,723 cases and 104,353 controls). This study demonstrated that a previous ACL injury is associated with 6.96-times increased odds of TKA (95% CI, 4.73-10.31) and that a meniscal injury is associated with a 15.24-times increased odds of TKA (95% CI, 13.88-16.69).83
In the study by Leroux et al,80 the median age of patients at the time of ACLR was 42 years and the median time from ACLR to subsequent TKA was 11 years. Other studies84-86 have reported longer times between ACLR and subsequent TKA. Hoxie et al84 reported a mean time of 19.1 years between ACLR and TKA in a study of 35 patients (36 knees) with an average age of 53 years at the time of TKA. In a study by Magnussen et al,85 a mean time of 25.7 years between ACLR and subsequent TKA was noted among 22 patients with a mean age of 58.1 years at the time of TKA. Similarly, Watters et al86 described a mean time of 22 years between ACLR and subsequent TKA in a study of 122 patients (average age, 58 years at the time of TKA).
Few studies84-86 have compared operative outcomes of patients undergoing TKA with versus without a prior history of ACLR on the ipsilateral knee. In a retrospective case-matched control study of 36 knees that underwent ACLR and a subsequent TKA, Hoxie et al84 reported that prior ACLR did not adversely affect knee range of motion (ROM), outcome scores, revision rates, infection, or patella baja following TKA, but no minimum follow-up was reported. This study84 also noted that there were no increased intra-operative complications related to prior ACLR, though 3 patients in the ACLR group required constrained-condylar designs.
In contrast, Magnussen et al85 found that, although there were no differences in final ROM, outcome scores, or knee alignment at 2 to 3 years following TKA, 6 of 22 patients (27%) in the post-ACLR group developed significant postoperative stiffness following TKA which required manipulation in 5 patients (23%) compared to none in the control group (p=0.048). Additionally, this study noted intra-operative tibial exposure difficulties requiring implant removal in 10 patients (45%) and tibial tubercle osteotomy in 3 patients (14%) in the post-ACLR group, which likely contributed to the increased mean TKA operative time in this group (84 minutes) compared to the control group (75 minutes).85
More recently, Watters et al86 reported results of a cohort of 122 patients undergoing TKA after ACLR compared to those of a control group undergoing routine TKA for OA. Similar functional outcomes following TKA were demonstrated in both groups, though there was a 5.5-times relative risk (95% CI, 1.2-24.3; p=0.01) of reoperation in the ACLR group at a mean follow-up of 3 years after TKA, with a 3.3% incidence of periprosthetic joint infection in the ACLR group.86 Furthermore, the mean operative time was significantly longer in the ACLR group (88 minutes) compared with the control group (73 minutes) (p<0.001), which is likely related to the need to remove preexisting implants in 50% of patients in the ACLR group to facilitate component placement and the greater complexity of the exposure due to prior ligament reconstruction and scar tissue formation.86 Based on these results, some authors have advocated for a two-stage procedure for these cases to remove metal components, given the unacceptably high rate of infection.87
Conclusion
Surgeons should acknowledge that ACLR is a patient-specific procedure and should be prepared to educate patients accordingly regarding ACL injury and what to expect after undergoing ACLR (Table 1). While younger age, male sex, and higher pre-injury level of sports participation significantly increase the rate of return to sport, these factors also predispose patients to risk of re-injury, either to the ipsilateral or contralateral knee. Clinicians should address both physical functioning and psychosocial factors to assist patients in their transition back to sports activity. Autograft use has been shown to result in the best overall outcomes with regard to knee function, stability, RTS, and risk of re-injury, particularly in younger patients. The literature does not support a prophylactic benefit of ACLR in preventing future knee OA, as the initial ACL injury is typically associated with cartilage damage, placing the patient at an increased risk for development of OA and eventual need for TKA. Surgeons performing TKA should be forewarned of potential intraoperative challenges that are associated with prior ACLR, and surgeons must adequately educate patients on these important postoperative topics.