Original Article
Predictors of Knee Arthrofibrosis and Outcomes after Arthroscopic Lysis of Adhesions following Ligamentous Reconstruction: A Retrospective Case–Control Study with Over Two Years’ Average Follow-Up Blake M. Bodendorfer, MD1 Laura E. Keeling, MD1 Evan M. Michaelson, MS2 Henry T. Shu3 Nicholas A. Apseloff, BS2 James D. Spratt, MS2 Patrick S. Malone, BS2 Evan H. Argintar, MD3 1 Department of Orthopaedic Surgery, Georgetown University
Medical Center, Washington, District of Columbia 2 School of Medicine, Georgetown University, Washington, District of Columbia 3 Department of Orthopaedic Surgery, MedStar Washington Hospital Center, Washington, District of Columbia
Address for correspondence Blake M. Bodendorfer, MD, Department of Orthopaedic Surgery, Georgetown University Medical Center, PHC Ground Floor, 3800 Reservoir Road NW, Washington, DC 20007 (e-mail:
[email protected]).
J Knee Surg
Abstract
Keywords
► arthrofibrosis ► lysis of adhesions ► manipulation under anesthesia ► ligamentous reconstruction ► knee
received February 14, 2018 accepted after revision April 21, 2018
Arthrofibrosis can be a devastating complication after ligamentous knee reconstruction. Beyond early range of motion (ROM), manipulation under anesthesia (MUA) and arthroscopic lysis of adhesions (LOAs) are the most frequently employed interventions for the condition. There is a paucity of data regarding predictive factors of arthrofibrosis requiring MUA and LOA, and even less data regarding changes in validated patient-reported outcome measures following the procedure. A retrospective case– control study was performed at an academic, urban Level I trauma center of patients that developed arthrofibrosis requiring MUA and LOA following ligamentous reconstruction. The indication for LOA was failure to achieve a 90° arc of ROM by 6 weeks. Seventeen cases and 141 controls were identified. Follow-up for cases was 26.9 17.1 months (mean standard deviation). Time from initial reconstruction to LOA was 75.2 27.9 days. Cases had higher body mass indices by a mean of 2.9 (p ¼ 0.024). The most significant risk factors for stiffness were concomitant anterior cruciate ligament, posterior cruciate ligament, and posterolateral corner/lateral collateral ligament injury (odds ratio [OR], 17.08), knee dislocation (OR, 12.84), and use of an external fixator (OR, 12.81, 95% confidence interval [CI], 3.03–54.20) (all p < 0.0026). Mean Knee Injury and Osteoarthritis Outcome Scores, Western Ontario and McMaster Universities Osteoarthritis Indices, and International Knee Documentation Committee scores improved by 47.5, 50.5, and 47.3% (all p < 0.0038), respectively. All patients reported improvement in pain, with maximum daily pain scores improving by a mean of 4.1 points on the Numeric Pain Rating Scale (p < 0.001). Mean ROM arc improved by 38.8° (p < 0.001). All 17 cases were satisfied with the procedure. Twelve cases (70.59%) reported a full return to preinjury level of activity. No factors were identified
Copyright © by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI https://doi.org/ 10.1055/s-0038-1655741. ISSN 1538-8506.
Predictors of Knee Arthrofibrosis and Outcomes after Arthroscopic Lysis
Bodendorfer et al.
that predicted success from the procedure, likely due to inadequate sample size. Arthrofibrosis following knee injury and ligamentous reconstruction can be predicted by the severity of injury and early intervention with MUA and arthroscopic LOA can lead to a satisfactory outcome for the patient.
