SPECIAL COMMUNICATIONS Methodological Advances
Developing a Mouse Model of Chronic Ankle Instability ERIK A. WIKSTROM1,2, TRICIA HUBBARD-TURNER1,2, SARA WOODS1, SOPHIE GUDERIAN3, and MICHAEL J. TURNER3 1
Biodynamics Research Laboratory, Department of Kinesiology, University of North Carolina at Charlotte, Charlotte, NC; Center for Biomedical Engineering and Science, University of North Carolina at Charlotte, Charlotte, NC; and 3Laboratory of Systems Physiology, Department of Kinesiology, University of North Carolina at Charlotte, Charlotte, NC
2
ABSTRACT WIKSTROM, E. A., T. HUBBARD-TURNER, S. WOODS, S. GUDERIAN, and M. J. TURNER. Developing a Mouse Model of Chronic Ankle Instability. Med. Sci. Sports Exerc., Vol. 47, No. 4, pp. 866–872, 2015. Introduction: Ankle sprains are the most common orthopedic pathology experienced during sport and physical activity and often result in chronic ankle instability (CAI). Understanding how to prevent CAI is difficult because of the costs and logistics associated with clinical trials aimed at preventing the heterogeneous symptoms associated with CAI. Thus, a need exists to develop an animal model that presents similar long-term consequences as CAI to assess preclinical data. Thus, the purpose was to determine whether surgically transecting the lateral ligaments of a mouse hind limb results in the development of CAI-like symptoms 12 months after injury. Methods: Thirty male mice (CBA/J) were randomly placed into a SHAM (control), CFL (calcaneofibular ligament; mild ankle sprain), or ATFL/CFL (anterior talofibular ligament/ CFL; severe) ankle sprain group and housed individually. Three days after surgically transecting the respective lateral ligaments, mice were given a solid surface running wheel and daily running wheel measurements were recorded. Outcome measures of balance and gait were obtained before and at 4, 48, 54, and 60 wk after injury. Results: The ATFL/CFL group had significantly more hind foot slips than the CFL and SHAM groups (P G 0.05). The CFL also had more hind foot slips relative to the SHAM group (P G 0.05). The ATFL/CFL group was significantly less physically active relative to the SHAM and CFL groups (P G 0.05). A cut score of 4.75 foot slips had a sensitivity of 0.68 and specificity of 1.00 and indicates that 70% (14/20) of mice with an ankle sprain had developed CAI. Conclusions: The results of this study indicate that an acute ankle sprain in mice can result in the development of CAI-like symptoms 12 months after injury. Key Words: MURINE, ANKLE INJURY, PHYSICAL ACTIVITY, BALANCE, GAIT
A
experience an ankle sprain for the first time develop chronic ankle instability (CAI); however, this number has been reported as high as 75% (1,23,24). CAI is characterized by lifelong residual symptoms (e.g., balance deficits) (1,7), recurrent injury (33), and decreased physical activity (29). More importantly, as many as 78% of those with CAI develop posttraumatic ankle osteoarthritis (9,27), for which there are no effective conservative treatments. Collectively, the high incidence of CAI and rising prevalence of posttraumatic osteoarthritis indicate that intervention effectiveness for acute lateral ankle sprains is poor. However, our inability to properly treat acute lateral ankle sprains and prevent CAI development is not surprising, given the multifactorial nature of CAI (7,32) and the heterogeneity of the pathology observed in the existing literature (4,8). The development of an animal model that mimics the symptoms seen in humans with CAI has the potential to lessen the challenges associated with prospective human ankle sprain research. For example, the multitude of available outcomes to investigate (e.g., balance, gait, strength, laxity, etc.) and differences in injury severity, healing rates,
nkle sprains are the most common injuries associated with physical activity and athletic participation, accounting for approximately 60% of all injuries that occur during interscholastic and intercollegiate sports (6,10). In addition, medical costs for ankle sprains have been estimated to be approximately $4 billion annually (25). Thus, ankle sprains, although often viewed as inconsequential injuries, represent a significant public health problem (25,29) and a major health care burden. Unfortunately, ankle sprains are not one-time injuries. Indeed, about 30% of those who
Address for correspondence: Erik A. Wikstrom, Ph.D., A.T.C., FACSM, Biodynamics Research Laboratory, Department of Kinesiology, University of North Carolina at Charlotte, 9201 University City Blvd., Charlotte, NC 28223; E-mail: [email protected]. Submitted for publication February 2014. Accepted for publication July 2014. 0195-9131/15/4704-0866/0 MEDICINE & SCIENCE IN SPORTS & EXERCISEÒ Copyright Ó 2014 by the American College of Sports Medicine DOI: 10.1249/MSS.0000000000000466
866
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
CHRONIC ANKLE INSTABILITY IN MICE
12 months was hypothesized to be sufficient for CAI-like symptoms to develop because this period is slightly greater than 50% of the lifespan for this mouse strain. Furthermore, we hypothesize that surgically induced ankle sprains will result in decreased physical activity levels but not spatial gait outcomes, on the basis of the human CAI literature.
METHODS Animals. Thirty male mice (CBA/J) were purchased from Jackson Laboratory (JAX, Bar Harbor, ME) at 5–6 wk old and were housed in the university vivarium with 12-h light/dark cycles and room temperature and relative humidity standardized to 18-C–22-C and 20%–40%, respectively (12,26). All mice were provided with water and standard chow (Teklad 8604 Rodent Diet; Harland, Madison, WI) ad libitum (12,26). The Institutional Animal Care and Use Committee at the University of North Carolina at Charlotte approved all study procedures as meeting the United States Department of Agriculture (USDA) and the Animal Welfare Act guidelines for the appropriate treatment of animal subjects. Surgical procedures. Using 4% isoflurane gas and supplemental oxygen, each mouse was anesthetized before having their right ankle shaved and cleaned with alcohol, followed by a chlorhexidine scrub (12). All mice remained under anesthesia while being moved to a sterile surgical field with a warming lamp. Using a microscope and sterile equipment, a small curvilinear incision was made behind the lateral malleolus for all mice. For the CFL group, the skin was then retracted and the CFL was transected (12). For the ATFL/CFL group, both the ATFL and CFL were identified and transected after retracting the skin (12). To transect the ATFL, the peroneal tendons were first identified and lifted using a surgical probe because the tendons were directly in line with and superficial to the ATFL in mice. For the SHAM group, no ligaments were damaged and the incision for all groups was closed using two drops of formulated cyanoacrylate surgical adhesive (12). Immediate postoperative care consisted of the following: 1) a subcutaneous injection of 5.0 mgIkgj1 of carprofen (Rimadyl) diluted with saline and 2) time under a warming lamp until each mouse was freely mobile (12). Additional postoperative care consisted of visual monitoring once at least every 24 h by the investigative team and by the university vivarium staff. In addition, 12.5-mg carprofen (Rimadyl) tablets were administered ad libitum for pain management for 72 h (12). Sensorimotor function. Balance was assessed by measuring the ability of the mice to cross an inclined (15-), narrow, round, wooden beam, 1 m long with a 19-mm diameter, that was elevated above a bench surface and connected with an enclosed box (20 cm2) for the mouse to escape into (3,12,22). All mice were trained to help ensure consistent performance. Training was complete when a mouse traversed the beam in less than 20 s for three consecutive attempts (3,12). During test trials, mice were allowed up to 60 s to cross the beam, with the duration to
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
867
SPECIAL COMMUNICATIONS
treatment plans, and patient adherence makes prospective investigations very difficult to conduct. As a result, a great deal of evidence related to contributing factors toward CAI is retrospective. However, using genetically identical mice that experience the same injury and are treated in the same manner could allow researchers the opportunity to gather preclinical data regarding how certain interventions can limit long-term functional loss and CAI development, potentially furthering our understanding of the mechanisms of CAI and/ or helping us identify effective therapeutic interventions for the prevention of CAI. Such an animal model could also support and/or further develop contemporary theoretical models associated with CAI. Although multiple models exist and overlap, some have hypothesized that the development of CAI is due to a cascade of events that stem from the initial ankle sprain consequences (7,30,31). Furthermore, an animal model could be used to address Hertel’s (7) call to focus on potential points of intervention to limit sensorimotor deficits (i.e., organismic constraints) via testing a range of outcomes at multiple time points after injury while also exploring the initial physiologic site of pathology. An animal model could test the hypothesis of Wikstrom et al. (31), using the dynamic systems theory as a framework, that limiting organismic constraints (i.e., residual symptoms) through effective initial treatments may inhibit the development of CAI more easily than a human model. Previously, Hubbard-Turner et al. (12) developed an acute ankle sprain mouse model. Transecting the calcaneofibular ligament (CFL) in the hind foot of a mouse seemed representative of a mild ankle sprain, as evidenced by the resolution of symptoms (i.e., altered balance, altered gait, and decreased physical activity) around 7 d after injury (12). Similarly, transecting the CFL and anterior talofibular ligament (ATFL) seemed representative of a moderate-to-severe ankle sprain on the basis of the presence of symptoms for up to 4 wk after injury (12). However, a pressing need to determine whether mice with a surgically induced ankle sprain will develop similar long-term consequences as humans diagnosed with CAI still exists. Therefore, the purpose of this investigation was to examine the long-term effects of surgically transecting the lateral ligaments of a mouse_s hind limb on physical activity and sensorimotor function. On the basis of the human CAI literature and the work by Hubbard-Turner et al. (12), we hypothesize that a surgically induced ankle sprain will result in development of CAI-like symptoms that mimic the presentation of CAI in humans. For the purposes of this investigation, CAI was operationally defined as the presence of balance impairments 1 yr after the injurious event. The time point of 12 months, although not directly translatable between humans and mice, was chosen for the following three reasons: 1) after 12 months, the risk of a recurrent lateral ankle sprain is similar to the risk of a first-time lateral ankle sprain in humans (28), 2) 12 months has been recommended as the time needed to satisfactorily determine whether an individual will become a coper (i.e., a person that has sprained their ankle but will not develop CAI in humans (30), and 3)
SPECIAL COMMUNICATIONS
