Reduced Anterior Cruciate Ligament Vascularization Is Associated With Chondral Knee Lesions

Transcription

1 Reduced Anterior Cruciate Ligament Vascularization Is Associated With Chondral Knee Lesions Iftach Hetsroni, MD; Amir Manor, MD; Alex Finsterbush, MD; Joseph Lowe, MD; Gideon Mann, MD; Ezequiel Palmanovich, MD abstract This study tested the association between periligamentous vascularization of the anterior cruciate ligament (ACL) and the presence of chondral knee lesions via retrospective analysis of prospectively collected data from 702 consecutive knee arthroscopic procedures. In each case, the ACL periligamentous envelope was documented as follows: (1) vascular, where the ACL was covered with blood vessels along its entire length; (2) centrally avascular, where the central third of the ACL was not covered but peripheral vascularized coverage was present; and (3) avascular, where there was no blood vessel coverage of the ACL. Inclusion criteria for the study were as follows: (1) age older than 18 years and (2) normal knee ligament laxity. Univariate analysis and multiple logistic regression were used to test the association between chondral lesions and each of the variables: sex, age, meniscus tear, decreased ACL vascularity, and concomitant chondral lesion in another knee compartment. The cohort included 516 knees. In the univariate analysis, all variables were associated with a chondral lesion, but only older age and decreased ACL vascularity were associated with chondral lesions in each knee compartment. In the regression model, only decreased ACL vascularity was associated with chondral lesions in each knee compartment. For avascular knees, the odds ratio was 2.84 for medial femoral condyle lesions (95% confidence interval, ; P=.000), 2.44 for lateral femoral condyle lesions (95% confidence interval, ; P=.015), and 2.48 for patellofemoral lesions (95% confidence interval, ; P=.000). The findings showed that decreased ACL periligamentous vascularization is associated with chondral lesions of the femoral condyles in knees with preserved ACL laxity. [Orthopedics. 2016; 39(4):e737-e743.] The anterior cruciate ligament (ACL) is a primary restraint to anterior and rotational translations of the tibia and a secondary restraint to valgus stress. 1 Although this role is fulfilled by passive action of properly oriented collagenous bundles of the ACL, neural mechanoreceptors within the periligamentous envelope 2,3 contribute to coordinating knee motion and provide proprioceptive function through periarticular muscle activation. 4-8 Previous studies showed that loss of knee proprioception was not necessarily associated with the amount of pathologic laxity after ACL tear 9 or ACL reconstruction. 10 These reports suggested that other lesions inside the knee, such as chondral lesions or meniscal tears, may have accounted for impaired proprioception. However, another possible explanation is that in some cases the ACL injury was not a pure mechanical lesion but also involved a neural lesion, with disruption The authors are from the Department of Orthopedic Surgery (IH, GM, EP), Meir General Hospital, Kfar Saba; Sackler Faculty of Medicine (IH, AM, GM, EP), Tel Aviv University, Tel Aviv; and the Department of Orthopedic Surgery (AF, JL), Hadassah University Hospital, Mount Scopus, Jerusalem, Israel. The authors have no relevant financial relationships to disclose. Correspondence should be addressed to: Iftach Hetsroni, MD, Department of Orthopedic Surgery, Meir General Hospital, Tsharnichovski St 59, Kfar Saba 44281, Israel (iftachhetsroni@ gmail.com). Received: November 30, 2015; Accepted: March 8, doi: / e737