Introduction Arthroscopic knee surgeries are among the most commonly performed procedures in orthopaedics, with anterior cruciate ligament (ACL) reconstruction performed more frequently in the United States annually than total knee arthroplasty.1 Less common, though often more devastating, are multiligamentous knee injuries, making up less than 0.2% of orthopaedic injuries.2 Despite technical advances in the treatment of these injuries, loss of knee motion remains a significant complication following both single ligament and multiligamentous injury and reconstruction. Arthrofibrosis following arthroscopic knee surgery represents a wide spectrum of disease, and is related to both preand postoperative factors including injury severity and postoperative mobilization.3 The implementation of enhanced postoperative rehabilitation protocols has led to a decrease in the incidence of arthrofibrosis following ligamentous reconstruction over the past 30 years.4 While early postoperative motion remains the mainstay of preventive treatment, those knees that do develop arthrofibrosis may require arthroscopic lysis of adhesions (LOAs), often in conjunction with manipulation under anesthesia (MUA). Recent studies have shown the incidence of arthrofibrosis necessitating LOA following ACL reconstruction to be between 0 and 6%, while the incidence of LOA following multiligamentous reconstruction has been demonstrated to be as high as 17%.4–7 While the incidence of arthrofibrosis necessitating LOAs has been described, ranges vary widely based on the pattern of injury. Werner et al performed an epidemiological study describing the absolute risk of LOA and MUA following various arthroscopic procedures, demonstrating a stepwise increase in risk corresponding to the number of ligaments repaired.8 This descriptive study offers insight into the correlation between injury severity and risk of arthrofibrosis, but did not examine key clinical factors relevant to injury severity, such as need for fixation or staging of surgery. In a more recent study, Hanley et al attempted to analyze increased risk for arthrofibrosis based on these factors, but confined their retrospective analysis to multiligamentous knee injury.6 Thus, the risk of arthrofibrosis necessitating LOA following ligamentous reconstruction in the context of a composite picture of initial injury has yet to be defined. The literature further demonstrates a lack of outcome data following LOAs in the knee. While several studies have quantified improvements in range of motion (ROM) following LOA with or without MUA, few have looked at patientreported outcome measures (PROMs) such as pain or validated knee scores. To our knowledge, no study has investigated factors that are predictive of success following LOAs The Journal of Knee Surgery
after ligamentous repair. The purpose of this study was to characterize risk factors for arthrofibrosis necessitating LOA following arthroscopic knee surgery based on the initial injury, change in PROMs, pain, ROM, satisfaction, and return to activity following LOA, and to determine predictors of success following LOA.
Methods The authors performed a retrospective case–control study of patients that developed knee stiffness requiring arthroscopic LOAs following ligamentous reconstruction, comparing them to a control group that underwent ligamentous reconstruction and did not require arthroscopic LOA. All procedures were performed by the senior author at an academic, urban Level I trauma center between October 2012 and March 2016. After approval from the Institutional Review Board, the hospital database was accessed. The database was queried for Current Procedural Terminology (CPT) code 29888 (arthroscopically assisted ACL repair/augmentation or reconstruction). All patients undergoing ACL repair or reconstruction during this time were identified, yielding a total of 156 knees in 154 patients. Two patients underwent bilateral ACL surgery; one concomitantly and one over a span of 2 years. Following identification, charts were reviewed to obtain demographic data including age, sex, and body mass index (BMI), as well as data relevant to initial injury including: laterality, multiligamentous or concomitant meniscal injury, knee dislocation (KD), and need for internal or external fixation. At our institution, external fixators were routinely applied to any patient with polytraumatic injury, extensive concurrent knee soft-tissue injury requiring plastic surgery intervention, or multiligamentous injuries with evidence of vascular compromise. KDs that failed reduction with a standard knee immobilizer or extension brace were similarly treated with external fixation. The authors also collected data relevant to the initial ligamentous reconstructive procedure including ACL graft type, performance of internal bracing, and performance of posterior cruciate ligament (PCL), medial collateral ligament (MCL), posterolateral corner (PLC)/lateral collateral ligament (LCL) complex, or meniscal repair versus reconstruction. Finally, charts were reviewed to determine which patients ultimately required arthroscopic LOA for knee stiffness following surgery. The database was queried for CPT code 29884 (arthroscopy of knee with LOAs) to ensure that no patients were missed. Of the 156 knees undergoing ligamentous reconstruction, 18 underwent arthroscopic LOA for stiffness during this period. The indication for arthroscopic
Predictors of Knee Arthrofibrosis and Outcomes after Arthroscopic Lysis LOA was failure to achieve a 90° arc of total ROM by 6 weeks following initial surgery. All patients requiring arthroscopic LOAs for knee stiffness underwent a standardized procedure performed by the senior author. Preoperatively, all patients received an indwelling regional nerve blockade. Once in the operating room, an examination under anesthesia was performed. Each patient was prepped and draped in the standard sterile fashion without the use of a tourniquet. An arthroscope was then introduced into the joint via an anterolateral arthrotomy, and an arthroscopically guided medial arthrotomy was performed to introduce a shaver or radiofrequency device in mid-flexion. Any adhesions surrounding the distal femur were primarily addressed to establish landmarks. Anterior adhesions were then lysed by sweeping medially and laterally, allowing visualization of the peripatellar region and the suprapatellar pouch. All adhesions within the suprapatellar pouch were removed. The medial and lateral gutter adhesions were then lysed or ablated via radiofrequency. Finally, the pretibial recess was exposed, and any remaining anterior adhesions were removed. At the conclusion of the procedure, the entire joint was inspected, and the knee was copiously irrigated to remove any remaining free bodies or debris. MUA was subsequently performed for each patient, with slight pressure on the tibial tubercle to avoid disruption of the extensor mechanism. Standard sterile closure and dressings were applied, and each patient began full active and passive ROM exercises immediately postoperatively. Planned physical therapy was continued on a daily basis for a minimum of 2 weeks. Continuous passive motion machines were not used in any case. In total, 18 cases underwent the described procedure, while 138 controls were identified based on initial chart review. Following identification, patients were surveyed to obtain subjective outcome data. Patients were excluded from the survey based on the following criteria: • Incarcerated at the time of or following surgery, limiting follow-up (2 patients). • Spanish-speaking only (4 patients). • Deceased at the time of attempted contact (2 patients). • Left country following surgery, limiting follow-up (1 patient). • Underwent subsequent total knee arthroplasty (2 patients). • Medical condition unrelated to knee, limiting follow-up (2 patients). Once exclusion criteria were applied, 17 cases (17 patients) and 124 controls (123 patients) remained eligible for the survey. All patients undergoing LOA were contacted to obtain Knee Injury and Osteoarthritis Outcome Scores (KOOS), Western Ontario and McMaster Universities Osteoarthritis Indices (WOMAC), and International Knee Documentation Committee (IKDC) scores prior to LOA and at most recent follow-up. Pre- and post-LOA flexion, flexion contracture, full arc of motion, return to activity, and patient-reported Numeric Pain Rating Scale (NPRS) scores were also collected.
Bodendorfer et al.
Patient satisfaction was measured by asking: “Are you satisfied with the surgery?” Outcome data were similarly collected for all control patients, in relation to initial ligamentous repair or reconstruction. All of the 17 cases (100%) were successfully contacted for collection of scores and satisfaction surveys, while outcome data were obtained for 104 of the 124 controls (83.87%).
Statistical Analysis Statistical analysis was performed using custom software written in Python, including use of the machine learning and scientific computing packages NumPy, scikit-learn, and pandas. Univariate analysis of suspected risk factors for knee stiffness was performed using Fischer’s exact test. A twotailed Student’s t-test was used to compare pre- and postoperative outcomes for LOA. Linear regression was used to determine factors predictive of successful outcomes following LOA. p-Values corrected for multiple comparisons (family-wise error rate) are reported, with p-value of < 0.05 considered statistically significant.
Results Univariate analysis was conducted to identify risk factors for the development of knee stiffness necessitating LOAs following ligamentous reconstruction. Subsequently, a two-tailed Student’s t-test was used to compare pre- versus postoperative outcomes following LOAs. The results of the risk factor analysis are presented first.
Risk Factor Analysis A total of 17 cases and 104 controls were included in the risk factor analysis. The average age of patients undergoing LOAs was 31.8 9.8 years (mean standard deviation [SD], range, 17–55 years), with 56% of the cohort being male. The average age of the control group was 30.1 9.3 years (mean SD, range, 13–55 years), with 56% of the cohort being male. Average follow-up in the LOA group was 26.9 17.1 months (range, 3.2–57.4 months), compared with 24.4 14.3 months in the control group (range, 4.0–53.8 months). The average time from initial reconstructive surgery to LOA was 75.2 27.9 days (range, 56– 154 days). Demographic data were therefore overall similar between the two groups. The exception was BMI, with a significant difference observed between the LOA group (29.7 6.7) and the control group (26.8 6.0, p ¼ 0.027).
Univariate Analysis Fischer’s exact tests were performed to identify risk factors for knee stiffness necessitating LOAs following ligamentous reconstruction. Odds ratios (ORs) are reported in ►Table 1. The most significant risk factors for stiffness were ACL with concomitant PCL and PLC/LCL injury (OR, 17.08, 95% confidence interval [CI], 4.57–63.70), KD (OR, 12.84, 95% CI, 4.07–40.44), and prior treatment with an external fixator (OR, 12.81, 95% CI, 3.03–54.20) (p < 0.0026 for all values, corrected for multiple comparisons). In the univariate analysis, there was no statistically significant difference in risk related to gender, BMI greater than 30, meniscal injury, The Journal of Knee Surgery
Predictors of Knee Arthrofibrosis and Outcomes after Arthroscopic Lysis
Bodendorfer et al.