cross the beam and the number of times the right hind foot slipped off the beam recorded as dependent variables (3,12). Each mouse completed two test trials per test session (baseline (presurgery) and 4, 48, 54, and 60 wk after surgery), with the two trial’s mean being used for further analysis. The primary balance assessor (EAW) conducted interrater reliability analysis with three separate individuals during the investigation. Two of these individuals where blinded to group assignment and had never participated in the previously described balance tests. The results of these analyses demonstrate good-to-excellent interrater reliability, with intraclass correlation coefficient values of 0.83, 0.92, and 0.84, respectively. Gait was assessed using the footprint test (3,12). The hind feet and forefeet of each mouse were painted with nontoxic red and green paints, respectively, before allowing them to traverse a 50-cm-long and 10-cm-wide runway (with 10-cmhigh walls) once during each test session (3,12). To paint the feet, mice were scruffed with their tail secured between the ring and pinky fingers of the same hand. This technique exposed all four paws for painting, which was typically completed with a single stroke using a narrow foam brush. Mice were then immediately placed in the runway and typically completed the footprint test within 10–15 s after their feet were painted. The dependent measures including stride length asymmetry, paw overlap asymmetry, hind foot base width, and forefoot base width were obtained as previously described (3,12). In brief, average stride length (cm) was measured between each stride using a heel-to-heel measuring technique before an asymmetry ratio was computed as right stride length divided by left stride length (12). Thus, a value 91 indicates a larger right stride length and a value G1 indicates a larger left stride length. Paw overlap was the average distance (cm) between the center of the left and right forefoot and hind foot prints (3,12). Asymmetry in average paw overlap was determined as the right paw overlap divided by the left paw overlap averages. Thus, a value 91 indicates less of a right forefoot/hind foot overlap and a value G1 indicates a larger left hind foot/forefoot overlap (12). The forefoot and hind foot width measures were defined as the average distance between hind foot and forefoot prints, respectively (3,12). Width outcomes (cm) were measured as the perpendicular distance from the inside of the paws among the left and right steps of each mouse. For each gait outcome, the maximum number of values was obtained from each test trial captured at each test session (baseline (presurgery) and 4, 48, 54, and 60 wk after surgery) while excluding the footprints made while the mouse was initiating and terminating gait. The mean value of each set of outcomes was used for statistical analysis. Intra- and interrater reliability of calculating these gait outcomes have been established as excellent (12). Data from 4 wk after injury were included in both the balance and gait analyses to contextualize current means with those observed acutely after injury and reported by Hubbard-Turner et al. (12). Data from 48 and 60 wk after injury were included to provide
868
Official Journal of the American College of Sports Medicine
confirmation of the 54-wk data because of the exploratory nature of the investigation. Physical activity measurement. Each mouse was individually housed in a cage containing a solid surface running wheel (127 mm; Ware Manufacturing, Phoenix, AZ), magnetic sensor, and digital odometer (BC600; Sigma Sport, Olney, IL) that recorded the number of running wheel revolutions (12,15,26). Daily running wheel measurements of duration (min) and distance (km) were recorded starting 4 d after surgery. Knab et al. (15) reported a high correlation and agreement between days of wheel running and wheel running measurements indicating that voluntary wheel running activity is repeatable and stable in mice. The daily average during the fourth, 48th, 54th, and 60th week after injury was then submitted for statistical analysis. Statistical analysis. Three separate between-group, twoway, multivariate ANOVA (MANOVA) (group–time) were performed to compare changes in physical activity (duration, distance), balance (time, slips), and gait (stride asymmetry, paw overlap asymmetry, hind foot width, and forefoot width) due to the independent variables. A between-group analysis was performed to overcome missing data points (i.e., allow unequal group comparisons). The Wilk lambda was used to interpret statistical differences, and Bonferroni post hoc comparisons were performed, when appropriate, to determine the location of differences when main effects and/or an interaction were observed. An alpha level of P G 0.05 was used to determine significant differences for all analyses. Because not all humans develop CAI after an acute lateral ankle sprain (see Wikstrom and Brown (30) for further review on copers), a secondary analysis was performed to determine the percentage of the current mice that had developed CAI on the basis of the current operational definition (i.e., balance deficits as 12 months after injury). To do this, we first submitted any dependent measure that demonstrated a significant group–time interaction or group main effect, during a univariate ANOVA, to a receiver operating characteristic (ROC) curve analysis. This analysis determines whether an outcome can accurately distinguish SHAM mice from mice that had a sprained ankle (i.e., mice in the ATFL/ CFL and CFL groups). An ROC curve illustrates the ‘‘tradeoff’’ between sensitivity and specificity throughout a continuous outcome’s entire range of values (32). The ROC curve not only calculates overall discriminatory accuracy with the area under the curve (AUC) statistic and associated 95% confidence intervals (CI) but also provides the data needed to calculate cutoff scores for outcomes that reach asymptotic significance. Cutoff scores, which were used as the classification tool, were determined by calculating the Youden index (J ) for each outcome value along significant ROC curves, with the largest J value representing the cutoff score (32). The formula used to calculate the Youden index (J ) is J = (sensitivity + specificity) j 1. Finally, calculated cutoff scores were applied to the appropriate 54-wk data to determine the percentage of mice that had CAI (i.e., percentage of mice whose outcome exceeds the calculated cut score for that outcome).
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
applied to the individual mice, 70% (14/20) with an ankle sprain exceeded 4.75 hind foot slips and this percentage was equal among the CFL (70%) and ATFL/CFL (70%) groups.
A significant group–time interaction (F16,262 = 3.25, P G 0.01) and group (F4,262 = 11.336, P G 0.01) and time main effects (F8,262 = 15.43, P G 0.01) were observed for the balance MANOVA. Subsequent univariate ANOVA indicated that significant group–time interactions were present for both the time to cross the balance beam (P = 0.01) and the number of hind foot slips (P G 0.01) but only revealed group and time main effects (P G 0.01) for the number of hind foot slips. Group means, SD, and the location of significant differences for balance data can be seen in Table 1. For the gait data, a significant time main effect (F16,367 = 2.95, P G 0.01) was noted but a group–time interaction (F32,444 = 1.34, P = 0.10) and group main effect (F8,240 = 1.44, P = 0.18) were not significant. Subsequent univariate ANOVA revealed no significant differences for any of the gait outcomes across time, indicating an overall change in gait outcomes when pooled but no significant change in any one outcome. Group means and SD of gait data can be seen in Table 2. Significant group (F4,208 = 2.72, P = 0.03) and time (F6,208 = 8.43, P G 0.01) main effects were revealed in the physical activity MANOVA, but the group–time interaction (F12,208 = 1.18, P = 0.29) failed to reach significance. Subsequent univariate ANOVA indicated that group and time main effects were present for both distance run and duration run (P G 0.03). Group means, SD, and the location of significant differences for physical activity data can be seen in Table 3. Three outcomes, hind foot slips, duration run, and distance run from 54 wk after injury were submitted for an ROC curve analysis. Only hind foot slips reached an asymptotic significance level (P G 0.01), with an AUC statistic of 0.90 and a 95% CI of 0.77–1.00. Distance run and duration run failed to reach asymptotic significance levels (P = 0.32), with an AUC statistic of 0.62 and a 95% CI of 0.42– 0.82 for both outcomes. Therefore, a J score (Youden index) was calculated for all values within the 54-wk hind foot slip data set and a cut score of 4.75 was identified (J = 0.68). This point had a corresponding sensitivity of 0.68 and a corresponding specificity of 1.00 as seen in Figure 1. When
DISCUSSION This study is an extension of the work done by HubbardTurner et al. (12), which developed an acute lateral ankle sprain model in mice. More specifically, we quantified the long-term consequences of a single lateral ankle sprain to physical activity and sensorimotor function. To date, the longest prospective tracking of an objective outcome after lateral ankle sprain in humans has been 12 wk (17), well short of the 12-month time frame recommended to confidently determine whether an individual will not develop CAI (30). Our results illustrate consequences 12 months after injury in mice that are comparable with that in humans who have been diagnosed with CAI and support our a priori hypothesis. Humans with CAI have been shown to have static and dynamic balance deficits across a range of postural control tasks and outcome measures (2). Furthermore, in the absence of patient-reported outcomes, dynamic balance was the best predictor of CAI status in a study by Wikstrom et al. (32). Our current results are consistent with the human literature and are comparable with those previously reported in mice (19). For example, the time needed to cross the beam (approximately 5–10 s) and the number of hind foot slips (0–5) observed for healthy control mice by Carter et al. (3) are comparable, albeit slightly lower in some cases than the values observed in the current investigation (Table 1). Long-term gait alterations identified in humans with CAI have been limited to kinematic variables obtained using motion analysis (7), but alterations in general spatiotemporal alterations (e.g., stride length or gait velocity) have yet to be reported. This lack of empirical data coincides with the clinical observations associated with CAI in humans, specifically no obvious limp and an ability to continue to participate in high-level activities, albeit with recurrent ankle sprains and/or giving-way episodes. However, future research using an animal model of CAI should attempt to capture the kinematics of gait to further validate that the
TABLE 1. Balance outcome group means, SD, and sample size per group at each test session. Baseline Time (s)
Slips
SHAM CFL ATFL/CFL Time means SHAM CFL ATFL/CFL Time means
4 wk after Injury
6.14 T 1.87* (n = 10) 12.31 T 5.10 (n = 10) 11.72 T 6.20 (n = 10) 11.12 T 6.05 (n = 10) 9.26 T 2.06 (n = 10) 9.68 T 3.56 (n = 10) 9.04 T 4.43 11.04 T 4.95 , , 1.30 T 1.18* ** *** 1.25 T 1.08*,**,*** 1.50 T 0.88*,**,*** 1.00 T 1.29*,**,*** 1.15 T 1.37*,**,*** 2.40 T 1.41*,** 1.15 T 1.24 1.71 T 1.21
48 wk after Injury 10.26 8.76 10.96 9.98 3.05 6.15 5.75 5.05
T 4.91 (n = 9) T 4.06 (n = 10) T 3.87 (n = 10) T 4.23 T 1.58* T 2.62 T 1.68 T 2.39******
54 wk after Injury 10.47 T 4.52 (n = 9) 10.98 T 2.79 (n = 10) 11.89 T 3.64 (n = 10) 11.14 T 4.43 3.00 T 0.86* 5.30 T 2.00 5.85 T 1.78 4.77 T 2.01******
60 wk after Injury
Group Mean
9.14 T 4.52 (n = 9) 9.64 T 5.06 9.93 T 2.29 (n = 10) 10.50 T 4.51 15.63 T 5.05 (n = 10) 11.48 T 4.27 11.65 T 4.94 2.44 T 2.02*,** 2.17 T 1.56 3.60 T 2.28 3.51 T 2.75**** 6.60 T 2.07 4.35 T 2.71****,***** 4.27 T 2.72******
An animal may not have been tested at each test session because of death, health, or behavioral issues. *Indicates a significant difference from ATFL/CFL at 60 wk after injury (P G 0.05). **Indicates a significant difference from CFL at 48 wk after injury and ATFL/CFL at 54 wk after injury (P G 0.05). ***Indicates a significant difference from CFL at 54 wk after injury and ATFL/CFL at 48 wk after injury (P G 0.05). ****Indicates a significant difference from the SHAM group (P G 0.05). *****Indicates a significant difference from the CFL group (P G 0.05). ******Indicates a significant difference from baseline and 4 wk after injury (P G 0.05).