2 Figure 1: Arthroscopic view through the anterolateral portal of the right knee showing vascular coverage along the entire ligament (A). Arthroscopic view through the anterolateral portal of the left knee showing vascular coverage proximally with subtle but still apparent vascular coverage distally (arrows) (B). A A Figure 2: Arthroscopic view through the anterolateral portal of the right knee showing lack of vascular coverage at the central third of the ligament (inside black oval) (A). Arthroscopic view through the anterolateral portal of the right knee showing lack of vascular coverage in the central third of the ligament (inside black oval) that may be related to a narrow notch caused by subtle osteophytes (B). of mechanoreceptors that did not recover after a tear or reconstruction, despite preservation of some of the mechanical properties of the ACL bundles. Although the importance of the proprioceptive role has been demonstrated, orthopedic surgeons do not routinely evaluate this aspect of ACL function during knee arthroscopy, mainly for practical reasons. This would require intraoperative nerve stimulation of the ACL 5 or periligamentous tissue biopsy. On the other hand, because the blood supply to the ACL that originates from branches of the middle geniculate artery and its neural innervation that originates from the tibial nerve 11 are distributed in close relationships within the ACL periligamentous tissue, 7,8 direct inspection of periligamentous ACL vascularity during arthroscopy could provide a simple indirect measure of the distribution of neural elements. Based on this and the observation that abnormalities in the distribution of ACL mechanoreceptors have been found in cadaveric dissections of endstage arthritic knees, 7 the authors hypothesized that interruption of periligamentous vascularization as an indirect measure of abnormal distribution of mechanoreceptors would be associated with chondral derangement of the knee despite normal ACL laxity. This study tested the association between interruptions in the integrity of periligamentous ACL vascularity and the presence of chondral lesions in knees with preserved ACL laxity. B B Materials and Methods The local hospital ethics committee approved the study protocol. Over 7 consecutive years ( ), 702 primary knee arthroscopy procedures were performed and documented by 1 senior surgeon (A.F.). Operative reports included intra-articular findings and the results of knee ligament laxity tests performed under anesthesia. In addition to standard assessment of chondral surfaces, meniscal integrity, and ligament laxity, in each case, the surgeon evaluated the ACL periligamentous envelope and documented the findings as part of the operative report. The following definitions were used to categorize ACL periligamentous vascularity: (1) vascular, when the ACL was covered with blood vessels along its entire (or nearly entire) length (Figure 1); (2) centrally avascular, when the central third of the ACL was not covered by vascularized tissue but there was peripheral (proximal and distal) coverage (Figure 2); and (3) avascular, when there was no blood vessel coverage for the entire (or nearly entire) ACL (Figure 3). For the purpose of this study, prospectively collected data from patient charts and operative reports were reviewed retrospectively. Inclusion criteria were as follows: (1) age older than 18 years; (2) normal findings on knee ligament laxity tests under anesthesia (Lachman, pivot shift, anterior drawer, posterior drawer, collateral, posteromedial, and posterolateral tests); and (3) intact ACL fibers on direct arthroscopic inspection. Articular chondral lesions were classified according to the Outerbridge classification system. 12 Statistical Analysis For the purpose of this study, a positive chondral lesion was defined as Outerbridge grade II or above (Figure 4), in accordance with previous reports. 13 For each chondral compartment (ie, medial femoral condyle, lateral femoral condyle, patellofemoral joint, lateral tibial condyle, and medial tibial condyle), univariate analysis with Pearson s chi-square test was performed to test the association between a positive chondral lesion and each of the e738 Copyright SLACK Incorporated

3 following potential associated factors: sex, increasing age, meniscus tear, decreased ACL vascularity, and concomitant positive chondral lesion in another knee compartment. In addition, because impairment in ACL vascularity may be closely associated with increasing age, the distribution of ACL vascularity was also compared in the different patient age groups with multinomial logistic regression. Because only 2 knees had positive medial tibial condyle lesions and only 16 knees had positive lateral tibial condyle lesions, a multiple logistic regression model was then built separately for the medial femoral condyle, lateral femoral condyle, and patellofemoral joint chondral compartments. This model included all associated factors that showed statistical significance in the preceding univariate analysis. Adjusted odds ratios and 95% confidence intervals were derived from the model to determine the degree of association between each factor and the occurrence of a positive chondral lesion. The threshold for statistical significance was set at P=.05. Analysis was performed with SPSS software version 22 (IBM, Armonk, New York). Interobserver variability of the classification of ACL vascularity (ie, vascular vs centrally avascular vs avascular) was tested between 2 observers (I.H., G.M.) by calculating kappa coefficients for a sample of 30 consecutive knee arthroscopy procedures (other than the study cohort) in patients who had an intact ACL and a negative pivot shift test result. This evaluation was performed intraoperatively in each of the 30 cases when the 2 observers separately inspected the ACL vascularity and indicated the status. Intraobserver variability was not determined because repeating the arthroscopic inspection was not justified for this purpose. Moreover, relying on arthroscopic images or videos, routinely obtained in these operations, 2 weeks after surgery without the possibility of repeating direct inspection of the ligament could introduce bias in the assessment of the appearance of the microvasculature. Figure 3: Example of avascular anterior cruciate ligament in the left knee. Arthroscopic view through the anterolateral portal of the left knee in a 40-year-old man who had medial knee pain without a clear traumatic episode. No vascular coverage is seen along the entire anterior cruciate ligament. Figure 4: Arthroscopic view through the anterolateral portal of the medial femoral condyle of the patient shown in Figure 3 showing a grade II chondral lesion. This patient also had a tear of the posterior horn of the medial meniscus. Chondral lesion margins are indicated by arrows. Figure 5: Flowchart of the study cohort. Abbreviation: ACL, anterior cruciate ligament. Results A total of 516 knees (366 men, 71%; 150 women, 29%) met the inclusion criteria (Figure 5). The right knee was involved in 48% of cases. Median patient age was 42 years (range, years). Duration of knee symptoms was 5.6±4.3 months (range, 2 weeks-10 years). The e739

4 Table 1 Characteristics of Medial Femoral Condyle, Lateral Femoral Condyle, and Patellofemoral Joint Lesions by Sex, Age, Meniscus Tear, Anterior Cruciate Ligament Vascularity, and Associated Chondral Lesion Variable Sex No. (%) of Medial Femoral Condyle Lesions No (n=333) Yes (n=183) P No. (%) of Lateral Femoral Condyle Lesions No (n=446) Yes (n=70) P No. (%) of Patellofemoral Joint Lesions No (n=334) Yes (n=182) Male 243 (73) 123 (67) (72) 42 (60) (72) 126 (69).6 Female 90 (27) 60 (33) 124 (28) 28 (40) 95 (28) 56 (31) Age group <30 y 88 (26) 11 (6) < (20) 12 (17) (24) 20 (11) < y 93 (28) 22 (12) 107 (24) 8 (11) 96 (29) 19 (10) y 91 (28) 50 (28) 123 (28) 18 (26) 84 (25) 57 (31) y 44 (13) 60 (34) 84 (19) 20 (29) 50 (15) 54 (30) 60+ y 17 (5) 40 (22) 45 (10) 12 (17) 25 (7) 32 (18) Meniscus tear None 130 (39) 49 (27) < (35) 23 (33) < (34) 67 (37).4 Medial meniscus 119 (36) 101 (55) 208 (47) 12 (17) 144 (43) 76 (42) Lateral meniscus 66 (20) 19 (10) 57 (13) 28 (40) 60 (18) 25 (14) Medial meniscus+lateral meniscus Anterior cruciate ligament vascularity 18 (5) 14 (8) 25 (6) 7 (10) 18 (5) 14 (8) Vascular 182 (55) 42 (23) < (47) 16 (23) < (53) 46 (25) <.01 Centrally avascular 54 (16) 49 (27) 91 (20) 12 (17) 59 (18) 44 (24) Avascular 97 (29) 92 (50) 147 (33) 42 (60) 97 (29) 92 (51) Chondral lesion a Medial femoral condyle 145 (33) 38 (54) < (28) 91 (50) <.01 Lateral femoral condyle 32 (10) 38 (21) < (11) 32 (18).05 Patellofemoral joint 91 (28) 91 (50) < (34) 32 (46).05 a There were only 2 cases of medial tibial condyle lesions and 14 cases of lateral tibial condyle lesions, and none of these had any significant association with medial femoral condyle, lateral femoral condyle, or patellofemoral joint lesions. P univariate model showed that all variables (ie, female sex, increasing age, meniscus tear, decreased ACL vascularity, and concomitant positive chondral lesion in another compartment) were associated with a positive chondral lesion in 1 or more of the 3 compartments (ie, medial femoral condyle, lateral femoral condyle, patellofemoral joint). However, only increasing age and decreased ACL vascularity were consistently associated with a positive chondral lesion in all 3 compartments (Table 1). The multiple logistic regression model showed that, of the variables, age older than 60 years vs age younger than 30 years had the strongest effect on the occurrence of a positive medial femoral condyle lesion (odds ratio, 10.82; 95% confidence interval, ; P<.01), but it was not associated with a positive lateral femoral condyle lesion (Table 2). Conversely, decreased ACL vascularity was the only variable associated with a positive chondral lesion in all 3 compartments. When the ACL was avascular, the odds ratio was 2.84 for a positive medial femoral condyle lesion (95% confidence interval, ; P=.000), 2.44 for a positive lateral femoral condyle lesion (95% confidence interval, ; P=.015), and 2.48 for a positive patellofemoral joint lesion (95% confidence interval, ; P=.000). The distribution of ACL vascularity was compared in patients in different age groups. Vascular e740 Copyright SLACK Incorporated

5 Table 2 Predicting Medial Femoral Condyle, Lateral Femoral Condyle, and Patellofemoral Joint Lesions by Relevant Independent Variables: Multiple Logistic Regression Model Medial Femoral Condyle Lesion Lateral Femoral Condyle Lesion Patellofemoral Joint Lesion Variable OR 95% CI P OR 95% CI P OR 95% CI P Sex Age group <30 y Reference <.01 Reference.6 Reference < y y < y < < y < <.01 Meniscus tear None Reference <.01 Reference <.01 Medial meniscus <.01 Lateral meniscus <.01 Medial meniscus+lateral meniscus Anterior cruciate ligament vascularity Vascular Reference <.01 Reference.05 Reference <.01 Centrally avascular < <.01 Avascular < <.01 Chondral lesion Medial femoral condyle < Lateral femoral condyle < Patellofemoral joint Abbreviations: CI, confidence interval; OR, odds ratio. ACL was the most prevalent type in patients younger than 30 years, centrally avascular ACL was the most prevalent type in patients 40 to 49 years, and avascular ACL was the most prevalent type in patients older than 60 years. Vascular ACL distribution was significantly different compared with the centrally avascular type in younger age groups up to patients 40 to 49 years (P<.01) and significantly different compared with the avascular type in all age groups (P<.01) (Table 3). Despite the higher prevalence of the centrally avascular type vs the avascular type in patients ranging in age from younger than 30 years to 40 to 49 years and the higher prevalence of the avascular type compared with the centrally vascular type in patients older than 60 years, differences in distributions of these 2 types of ACL vascularity were not statistically significant (ie, centrally avascular and avascular) in any age group (P=not significant). Interobserver assessment of the categorization of ACL vascularity between the 2 investigators in the 30 consecutive knee arthroscopy procedures showed almost perfect agreement (kappa=0.89). Discussion The main finding of this study was that decreased ACL periligamentous vascularization was significantly associated with chondral lesions in all 3 knee compartments. This finding supported the hypothesis. This association was indicated in the multiple logistic regression model that accounted for all other associated factors for chondral lesions (ie, increasing age, meniscal injuries, and existing chondral lesion in another knee compartment). In the avascular ACL knee, the risk of a chondral lesion at the medial femoral condyle was almost 3-fold compared with the vascular ACL e741

6 Age Group Table 3 Age Group vs Distribution of Anterior Cruciate Ligament Vascularity Vascular No. (%) of Cases Centrally Avascular Avascular Total <30 y 60 (26.8) 15 (14.6) 24 (12.7) 99 (19.2) y 70 (31.3) 18 (17.5) 27 (14.3) 115 (22.3) y 50 (22.3) 35 (34.0) 56 (29.6) 141 (27.3) y 33 (14.7) 25 (24.3) 46 (24.3) 104 (20.2) 60+ y 11 (4.9) 10 (9.7) 36 (19.0) 57 (11.0) Total 224 (100.0) 103 (100.0) 189 (100.0) 516 (100.0) Figure 6: Suggested interrelationships between anterior cruciate ligament (ACL) vascularity and knee chondral lesions. *Based on Amir et al. 2 knee. Nevertheless, the current findings cannot determine the reason for reduced vascularized envelope of the ACL in knees with mild chondral lesions and whether this finding represented a cause-and-effect relationship in which ACL dysfunction as a result of deficient mechanoreceptors in the avascular ACL preceded and promoted chondral lesions as a result of minor uncoordinated knee motions that lead to cartilage overload. Moreover, because increasing age was associated with reduced ACL vascularity, it could be argued that aging per se and not reduced ACL vascularity was the main cause of the increasing prevalence of chondral lesions in these cases. However, the multivariate logistic regression model showed that reduction in ACL vascularity by itself contributed to the occurrence of chondral lesions. Further, a substantial number of avascular and centrally avascular categories of ACL occurred in the younger age groups as well. In these cases, impairment of ACL vascularity is less likely to be related to age and could be related to acute traumatic events, recurrent minor traumatic episodes, or disruption of the anterior fat pad that provides some of the microvasculature of the ACL. 14,15 Alternatively, narrow notch impingement could have led to repeated compression of the periligamentous tissue, resulting in decreased vascularization of the ACL coverage. In other cases, reduced vascularization may have been the natural histomorphology (ie, genetic predisposition). Other in vivo studies of the integrity of the ACL periligamentous tissue are lacking, but a cadaveric study showed that knees with severe osteoarthritis had an increase in neural elements within the ACL periligamentous coverage. 2 The authors suggested that this finding may have represented late reactive synovial proliferation with neuronal sprouting to improve knee proprioception in patients with advanced arthritis. 2 Although this may appear contradictory to the current findings, the current series included knees with mild chondral lesions and not knees with advanced arthritis. Therefore, these knees may have represented an earlier stage in which lack of a vascularized periligamentous envelope was associated with only mild to moderate chondral lesions, a situation that is distinctive from knees with advanced 3-compartmental arthritis that were investigated in the other study. 2 In the current study, there were no grade IV chondral lesions, which is consistent with the preferred clinical approach of the senior author to perform arthroscopic interventions for meniscal lesions or unstable chondral flaps primarily in joints without advanced arthritis. Both studies showed that alterations in ACL periligamentous coverage are related to chondral derangement of the knee. The suggested interrelationships between ACL vascularity and knee chondral lesions are summarized in Figure 6. The findings of the current observational study should be interpreted cau- e742 Copyright SLACK Incorporated

7 tiously. Nevertheless, these findings suggest several clinical implications. First, an additional simple step could be added to intra-articular systematic assessment during arthroscopy in knees with preserved ACL laxity (ie, documenting ACL periligamentous vascular status because this could have clinical significance). The high interobserver agreement in this study supports the reliability of this assessment step. Second, this observation provides a rationale for investigating re-creation of the neurovascularized ACL envelope after ACL reconstruction, either by direct inspection or, preferably, by neurostimulation or tissue biopsy. Interruption in the completion of this specific biologic step after ligament reconstruction may improve the current understanding of clinical failure after allograft vs autograft ligament transplantations. Allografts may have poorer biologic re-establishment of the neurovascularized graft envelope, resulting in proprioceptive graft dysfunction, graft overload, stretch, and failure. Limitations Limitations of this study include interpreting decreased ACL vascularity as a measure of a decrease in neural mechanoreceptor distribution instead of performing direct ACL neural stimulation or periligamentous biopsy. In addition, there were no data on other potential associated factors for chondral lesions, such as knee malalignment and genetic predisposition to knee arthritis. Moreover, the mechanism of injury and duration of symptoms, which were not included in the multivariate model, also may be associated with chondral injuries. Nevertheless, in some cases, it may be difficult to define exactly when symptoms started and whether there was an acute event in a normal knee or in a symptomatic knee. It also may be difficult to define whether some events, such as an acute rise from a deep knee-bent position, should be defined as acute injuries or whether they represent symptoms of a chronically unhealthy knee. Consequently, because both variables are subject to distinct subjective interpretations and may not be clear-cut, they were not included in the model. Also, traumatic events that involved severe ligamentous tears were, by definition, excluded. Conclusion Decreased ACL periligamentous vascularization is associated with chondral lesions of the femoral condyles in knees with preserved ACL laxity. Thus, ACL vascularity by itself, not just normal ligament laxity, may play a chondroprotective role. Further, this finding supports the addition of direct inspection of the ACL periligamentous tissue to systematic arthroscopic assessment and may provide a rationale for future studies of the role of this tissue in recreating ACL proprioception after ACL reconstructive surgery. References 1. Markolf KL, Mensch JS, Amstutz HC. Stiffness and laxity of the knee: the contributions of the supporting structures. A quantitative in vitro study. J Bone Joint Surg Am. 1976; 58(5): Amir G, Lowe J, Finsterbush A. Histomorphometric analysis of innervation of the anterior cruciate ligament in osteoarthritis. J Orthop Res. 1995; 13(1): Koch B, Kurriger G, Brand RA. Characterisation of the neurosensory elements of the feline cranial cruciate ligament. J Anat. 1995; 187(pt 2): Duthon VB, Barea C, Abrassart S, Fasel JH, Fritschy D, Menetrey J. Anatomy of the anterior cruciate ligament. Knee Surg Sports Traumatol Arthrosc. 2006; 14(3): Dyhre-Poulsen P, Krogsgaard MR. Muscular reflexes elicited by electrical stimulation of the anterior cruciate ligament in humans. J Appl Physiol (1985). 2000; 89(6): Halata Z, Wagner C, Baumann KI. Sensory nerve endings in the anterior cruciate ligament (Lig. cruciatum anterius) of sheep. Anat Rec. 1999; 254(1): Raunest J, Sager M, Burgener E. Proprioception of the cruciate ligaments: receptor mapping in an animal model. Arch Orthop Trauma Surg. 1998; 118(3): Reider B, Arcand MA, Diehl LH, et al. Proprioception of the knee before and after anterior cruciate ligament reconstruction. Arthroscopy. 2003; 19(1): Friden T, Roberts D, Zatterstrom R, Lindstrand A, Moritz U. Proprioceptive defects after an anterior cruciate ligament rupture: the relation to associated anatomical lesions and subjective knee function. Knee Surg Sports Traumatol Arthrosc. 1999; 7(4): Barrett DS. Proprioception and function after anterior cruciate reconstruction. J Bone Joint Surg Br. 1991; 73(5): Kennedy JC, Alexander IJ, Hayes KC. Nerve supply of the human knee and its functional importance. Am J Sports Med. 1982; 10(6): Outerbridge RE. The etiology of chondromalacia patella. J Bone Joint Surg Br. 1961; 43: Spindler KP, Huston LJ, Wright RW, et al. The prognosis and predictors of sports function and activity at minimum 6 years after anterior cruciate ligament reconstruction: a population cohort study. Am J Sports Med. 2011; 39(2): Arnoczky SP, Rubin RM, Marshall JL. Microvasculature of the cruciate ligaments and its response to injury: an experimental study in dogs. J Bone Joint Surg Am. 1979; 61(8): Davies DV, Edwards DA. The blood supply of the synovial membrane and intraarticular structures. Ann R Coll Surg Engl. 1948; 2(3): e743