Table 1 Risk factors for arthroscopic LOA identified using Fischer’s exact tests Odds ratio
95% CI
p-Value
ACL
0.06
0.02–0.22
< 0.001a
ACL þ PCL
7.88
0.47–132.15
0.223
ACL þ MCL
Insufficient data
ACL þ PLC
1.98
0.38–10.22
0.335
ACL þ PCL þ MCL
5.51
0.85–35.68
0.106
ACL þ PCL þ PLC/LCL
17.08
4.57–63.70
< 0.001a
Risk factor
ACL þ MCL þ PLC/LCL
Insufficient data
ACL þ PCL þ MCL þ PLC/LCL
Insufficient data
Medial meniscus
2.34
0.79–6.96
0.124
Lateral meniscus
1.86
0.67–5.18
0.278
Either meniscus
2.95
0.98–8.86
0.069
Both menisci
1.98
0.38–10.22
0.335
Cartilage/chondral injury of grade II or higher
1.60
0.52–4.96
0.530
Dislocation
12.84
4.07–40.44
< 0.001a
External fixator
12.81
3.03–54.20
< 0.001a
ORIF
2.29
0.43–12.03
0.287
Either external fixator or ORIF
7.05
2.67–18.59
< 0.001a
Internal bracing
0.75
0.27–2.09
0.608
Revision ACL
0.57
0.12–2.67
0.738
Chronicity of ACL reconstruction (< 6 wk)
1.16
0.38–3.55
0.776
Chronicity of ACL reconstruction (6 wk–3 mo)
0.94
0.24–3.52
1.000
Chronicity of ACL reconstruction (> 3 mo)
Insufficient data
ACL reconstruction with autograft
0.96
0.35–2.65
1.00
ACL reconstruction with allograft
1.40
0.50–3.93
0.59
ACL repair
0.41
0.05–3.31
0.70
Gender
1.17
0.42–3.26
0.80
BMI > 30
2.45
0.81–7.44
0.116
Abbreviations: ACL, anterior cruciate ligament; BMI, body mass index; CI, confidence interval; LCL, lateral collateral ligament; LOA, lysis of adhesion; MCL, medial collateral ligament; ORIF, open reduction internal fixation; PCL, posterior cruciate ligament; PLC, posterolateral corner. a Statistically significant p-values corrected for multiple comparisons (p < 0.0026).
chronicity of reconstruction, graft type, or high-grade chondral injury. Of note, there was insufficient data to determine the OR for several risk factors including chronicity of reconstruction greater than 3 months, as well as several patterns of ligamentous injury including ACL þ MCL, ACL þ MCL þ PLC/LCL, and ACLþ PCLþ MCL þ PLC/LCL (KDIV).
Outcome Analysis Two-tailed Student’s t-tests were used to compare pre- and postoperative outcomes. Outcome data are reported in ►Table 2, and is detailed below.
Knee Scores Patients were surveyed to obtain KOOS (including all five subscores), WOMAC, and IKDC scores, both pre- and postLOA. The mean KOOS composite score improved by 47.5 points postoperatively, from a mean preoperative score of The Journal of Knee Surgery
26.0 19.0 to a mean postoperative score of 73.5 21.1. Similar trends were observed for all five KOOS subscores, with significant improvements noted for pain, symptoms, activities of daily living (ADL), sport/recreation, and quality of life (QOL). The mean WOMAC score improved by 50.5% postoperatively. The mean preoperative WOMAC score was 64.6 23.0%, while the mean postoperative WOMAC score was 14.1 15.1%. Finally, the IKDC also improved postoperatively by 47.3 points. The mean preoperative IKDC score was 16.3 12.1, while the mean postoperative score was 63.6 23.4. All improvements in knee scores and subscores were statistically significant when corrected for multiple comparisons (p < 0.0038).
Pain Patient-reported maximum and average daily pain levels were measured via the NPRS. Maximum daily pain scores
Predictors of Knee Arthrofibrosis and Outcomes after Arthroscopic Lysis
Bodendorfer et al.
Table 2 Comparison of pre- versus postoperative knee scores for patients undergoing LOA Pre-LOA
Post-LOA
p-Value
KOOS composite
26.0 19.0
73.5 21.1
< 0.001a
KOOS pain subscore
28.0 23.4
81.8 17.0
< 0.001a
KOOS symptoms subscore
35.0 23.8
76.0 15.8
< 0.001a
KOOS ADL subscore
31.7 25.0
85.5 15.2
< 0.001a
KOOS sport/Rec subscore
3.6 5.1
48.6 34.0
< 0.001a
KOOS QOL subscore
10.8 16.1
52.7 24.4
< 0.001a
WOMAC
64.6 23.0%
14.1 15.1%
< 0.001a
IKDC
16.3 12.1
63.6 23.4
< 0.001a
Maximum daily pain (NPRS)
9.0 1.6
4.9 2.4
< 0.001a
Average daily pain (NPRS)
3.0 3.0
2.4 2.4
0.407
Flexion
67.0 24.1°
103.4 23.6°
< 0.001a
Flexion contracture
8.0 10.1°
5.6 8.9°
0.174
Full ROM arc
59.0 28.2°
97.8 27.8°
< 0.001a
Return to activity
–
69.2%
–
Satisfaction
–
100.0%
–
Outcome measure
Abbreviations: ADL, activities of daily living; IKDC, International Knee Documentation Committee; KOOS, Knee Injury and Osteoarthritis Outcome Scores; LOA, lysis of adhesion; NPRS, Numeric Pain Rating Scale; QOL, quality of life; ROM, range of motion; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Indices. Note: Values are represented as mean standard deviation. a Statistically significant p-values corrected for multiple comparisons (p < 0.0038).
improved by a mean of 4.1 (p < 0.001), from 9.0 1.6 preoperatively to 4.9 2.4 postoperatively. Average daily pain scores improved by a mean of 0.6, from 3.0 3.0 to 2.4 2.4, although this change was not significant (p ¼ 0.407). All 17 patients reported an improvement in knee pain following LOA.
