Human Movement Science 39 (2015) 73–87
Contents lists available at ScienceDirect
Human Movement Science journal homepage: www.elsevier.com/locate/humov
Muscle strength and functional performance is markedly impaired at the recommended time point for sport return after anterior cruciate ligament reconstruction in recreational athletes Jesper Bie Larsen a,c,⇑, Jean Farup a, Martin Lind b, Ulrik Dalgas a a
Dep. Public Health, Section of Sport Science, Aarhus University, Dalgas Avenue 4, 8000 Aarhus C, Denmark Department of Orthopedics, Aarhus University Hospital, Tage Hansensgade 2, 8000 Aarhus C, Denmark c University College of Northern Denmark, Selma Lagerløfs Vej 2, 9220 Aalborg Øst, Denmark b
a r t i c l e
i n f o
Article history:
PsycINFO classification: 2220 2221 2330 3720 Keywords: Eccentric Isometric and concentric muscle strength Rate of force development Control group
a b s t r a c t Purpose: To examine potential deficits in muscle strength or functional capacity when comparing (1) an ACL reconstructed group to matched healthy controls, (2) the ACL reconstructed leg to the non-injured leg and (3) the non-injured leg to matched healthy controls, at the time-point of recommended sport return 9–12 months post-surgery. Methods: Sixteen patients (male-female ratio: 9:7) 9–12 months post ACL reconstruction and sixteen age and sex matched healthy controls were included. Outcome measures included maximal knee extensor (KE) and knee flexor (KF) dynamometry, including measurement of rate of force development, functional capacity (counter movement jump (CMJ) and single distance hop (SDH)) and the Lysholm score. Results: Compared to the control group, maximal KE and KF muscle strength were impaired in the ACL reconstructed leg by 27–39% and 16–35%, respectively (p < .001). Also, impairments of both CMJ (38%) and SDH (33%) were observed (p < .001). Rate of force development for KE were reduced in the ACL group compared to the control group (p < .001). Similarly, the KE and KF muscle strength, CMJ and SDH of the ACL reconstructed leg were impaired, when compared to the non-injured leg by 15–23%, 8–20%, 23% and 20%, respectively (p < .05).
⇑ Corresponding author at: Dep. Public Health, Section of Sport Science, Aarhus University, Dalgas Avenue 4, 8000 Aarhus C, Denmark. Tel.: +45 28 26 84 59. E-mail address: [email protected] (J.B. Larsen). http://dx.doi.org/10.1016/j.humov.2014.10.008 0167-9457/Ó 2014 Elsevier B.V. All rights reserved.
74
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
Conclusion: Muscle strength and functional capacity are markedly impaired in the ACL reconstructed leg of recreationally active people 9–12 months post-surgery when compared to healthy matched controls and to their non-injured leg. This suggests that objective criteria rather than ‘‘time-since-surgery’’ criteria should guide return to sport. Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction Anterior cruciate ligament (ACL) injures occurs frequently and the incidence of ACL reconstruction is 38 pr. 100.000 citizens in Denmark (Lind, Menhert, & Pedersen, 2009). Typically, ACL injuries occur during sports activities, such as soccer, handball or alpine skiing, and usually the injured person wishes to return to sports activities after the ACL reconstruction (Granan, Forssblad, Lind, & Engebretsen, 2009; Lind et al., 2009). There are no evidence based guidelines describing when to return to sport after ACL reconstruction, although a time-point 9–12 months post-surgery is usually recommended before returning to contact sports (Ardern, Webster, Taylor, & Feller, 2011; Gobbi & Francisco, 2006; Thomee et al., 2011). However, a narrow fixed timeframe may not allow for individual variation, nor give any quantitative evaluation establishing, whether the ACL reconstructed knee is ready to reenter sports activity. It has, therefore, been suggested, that muscle strength and functional capacity of the ACL reconstructed leg is ‘‘normal’’ and ready for sports return, when corresponding to 90% of the non-injured leg (Risberg, Holm, & Ekeland, 1995; Thomee et al., 2011). In the literature there is a vast amount of studies comparing muscle function and functional capacity of the ACL reconstructed leg to the non-injured leg at different time-points after ACL reconstruction. At 9–12 months after ACL reconstruction studies have shown knee extensor (KE) (Burks, Crim, Fink, Boylan, & Greis, 2005; de Jong, van Caspel, van Haeff, & Saris, 2007; Kobayashi et al., 2004; Palmieri-Smith, Thomas, & Wojtys, 2008; Yosmaoglu, Baltaci, Kaya, & Ozer, 2011) and knee flexor (KF) (Burks et al., 2005; de Jong et al., 2007; Kobayashi et al., 2004; Yosmaoglu et al., 2011) muscle strength deficits of 5–27% and +1–21%, respectively, when comparing the ACL reconstructed leg to the non-injured leg. Generally subjects with a Bone Patellar Tendon Bone (BPTB) graft exhibits a larger deficit in KE muscle strength, whereas subjects with a semitendinosus-gracilis (STG) reveals larger deficits in KF muscle strength (Thomee et al., 2011). Also, functional capacity, in terms of one-legged hop tests such as Counter Movement Jump (CMJ) and Single Hop for Distance (SHD), have shown deficits varying from 5-11%, when compared to the non-injured leg 9–12 months post ACL reconstruction (Augustsson, Thomee, & Karlsson, 2004; de Jong et al., 2007; Hopper et al., 2002; Yosmaoglu et al., 2011). It appears that some studies exhibit LSI scores within the acceptable range of >90% and others reveal larger deficits at the time of recommended return to sport. These results could be biased because the non-injured leg, which is used as reference, seems to be influenced by the change in activity level that occurs when injured. Several authors have reported that a unilateral ACL injury results in bilateral muscle strength deficits, making the non-injured leg sub-optimal as a ‘‘healthy’’ reference (Hiemstra, Webber, MacDonald, & Kriellaars, 2000, 2007; Krishnan & Williams, 2011; PalmieriSmith et al., 2008; Urbach, Nebelung, Becker, & Awiszus, 2001). Several studies exhibit somewhat larger deficit in the range of 16–27% when comparing the ACL reconstructed leg to a healthy control group (Hiemstra et al., 2000, 2007; Mattacola et al., 2002; Xergia, Pappas, Zampeli, Georgiou, & Georgoulis, 2013). Yet, none of the existing studies made their examination within the timeframe for recommended return to sports and therefore, it is unknown to what degree deficits exists regarding muscle strength and functional capacity at this timeframe. Hence, knowledge is warranted in order to evaluate if the timeframe 9–12 months is appropriate or to promote the need for objective measurements to evaluate readiness for return to sport. Another important aspect of muscle function is the rate of force development (RFD), which quantifies the ability to rapidly generate muscle force.
75
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
RFD is related to performance in sports that require the ability to generate muscle strength quickly and furthermore, hamstring RFD is important for protection against ACL injuries (Angelozzi et al., 2012; Zebis, Andersen, Ellingsgaard, & Aagaard, 2011). However, RFD has only been sparsely investigated after ACL reconstruction. Angelozzi and colleagues (Angelozzi et al., 2012) showed a KE RFD deficit in professional soccer players 6 months post-surgery compared to a routine evaluation performed before ACL injury. Twelve months post-surgery the RFD level of the ACL reconstructed leg was recovered. Although RFD remains an important factor in sports performance and an aspect in injury prevention, studies’ examining this area in recreational athletes is non-existing and therefore warranted. Thus, the purpose of the present study was to examine muscle strength, including RFD, and functional capacity when comparing an ACL reconstructed group to a matched healthy control group at the time of recommended sport return, 9–12 months post-surgery. A secondary purpose was to compare muscle strength and functional capacity between the ACL reconstructed leg and the non-injured leg. It was hypothesized that muscle strength and functional capacity would be reduced (LSI < 90%) in (1) the ACL group compared to the matched healthy control group, (2) the ACL reconstructed leg compared to the ACL patients non-injured leg and (3) the non-injured leg compared to the matched healthy control group. 2. Methods 2.1. Study design and approvals A cross-sectional study was conducted at Department of Public Health, Section of Sports Science, Aarhus University, Denmark. Data on muscle strength and functional capacity was evaluated from ACL reconstructed subjects 9–12 months post-surgery and from healthy matched controls. The study was approved by the local scientific ethics committee (Videnskabsetisk Komite, Region Midtjylland) and performed according to the Helsinki Declaration. Written informed consent was obtained from all participants. 2.2. Study population All patients undergoing ACL reconstruction between November 2010 and March 2011 at the Division of Sportstrauma, Orthopedic Department, Aarhus University Hospital, were invited by mail Table 1 Subjects characteristics.
a
Age (years) Weight (kg)a Height (cm)a Body Mass Indexa Male-female ratio, n (%) Dominant leg, right, n (%) Occupational physical activitya Spare-time physical activitya Surgical leg, right, n (%) Graft type, BPTB, n, (%) Graft type, STG, n, (%) Time from injury to ACL reconstruction (months)a Time from ACL reconstruction to test (months)a Isolated ACL reconstruction, n Partial meniscectomy, n Cartilage debridement, n
ACL (n = 16)
CON (n = 16)
p-Value
29.6 (9.7) 78.8 (15) 176 (8.8) 22.3 (3.9) 9:7 (56%/44%) 14 (88%) 1.8 (1.0) 2.7 (1.0) 9 (56%) 2 (13%) 14 (87%) 8.4 (6.3) 11.1 (0.8) 7 8 3
29.9 (9) 72.2 (11.1) 179 (8) 20.1 (2.3) 9:7 (56%/44%) 15 (94%) 2.3 (0.9) 3.4 (0.6)
.94 .17 .36 .06
BPTB: Bone Patella Tendon Bone graft, STG: Semitendinosus/gracilis graft. a Mean ± standard deviation. * p 6 .05.
.14 .03*
76
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
to participate in the study (Table 1). Patients were ACL reconstructed with a BPTB autograft or STG autograft and the surgeries were performed by six different surgeons. After surgery, standardized rehabilitation was initiated by the hospital physiotherapy department before patients were discharged on the same day. During rehabilitation, weight bearing was allowed and encouraged and initiation of range of motion exercises, as well as neuromuscular exercises, was introduced. After discharge, the patients continued their rehabilitation under supervision in their local community for 3 months, before continuing training unsupervised. Generally, rehabilitation included progressive strength training, neuromuscular exercises and balance training in accordance with current rehabilitation concepts (Adams, Logerstedt, Hunter-Giordano, Axe, & Snyder-Mackler, 2012) (Rehabilitation milestones are outlined in Appendix A). Included patients were recreational athletes between 18–50 years old and had to be 9–12 months post-surgery. Exclusion criteria were: (1) additional surgical procedures than ACL reconstruction except meniscal resections or cartilage debridement, (2) previous ACL reconstruction, (3) other musculoskeletal injuries that could interfere with the tests performed and (4) participation in other clinical trials. Details on the inclusion process can be seen in the flowchart depicted in Fig. 1. The healthy control subjects were sex- and age matched (within ±5 years) in a 1:1 ratio to the ACL reconstructed subjects, and were recruited among teammates, if possible and otherwise among friends, colleagues and relatives of the ACL group or of the authors. Testing was performed between November 2011 and April 2012.
2.3. Sample size calculation and primary outcome measure The recruited number of participants was based on a sample size calculation where the primary outcome was KE muscle strength. The sample size estimate revealed that 10 subjects per group would be sufficient. In the calculation an expected difference in mean eccentric KE muscle strength of 0.88 N m kg 1 (alpha level = 0.05, statistical power of 80%) was applied, based on data from Hiemstra et al. (2007). Hence, 58 patients were invited to participate due to an expected participation rate of approximately 20%.
Contacted: n=58
Unwillingly to parcipate: n=2 No response: n=34
Exclusions: n=6: Previous ACL injury: n=2 Unavailable during the tesng period n=2 Parcipang in other clinical trial: n=1 Mulligament injury: n=1
Willingly to parcipate: n=22
Inclusions: n=16
Fig. 1. Study flowchart.
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
77
2.4. Outcome measures The test battery consisted of 3 muscle strength tests for the KE and the KF, 2 hop tests and two questionnaires. The order of testing was (1) completion of questionnaires, (2) CMJ, (3) SHD, (4) isokinetic muscle strength and (5) maximal voluntary isometric contraction (MVIC). 2.4.1. Questionnaires A subjective measurement of knee performance was obtained through the Lysholm Score (Briggs, Steadman et al., 2009). The Lysholm score provides an overall score of 0–100 based on 8 domains: limp, locking, pain, stair climbing, support, instability, swelling and squatting. Previous studies (Ahlen & Liden, 2011; Briggs, Steadman, Hay, & Hines, 2009; Risberg, Holm, Steen, & Beynnon, 1999) have interpreted a score of 95–100 as ‘‘Excellent’’, 84–94 as ‘‘Good’’, 65–83 as ‘‘Fair’’ and 90% was deemed acceptable. 2.6. Statistical analysis Data were checked for normal distribution using Bland–Altman plots, scatterplot, histogram and QQ-plots. Data is presented as mean ± standard deviation (SD) unless otherwise stated. For comparisons of demographic data between the ACL group and the control group Student´s unpaired t-tests or the Mann Whitney test were applied. The study´s primary outcome measure was comparison of muscle strength, and functional capacity between the ACL reconstructed leg and the matched control group (ACLOP vs. CONLeg1). A mixed linear model was used for the comparisons between the ACL group and the control group. The mixed linear model used ‘‘subject’’ as random effect and ‘‘leg’’ (ACL reconstructed leg, non-injured leg or control leg) ‘‘side’’ (dominant or non-dominant leg), ‘‘weight’’, ‘‘sex’’ and ‘‘age’’ as fixed effects (with weight and age applied as continues covariates). Side, weight, age and sex were added to the model to account for possible confounding. A Mann Whitney test was used to compare the Lysholm Score between the ACL group and the control group. A Pearson’s correlation analysis was performed to evaluate possible linear relationships between muscle strength and functional capacity in the ACL group. Furthermore, a Spearman rank correlation analysis was performed to evaluate possible linear relationships between Lysholm score, muscle strength and functional capacity. The correlation analysis included Lysholm score, CMJ, SHD, concentric and eccentric KE and KF strength at 60°/s and isometric KE and KF strength. A correlation above 0.90 was interpreted as very high, 0.70–0.89 as high, 0.50–0.69 as moderate, 0.30–0.49 as low, and less than 0.29 as little, if any, relation (Gijbels et al., 2010). Stata software, version 11.2 (StataCorp) was used for all statistical analyses, with the significance level set at p 6 .05. 3. Results The ACL reconstructed group and the control group did not differ in terms of age (p = .94), weight (p = .17), height (p = .36) or occupational physical activity (p = .14). There was a difference between the groups concerning spare-time physical activity (p = .03) (Table 1). 3.1. Muscle strength The ACLOP group had significant lower muscle strength in all types of measurements compared to the control group (Fig. 2A and B). Differences in LSI varied between 16% and 39%, and were, therefore, not within the acceptable LSI limit (Table 2). Regarding the two ACL legs there were significant lower muscle strength in all types of muscle measurements in the ACL reconstructed leg. Differences in LSI varied between 8% and 23%, and were not within the acceptable LSI limit, except for the KF MVIC measurement (Table 2). For the comparison between ACLHealthy and the control group all types of muscle strength measurements were in favor of the control group, although not all differences were significant. Differences in LSI varied between 3% and 16%, and were not within the acceptable LSI limit, except for the KF MVIC measurement (Table 2). 3.2. Rate of force development Significant differences were observed when comparing ACLOP and CON for KE RFD at 30 ms, 50 ms, 100 ms and 200 ms and KF RFD at 200 ms (Fig. 3A and B). In all other RFD measurements ACLOP revealed lower scores, although not significant (Table 3).
80
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
Fig. 2. Comparison of the force velocity curves of the maximal (A) knee extensor muscle strength between groups and (B) knee flexor muscle strength between groups in the operated leg of an ACL reconstructed group (ACLOP) 9–12 months post-surgery to the leg of a of sex and age matched group of healthy subjects (CONLeg1). ⁄p 6 .05.
Table 2 Muscle strength and LSI in the operated and non-injured leg of ACL patients and in healthy matched controls. Measured in Newton-meter.
KE Conc. 60°/sa KE Conc 180°/sa KE Ecc. 60°/sa KE Ecc. 180°/sa KE MVICa KF Conc. 60°/sa KF Conc. 180°/sa KF Ecc. 60°/sa KF Ecc. 180°/sa KF MVICa
ACLOP (n = 16)
LSI difference ACLOP vs. CONa
ACLHealthy (n = 16)
LSI difference ACLOP vs. ACLHealthya
CON (n = 16, 32 legs)
LSI difference ACLHealthy vs. CONa
126.1 (48.6) 77.3 (28.7) 163.7 (56.7) 146.6 (53.7) 188.5 (65.1) 89.9 (30.8) 67.6 (23.5) 117.7 (34.1) 110.4 (36.0) 133.7 (31.9)
29 ± 26% 33 ± 21% 39 ± 18% 31 ± 25% 27 ± 25% 35 ± 21% 34 ± 20% 32 ± 16% 34 ± 18% 16 ± 27%
158.3 (52.5) 94.8 (37.3) 213.6 (66.9) 182.3 (63.3) 234.2 (72.2) 112.4 (31.2) 82.2 (25.7) 137.2 (32.7) 135.4 (38.3) 146.8 (33.6)
17 ± 25% 15 ± 26% 23 ± 15% 20 ± 14% 17 ± 19% 20 ± 21% 18 ± 17% 14 ± 15% 19 ± 20% 8 ± 13%
172.5 (54.1)* 106.9 (34.4)*,§ 248.8 (79.6)* 205.4 (74.2)* 249.0 (73.6)* 125.4 (33.2)*,§ 92.4 (25.9)*,§ 157.5 (41.4)*,§ 153.4 (39.3)*,§ 151.0 (43.5)*
13 ± 28% 13 ± 27% 16 ± 21% 13 ± 29% 12 ± 18% 12 ± 21% 12 ± 25% 14 ± 20% 15 ± 22% 3 ± 29%
a Mean ± standard deviation. KE: Knee extensor. KF: Knee flexor. Conc: Concentric. Ecc: Eccentric. MVIC: Maximal Voluntary Isometric Contraction. s: Seconds. * Significant difference between ACLOP vs. CON, p 6 .05. Significant difference between ACLOP vs. ACLHealthy, p 6 .05. § Significant difference between ACLHealthy vs. CON, p 6 .05.
Significant differences between the ACL legs were observed for KE RFD at 30 ms, 50 ms and 100 ms. For all other RFD measurements ACLOP elicited lower scores, although not significant (Table 3). Significant differences between ACLHealthy and CON were observed for KE RFD at 30 ms, 50 ms and 100 ms, as well as for KF RFD at 200 ms. For all other RFD measurements ACLHealthy had the lowest scores, although not significant (Table 3).
3.3. Functional capacity The ACLOP group performed significantly lower in the CMJ and SHD tests compared to the control group. Differences in LSI were 38% for CMJ and 33% for SHD, and were therefore not within the acceptable LSI limit (Table 4). The ACLOP leg showed significant lower scores in the CMJ and SHD tests compared to the ACLHealthy leg. Differences in LSI were 23% for CMJ and 20% for SHD, and therefore not within the acceptable LSI limit (Table 4).
81
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
Fig. 3. Comparison of the rate of force development for maximal (A) knee extensor muscle strength between groups and (B) knee flexor muscle strength between groups in the operated leg of an ACL reconstructed group (ACLOP) 9–12 months postsurgery to the leg of a of sex and age matched group of healthy subjects (CONLeg1). ⁄p 6 .05.
Table 3 Rate of force development in the operated and non-injured leg of ACL patients and in healthy matched controls. Measured in Newton-meter.
KE RFD 30 msa KE RFD 50 msa KE RFD 100 msa KE RFD 200 msa KF RFD 30 msa KF RFD 50 msa KF RFD 100 msa KF RFD 200 msa
ACLOP (n = 16)
ACLHealthy (n = 16)
CON (n = 16, 32 legs)
1174 (294) 1043 (332) 989 (358) 870 (322) 992 (218) 843 (248) 758 (273) 615 (199)
1485 (394) 1404 (502) 1377 (644) 1107 (490) 1041 (245) 892 (286) 811 (329) 673 (235)
1613 (416)*,§ 1588 (535)*,§ 1615 (688)*,§ 1277 (523)* 1138 (226) 1019 (277) 978 (357) 773 (279)*,§
a Mean ± standard deviation. KE: Knee extensor. KF: Knee flexor. Conc: Concentric. Ecc: Eccentric. RFD: Rate of Force Development. ms: Milliseconds. * Significant difference between ACLOP vs. CON, p 6 .005. Significant difference between ACLOP vs. ACLHealthy, p 6 .05. § Significant difference between ACLHealthy vs. CON, p 6 .05.
Table 4 Functional capacity and LSI in the reconstructed and non-injured leg of ACL patients and in healthy matched controls.
CMJ (cm)a SHD (cm)a
ACLOP (n = 16)
LSI difference ACLOP vs. CONa
ACLHealthy (n = 16)
LSI difference ACLOP vs. ACLHealthya
CON (n = 16, 32 legs)
LSI difference ACLHealthy vs. CONa
9.4 (5.9) 78.7 (40.0)
38 ± 32% 33 ± 29%
11.9 (5.9) 96.6 (36.9)
23 ± 19% 20 ± 22%
15.7 (6.6)*,§ 120.8 (42.9)*,§
25 ± 23% 18 ± 22%
a
Mean ± standard deviation. CMJ: Counter Movement Jump. SHD: Single Hop for Distance. Significant difference between ACLOP vs. CON, p 6 .0001. Significant difference between ACLOP vs. ACLHealthy, p 6 .0001. Significant difference between ACLHealthy vs. CON, p 6 .001.
* §
The ACLHealthy leg showed significant lower scores in the CMJ and SHD tests compared to the control group. Differences in LSI were 25% for CMJ and 18% for SHD, and thus not within the acceptable LSI limit (Table 4).
82
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
3.4. Questionnaires As shown in Table 5 a significant difference was observed in the Lysholm Score when ACLOP was compared to the control group. Concerning return to sports, 2(13%) ACL reconstructed participants had returned fully to sport, 2(13%) had returned partly to sport and 12(74%) had not returned to sport. 3.5. Correlations Results from the Spearman rank correlation and Pearson correlation analysis can be seen in Table 6. Significant moderate to high correlations between all measures of muscle strength and both CMJ and SHD were found ranging from r = .53–.80 and r = .63–.75, respectively. Furthermore, high correlations between the Lysholm score and both CMJ (r = .78) and SHD (r = .68) were observed. For the Lysholm score and muscle strength significant high to moderate correlations between all variables, except KF MVIC, were found ranging from r = .51–.78. 4. Discussion To our knowledge this is the first study to compare muscle strength and functional capacity of ACL reconstructed patients 9–12 months post-surgery to a healthy age and sex matched control group. The present study observed significant deficits of maximal muscle strength, RFD and functional capacity between the ACL reconstructed group and the control group. This is of interest since 9–12 months post-surgery is typically the time-point recommended for sport return after ACL reconstruction for recreational athletes. Furthermore, significant differences between maximal muscle strength, RFD and functional capacity of the ACL reconstructed leg and the non-injured leg of the ACL patients, were found. The overall pattern is that the ACL reconstructed leg is weaker than the healthy leg, which itself is weaker than that seen in matched healthy controls. Consequently, one should be cautious when evaluating readiness for sport return by comparing the ACL reconstructed leg to the healthy leg. This recommendation has been pointed out in previous studies (Hiemstra et al., 2000, 2007; Krishnan & Williams, 2011; Palmieri-Smith et al., 2008; Urbach et al., 2001) and is further confirmed by the present study. The present study expands previous research by assessing muscle strength at the recommended time-point for sport return and by including measurements of all types of muscle contraction as well as RFD. Our results are consistent with other studies comparing an ACL reconstructed group to a control group, exhibiting a deficit markedly higher than within the accepted LSI (>90%) (Hiemstra et al., 2000, 2007; Mattacola et al., 2002). The long lasting muscle strength deficit after ACL reconstruction may be due to inadequate rehabilitation, but other factors, such as graft type, may also affect longterm recovery (Ageberg, Roos, Silbernagel, Thomee, & Roos, 2009; Hiemstra et al., 2000, 2007; Hopper et al., 2002; Landes, Nyland, Elmlinger, Tillett, & Caborn, 2010; Moisala, Jarvela, Kannus, & Jarvinen, 2007). Impaired quadriceps muscle strength has been shown to correlate with a poor outcome after ACL reconstruction (Wilk, Romaniello, Soscia, Arrigo, & Andrews, 1994). This further
Table 5 Lysholm score for ACL patients and matched healthy controls.
Mean Lysholm Score (arbitrary unit)a Median (arbitrary unit) Range ’’Excellent’’, n (%) ’’Good’’, n (%) ’’Fair’’, n (%) ’’Poor’’, n (%) a *
Standard Deviation p 6 0.0001
ACL (n = 16)
CON (n = 16)
77 (21) 85 22-100 3 (19%) 7 (43%) 3 (19%) 3 (19%)
97* (4) 97.5 89-100 13 (81%) 3 (19%)
83
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
Table 6 Correlations between muscle strength, functional capacity and Lysholm score in the operated leg of ACL patients.
Lysholm CMJ SHD KE Conc. 60°/s KE Ecc. 60°/s KE MVIC KF Conc. 60°/s KF Ecc. 60°/s KF MVIC
Lysholma
CMJb
SHDb
1.00 0.78* 0.68* 0.67* 0.78* 0.69* 0.77* 0.53* 0.51
1.00 0.83* 0.58* 0.57* 0.67* 0.80* 0.53* 0.56*
1.00 0.63* 0.64* 0.75* 0.71* 0.68* 0.67*
KE konc. 60°/sb
KE Ecc. 60°/sb
KE MVICb
KF Conc. 60°/sb
KF Ecc. 60°/sb
1.00 0.88* 0.83* 0.68* 0.74* 0.63*
1.00 0.92* 0.59* 0.77* 0.61*
1.00 0.53* 0.63* 0.58*
1.00 0.72* 0.74*
1.00 0.80*
KFMVICb
1.00
*
p 6 .05. a Spearman Rank Correlation. b Pearsons Correlation CMJ: Counter Movement Jump. SHD: Single Hop for Distance. KE: Knee extensor. KF: Knee flexor. Conc: Concentric. Ecc: Eccentric. MVIC: Maximal Voluntary Isometric Contraction. s: Seconds.
highlights the need for adequate rehabilitation, especially before commencing sports participation in order to minimize risk for re-injury. Yet, there is no consensus to what extent deficits remains acceptable, as different authors have suggested various deficit rates between 10% and 35% (Kyritsis & Witvrouw, 2014). Schmidt et al. found that deficits of 15% or larger was negatively correlated to function and performance after ACL reconstruction, which could be used as an argument for not accepting deficits larger than 10% and 15% (Schmitt, Paterno, & Hewett, 2012). The reported ranges in deficits usually restricts to comparison between the ACL reconstructed leg and the non-injured leg. The present study has shown that there is a discrepancy, when evaluating muscle strength in the ACL reconstructed leg compared to either the non-injured leg or a control group. It appears that deficits are underestimated when using the non-injured leg as reference. Our results shows that, not only are deficits larger when comparing ACL reconstructed patients to a control group instead of against the non-injured leg, but deficits persist when comparing the non-injured leg to the control group. This highly implies that the non-injured leg is significantly affected by the post-surgery deconditioning phase, causing a biased comparison, when using the non-injured leg as reference for return to sport. Typically the non-injured leg is being used as reference for practical reasons and because there, to the authors knowledge, does not exist normative data for different sports, which could be used as reference values. As a consequence, the authors suggest that marked deficits between the ACL reconstructed leg and the non-injured leg should not be accepted before return to sport. This would minimize the deficits compared to a control group. RFD is only sparsely investigated in patients undergoing ACL reconstruction. The present study demonstrated that RFD of both KE and KF were impaired when compared to matched healthy controls. RFD has been shown to be related to performance in many sports disciplines in which fast movements and muscle contractions are essential (Aagaard, Simonsen, Andersen, Magnusson, & Dyhre-Poulsen, 2002). Angelozzi and colleagues (Angelozzi et al., 2012) targeted a population of professional soccer players who had received an ACL reconstruction. At 6 months post-surgery they observed significant deficits in the subjects KE RFD. After an additional 20 week training period specifically aiming at improving RFD the study found no significant deficits in RFD, when reassessed 12 months postsurgery. The present study observed significant deficits in KE RFD at 30, 50, 100 and 200 ms and in KF RFD at 200 ms 9–12 months post-surgery. The diverging results are probably explained by different study populations (professional athletes vs. recreational athletes) as well as different post-surgery rehabilitation programs (specific RFD program vs. standard rehabilitation program) applied in the two studies. The majority of patients in the present study were reconstructed using a STG autograft and therefore, it is somewhat surprisingly that the KE, and not the KF, exhibits the most significant deficits in RFD as subjects generally exhibits larger deficits from the graft donor site (Thomee et al.,
84
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
2011). Furthermore, when evaluating the results concerning concentric, eccentric and MVIC between the ACL reconstructed legs and the control group, deficits seems to be of the same magnitude (KE: 27– 39% vs. KF: 16–35%) and all differences were significant, but when evaluating RFD, primarily KE exhibits significant differences. Therefore, it appears that an ACL reconstruction has a marked impact on the KE regarding the ability to rapidly generate muscle force. The findings suggest that ACL rehabilitation programs in recreationally active athletes should also target RFD, especially the KE, to normalize muscle function. Further research on this area is warranted. The present study showed impairments of the functional capacity during both CMJ and SHD in the ACL reconstructed leg when compared to the matched control group. Furthermore, findings revealed a large deficit in functional capacity for the comparison between the non-injured leg and the control group (CMJ: 25% & SHD: 18%). Once again, these results highlight that the non-injured leg is far from optimal as a reference. Outcomes from functional testing can be influenced by the methods used during tests. The present study did not allow any arm involvement to improve the balance during the landing phase in the tests, which is consistent with other studies (Ageberg et al., 2008; Gustavsson et al., 2006; Hopper et al., 2002; Keays, Bullock-Saxton, Newcombe, & Keays, 2003). Yet, no consensus exists since other studies allow involvement of arms during functional tests (de Jong et al., 2007; Mattacola et al., 2002; Yosmaoglu et al., 2011). Therefore, diverging results between studies could be explained by different measurement methods. The current findings underline that the time-since-surgery is less than ideal as a measure for return to sport. Our results show that major deficits still occur regarding muscle strength, RFD and functional capacity at the timeframe 9–12 months postoperatively. Furthermore, only 4 participants (25%) had returned to sport to some extent. It is unknown whether the lack of return to sport in this population is because of physiological, psychological or other reasons. Evaluating, when ready to return to sport, should be made on the background of objective measures, such as LSI scores, concerning muscle strength, RFD and functional capacity, although, no objective guidelines that permits for safe return to sports currently exist. A LSI within acceptable limit does not give any insurance for safe return to sport, but still allows objective measures to be taken into consideration, contrary to the timesince-surgery, which gives no information to what extent muscle strength and functional capacity has recovered. Positive correlations were found between functional capacity and different measures of muscle strength, suggesting that functional capacity can be improved by improving muscle strength, in agreement with previous studies (Petschnig, Baron, & Albrecht, 1998; Wilk et al., 1994). This study has shown that the non-injured leg is less than ideal to act as a control when evaluating muscle strength and functional capacity in the ACL reconstructed leg due to underestimation of deficits. This issue has previously been highlighted by other studies (Hiemstra et al., 2000; Krishnan & Williams, 2011; Palmieri-Smith et al., 2008), but Lautamies, Harilainen, Kettunen, Sandelin, and Kujala, (2008) argues that the non-injured leg is useful as a control when evaluating non-athletes (participating in exercise or sport
Contents lists available at ScienceDirect
Human Movement Science journal homepage: www.elsevier.com/locate/humov
Muscle strength and functional performance is markedly impaired at the recommended time point for sport return after anterior cruciate ligament reconstruction in recreational athletes Jesper Bie Larsen a,c,⇑, Jean Farup a, Martin Lind b, Ulrik Dalgas a a
Dep. Public Health, Section of Sport Science, Aarhus University, Dalgas Avenue 4, 8000 Aarhus C, Denmark Department of Orthopedics, Aarhus University Hospital, Tage Hansensgade 2, 8000 Aarhus C, Denmark c University College of Northern Denmark, Selma Lagerløfs Vej 2, 9220 Aalborg Øst, Denmark b
a r t i c l e
i n f o
Article history:
PsycINFO classification: 2220 2221 2330 3720 Keywords: Eccentric Isometric and concentric muscle strength Rate of force development Control group
a b s t r a c t Purpose: To examine potential deficits in muscle strength or functional capacity when comparing (1) an ACL reconstructed group to matched healthy controls, (2) the ACL reconstructed leg to the non-injured leg and (3) the non-injured leg to matched healthy controls, at the time-point of recommended sport return 9–12 months post-surgery. Methods: Sixteen patients (male-female ratio: 9:7) 9–12 months post ACL reconstruction and sixteen age and sex matched healthy controls were included. Outcome measures included maximal knee extensor (KE) and knee flexor (KF) dynamometry, including measurement of rate of force development, functional capacity (counter movement jump (CMJ) and single distance hop (SDH)) and the Lysholm score. Results: Compared to the control group, maximal KE and KF muscle strength were impaired in the ACL reconstructed leg by 27–39% and 16–35%, respectively (p < .001). Also, impairments of both CMJ (38%) and SDH (33%) were observed (p < .001). Rate of force development for KE were reduced in the ACL group compared to the control group (p < .001). Similarly, the KE and KF muscle strength, CMJ and SDH of the ACL reconstructed leg were impaired, when compared to the non-injured leg by 15–23%, 8–20%, 23% and 20%, respectively (p < .05).
⇑ Corresponding author at: Dep. Public Health, Section of Sport Science, Aarhus University, Dalgas Avenue 4, 8000 Aarhus C, Denmark. Tel.: +45 28 26 84 59. E-mail address: [email protected] (J.B. Larsen). http://dx.doi.org/10.1016/j.humov.2014.10.008 0167-9457/Ó 2014 Elsevier B.V. All rights reserved.
74
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
Conclusion: Muscle strength and functional capacity are markedly impaired in the ACL reconstructed leg of recreationally active people 9–12 months post-surgery when compared to healthy matched controls and to their non-injured leg. This suggests that objective criteria rather than ‘‘time-since-surgery’’ criteria should guide return to sport. Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction Anterior cruciate ligament (ACL) injures occurs frequently and the incidence of ACL reconstruction is 38 pr. 100.000 citizens in Denmark (Lind, Menhert, & Pedersen, 2009). Typically, ACL injuries occur during sports activities, such as soccer, handball or alpine skiing, and usually the injured person wishes to return to sports activities after the ACL reconstruction (Granan, Forssblad, Lind, & Engebretsen, 2009; Lind et al., 2009). There are no evidence based guidelines describing when to return to sport after ACL reconstruction, although a time-point 9–12 months post-surgery is usually recommended before returning to contact sports (Ardern, Webster, Taylor, & Feller, 2011; Gobbi & Francisco, 2006; Thomee et al., 2011). However, a narrow fixed timeframe may not allow for individual variation, nor give any quantitative evaluation establishing, whether the ACL reconstructed knee is ready to reenter sports activity. It has, therefore, been suggested, that muscle strength and functional capacity of the ACL reconstructed leg is ‘‘normal’’ and ready for sports return, when corresponding to 90% of the non-injured leg (Risberg, Holm, & Ekeland, 1995; