- Email: [email protected]
Altered biomechanics and cartilage health changes in bilateral knees following unilateral ACL reconstruction: a 2-year follow-up
S406
Abstracts / Osteoarthritis and Cartilage 24 (2016) S63eS534
681 USING ENDOPLASMIC RETICULUM STRESS TO MODEL OA IN MICE K. Posey, J. Alcorn, J. Hecht. Univ. of Texas Med. Sch. at Houston, Houston, TX, USA Purpose: There is growing evidence that endoplasmic reticulum (ER) stress plays a role in osteoarthritis (OA). In order to evaluate the role of ER stress in idiopathic adult-onset OA, we utilized ER stress as an OA inducer in a mouse model. Methods: In this system, ER stress is induced by expression of mutant cartilage oligomeric matrix protein (COMP) in the chondrocytes of adult mice between the ages of 16 to 20 weeks. Mutant COMP expression is dependent on presence of doxycycline (Tet-On system) and is restricted to chondrocytes by a type II collagen promoter. Results: We have preciously shown that mutant COMP causes ER and oxidative stress and inflammation in growth plate and articular chondrocytes. Our data further shows additional characteristics of OA including reduction in proteoglycans and glycosaminoglycans, presence of inflammatory markers and thinning of the articular cartilage in adult mice. We are now assessing classic OA markers in the articular cartilage, evaluating articular cartilage health using the OARSI scoring system, gait as a measure of limb function and pain. Conclusions: The strengths of this new model system of OA are the ability to induce ER stress at different time points and a system to investigate and define the role that ER stress plays in OA. This inducible model of adult-onset idiopathic OA will be used to assess recovery of the joint after damage, the contribution of multiple OA risk factors, identify early markers of OA and evaluate new treatments directed at dampening ER stress.
OA: Ligament/Meniscus/ Tendon/Muscle 682 ALTERED BIOMECHANICS AND CARTILAGE HEALTH CHANGES IN BILATERAL KNEES FOLLOWING UNILATERAL ACL RECONSTRUCTION: A 2-YEAR FOLLOW-UP
months (6M), 1 year (1Y) and 2 years (2Y) post reconstruction were included in this analysis. Images of the bilateral knees were acquired using a 3-Tesla MRI-scanner. High-resolution 3D FSE and quantitative T1r sequences, and sagittal T2 FSE images with 25% of the patient’s body weight applied axially were obtained. Using an in-house Matlab program, image segmentation was performed to define the biomechanical parameters of tibial position (TP) and internal tibial rotation (ITR) with respect to the femur in the extended position. T1r relaxation times were quantified for 6 compartments which were then divided into 14 sub-compartments (Fig. 1). Paired t-tests were used for comparing the biomechanical parameters and T1r between the injured and uninjured knees. Paired t-tests were also used for longitudinal comparisons of the four time points. Partial correlations were used to determine correlation coefficients between the biomechanical parameters and T1r relaxation times while controlling for age, gender, BMI and knee flexion angle. Significance was defined as p <0.05. Results: Biomechanics: TP of the injured knee was significantly more anterior than the contralateral uninjured knee at BL (1.65mm, p<0.001), 6M (1.02mm, p¼0.003), 1Y (0.73mm, p¼0.016), and 2Y (1.46mm, p<0.001). TP of the injured knee was also more anterior at 2Y than at 1Y (0.69mm, p¼0.006). TP of the contralateral knee was significantly more anterior at 2Y compared to BL (0.85mm, p<0.001). Both injured and contralateral knees also had higher ITR at 2Y than at BL (p¼0.045, p<0.001). T1r: At 2Y, both injured and contralateral knees had elevated T1r relaxation times compared to BL in the medial femur (MF)(p<0.001) and lateral femur (LF)(p<0.001). At 2Y, the injured knee had significantly higher T1r relaxation times in the MF than the contralateral knee (p<0.001). Correlations at 2Y: ITR was correlated with T1r relaxation times in the region overlying the posterior horn of the meniscus (cLF-p) in both the injured (r¼-0.583, p¼0.001)(Fig.2) and contralateral knees (r¼0.454, p¼0.015). ITR in both injured and contralateral knees were also correlated with T1r relaxation times in the anterolateral tibia (aLT)(r¼0.410, p¼0.030 and r¼0.580, p¼0.001). Lastly, TP of the contralateral knee was correlated with T1r relaxation times in the anteromedial tibia (aMT) (r¼0.432, p¼0.022). Conclusions: These results suggest that ACL reconstruction was not able to completely restore the TP and ITR. The anterior translation of the tibia in the injured knee at 2Y compared to 1Y may be indicative of worsening biomechanics that were not present at 6M or 1Y. The larger ITR and more anterior TP of the contralateral knee at 2Y compared to BL suggest that the uninjured knee has undergone significant biomechanical changes over the two years. This may be a result of shifts in loading patterns on the contralateral knee due to instability in the injured knee. The rise in T1r relaxation times in both knees at 2Y compared to BL is indicative of signs of cartilage matrix changes in both knees. The correlations between ITR and T1r relaxation times in the same two compartments in both knees suggest that the similar altered biomechanics in both knees may be contributing to these cartilage health changes. These negative correlations, which were not present at 6M or 1Y, may indicate that tighter knees may be doing worse at 2Y. 3 year follow up scans will provide more data on the interrelationship between altered biomechanics in both injured and contralateral knees and cartilage health. Acknowledgments: Funding was provided by NIH/ NIAMS and AOSSM.
J.K. Ochoa, K. Amano, M. Tanaka, V. Pedoia, R. Souza, X. Li, C. Ma. Univ. of California, San Francisco, San Francisco, CA, USA Purpose: Anterior cruciate ligament (ACL) injuries are a common injury of the knee that can lead to post-traumatic osteoarthritis. The effectiveness of ACL reconstruction in preventing osteoarthritis has been debated. While ACL reconstruction aims to restore normal knee biomechanics, subtle kinematic changes detectable only with MRI may still persist following surgery. The purpose of this study was to analyze knee biomechanics at baseline (BL) (at the time of injury and prior to surgery) and up to 2-years post-reconstruction, and correlate the biomechanical findings with cartilage matrix changes using T1r quantitative MRI, a method that has shown to characterize cartilage health by detecting degenerative changes. Methods: 32 patients (age 29.9 ± 8.4 years) with isolated, acute, unilateral ACL injury who completed scans at BL and follow-ups at 6
Fig. 1. (A) Lateral side compartments (LF and LT) and their sub-compartments. Also includes patella and trochlear compartment. (B) Medial side compartments (MF and MT) and their sub-compartments.
Abstracts / Osteoarthritis and Cartilage 24 (2016) S63eS534
S407
S.R. Filbay y, I.N. Ackerman z, T.G. Russell y, K.M. Crossley x. y The Univ. of Queensland, Brisbane, Australia; z The Univ. of Melbourne, Melbourne, Australia; x La Trobe Univ., Melbourne, Australia
in Figure 2. A delay from injury to ACLR > 6 months was associated with a 6.9 times greater odds of having OA, compared to a delay of < 6 months (OR 6.9, 95% CI 1.5 to 18.8). Having at least one additional knee surgery since ACLR was associated with a 5.1 fold increased odds of having knee OA (OR 5.1, 95% CI 1.7 to 15.4). Reporting a contact mechanism of ACL injury reduced the odds of having OA by 29% compared to reporting a non-contact mechanism of injury (OR 0.29, 95% CI 0.1 to 0.9). A 1 point increase in ACL-QOL (better score on a 0100 scale) corresponded to a 3% decrease in the odds of having OA (Figure 2B). Participants who reported a moderate to significant impact of their knee on their QOL had 8.9 times greater odds of having knee OA, compared to participants reporting no or slight impact on QOL (OR 8.9, 95% CI 2.3 to 35.1). Dissatisfaction with knee function was associated with 6.9 times greater odds of having OA, compared to satisfaction with knee function (OR 6.9, 95 % CI 1.4 to 32.7). Wide CIs were observed for all analyses, likely related to the small sample size. Time since ACLR, sex, anxiety and depression, pain, symptoms, work limitations and return to sport status were not associated with the odds of having radiographic OA (Figure 2). Conclusions: Five key factors were strongly related to increased odds of radiographic OA in individuals with knee difficulties at a median 8 years following ACLR. Reporting more knee-related QOL impairment or dissatisfaction with knee function was associated with increased odds of having radiographic OA. Receiving one or more additional knee surgeries since ACLR, reporting a non-contact mechanism of ACL injury and a delay longer than six months from injury to ACLR were associated with a heightened risk of radiographic OA. Considering pain and symptoms were not related to OA in this sample, there may be other factors contributing to dissatisfaction and impaired QOL in people with OA and knee difficulties after ACLR.
Purpose: The high rates of early-onset knee osteoarthritis (OA) after anterior cruciate ligament reconstruction (ACLR) are alarming considering the young, active population in which ACLR is most prevalent. The relevance and impact of the presence or absence of radiographic OA in people with knee pain, symptoms or functional limitations after ACLR is poorly understood. The purpose of this study was to compare (i) participant characteristics; (ii) QOL and psychological factors, and (iii) physical and activity-related factors, in ACL-reconstructed individuals with knee difficulties with or without radiographic tibiofemoral and/or patellofemoral OA. Methods: Knee radiographs were obtained from 81 individuals recruited from a larger cross-sectional study of 162 people with knee difficulties 5 to 20 years following ACLR. Knee difficulties were defined as reporting a less than optimal score for at least 50% of items in any two Knee Injury and Osteoarthritis Outcome Score (KOOS) subscales (pain, symptoms, ADL, sport/rec, QOL). Three radiograph views were requested: weight bearing postero-anterior (PA) erect in 15 degrees knee flexion, weight-bearing lateral in 30 degrees knee flexion, and non-weight bearing skyline in 45 degrees knee flexion. All radiographs were graded by an experienced radiologist using the Kellgren & Lawrence criteria and a score of grade 2 for the tibiofemoral or patellofemoral joint was used to define the presence of radiographic OA. All exploratory factors (described in Figure 2) were selected based on literature review and clinical reasoning and were collected by questionnaire at a median 8 (IQR 7 to 12) years after ACLR. Direct acyclic graphs were used to minimise conditional associations and bias in multivariable analyses. Binary logistic regression was utilised to investigate the relationship between exploratory factors and the presence or absence of radiographic OA. Results are reported as odds ratios (ORs) and 95% confidence intervals (CIs), with adjustment for potential confounders (BMI, age, time since ACLR, additional knee surgery). Results: Knee radiographs were undertaken at a median 9 (IQR 8 to 11) months after questionnaire completion. Of the 50 people (62%) with radiographic OA, 38% had OA in one compartment, 34% had OA in 2 compartments and 28% had radiographic OA in all 3 knee compartments (Figure 1). Participant characteristics and patient-reported outcomes for people with and without radiographic OA are described
Figure 1. Knee osteoarthritis prevalence by severity and compartment. Figure 1A: n¼81; Figure 1B: n¼50; TFJ: tibiofemoral joint; PFJ: patellofemoral joint; All compartments: presents the highest grade of osteoarthritis from any compartment for each participant; Grade 0e4: Graded using the Kellgren and Lawrence score.
Fig. 2. Partial correlation graphing the adjusted values of ITR in the injured knees and T1r relaxation times in the cLF-p. Adjusted values have been used as they have been corrected for age, gender, BMI and knee flexion angle.
683 FACTORS ASSOCIATED WITH RADIOGRAPHIC OSTEOARTHRITIS IN PEOPLE WITH KNEE PAIN, SYMPTOMS OR FUNCTIONAL LIMITATIONS AFTER ANTERIOR CRUCIATE LIGAMENT RECONSTRUCTION
Abstracts / Osteoarthritis and Cartilage 24 (2016) S63eS534
681 USING ENDOPLASMIC RETICULUM STRESS TO MODEL OA IN MICE K. Posey, J. Alcorn, J. Hecht. Univ. of Texas Med. Sch. at Houston, Houston, TX, USA Purpose: There is growing evidence that endoplasmic reticulum (ER) stress plays a role in osteoarthritis (OA). In order to evaluate the role of ER stress in idiopathic adult-onset OA, we utilized ER stress as an OA inducer in a mouse model. Methods: In this system, ER stress is induced by expression of mutant cartilage oligomeric matrix protein (COMP) in the chondrocytes of adult mice between the ages of 16 to 20 weeks. Mutant COMP expression is dependent on presence of doxycycline (Tet-On system) and is restricted to chondrocytes by a type II collagen promoter. Results: We have preciously shown that mutant COMP causes ER and oxidative stress and inflammation in growth plate and articular chondrocytes. Our data further shows additional characteristics of OA including reduction in proteoglycans and glycosaminoglycans, presence of inflammatory markers and thinning of the articular cartilage in adult mice. We are now assessing classic OA markers in the articular cartilage, evaluating articular cartilage health using the OARSI scoring system, gait as a measure of limb function and pain. Conclusions: The strengths of this new model system of OA are the ability to induce ER stress at different time points and a system to investigate and define the role that ER stress plays in OA. This inducible model of adult-onset idiopathic OA will be used to assess recovery of the joint after damage, the contribution of multiple OA risk factors, identify early markers of OA and evaluate new treatments directed at dampening ER stress.
OA: Ligament/Meniscus/ Tendon/Muscle 682 ALTERED BIOMECHANICS AND CARTILAGE HEALTH CHANGES IN BILATERAL KNEES FOLLOWING UNILATERAL ACL RECONSTRUCTION: A 2-YEAR FOLLOW-UP
months (6M), 1 year (1Y) and 2 years (2Y) post reconstruction were included in this analysis. Images of the bilateral knees were acquired using a 3-Tesla MRI-scanner. High-resolution 3D FSE and quantitative T1r sequences, and sagittal T2 FSE images with 25% of the patient’s body weight applied axially were obtained. Using an in-house Matlab program, image segmentation was performed to define the biomechanical parameters of tibial position (TP) and internal tibial rotation (ITR) with respect to the femur in the extended position. T1r relaxation times were quantified for 6 compartments which were then divided into 14 sub-compartments (Fig. 1). Paired t-tests were used for comparing the biomechanical parameters and T1r between the injured and uninjured knees. Paired t-tests were also used for longitudinal comparisons of the four time points. Partial correlations were used to determine correlation coefficients between the biomechanical parameters and T1r relaxation times while controlling for age, gender, BMI and knee flexion angle. Significance was defined as p <0.05. Results: Biomechanics: TP of the injured knee was significantly more anterior than the contralateral uninjured knee at BL (1.65mm, p<0.001), 6M (1.02mm, p¼0.003), 1Y (0.73mm, p¼0.016), and 2Y (1.46mm, p<0.001). TP of the injured knee was also more anterior at 2Y than at 1Y (0.69mm, p¼0.006). TP of the contralateral knee was significantly more anterior at 2Y compared to BL (0.85mm, p<0.001). Both injured and contralateral knees also had higher ITR at 2Y than at BL (p¼0.045, p<0.001). T1r: At 2Y, both injured and contralateral knees had elevated T1r relaxation times compared to BL in the medial femur (MF)(p<0.001) and lateral femur (LF)(p<0.001). At 2Y, the injured knee had significantly higher T1r relaxation times in the MF than the contralateral knee (p<0.001). Correlations at 2Y: ITR was correlated with T1r relaxation times in the region overlying the posterior horn of the meniscus (cLF-p) in both the injured (r¼-0.583, p¼0.001)(Fig.2) and contralateral knees (r¼0.454, p¼0.015). ITR in both injured and contralateral knees were also correlated with T1r relaxation times in the anterolateral tibia (aLT)(r¼0.410, p¼0.030 and r¼0.580, p¼0.001). Lastly, TP of the contralateral knee was correlated with T1r relaxation times in the anteromedial tibia (aMT) (r¼0.432, p¼0.022). Conclusions: These results suggest that ACL reconstruction was not able to completely restore the TP and ITR. The anterior translation of the tibia in the injured knee at 2Y compared to 1Y may be indicative of worsening biomechanics that were not present at 6M or 1Y. The larger ITR and more anterior TP of the contralateral knee at 2Y compared to BL suggest that the uninjured knee has undergone significant biomechanical changes over the two years. This may be a result of shifts in loading patterns on the contralateral knee due to instability in the injured knee. The rise in T1r relaxation times in both knees at 2Y compared to BL is indicative of signs of cartilage matrix changes in both knees. The correlations between ITR and T1r relaxation times in the same two compartments in both knees suggest that the similar altered biomechanics in both knees may be contributing to these cartilage health changes. These negative correlations, which were not present at 6M or 1Y, may indicate that tighter knees may be doing worse at 2Y. 3 year follow up scans will provide more data on the interrelationship between altered biomechanics in both injured and contralateral knees and cartilage health. Acknowledgments: Funding was provided by NIH/ NIAMS and AOSSM.
J.K. Ochoa, K. Amano, M. Tanaka, V. Pedoia, R. Souza, X. Li, C. Ma. Univ. of California, San Francisco, San Francisco, CA, USA Purpose: Anterior cruciate ligament (ACL) injuries are a common injury of the knee that can lead to post-traumatic osteoarthritis. The effectiveness of ACL reconstruction in preventing osteoarthritis has been debated. While ACL reconstruction aims to restore normal knee biomechanics, subtle kinematic changes detectable only with MRI may still persist following surgery. The purpose of this study was to analyze knee biomechanics at baseline (BL) (at the time of injury and prior to surgery) and up to 2-years post-reconstruction, and correlate the biomechanical findings with cartilage matrix changes using T1r quantitative MRI, a method that has shown to characterize cartilage health by detecting degenerative changes. Methods: 32 patients (age 29.9 ± 8.4 years) with isolated, acute, unilateral ACL injury who completed scans at BL and follow-ups at 6
Fig. 1. (A) Lateral side compartments (LF and LT) and their sub-compartments. Also includes patella and trochlear compartment. (B) Medial side compartments (MF and MT) and their sub-compartments.
Abstracts / Osteoarthritis and Cartilage 24 (2016) S63eS534
S407
S.R. Filbay y, I.N. Ackerman z, T.G. Russell y, K.M. Crossley x. y The Univ. of Queensland, Brisbane, Australia; z The Univ. of Melbourne, Melbourne, Australia; x La Trobe Univ., Melbourne, Australia
in Figure 2. A delay from injury to ACLR > 6 months was associated with a 6.9 times greater odds of having OA, compared to a delay of < 6 months (OR 6.9, 95% CI 1.5 to 18.8). Having at least one additional knee surgery since ACLR was associated with a 5.1 fold increased odds of having knee OA (OR 5.1, 95% CI 1.7 to 15.4). Reporting a contact mechanism of ACL injury reduced the odds of having OA by 29% compared to reporting a non-contact mechanism of injury (OR 0.29, 95% CI 0.1 to 0.9). A 1 point increase in ACL-QOL (better score on a 0100 scale) corresponded to a 3% decrease in the odds of having OA (Figure 2B). Participants who reported a moderate to significant impact of their knee on their QOL had 8.9 times greater odds of having knee OA, compared to participants reporting no or slight impact on QOL (OR 8.9, 95% CI 2.3 to 35.1). Dissatisfaction with knee function was associated with 6.9 times greater odds of having OA, compared to satisfaction with knee function (OR 6.9, 95 % CI 1.4 to 32.7). Wide CIs were observed for all analyses, likely related to the small sample size. Time since ACLR, sex, anxiety and depression, pain, symptoms, work limitations and return to sport status were not associated with the odds of having radiographic OA (Figure 2). Conclusions: Five key factors were strongly related to increased odds of radiographic OA in individuals with knee difficulties at a median 8 years following ACLR. Reporting more knee-related QOL impairment or dissatisfaction with knee function was associated with increased odds of having radiographic OA. Receiving one or more additional knee surgeries since ACLR, reporting a non-contact mechanism of ACL injury and a delay longer than six months from injury to ACLR were associated with a heightened risk of radiographic OA. Considering pain and symptoms were not related to OA in this sample, there may be other factors contributing to dissatisfaction and impaired QOL in people with OA and knee difficulties after ACLR.
Purpose: The high rates of early-onset knee osteoarthritis (OA) after anterior cruciate ligament reconstruction (ACLR) are alarming considering the young, active population in which ACLR is most prevalent. The relevance and impact of the presence or absence of radiographic OA in people with knee pain, symptoms or functional limitations after ACLR is poorly understood. The purpose of this study was to compare (i) participant characteristics; (ii) QOL and psychological factors, and (iii) physical and activity-related factors, in ACL-reconstructed individuals with knee difficulties with or without radiographic tibiofemoral and/or patellofemoral OA. Methods: Knee radiographs were obtained from 81 individuals recruited from a larger cross-sectional study of 162 people with knee difficulties 5 to 20 years following ACLR. Knee difficulties were defined as reporting a less than optimal score for at least 50% of items in any two Knee Injury and Osteoarthritis Outcome Score (KOOS) subscales (pain, symptoms, ADL, sport/rec, QOL). Three radiograph views were requested: weight bearing postero-anterior (PA) erect in 15 degrees knee flexion, weight-bearing lateral in 30 degrees knee flexion, and non-weight bearing skyline in 45 degrees knee flexion. All radiographs were graded by an experienced radiologist using the Kellgren & Lawrence criteria and a score of grade 2 for the tibiofemoral or patellofemoral joint was used to define the presence of radiographic OA. All exploratory factors (described in Figure 2) were selected based on literature review and clinical reasoning and were collected by questionnaire at a median 8 (IQR 7 to 12) years after ACLR. Direct acyclic graphs were used to minimise conditional associations and bias in multivariable analyses. Binary logistic regression was utilised to investigate the relationship between exploratory factors and the presence or absence of radiographic OA. Results are reported as odds ratios (ORs) and 95% confidence intervals (CIs), with adjustment for potential confounders (BMI, age, time since ACLR, additional knee surgery). Results: Knee radiographs were undertaken at a median 9 (IQR 8 to 11) months after questionnaire completion. Of the 50 people (62%) with radiographic OA, 38% had OA in one compartment, 34% had OA in 2 compartments and 28% had radiographic OA in all 3 knee compartments (Figure 1). Participant characteristics and patient-reported outcomes for people with and without radiographic OA are described
Figure 1. Knee osteoarthritis prevalence by severity and compartment. Figure 1A: n¼81; Figure 1B: n¼50; TFJ: tibiofemoral joint; PFJ: patellofemoral joint; All compartments: presents the highest grade of osteoarthritis from any compartment for each participant; Grade 0e4: Graded using the Kellgren and Lawrence score.
Fig. 2. Partial correlation graphing the adjusted values of ITR in the injured knees and T1r relaxation times in the cLF-p. Adjusted values have been used as they have been corrected for age, gender, BMI and knee flexion angle.
683 FACTORS ASSOCIATED WITH RADIOGRAPHIC OSTEOARTHRITIS IN PEOPLE WITH KNEE PAIN, SYMPTOMS OR FUNCTIONAL LIMITATIONS AFTER ANTERIOR CRUCIATE LIGAMENT RECONSTRUCTION