Range of Motion Flexion, flexion contracture, and full arc of motion were examined. Mean flexion prior to LOA was 67.0 24.1°, while mean flexion postoperatively was 103.4 23.6°. A significant improvement of 36.4° was therefore observed (p < 0.001). Flexion contracture improved by 2.4° from 8.0° 10.1° to 5.6° 8.9° postoperatively, although this result was not significant (p ¼ 0.174). Mean ROM arc prior to LOA was 59.0° 28.2°, and improved to 97.8° 27.8° following LOA for an improvement of 38.8° (p < 0.001).
Satisfaction All 17 patients (100%) reported satisfaction with the procedure.
Return to Activity Twelve of the 17 patients (70.59%) reported a full return to preinjury level of activity following LOAs. Of those that did not, one was a competitive athlete who reported a return to recreational levels of activity. One competitive athlete reported a minimal level of activity following LOA. The remaining three patients had previously been recreational athletes, and reported minimal levels of activity.
Predictors of Success Demographic data including age, BMI, gender, time to LOA, and compliance with physical therapy were analyzed to determine predictors of successful outcomes following LOA. No factor was found to be predictive of higher knee scores, decreased pain, or increased ROM postoperatively.
Discussion Arthrofibrosis represents a debilitating outcome of ligamentous knee reconstruction, often necessitating surgical intervention in the form of LOA or MUA. Our retrospective review found the overall incidence of arthrofibrosis necessitating LOA following arthroscopic knee surgery to be 12.1% (17 out of 141). Of the patients with isolated ACL injury, 3.7% (4 out of 107) underwent LOAs, consistent with previous studies.7,8 Of the patients with multiligamentous injury, 38.2% (13 out of 34) underwent LOAs, a higher proportion than the 17% reported by Mook et al in their 2009 systematic review.5 This may be because as an urban Level I trauma center, our institution sees increased rates of high-energy injury and traumatic KD compared with other hospitals. Reported rates of arthrofibrosis necessitating LOA are higher in cases of traumatic KD. Sisto and Warren reported a 30% rate of arthrofibrosis following such injury, while Shapiro and Freedman found the incidence of arthrofibrosis following traumatic KD to be 57%.9,10 Of the 17 patients in our study who underwent LOA, 12 (70.59%) had experienced a KD. It should be noted, however, that Mook et al excluded studies for which there was not at least a 75% rate of injuries The Journal of Knee Surgery
Predictors of Knee Arthrofibrosis and Outcomes after Arthroscopic Lysis classified as KDIIIM, KDIIIL, or KDIV according to the Schenck classification system.11 Such injuries consist of damage to both cruciate ligaments and either the MCL or LCL (KDIIIM and KDIIIL, respectively), or those with damage to all four ligaments (KDIV). Due to the stringency of criteria by which this systematic review selected high-level knee injury, it is more likely that the discrepancy in incidence of arthrofibrosis was due to our smaller sample size. Despite this discrepancy, the findings of our study were overall consistent with what has been reported in the literature. Of the risk factors for LOA following arthroscopic knee repair that were examined, KD was highly significant, with an OR of 12.84 in the univariate analysis. Numerous factors likely contribute to knee stiffness following dislocation, including soft tissue injury, hemarthrosis with resulting capsulitis, scarring due to immobilization, and muscle contractures.12 KDs are more common following high-energy blunt trauma than athletic injury, and as a result are frequently accompanied by polytraumatic injury (such as acetabular and tibial plateau fractures) that may limit postoperative mobility.13 Given the combination of intraarticular inflammatory processes and the frequent necessity of postoperative immobilization, it is unsurprising that KD should portend a high risk of arthrofibrosis. Interestingly, while multiligamentous injuries involving the PLC/LCL complex were a significant risk factor for the development of arthrofibrosis necessitating LOA, those involving the MCL (Schenck KDIIIM) were not. We hypothesize that this is due the fact that the MCL is extra-articular with its injury generally involving less soft-tissue disruption compared with the PLC/LCL complex which is partially intraarticular and both the injury reconstruction typically involve more soft-tissue disruption. We also found the use of a knee-spanning external fixator preoperatively to be a significant risk factor for the development of arthrofibrosis. This contrasts with a recent study by Hanley et al, who found KD but not the use of external fixation to be a risk factor for knee stiffness.6 Hanley et al cite a 2010 study by Levy et al, who report clinical outcomes for nine patients treated with spanning external fixation following KD. In their study, one patient (11%) went on to require MUA for development of arthrofibrosis. The authors conclude that because this number is lower than the incidence of MUA/LOA following KD reported in the literature, external fixation does not appear to be a risk factor for postoperative stiffness.14 However, as this is an observational rather than a comparative study, it is difficult to objectively make this conclusion. Further, the authors do not specify their criteria for performing MUA/LOA following ligamentous repair. It is possible that our criteria for operative intervention for arthrofibrosis were more aggressive, explaining the discrepancy in outcomes. Another study by Stannard et al reported excellent postoperative knee ROM in patients treated with hinged external fixation following KD. This randomized controlled trial compared patients undergoing multiligamentous knee reconstructions treated with hinged external fixation to those with hinged knee braces, and found lower rates of The Journal of Knee Surgery
Bodendorfer et al.
failure in ligamentous reconstruction failure in the hinged external fixator group.13 In this study, the use of a hinged external fixator provided stability without compromising sagittal plane motion, likely accounting for the high postoperative ROM (range, 2–124°). Numerous studies have suggested that prolonged immobilization is a risk factor for development of arthrofibrosis following ligamentous knee injury.4,15,16 In our study, the use of static spanning external fixation likely caused sufficient immobilization to contribute to its significance as a risk factor. Notably, there were several factors in our study that were not significantly associated with the development of postoperative arthrofibrosis. Among these were the presence of meniscal injury, BMI greater than 30, use of internal bracing, grade II or higher chondral injury, and the timing of intervention. Timing of arthroscopic surgery has been debated, with some authors advocating for early intervention and others supporting delayed reconstruction. In a study of 48 patients with multiligamentous knee injuries, Tzurbakis et al found no significant difference in functional outcomes between patients treated in the acute versus chronic phase.17 Harner et al similarly found no significant difference in postoperative ROM between acutely and chronically treated patients, but did find acutely treated patients to have significantly higher subjective knee scores.18 In their more recent and comprehensive systematic review, Mook et al found that patients managed in a staged fashion (with both acute and chronic repair or reconstruction) had better subjective outcomes than those managed acutely or chronically, but were still at high risk of arthrofibrosis necessitating surgical intervention.5 Based on our review of the literature as well as the results of our study, it is evident that further studies are warranted to elucidate the optimal timing of surgical intervention following ligamentous knee injury. Despite the risk of postoperative arthrofibrosis demonstrated in our study, we found arthroscopic LOA with MUA to be successful in the treatment of knee stiffness. Patients were surveyed via multiple validated knee-specific PROMs, including KOOS, WOMAC, and IKDC.19,20 Significant improvements were observed in all three scores following LOA, as well as in all five KOOS subscores including pain, symptoms, ADL, sport/recreation, and QOL. Of greater importance than statistical significance is the clinical significance of these results. All improvements exceeded reported values of minimum clinically important difference (MCID) for each score. An increase in 8 to 10 points has been tentatively suggested as a clinically significant improvement in the KOOS following ACL reconstruction.21 A 47.5 point improvement in KOOS was observed in our study following LOA, and thus far exceeds this measure of clinical significance. The postoperative increase of 47.3 points for the IKDC, similarly, is well above the MCID of 11.5 to 20.5 reported for patients who have undergone various surgical procedures for mixed knee pathologies.22 Although MCID has not been established for the WOMAC in patients undergoing ligamentous knee surgery, a frequently cited study by Escobar et al finds the MCID for stiffness in patients undergoing total knee arthroplasty to be a decrease in 14 percentage points.23 Further, the
Predictors of Knee Arthrofibrosis and Outcomes after Arthroscopic Lysis patient-acceptable symptom state in patients with knee osteoarthritis has been determined to be 31% according to the WOMAC.24 If these values are extrapolated to our study population, the observed decrease in WOMAC from 64.6% to 14.1% suggests that LOA produces clinically significant results. These conclusions are supported by the fact that all patients in our study endorsed satisfaction with the procedure, and 70.59% of patients reported a full return to preinjury level of activity. We further found LOA to yield substantial improvements in ROM, as evidenced by statistically significant increases in both flexion and total arc of motion. To our knowledge, this is the first study to quantify functional improvement in terms of both ROM and validated knee scores following LOA for arthrofibrosis secondary to ligamentous injury. Previous studies have demonstrated the efficacy of both LOA and MUA in the treatment of posttraumatic arthrofibrosis of the knee, but each study has focused solely on ROM, to the exclusion of PROMs.25–27 Gittings et al found an improvement in mean total ROM from 72° to 127° following LOA for arthrofibrosis secondary to open reduction internal fixation of fractures about the knee.25 Similarly, Cosgarea et al found LOA to improve ROM from 83 to 97% of the contralateral side in patients with arthrofibrosis following previous ligamentous injury.26 We noted similar outcomes in our cohort, with a mean improvement in total ROM of 38.8° following LOA. This progression in ROM mirrored the significant improvements we found across all three knee scores surveyed. Other studies have reported improvements in knee scores following LOA for the stiff total knee arthroplasty.28,29 Despite these findings, no previous study has examined improvements in functional knee scores following LOA in the stiff knee after ligamentous reconstruction. In addition to quantifying improvements in functional outcomes following LOA for the stiff knee, we attempted to determine factors that would be predictive of success. Despite analyzing multiple demographic factors including age, BMI, gender, time to LOA, and compliance with physical therapy, no factor was found to be significantly predictive of success following LOA. This was likely due to our small sample size, and highlights one of the weaknesses of our study. The failure of our analysis to detect a significant predictive factor was likely due to diminished power from our small cohort. This was also noted in our analysis of multiple combinations of multiligamentous injury as risk factors for arthrofibrosis. A further weakness of our study was its retrospective nature. It is likely that a larger study, either retrospective or prospective in nature, would successfully discern trends of significance that eluded our analysis.
Conclusion This retrospective case–control study examined both risk factors for, and outcomes following, arthrofibrosis necessitating arthroscopic LOA and MUA in the ligamentously injured knee. Our results were overall consistent with previous literature, implicating KD and preoperative use of spanning external fixation as risk factors for arthrofibrosis,
Bodendorfer et al.
with higher levels of ligamentous injury portending higher risk of arthrofibrosis. Arthroscopic LOA and MUA were found to result in significant improvements in ROM in the stiff knee, in keeping with previous studies. Despite being retrospective and relatively small in nature, this study contributes to the current body of data on this topic by introducing PROMs following arthroscopic LOA in cases of traumatically induced arthrofibrosis. We found arthroscopic LOA to be efficacious in improving three validated knee scores, as well as pain scores, patient satisfaction, and return to activity. Future studies are warranted to elucidate those factors that are predictive of success following LOA, so that physicians may better stratify and counsel their patients. Funding None. Conflict of Interest None.
References 1 Garrett WE Jr, Swiontkowski MF, Weinstein JN, et al. American
2
3
4
5
6
7
8
9 10
11 12
13
Board of Orthopaedic Surgery practice of the orthopaedic surgeon, part II: certification examination case mix. J Bone Joint Surg Am 2006;88(03):660–667 Levy BA, Fanelli GC, Whelan DB, et al; Knee Dislocation Study Group. Controversies in the treatment of knee dislocations and multiligament reconstruction. J Am Acad Orthop Surg 2009;17(04):197–206 Hooper GJ, Walton DI. Reconstruction of the anterior cruciate ligament using the bone-block iliotibial-tract transfer. J Bone Joint Surg Am 1987;69(08):1150–1154 Noyes FR, Berrios-Torres S, Barber-Westin SD, Heckmann TP. Prevention of permanent arthrofibrosis after anterior cruciate ligament reconstruction alone or combined with associated procedures: a prospective study in 443 knees. Knee Surg Sports Traumatol Arthrosc 2000;8(04):196–206 Mook WR, Miller MD, Diduch DR, Hertel J, Boachie-Adjei Y, Hart JM. Multiple-ligament knee injuries: a systematic review of the timing of operative intervention and postoperative rehabilitation. J Bone Joint Surg Am 2009;91(12):2946–2957 Hanley J, Westermann R, Cook S, et al. Factors associated with knee stiffness following surgical management of multiligament knee injuries. J Knee Surg 2017;30(06):549–554 Freedman KB, D’Amato MJ, Nedeff DD, Kaz A, Bach BR Jr. Arthroscopic anterior cruciate ligament reconstruction: a meta-analysis comparing patellar tendon and hamstring tendon autografts. Am J Sports Med 2003;31(01):2–11 Werner BC, Cancienne JM, Miller MD, Gwathmey FW. Incidence of manipulation under anesthesia or lysis of adhesions after arthroscopic knee surgery. Am J Sports Med 2015;43(07):1656–1661 Sisto DJ, Warren RF. Complete knee dislocation. A follow-up study of operative treatment. Clin Orthop Relat Res 1985;(198):94–101 Shapiro MS, Freedman EL. Allograft reconstruction of the anterior and posterior cruciate ligaments after traumatic knee dislocation. Am J Sports Med 1995;23(05):580–587 Schenck R Jr. Classification of knee dislocations. Oper Tech Sports Med 2003;11:193–198 Manske RC, Hosseinzadeh P, Giangarra CE. Multiple ligament knee injury: complications. N Am J Sports Phys Ther 2008;3(04): 226–233 Stannard JP, Nuelle CW, McGwin G, Volgas DA. Hinged external fixation in the treatment of knee dislocations: a prospective randomized study. J Bone Joint Surg Am 2014;96(03):184–191 The Journal of Knee Surgery
Predictors of Knee Arthrofibrosis and Outcomes after Arthroscopic Lysis
Bodendorfer et al.
14 Levy BA, Krych AJ, Shah JP, Morgan JA, Stuart MJ. Staged
21 Roos EM, Lohmander LS. The Knee injury and Osteoarthritis
protocol for initial management of the dislocated knee. Knee Surg Sports Traumatol Arthrosc 2010;18(12):1630– 1637 Wascher DC, Becker JR, Dexter JG, Blevins FT. Reconstruction of the anterior and posterior cruciate ligaments after knee dislocation. Results using fresh-frozen nonirradiated allografts. Am J Sports Med 1999;27(02):189–196 Roman PD, Hopson CN, Zenni EJ Jr. Traumatic dislocation of the knee: a report of 30 cases and literature review. Orthop Rev 1987; 16(12):917–924 Tzurbakis M, Diamantopoulos A, Xenakis T, Georgoulis A. Surgical treatment of multiple knee ligament injuries in 44 patients: 2-8 years follow-up results. Knee Surg Sports Traumatol Arthrosc 2006;14(08):739–749 Harner CD, Waltrip RL, Bennett CH, Francis KA, Cole B, Irrgang JJ. Surgical management of knee dislocations. J Bone Joint Surg Am 2004;86-A(02):262–273 Bellamy N. WOMAC: a 20-year experiential review of a patientcentered self-reported health status questionnaire. J Rheumatol 2002;29(12):2473–2476 Collins NJ, Misra D, Felson DT, Crossley KM, Roos EM. Measures of knee function: International Knee Documentation Committee (IKDC) Subjective Knee Evaluation Form, Knee Injury and Osteoarthritis Outcome Score (KOOS), Knee Injury and Osteoarthritis Outcome Score Physical Function Short Form (KOOS-PS), Knee Outcome Survey Activities of Daily Living Scale (KOS-ADL), Lysholm Knee Scoring Scale, Oxford Knee Score (OKS), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), Activity Rating Scale (ARS), and Tegner Activity Score (TAS). Arthritis Care Res (Hoboken) 2011;63(Suppl 11): S208–S228
Outcome Score (KOOS): from joint injury to osteoarthritis. Health Qual Life Outcomes 2003;1:64 Irrgang JJ, Anderson AF, Boland AL, et al; International Knee Documentation Committee. Responsiveness of the International Knee Documentation Committee subjective knee form. Am J Sports Med 2006;34(10):1567–1573 Escobar A, Quintana JM, Bilbao A, Aróstegui I, Lafuente I, Vidaurreta I. Responsiveness and clinically important differences for the WOMAC and SF-36 after total knee replacement. Osteoarthritis Cartilage 2007;15(03):273–280 Tubach F, Ravaud P, Baron G, et al. Evaluation of clinically relevant states in patient reported outcomes in knee and hip osteoarthritis: the patient acceptable symptom state. Ann Rheum Dis 2005; 64(01):34–37 Gittings D, Hesketh P, Dattilo J, Zgonis M, Kelly J, Mehta S. Arthroscopic lysis of adhesions improves knee range of motion after fixation of intra-articular fractures about the knee. Arch Orthop Trauma Surg 2016;136(12):1631–1635 Cosgarea AJ, DeHaven KE, Lovelock JE. The surgical treatment of arthrofibrosis of the knee. Am J Sports Med 1994;22(02):184–191 Sassoon AA, Adigweme OO, Langford J, Koval KJ, Haidukewych GJ. Manipulation under anesthesia: a safe and effective treatment for posttraumatic arthrofibrosis of the knee. J Orthop Trauma 2015; 29(12):e464–e468 Bodendorfer BM, Kotler JA, Zelenty WD, Termanini K, Sanchez R, Argintar EH. Outcomes and predictors of success for arthroscopic lysis of adhesions for the stiff total knee arthroplasty. Orthopedics 2017;40(06):e1062–e1068 Schwarzkopf R, William A, Deering RM, Fitz W. Arthroscopic lysis of adhesions for stiff total knee arthroplasty. Orthopedics 2013; 36(12):e1544–e1548
15
16
17
18
19
20
The Journal of Knee Surgery
22
23
24
25
26 27
28
29