CHRONIC ANKLE INSTABILITY IN MICE
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
869
SPECIAL COMMUNICATIONS
RESULTS
SPECIAL COMMUNICATIONS
TABLE 2. Gait outcome group means, SD, and sample size per group at each test session. Stride asymmetry
Overlap asymmetry
Hind foot width (cm)
Forefoot width (cm)
SHAM CFL ATFL/CFL Time means SHAM CFL ATFL/CFL Time means SHAM CFL ATFL/CFL Time means SHAM CFL ATFL/CFL Time means
Baseline
4 wk after Injury
48 wk after Injury
54 wk after Injury
60 wk after Injury
0.99 T 0.03 (n = 10) 1.00 T 0.07 (n = 10) 1.04 T 0.07 (n = 10) 1.01 T 0.06 1.03 T 0.48 1.38 T 1.01 0.96 T 0.27 1.12 T 0.67 1.93 T 0.30 1.73 T 0.17 2.14 T 0.44 1.93 T 0.35 1.45 T 0.27 1.23 T 0.25 1.55 T 0.36 1.41 T 0.32
0.99 T 0.07 (n = 9) 1.03 T 0.07 (n = 9) 0.96 T 0.11 (n = 8) 1.00 T 0.09 1.27 T 0.56 1.34 T 0.71 0.98 T 0.66 1.20 T 0.64 2.22 T 0.41 2.19 T 0.29 2.17 T 0.44 2.19 T 0.37 1.35 T 0.29 1.47 T 0.22 1.47 T 0.17 1.43 T 0.26
1.03 T 0.07 (n = 8) 1.00 T 0.05 (n = 10) 1.05 T 0.14 (n = 8) 1.03 T 0.09 1.25 T 1.02 1.07 T 1.38 0.88 T 0.59 1.06 T 1.05 2.3 T 0.53 2.44 T 0.39 2.15 T 0.47 2.30 T 0.46 1.46 T 0.27 1.67 T 0.29 1.43 T 0.32 1.53 T 0.31
0.99 T 0.09 (n = 9) 0.92 T 0.11 (n = 10) 1.00 T 0.13 (n = 9) 0.97 T 0.11 1.43 T 0.97 0.92 T 0.45 1.41 T 1.59 1.24 T 1.07 2.05 T 0.41 2.32 T 0.35 2.09 T 0.42 2.16 T 0.40 1.09 T 0.44 1.43 T 0.34 1.35 T 0.19 1.30 T 0.36
1.03 T 0.15 (n = 9) 0.99 T 0.10 (n = 10) 0.94 T 0.16 (n = 9) 0.99 T 0.14 2.45 T 1.58 3.52 T 3.66 1.14 T 0.31 2.41 T 2.49 2.44 T 0.41 2.10 T 0.32 2.21 T 0.19 2.25 T 0.42 1.40 T 0.34 1.40 T 0.21 1.47 T 0.59 1.42 T 0.39
Group Mean 1.01 0.99 1.00 1.00 1.48 1.65 1.08 1.41 2.18 2.16 2.15 2.16 1.35 1.44 1.46 1.42
T 0.09 T 0.09 T 0.13 T 0.10 T 1.07 T 2.03 T 0.82 T 1.44 T 0.43 T 0.39 T 0.43 T 0.41 T 0.34 T 0.29 T 0.35 T 0.33
An animal may not have been tested at each test session because of death, health, or behavioral issues.
model presents in a manner similar to CAI in humans. The observed stride lengths (6–8 cm), paw overlap (0–0.5 cm), hind foot width (2–3 cm), and forefoot width (1–1.5 cm) for healthy control mice made by Carter et al. (3) are also comparable with those observed during the current study (Table 2) in similarly age mice. Although physical activity data remain relatively sparse in the human CAI literature, our results do coincide with available empirical data (11,29), providing further evidence that an acute ankle sprain in mice can lead to symptoms that mimic the presentation of CAI in humans. In addition, the physical activity levels of the current SHAM mice are lower than those observed by Lightfoot et al. (20), but the mice in the current investigation were about 3–4 wk older and did undergo a SHAM surgery. Although the agerelated difference seems consistent with the natural decline in physical activity in mice as they age, as reported by Turner et al. (26), it is important to note that different strains of mice were used and that the effect of the SHAM surgery of physical activity levels remains unknown at this time. Given the consistency of the chosen outcome measures among the assessments 48–60 wk after injury, the current data provide evidence that surgically transecting lateral ligaments of a mouse’s ankle can result in symptoms that mimic the presentation of CAI in humans and thus may be a valid animal model of CAI. Although the percentage of mice (70%) classified as having CAI is high, it is within the range of values reported in humans (1,23,24). Furthermore, the cutoff score of 4.75
hind foot slips is reasonable because it falls well within the range of scores observed during the balance assessment of mice with a surgically induced ankle sprain. Interestingly, both ankle sprain groups (CFL and ATFL/CFL) demonstrated a high but equal percentage of being classified as having CAI (70%). Although not an anticipated result, there is no empirical data suggesting a correlation between initial ankle sprain severity and the likelihood of developing CAI in humans (16). Thus, the equal rates of CAI development in both the mild and severe ankle sprain groups of the current investigation seem to support the hypothesis that the sensorimotor system’s ability to successfully compensate for structural damage is critical for the prevention of CAI regardless of the severity of the initial sprain induced (21,31). However, the current results would also suggest that the greater the magnitude of structural damage, the greater the difficulty an organism will have in physical activity over its lifespan. For example, only the ATFL/CFL group ran significantly less (distance and duration) relative to the SHAM group over the course of the entire investigation. Because the group differences in physical activity never fully resolved but were greatest shortly after the injury, further research is needed to determine whether surgical repair of severe acute ankle sprains may be an appropriate treatment option and/or whether more conservative management (e.g., cast immobilization) of severe acute ankle sprains may be more beneficial than the current standard of care (i.e., early mobilization). Unfortunately, more high-quality randomized
TABLE 3. Physical activity outcome group means, SD, and sample size per group at each test session. Duration (min)
Distance (km)
SHAM CFL ATFL/CFL Time means SHAM CFL ATFL/CFL Time means
4-wk Post Injury
48-wk Post Injury
54-wk Post Injury
60-wk Post Injury
207.00 T 129.26 (n = 10) 217.53 T 108.61 (n = 10) 141.33 T 112.69 (n = 10) 188.62 T 118.16 4.97 T 3.47 4.19 T 2.29 2.85 T 2.77 4.00 T 2.92
158.00 T 49.55 (n = 10) 156.74 T 50.24 (n = 10) 108.74 T 35.76 (n = 10) 140.58 T 49.77 2.20 T 0.79 2.43 T 1.00 1.55 T 0.74 2.05 T 0.91***
128.17 T 37.37 (n = 10) 140.71 T 62.11 (n = 10) 95.77 T 37.25 (n = 10) 121.32 T 49.77*** 1.82 T 0.63 2.15 T 1.09 1.32 T 0.63 1.76 T 0.86***
134.86 T 64.55 (n = 10) 121.31 T 46.04 (n = 10) 99.49 T 73.02 (n = 10) 117.99 T 61.67*** 1.86 T 1.04 1.76 T 0.73 1.50 T 1.03 1.70 T 0.92***
Group Mean 158.36 T 83.56 159.07 T 77.55 111.33 T 71.46*,** 2.77 T 2.31 2.63 T 1.65 1.81 T 1.62*
*Indicates a significant difference from the SHAM group (P G 0.05). **Indicates a significant difference from the CFL group (P G 0.05). ***Indicates a significant difference from 4 wk after injury (P G 0.05).
870
Official Journal of the American College of Sports Medicine
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
controlled trials are needed to determine whether surgical interventions are more effective than conservative interventions for acute lateral ankle sprains in humans (13). However, within existing conservative treatments of humans, longer duration and more stringent forms of immobilization seem to be advantageous for severe ankle sprains (19). Given the high percentage of mice that were classified as having CAI, on the basis of the current operational definition, both the CFL and ATLF/CFL mice models seem to be ideal candidates to use when examining the efficacy of various therapeutic interventions delivered acutely or subacutely, as called for by Hertel (7), at limiting the introduction of organismic constraints (31) and preventing CAI development. On the contrary, these models do not seem ideal to investigate characteristics that differ between those that do and do not develop CAI (30). Despite this limitation, the model can be used in future research to capture the proposed cascade of events (7,30,31) through repeated testing of a breadth of outcomes that represent the full spectrum of structural and sensorimotor adaptations seen in humans. Although this model seems to fit within the framework of contemporary theoretical models of CAI, future research is also needed to explore the physiologic response to the
CONCLUSIONS The transection of the lateral ligaments of a mouse’s hind foot results in long-term balance impairments that exceed the age-related alterations observed in the SHAM group. Furthermore, a high percentage of CAI developed regardless of the number of lateral ligaments transected (CFL vs ATFL/ CFL). Transecting the ATFL/CFL also seems to affect a mouse’s ability to remain physically active over the subsequent 12 months, but further research is needed to confirm the current results. Given the cost and logistics of quality clinical trials designed at developing CAI prevention interventions, either of the current mouse models seems ideal to allow researchers the opportunity to gather preclinical data regarding how certain interventions can limit long-term functional loss and CAI development. This study was funded by the University of North Carolina at Charlotte Faculty Research Grants program. There are no conflicts of interest to report. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
REFERENCES 1. Anandacoomarasamy A, Barnsley L. Long term outcomes of inversion ankle injuries. Br J Sports Med. 2005;39(3):e14. 2. Arnold BL, De La Motte S, Linens S, Ross SE. Ankle instability is associated with balance impairments: a meta-analysis. Med Sci Sports Exerc. 2009;41(5):1048–62. 3. Carter RJ, Lione LA, Humbry T, et al. Characterization of progresive motor deficits in mice transgenic for the human Huntington’s disease mutation. J Neurosci. 1999;19(8):3248–57. 4. Delahunt E, Coughlin GF, Caulfield B, Nightingale EJ, Lin CC, Hiller CE. Inclusion criteria when investigating insuficiencies in
CHRONIC ANKLE INSTABILITY IN MICE
chronic ankle instability. Med Sci Sports Exerc. 2010;42(11): 2106–21. 5. Delahunt E, d’Helft C, Brama PA. Development of an animal model of chronic ankle instability. J Orthop Sports Phys Ther. 2013;43(3):A14. 6. Fernandez WG, Yard EE, Comstock RD. Epidemiology of lower extremity injuries among U.S. high school athletes. Acad Emerg Med. 2007;14(7):641–5. 7. Hertel J. Sensorimotor deficits with ankle sprains and chronic ankle instability. Clin Sports Med. 2008;27(3):353–70.
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
871
SPECIAL COMMUNICATIONS
FIGURE 1—Hind foot slips ROC curve indicating sensitivity and 1specificity trade-off. The diagnostic accuracy of hind foot slips (solid black line) across the range of recorded scores is shown relative to the reference line (dashed black line), which indicates that a test performed no better than random. The large diamond on the ROC curve represent the location of the cutoff score.
current model (e.g., RNA expression in various tissues, histology of tissue, etc.), the consistency of the model across different mouse strains, and how other injury and animal models compare with the existing results (5,14,18). Consistency across mouse strains is important because each strain has unique genetic characteristics that may affect the results. In this investigation, the CBA/J strain was selected despite the presence of a retinal degeneration allele because of the following: 1) it is a moderately to highly active mouse, which minimized potential floor effects with regard to physical activity after injury and 2) the strain is not predisposed to development of osteoarthritis. Although we acknowledge that retinal degeneration may have affected our results over time, the only difference between the groups was the presence and severity of a surgically induced ankle sprain. Although we do not believe that the use of the CBA/J strain compromised the results of this investigation, using different strains in future research will only strengthen the validity of the mouse ankle sprain model. With regard to other injury and animal models, several researchers have induced both a surgical and manual ankle sprain in rats but have only studied the immediate consequences of those models to date (14,18).
SPECIAL COMMUNICATIONS
8. Hiller CE, Kilbreath SL, Refshauge KM. Chronic ankle instability: evolution of the model. J Athl Train. 2011;46(2):133–41. 9. Hirose K, Murakami G, Minowa T, Kura H, Yamashita T. Lateral ligament injury of the ankle and associated articular cartilage degeneration in the talocrural joint: anatomic study using elderly cadavers. J Orthop Sci. 2004;9(1):37–43. 10. Hootman JM, Dick R, Agel J. Epidemiology of collegiate injuries for 15 sports: summary and recommendations for injury prevention initiatives. J Athl Train. 2007;42(2):311–9. 11. Hubbard-Turner T, Turner MJ. Physical activity levels in those with chronic ankle instability. J Orthop Sports Phys Ther. 2013; 43(3):A23. 12. Hubbard-Turner T, Wikstrom EA, Guderian S, Turner MJ. Acute ankle sprain in a mouse model. Med Sci Sports Exerc. 2013;45(8): 1623–8. 13. Kerkhoffs GM, Handoll HH, de Bie R, Rower BH, Struijs PA. Surgical versus conservative treatment for acute injuires of the lateral ligament complex of the ankle in adults. Cochrane Database Syst Rev. 2007;18(2):CD000380. 14. Kim HY, Wang J, Chung K, Chung JM. A surgical ankle sprain pain model in the rat: effects of morphine and idomethacin. Neurosci Lett 2008;442:161–4. 15. Knab A, Bowen RS, Moore-Harrison T, Hamilton AT, Turner MJ, Lightfoot JT. Repeatability of exercise behaviors in mice. Physiol Behav. 2009;98(4):433–40. 16. Konradsen L, Bech L, Ehrenbjerg M, Nickelsen T. Seven years follow-up after ankle inversion trauma. Scand J Med Sci Sports. 2002;12(3):129–35. 17. Konradsen L, Olesen S, Hansen H. Ankle sensorimotor control and eversion strength after acute ankle inversion injuries. Am J Sports Med. 1998;26:72–7. 18. Koo ST, Park YI, Lim KS, Chung K, Chung JM. Acupuncture analgesis in a new rat model of ankle sprain pain. Pain. 2002;99: 423–31. 19. Lamb SE, Marsh JL, Nakash R, Cooke MW. Mechanical supports for acute, severe ankle sprain: a pragmatic, multicentre, randomised controlled trial. Lancet. 2009;373:575–81. 20. Lightfoot JT, Leamy L, Pomp D, et al. Strain screen and haplotype association mapping of wheel running in inbred mouse strains. J Appl Physiol (1985). 2010;109(1):623–34.
872
Official Journal of the American College of Sports Medicine
21. McKeon PO, Hubbard-Turner TJ, Wikstrom EA. Consequences of ankle inversion trauma: a novel recognition and treatment paradigm. In: Zaslav KR, editor. An International Perspective on Topics in Sports Medicine and Sports Injury. 1st ed. Rijeka (Croatia): InTech; 2012. p. 457–80. 22. Perry TA, Torres EM, Czech C, Beyreuther K, Richards S, Dunnett SB. Cognitive and motor function in transgenic mice carrying excess copies of the 695 and 751 amino acid isoforms of the amyloid precursor protein gene. Alzheimers Res. 1995;1:5–14. 23. Peters JW, Trevino SG, Renstrom PA. Chronic lateral ankle instability. Foot Ankle. 1991;12(3):182–91. 24. Smith RW, Reischl SF. Treatment of ankle sprains in young athletes. Am J Sports Med. 1986;14(6):465–71. 25. Soboroff SH, Pappius EM, Komaroff AL. Benefits, risks, and costs of alternative approaches to the evaluation and treatment of severe ankle sprain. Clin Orthop Relat Res. 1984;183:160–8. 26. Turner MJ, Chavis MN, Turner TH. Enhanced diastolic filling performance with lifelong physical activity in aging mice. Med Sci Sports Exerc. 2013;45(10):1933–40. 27. Valderrabano V, Hintermann B, Horisberger M, Fung TS. Ligamentous posttraumatic ankle osteoarthritis. Am J Sports Med. 2006;34(4):612–20. 28. Verhagen E, Van der Beek A, Twisk J, Bouter L, Bahr R, Van Mechelen W. The effect of a proprioceptive balance board training program for the prevention of ankle sprains: a prospective controlled trial. Am J Sports Med. 2004;32(6):1385–93. 29. Verhagen RA, de Keizer G, Van Dijk CN. Long-term follow-up of inversion trauma of the ankle. Arch Orthop Trauma Surg. 1995;114: 92–6. 30. Wikstrom EA, Brown CN. Minimum reporting standards for copers in chronic ankle instability research. Sports Med. 2014;44(4): 251–68. 31. Wikstrom EA, Hubbard-Turner T, McKeon PO. Understanding and treating lateral ankle sprains and their consequences: a constraints-based approach. Sports Med. 2013;43(6):385–93. 32. Wikstrom EA, Tillman MD, Chmielewski TL, Cauraugh JH, Naugle KE, Borsa PA. Discriminating between copers and people with chronic ankle instability. J Athl Train. 2012;47(2):136–42. 33. Yeung MS, Chan KM, So CH, Yuan WY. An epidemiological survey on ankle sprain. Br J Sports Med. 1994;28(2):112–6.
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Copyright of Medicine & Science in Sports & Exercise is the property of Lippincott Williams & Wilkins and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Developing a Mouse Model of Chronic Ankle Instability ERIK A. WIKSTROM1,2, TRICIA HUBBARD-TURNER1,2, SARA WOODS1, SOPHIE GUDERIAN3, and MICHAEL J. TURNER3 1
Biodynamics Research Laboratory, Department of Kinesiology, University of North Carolina at Charlotte, Charlotte, NC; Center for Biomedical Engineering and Science, University of North Carolina at Charlotte, Charlotte, NC; and 3Laboratory of Systems Physiology, Department of Kinesiology, University of North Carolina at Charlotte, Charlotte, NC
2
ABSTRACT WIKSTROM, E. A., T. HUBBARD-TURNER, S. WOODS, S. GUDERIAN, and M. J. TURNER. Developing a Mouse Model of Chronic Ankle Instability. Med. Sci. Sports Exerc., Vol. 47, No. 4, pp. 866–872, 2015. Introduction: Ankle sprains are the most common orthopedic pathology experienced during sport and physical activity and often result in chronic ankle instability (CAI). Understanding how to prevent CAI is difficult because of the costs and logistics associated with clinical trials aimed at preventing the heterogeneous symptoms associated with CAI. Thus, a need exists to develop an animal model that presents similar long-term consequences as CAI to assess preclinical data. Thus, the purpose was to determine whether surgically transecting the lateral ligaments of a mouse hind limb results in the development of CAI-like symptoms 12 months after injury. Methods: Thirty male mice (CBA/J) were randomly placed into a SHAM (control), CFL (calcaneofibular ligament; mild ankle sprain), or ATFL/CFL (anterior talofibular ligament/ CFL; severe) ankle sprain group and housed individually. Three days after surgically transecting the respective lateral ligaments, mice were given a solid surface running wheel and daily running wheel measurements were recorded. Outcome measures of balance and gait were obtained before and at 4, 48, 54, and 60 wk after injury. Results: The ATFL/CFL group had significantly more hind foot slips than the CFL and SHAM groups (P G 0.05). The CFL also had more hind foot slips relative to the SHAM group (P G 0.05). The ATFL/CFL group was significantly less physically active relative to the SHAM and CFL groups (P G 0.05). A cut score of 4.75 foot slips had a sensitivity of 0.68 and specificity of 1.00 and indicates that 70% (14/20) of mice with an ankle sprain had developed CAI. Conclusions: The results of this study indicate that an acute ankle sprain in mice can result in the development of CAI-like symptoms 12 months after injury. Key Words: MURINE, ANKLE INJURY, PHYSICAL ACTIVITY, BALANCE, GAIT
A
experience an ankle sprain for the first time develop chronic ankle instability (CAI); however, this number has been reported as high as 75% (1,23,24). CAI is characterized by lifelong residual symptoms (e.g., balance deficits) (1,7), recurrent injury (33), and decreased physical activity (29). More importantly, as many as 78% of those with CAI develop posttraumatic ankle osteoarthritis (9,27), for which there are no effective conservative treatments. Collectively, the high incidence of CAI and rising prevalence of posttraumatic osteoarthritis indicate that intervention effectiveness for acute lateral ankle sprains is poor. However, our inability to properly treat acute lateral ankle sprains and prevent CAI development is not surprising, given the multifactorial nature of CAI (7,32) and the heterogeneity of the pathology observed in the existing literature (4,8). The development of an animal model that mimics the symptoms seen in humans with CAI has the potential to lessen the challenges associated with prospective human ankle sprain research. For example, the multitude of available outcomes to investigate (e.g., balance, gait, strength, laxity, etc.) and differences in injury severity, healing rates,
nkle sprains are the most common injuries associated with physical activity and athletic participation, accounting for approximately 60% of all injuries that occur during interscholastic and intercollegiate sports (6,10). In addition, medical costs for ankle sprains have been estimated to be approximately $4 billion annually (25). Thus, ankle sprains, although often viewed as inconsequential injuries, represent a significant public health problem (25,29) and a major health care burden. Unfortunately, ankle sprains are not one-time injuries. Indeed, about 30% of those who
Address for correspondence: Erik A. Wikstrom, Ph.D., A.T.C., FACSM, Biodynamics Research Laboratory, Department of Kinesiology, University of North Carolina at Charlotte, 9201 University City Blvd., Charlotte, NC 28223; E-mail: [email protected]. Submitted for publication February 2014. Accepted for publication July 2014. 0195-9131/15/4704-0866/0 MEDICINE & SCIENCE IN SPORTS & EXERCISEÒ Copyright Ó 2014 by the American College of Sports Medicine DOI: 10.1249/MSS.0000000000000466
866
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
CHRONIC ANKLE INSTABILITY IN MICE
12 months was hypothesized to be sufficient for CAI-like symptoms to develop because this period is slightly greater than 50% of the lifespan for this mouse strain. Furthermore, we hypothesize that surgically induced ankle sprains will result in decreased physical activity levels but not spatial gait outcomes, on the basis of the human CAI literature.
METHODS Animals. Thirty male mice (CBA/J) were purchased from Jackson Laboratory (JAX, Bar Harbor, ME) at 5–6 wk old and were housed in the university vivarium with 12-h light/dark cycles and room temperature and relative humidity standardized to 18-C–22-C and 20%–40%, respectively (12,26). All mice were provided with water and standard chow (Teklad 8604 Rodent Diet; Harland, Madison, WI) ad libitum (12,26). The Institutional Animal Care and Use Committee at the University of North Carolina at Charlotte approved all study procedures as meeting the United States Department of Agriculture (USDA) and the Animal Welfare Act guidelines for the appropriate treatment of animal subjects. Surgical procedures. Using 4% isoflurane gas and supplemental oxygen, each mouse was anesthetized before having their right ankle shaved and cleaned with alcohol, followed by a chlorhexidine scrub (12). All mice remained under anesthesia while being moved to a sterile surgical field with a warming lamp. Using a microscope and sterile equipment, a small curvilinear incision was made behind the lateral malleolus for all mice. For the CFL group, the skin was then retracted and the CFL was transected (12). For the ATFL/CFL group, both the ATFL and CFL were identified and transected after retracting the skin (12). To transect the ATFL, the peroneal tendons were first identified and lifted using a surgical probe because the tendons were directly in line with and superficial to the ATFL in mice. For the SHAM group, no ligaments were damaged and the incision for all groups was closed using two drops of formulated cyanoacrylate surgical adhesive (12). Immediate postoperative care consisted of the following: 1) a subcutaneous injection of 5.0 mgIkgj1 of carprofen (Rimadyl) diluted with saline and 2) time under a warming lamp until each mouse was freely mobile (12). Additional postoperative care consisted of visual monitoring once at least every 24 h by the investigative team and by the university vivarium staff. In addition, 12.5-mg carprofen (Rimadyl) tablets were administered ad libitum for pain management for 72 h (12). Sensorimotor function. Balance was assessed by measuring the ability of the mice to cross an inclined (15-), narrow, round, wooden beam, 1 m long with a 19-mm diameter, that was elevated above a bench surface and connected with an enclosed box (20 cm2) for the mouse to escape into (3,12,22). All mice were trained to help ensure consistent performance. Training was complete when a mouse traversed the beam in less than 20 s for three consecutive attempts (3,12). During test trials, mice were allowed up to 60 s to cross the beam, with the duration to
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
867
SPECIAL COMMUNICATIONS
treatment plans, and patient adherence makes prospective investigations very difficult to conduct. As a result, a great deal of evidence related to contributing factors toward CAI is retrospective. However, using genetically identical mice that experience the same injury and are treated in the same manner could allow researchers the opportunity to gather preclinical data regarding how certain interventions can limit long-term functional loss and CAI development, potentially furthering our understanding of the mechanisms of CAI and/ or helping us identify effective therapeutic interventions for the prevention of CAI. Such an animal model could also support and/or further develop contemporary theoretical models associated with CAI. Although multiple models exist and overlap, some have hypothesized that the development of CAI is due to a cascade of events that stem from the initial ankle sprain consequences (7,30,31). Furthermore, an animal model could be used to address Hertel’s (7) call to focus on potential points of intervention to limit sensorimotor deficits (i.e., organismic constraints) via testing a range of outcomes at multiple time points after injury while also exploring the initial physiologic site of pathology. An animal model could test the hypothesis of Wikstrom et al. (31), using the dynamic systems theory as a framework, that limiting organismic constraints (i.e., residual symptoms) through effective initial treatments may inhibit the development of CAI more easily than a human model. Previously, Hubbard-Turner et al. (12) developed an acute ankle sprain mouse model. Transecting the calcaneofibular ligament (CFL) in the hind foot of a mouse seemed representative of a mild ankle sprain, as evidenced by the resolution of symptoms (i.e., altered balance, altered gait, and decreased physical activity) around 7 d after injury (12). Similarly, transecting the CFL and anterior talofibular ligament (ATFL) seemed representative of a moderate-to-severe ankle sprain on the basis of the presence of symptoms for up to 4 wk after injury (12). However, a pressing need to determine whether mice with a surgically induced ankle sprain will develop similar long-term consequences as humans diagnosed with CAI still exists. Therefore, the purpose of this investigation was to examine the long-term effects of surgically transecting the lateral ligaments of a mouse_s hind limb on physical activity and sensorimotor function. On the basis of the human CAI literature and the work by Hubbard-Turner et al. (12), we hypothesize that a surgically induced ankle sprain will result in development of CAI-like symptoms that mimic the presentation of CAI in humans. For the purposes of this investigation, CAI was operationally defined as the presence of balance impairments 1 yr after the injurious event. The time point of 12 months, although not directly translatable between humans and mice, was chosen for the following three reasons: 1) after 12 months, the risk of a recurrent lateral ankle sprain is similar to the risk of a first-time lateral ankle sprain in humans (28), 2) 12 months has been recommended as the time needed to satisfactorily determine whether an individual will become a coper (i.e., a person that has sprained their ankle but will not develop CAI in humans (30), and 3)
SPECIAL COMMUNICATIONS
cross the beam and the number of times the right hind foot slipped off the beam recorded as dependent variables (3,12). Each mouse completed two test trials per test session (baseline (presurgery) and 4, 48, 54, and 60 wk after surgery), with the two trial’s mean being used for further analysis. The primary balance assessor (EAW) conducted interrater reliability analysis with three separate individuals during the investigation. Two of these individuals where blinded to group assignment and had never participated in the previously described balance tests. The results of these analyses demonstrate good-to-excellent interrater reliability, with intraclass correlation coefficient values of 0.83, 0.92, and 0.84, respectively. Gait was assessed using the footprint test (3,12). The hind feet and forefeet of each mouse were painted with nontoxic red and green paints, respectively, before allowing them to traverse a 50-cm-long and 10-cm-wide runway (with 10-cmhigh walls) once during each test session (3,12). To paint the feet, mice were scruffed with their tail secured between the ring and pinky fingers of the same hand. This technique exposed all four paws for painting, which was typically completed with a single stroke using a narrow foam brush. Mice were then immediately placed in the runway and typically completed the footprint test within 10–15 s after their feet were painted. The dependent measures including stride length asymmetry, paw overlap asymmetry, hind foot base width, and forefoot base width were obtained as previously described (3,12). In brief, average stride length (cm) was measured between each stride using a heel-to-heel measuring technique before an asymmetry ratio was computed as right stride length divided by left stride length (12). Thus, a value 91 indicates a larger right stride length and a value G1 indicates a larger left stride length. Paw overlap was the average distance (cm) between the center of the left and right forefoot and hind foot prints (3,12). Asymmetry in average paw overlap was determined as the right paw overlap divided by the left paw overlap averages. Thus, a value 91 indicates less of a right forefoot/hind foot overlap and a value G1 indicates a larger left hind foot/forefoot overlap (12). The forefoot and hind foot width measures were defined as the average distance between hind foot and forefoot prints, respectively (3,12). Width outcomes (cm) were measured as the perpendicular distance from the inside of the paws among the left and right steps of each mouse. For each gait outcome, the maximum number of values was obtained from each test trial captured at each test session (baseline (presurgery) and 4, 48, 54, and 60 wk after surgery) while excluding the footprints made while the mouse was initiating and terminating gait. The mean value of each set of outcomes was used for statistical analysis. Intra- and interrater reliability of calculating these gait outcomes have been established as excellent (12). Data from 4 wk after injury were included in both the balance and gait analyses to contextualize current means with those observed acutely after injury and reported by Hubbard-Turner et al. (12). Data from 48 and 60 wk after injury were included to provide
868
Official Journal of the American College of Sports Medicine
confirmation of the 54-wk data because of the exploratory nature of the investigation. Physical activity measurement. Each mouse was individually housed in a cage containing a solid surface running wheel (127 mm; Ware Manufacturing, Phoenix, AZ), magnetic sensor, and digital odometer (BC600; Sigma Sport, Olney, IL) that recorded the number of running wheel revolutions (12,15,26). Daily running wheel measurements of duration (min) and distance (km) were recorded starting 4 d after surgery. Knab et al. (15) reported a high correlation and agreement between days of wheel running and wheel running measurements indicating that voluntary wheel running activity is repeatable and stable in mice. The daily average during the fourth, 48th, 54th, and 60th week after injury was then submitted for statistical analysis. Statistical analysis. Three separate between-group, twoway, multivariate ANOVA (MANOVA) (group–time) were performed to compare changes in physical activity (duration, distance), balance (time, slips), and gait (stride asymmetry, paw overlap asymmetry, hind foot width, and forefoot width) due to the independent variables. A between-group analysis was performed to overcome missing data points (i.e., allow unequal group comparisons). The Wilk lambda was used to interpret statistical differences, and Bonferroni post hoc comparisons were performed, when appropriate, to determine the location of differences when main effects and/or an interaction were observed. An alpha level of P G 0.05 was used to determine significant differences for all analyses. Because not all humans develop CAI after an acute lateral ankle sprain (see Wikstrom and Brown (30) for further review on copers), a secondary analysis was performed to determine the percentage of the current mice that had developed CAI on the basis of the current operational definition (i.e., balance deficits as 12 months after injury). To do this, we first submitted any dependent measure that demonstrated a significant group–time interaction or group main effect, during a univariate ANOVA, to a receiver operating characteristic (ROC) curve analysis. This analysis determines whether an outcome can accurately distinguish SHAM mice from mice that had a sprained ankle (i.e., mice in the ATFL/ CFL and CFL groups). An ROC curve illustrates the ‘‘tradeoff’’ between sensitivity and specificity throughout a continuous outcome’s entire range of values (32). The ROC curve not only calculates overall discriminatory accuracy with the area under the curve (AUC) statistic and associated 95% confidence intervals (CI) but also provides the data needed to calculate cutoff scores for outcomes that reach asymptotic significance. Cutoff scores, which were used as the classification tool, were determined by calculating the Youden index (J ) for each outcome value along significant ROC curves, with the largest J value representing the cutoff score (32). The formula used to calculate the Youden index (J ) is J = (sensitivity + specificity) j 1. Finally, calculated cutoff scores were applied to the appropriate 54-wk data to determine the percentage of mice that had CAI (i.e., percentage of mice whose outcome exceeds the calculated cut score for that outcome).
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
applied to the individual mice, 70% (14/20) with an ankle sprain exceeded 4.75 hind foot slips and this percentage was equal among the CFL (70%) and ATFL/CFL (70%) groups.
A significant group–time interaction (F16,262 = 3.25, P G 0.01) and group (F4,262 = 11.336, P G 0.01) and time main effects (F8,262 = 15.43, P G 0.01) were observed for the balance MANOVA. Subsequent univariate ANOVA indicated that significant group–time interactions were present for both the time to cross the balance beam (P = 0.01) and the number of hind foot slips (P G 0.01) but only revealed group and time main effects (P G 0.01) for the number of hind foot slips. Group means, SD, and the location of significant differences for balance data can be seen in Table 1. For the gait data, a significant time main effect (F16,367 = 2.95, P G 0.01) was noted but a group–time interaction (F32,444 = 1.34, P = 0.10) and group main effect (F8,240 = 1.44, P = 0.18) were not significant. Subsequent univariate ANOVA revealed no significant differences for any of the gait outcomes across time, indicating an overall change in gait outcomes when pooled but no significant change in any one outcome. Group means and SD of gait data can be seen in Table 2. Significant group (F4,208 = 2.72, P = 0.03) and time (F6,208 = 8.43, P G 0.01) main effects were revealed in the physical activity MANOVA, but the group–time interaction (F12,208 = 1.18, P = 0.29) failed to reach significance. Subsequent univariate ANOVA indicated that group and time main effects were present for both distance run and duration run (P G 0.03). Group means, SD, and the location of significant differences for physical activity data can be seen in Table 3. Three outcomes, hind foot slips, duration run, and distance run from 54 wk after injury were submitted for an ROC curve analysis. Only hind foot slips reached an asymptotic significance level (P G 0.01), with an AUC statistic of 0.90 and a 95% CI of 0.77–1.00. Distance run and duration run failed to reach asymptotic significance levels (P = 0.32), with an AUC statistic of 0.62 and a 95% CI of 0.42– 0.82 for both outcomes. Therefore, a J score (Youden index) was calculated for all values within the 54-wk hind foot slip data set and a cut score of 4.75 was identified (J = 0.68). This point had a corresponding sensitivity of 0.68 and a corresponding specificity of 1.00 as seen in Figure 1. When
DISCUSSION This study is an extension of the work done by HubbardTurner et al. (12), which developed an acute lateral ankle sprain model in mice. More specifically, we quantified the long-term consequences of a single lateral ankle sprain to physical activity and sensorimotor function. To date, the longest prospective tracking of an objective outcome after lateral ankle sprain in humans has been 12 wk (17), well short of the 12-month time frame recommended to confidently determine whether an individual will not develop CAI (30). Our results illustrate consequences 12 months after injury in mice that are comparable with that in humans who have been diagnosed with CAI and support our a priori hypothesis. Humans with CAI have been shown to have static and dynamic balance deficits across a range of postural control tasks and outcome measures (2). Furthermore, in the absence of patient-reported outcomes, dynamic balance was the best predictor of CAI status in a study by Wikstrom et al. (32). Our current results are consistent with the human literature and are comparable with those previously reported in mice (19). For example, the time needed to cross the beam (approximately 5–10 s) and the number of hind foot slips (0–5) observed for healthy control mice by Carter et al. (3) are comparable, albeit slightly lower in some cases than the values observed in the current investigation (Table 1). Long-term gait alterations identified in humans with CAI have been limited to kinematic variables obtained using motion analysis (7), but alterations in general spatiotemporal alterations (e.g., stride length or gait velocity) have yet to be reported. This lack of empirical data coincides with the clinical observations associated with CAI in humans, specifically no obvious limp and an ability to continue to participate in high-level activities, albeit with recurrent ankle sprains and/or giving-way episodes. However, future research using an animal model of CAI should attempt to capture the kinematics of gait to further validate that the
TABLE 1. Balance outcome group means, SD, and sample size per group at each test session. Baseline Time (s)
Slips
SHAM CFL ATFL/CFL Time means SHAM CFL ATFL/CFL Time means
4 wk after Injury
6.14 T 1.87* (n = 10) 12.31 T 5.10 (n = 10) 11.72 T 6.20 (n = 10) 11.12 T 6.05 (n = 10) 9.26 T 2.06 (n = 10) 9.68 T 3.56 (n = 10) 9.04 T 4.43 11.04 T 4.95 , , 1.30 T 1.18* ** *** 1.25 T 1.08*,**,*** 1.50 T 0.88*,**,*** 1.00 T 1.29*,**,*** 1.15 T 1.37*,**,*** 2.40 T 1.41*,** 1.15 T 1.24 1.71 T 1.21
48 wk after Injury 10.26 8.76 10.96 9.98 3.05 6.15 5.75 5.05
T 4.91 (n = 9) T 4.06 (n = 10) T 3.87 (n = 10) T 4.23 T 1.58* T 2.62 T 1.68 T 2.39******
54 wk after Injury 10.47 T 4.52 (n = 9) 10.98 T 2.79 (n = 10) 11.89 T 3.64 (n = 10) 11.14 T 4.43 3.00 T 0.86* 5.30 T 2.00 5.85 T 1.78 4.77 T 2.01******
60 wk after Injury
Group Mean
9.14 T 4.52 (n = 9) 9.64 T 5.06 9.93 T 2.29 (n = 10) 10.50 T 4.51 15.63 T 5.05 (n = 10) 11.48 T 4.27 11.65 T 4.94 2.44 T 2.02*,** 2.17 T 1.56 3.60 T 2.28 3.51 T 2.75**** 6.60 T 2.07 4.35 T 2.71****,***** 4.27 T 2.72******
An animal may not have been tested at each test session because of death, health, or behavioral issues. *Indicates a significant difference from ATFL/CFL at 60 wk after injury (P G 0.05). **Indicates a significant difference from CFL at 48 wk after injury and ATFL/CFL at 54 wk after injury (P G 0.05). ***Indicates a significant difference from CFL at 54 wk after injury and ATFL/CFL at 48 wk after injury (P G 0.05). ****Indicates a significant difference from the SHAM group (P G 0.05). *****Indicates a significant difference from the CFL group (P G 0.05). ******Indicates a significant difference from baseline and 4 wk after injury (P G 0.05).
CHRONIC ANKLE INSTABILITY IN MICE
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
869
SPECIAL COMMUNICATIONS
RESULTS
SPECIAL COMMUNICATIONS
TABLE 2. Gait outcome group means, SD, and sample size per group at each test session. Stride asymmetry
Overlap asymmetry
Hind foot width (cm)
Forefoot width (cm)
SHAM CFL ATFL/CFL Time means SHAM CFL ATFL/CFL Time means SHAM CFL ATFL/CFL Time means SHAM CFL ATFL/CFL Time means
Baseline
4 wk after Injury
48 wk after Injury
54 wk after Injury
60 wk after Injury
0.99 T 0.03 (n = 10) 1.00 T 0.07 (n = 10) 1.04 T 0.07 (n = 10) 1.01 T 0.06 1.03 T 0.48 1.38 T 1.01 0.96 T 0.27 1.12 T 0.67 1.93 T 0.30 1.73 T 0.17 2.14 T 0.44 1.93 T 0.35 1.45 T 0.27 1.23 T 0.25 1.55 T 0.36 1.41 T 0.32
0.99 T 0.07 (n = 9) 1.03 T 0.07 (n = 9) 0.96 T 0.11 (n = 8) 1.00 T 0.09 1.27 T 0.56 1.34 T 0.71 0.98 T 0.66 1.20 T 0.64 2.22 T 0.41 2.19 T 0.29 2.17 T 0.44 2.19 T 0.37 1.35 T 0.29 1.47 T 0.22 1.47 T 0.17 1.43 T 0.26
1.03 T 0.07 (n = 8) 1.00 T 0.05 (n = 10) 1.05 T 0.14 (n = 8) 1.03 T 0.09 1.25 T 1.02 1.07 T 1.38 0.88 T 0.59 1.06 T 1.05 2.3 T 0.53 2.44 T 0.39 2.15 T 0.47 2.30 T 0.46 1.46 T 0.27 1.67 T 0.29 1.43 T 0.32 1.53 T 0.31
0.99 T 0.09 (n = 9) 0.92 T 0.11 (n = 10) 1.00 T 0.13 (n = 9) 0.97 T 0.11 1.43 T 0.97 0.92 T 0.45 1.41 T 1.59 1.24 T 1.07 2.05 T 0.41 2.32 T 0.35 2.09 T 0.42 2.16 T 0.40 1.09 T 0.44 1.43 T 0.34 1.35 T 0.19 1.30 T 0.36
1.03 T 0.15 (n = 9) 0.99 T 0.10 (n = 10) 0.94 T 0.16 (n = 9) 0.99 T 0.14 2.45 T 1.58 3.52 T 3.66 1.14 T 0.31 2.41 T 2.49 2.44 T 0.41 2.10 T 0.32 2.21 T 0.19 2.25 T 0.42 1.40 T 0.34 1.40 T 0.21 1.47 T 0.59 1.42 T 0.39
Group Mean 1.01 0.99 1.00 1.00 1.48 1.65 1.08 1.41 2.18 2.16 2.15 2.16 1.35 1.44 1.46 1.42
T 0.09 T 0.09 T 0.13 T 0.10 T 1.07 T 2.03 T 0.82 T 1.44 T 0.43 T 0.39 T 0.43 T 0.41 T 0.34 T 0.29 T 0.35 T 0.33
An animal may not have been tested at each test session because of death, health, or behavioral issues.
model presents in a manner similar to CAI in humans. The observed stride lengths (6–8 cm), paw overlap (0–0.5 cm), hind foot width (2–3 cm), and forefoot width (1–1.5 cm) for healthy control mice made by Carter et al. (3) are also comparable with those observed during the current study (Table 2) in similarly age mice. Although physical activity data remain relatively sparse in the human CAI literature, our results do coincide with available empirical data (11,29), providing further evidence that an acute ankle sprain in mice can lead to symptoms that mimic the presentation of CAI in humans. In addition, the physical activity levels of the current SHAM mice are lower than those observed by Lightfoot et al. (20), but the mice in the current investigation were about 3–4 wk older and did undergo a SHAM surgery. Although the agerelated difference seems consistent with the natural decline in physical activity in mice as they age, as reported by Turner et al. (26), it is important to note that different strains of mice were used and that the effect of the SHAM surgery of physical activity levels remains unknown at this time. Given the consistency of the chosen outcome measures among the assessments 48–60 wk after injury, the current data provide evidence that surgically transecting lateral ligaments of a mouse’s ankle can result in symptoms that mimic the presentation of CAI in humans and thus may be a valid animal model of CAI. Although the percentage of mice (70%) classified as having CAI is high, it is within the range of values reported in humans (1,23,24). Furthermore, the cutoff score of 4.75
hind foot slips is reasonable because it falls well within the range of scores observed during the balance assessment of mice with a surgically induced ankle sprain. Interestingly, both ankle sprain groups (CFL and ATFL/CFL) demonstrated a high but equal percentage of being classified as having CAI (70%). Although not an anticipated result, there is no empirical data suggesting a correlation between initial ankle sprain severity and the likelihood of developing CAI in humans (16). Thus, the equal rates of CAI development in both the mild and severe ankle sprain groups of the current investigation seem to support the hypothesis that the sensorimotor system’s ability to successfully compensate for structural damage is critical for the prevention of CAI regardless of the severity of the initial sprain induced (21,31). However, the current results would also suggest that the greater the magnitude of structural damage, the greater the difficulty an organism will have in physical activity over its lifespan. For example, only the ATFL/CFL group ran significantly less (distance and duration) relative to the SHAM group over the course of the entire investigation. Because the group differences in physical activity never fully resolved but were greatest shortly after the injury, further research is needed to determine whether surgical repair of severe acute ankle sprains may be an appropriate treatment option and/or whether more conservative management (e.g., cast immobilization) of severe acute ankle sprains may be more beneficial than the current standard of care (i.e., early mobilization). Unfortunately, more high-quality randomized
TABLE 3. Physical activity outcome group means, SD, and sample size per group at each test session. Duration (min)
Distance (km)
SHAM CFL ATFL/CFL Time means SHAM CFL ATFL/CFL Time means
4-wk Post Injury
48-wk Post Injury
54-wk Post Injury
60-wk Post Injury
207.00 T 129.26 (n = 10) 217.53 T 108.61 (n = 10) 141.33 T 112.69 (n = 10) 188.62 T 118.16 4.97 T 3.47 4.19 T 2.29 2.85 T 2.77 4.00 T 2.92
158.00 T 49.55 (n = 10) 156.74 T 50.24 (n = 10) 108.74 T 35.76 (n = 10) 140.58 T 49.77 2.20 T 0.79 2.43 T 1.00 1.55 T 0.74 2.05 T 0.91***
128.17 T 37.37 (n = 10) 140.71 T 62.11 (n = 10) 95.77 T 37.25 (n = 10) 121.32 T 49.77*** 1.82 T 0.63 2.15 T 1.09 1.32 T 0.63 1.76 T 0.86***
134.86 T 64.55 (n = 10) 121.31 T 46.04 (n = 10) 99.49 T 73.02 (n = 10) 117.99 T 61.67*** 1.86 T 1.04 1.76 T 0.73 1.50 T 1.03 1.70 T 0.92***
Group Mean 158.36 T 83.56 159.07 T 77.55 111.33 T 71.46*,** 2.77 T 2.31 2.63 T 1.65 1.81 T 1.62*
*Indicates a significant difference from the SHAM group (P G 0.05). **Indicates a significant difference from the CFL group (P G 0.05). ***Indicates a significant difference from 4 wk after injury (P G 0.05).
870
Official Journal of the American College of Sports Medicine
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
controlled trials are needed to determine whether surgical interventions are more effective than conservative interventions for acute lateral ankle sprains in humans (13). However, within existing conservative treatments of humans, longer duration and more stringent forms of immobilization seem to be advantageous for severe ankle sprains (19). Given the high percentage of mice that were classified as having CAI, on the basis of the current operational definition, both the CFL and ATLF/CFL mice models seem to be ideal candidates to use when examining the efficacy of various therapeutic interventions delivered acutely or subacutely, as called for by Hertel (7), at limiting the introduction of organismic constraints (31) and preventing CAI development. On the contrary, these models do not seem ideal to investigate characteristics that differ between those that do and do not develop CAI (30). Despite this limitation, the model can be used in future research to capture the proposed cascade of events (7,30,31) through repeated testing of a breadth of outcomes that represent the full spectrum of structural and sensorimotor adaptations seen in humans. Although this model seems to fit within the framework of contemporary theoretical models of CAI, future research is also needed to explore the physiologic response to the
CONCLUSIONS The transection of the lateral ligaments of a mouse’s hind foot results in long-term balance impairments that exceed the age-related alterations observed in the SHAM group. Furthermore, a high percentage of CAI developed regardless of the number of lateral ligaments transected (CFL vs ATFL/ CFL). Transecting the ATFL/CFL also seems to affect a mouse’s ability to remain physically active over the subsequent 12 months, but further research is needed to confirm the current results. Given the cost and logistics of quality clinical trials designed at developing CAI prevention interventions, either of the current mouse models seems ideal to allow researchers the opportunity to gather preclinical data regarding how certain interventions can limit long-term functional loss and CAI development. This study was funded by the University of North Carolina at Charlotte Faculty Research Grants program. There are no conflicts of interest to report. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
REFERENCES 1. Anandacoomarasamy A, Barnsley L. Long term outcomes of inversion ankle injuries. Br J Sports Med. 2005;39(3):e14. 2. Arnold BL, De La Motte S, Linens S, Ross SE. Ankle instability is associated with balance impairments: a meta-analysis. Med Sci Sports Exerc. 2009;41(5):1048–62. 3. Carter RJ, Lione LA, Humbry T, et al. Characterization of progresive motor deficits in mice transgenic for the human Huntington’s disease mutation. J Neurosci. 1999;19(8):3248–57. 4. Delahunt E, Coughlin GF, Caulfield B, Nightingale EJ, Lin CC, Hiller CE. Inclusion criteria when investigating insuficiencies in
CHRONIC ANKLE INSTABILITY IN MICE
chronic ankle instability. Med Sci Sports Exerc. 2010;42(11): 2106–21. 5. Delahunt E, d’Helft C, Brama PA. Development of an animal model of chronic ankle instability. J Orthop Sports Phys Ther. 2013;43(3):A14. 6. Fernandez WG, Yard EE, Comstock RD. Epidemiology of lower extremity injuries among U.S. high school athletes. Acad Emerg Med. 2007;14(7):641–5. 7. Hertel J. Sensorimotor deficits with ankle sprains and chronic ankle instability. Clin Sports Med. 2008;27(3):353–70.
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
871
SPECIAL COMMUNICATIONS
FIGURE 1—Hind foot slips ROC curve indicating sensitivity and 1specificity trade-off. The diagnostic accuracy of hind foot slips (solid black line) across the range of recorded scores is shown relative to the reference line (dashed black line), which indicates that a test performed no better than random. The large diamond on the ROC curve represent the location of the cutoff score.
current model (e.g., RNA expression in various tissues, histology of tissue, etc.), the consistency of the model across different mouse strains, and how other injury and animal models compare with the existing results (5,14,18). Consistency across mouse strains is important because each strain has unique genetic characteristics that may affect the results. In this investigation, the CBA/J strain was selected despite the presence of a retinal degeneration allele because of the following: 1) it is a moderately to highly active mouse, which minimized potential floor effects with regard to physical activity after injury and 2) the strain is not predisposed to development of osteoarthritis. Although we acknowledge that retinal degeneration may have affected our results over time, the only difference between the groups was the presence and severity of a surgically induced ankle sprain. Although we do not believe that the use of the CBA/J strain compromised the results of this investigation, using different strains in future research will only strengthen the validity of the mouse ankle sprain model. With regard to other injury and animal models, several researchers have induced both a surgical and manual ankle sprain in rats but have only studied the immediate consequences of those models to date (14,18).
SPECIAL COMMUNICATIONS
8. Hiller CE, Kilbreath SL, Refshauge KM. Chronic ankle instability: evolution of the model. J Athl Train. 2011;46(2):133–41. 9. Hirose K, Murakami G, Minowa T, Kura H, Yamashita T. Lateral ligament injury of the ankle and associated articular cartilage degeneration in the talocrural joint: anatomic study using elderly cadavers. J Orthop Sci. 2004;9(1):37–43. 10. Hootman JM, Dick R, Agel J. Epidemiology of collegiate injuries for 15 sports: summary and recommendations for injury prevention initiatives. J Athl Train. 2007;42(2):311–9. 11. Hubbard-Turner T, Turner MJ. Physical activity levels in those with chronic ankle instability. J Orthop Sports Phys Ther. 2013; 43(3):A23. 12. Hubbard-Turner T, Wikstrom EA, Guderian S, Turner MJ. Acute ankle sprain in a mouse model. Med Sci Sports Exerc. 2013;45(8): 1623–8. 13. Kerkhoffs GM, Handoll HH, de Bie R, Rower BH, Struijs PA. Surgical versus conservative treatment for acute injuires of the lateral ligament complex of the ankle in adults. Cochrane Database Syst Rev. 2007;18(2):CD000380. 14. Kim HY, Wang J, Chung K, Chung JM. A surgical ankle sprain pain model in the rat: effects of morphine and idomethacin. Neurosci Lett 2008;442:161–4. 15. Knab A, Bowen RS, Moore-Harrison T, Hamilton AT, Turner MJ, Lightfoot JT. Repeatability of exercise behaviors in mice. Physiol Behav. 2009;98(4):433–40. 16. Konradsen L, Bech L, Ehrenbjerg M, Nickelsen T. Seven years follow-up after ankle inversion trauma. Scand J Med Sci Sports. 2002;12(3):129–35. 17. Konradsen L, Olesen S, Hansen H. Ankle sensorimotor control and eversion strength after acute ankle inversion injuries. Am J Sports Med. 1998;26:72–7. 18. Koo ST, Park YI, Lim KS, Chung K, Chung JM. Acupuncture analgesis in a new rat model of ankle sprain pain. Pain. 2002;99: 423–31. 19. Lamb SE, Marsh JL, Nakash R, Cooke MW. Mechanical supports for acute, severe ankle sprain: a pragmatic, multicentre, randomised controlled trial. Lancet. 2009;373:575–81. 20. Lightfoot JT, Leamy L, Pomp D, et al. Strain screen and haplotype association mapping of wheel running in inbred mouse strains. J Appl Physiol (1985). 2010;109(1):623–34.
872
Official Journal of the American College of Sports Medicine
21. McKeon PO, Hubbard-Turner TJ, Wikstrom EA. Consequences of ankle inversion trauma: a novel recognition and treatment paradigm. In: Zaslav KR, editor. An International Perspective on Topics in Sports Medicine and Sports Injury. 1st ed. Rijeka (Croatia): InTech; 2012. p. 457–80. 22. Perry TA, Torres EM, Czech C, Beyreuther K, Richards S, Dunnett SB. Cognitive and motor function in transgenic mice carrying excess copies of the 695 and 751 amino acid isoforms of the amyloid precursor protein gene. Alzheimers Res. 1995;1:5–14. 23. Peters JW, Trevino SG, Renstrom PA. Chronic lateral ankle instability. Foot Ankle. 1991;12(3):182–91. 24. Smith RW, Reischl SF. Treatment of ankle sprains in young athletes. Am J Sports Med. 1986;14(6):465–71. 25. Soboroff SH, Pappius EM, Komaroff AL. Benefits, risks, and costs of alternative approaches to the evaluation and treatment of severe ankle sprain. Clin Orthop Relat Res. 1984;183:160–8. 26. Turner MJ, Chavis MN, Turner TH. Enhanced diastolic filling performance with lifelong physical activity in aging mice. Med Sci Sports Exerc. 2013;45(10):1933–40. 27. Valderrabano V, Hintermann B, Horisberger M, Fung TS. Ligamentous posttraumatic ankle osteoarthritis. Am J Sports Med. 2006;34(4):612–20. 28. Verhagen E, Van der Beek A, Twisk J, Bouter L, Bahr R, Van Mechelen W. The effect of a proprioceptive balance board training program for the prevention of ankle sprains: a prospective controlled trial. Am J Sports Med. 2004;32(6):1385–93. 29. Verhagen RA, de Keizer G, Van Dijk CN. Long-term follow-up of inversion trauma of the ankle. Arch Orthop Trauma Surg. 1995;114: 92–6. 30. Wikstrom EA, Brown CN. Minimum reporting standards for copers in chronic ankle instability research. Sports Med. 2014;44(4): 251–68. 31. Wikstrom EA, Hubbard-Turner T, McKeon PO. Understanding and treating lateral ankle sprains and their consequences: a constraints-based approach. Sports Med. 2013;43(6):385–93. 32. Wikstrom EA, Tillman MD, Chmielewski TL, Cauraugh JH, Naugle KE, Borsa PA. Discriminating between copers and people with chronic ankle instability. J Athl Train. 2012;47(2):136–42. 33. Yeung MS, Chan KM, So CH, Yuan WY. An epidemiological survey on ankle sprain. Br J Sports Med. 1994;28(2):112–6.
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Copyright of Medicine & Science in Sports & Exercise is the property of Lippincott Williams & Wilkins and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
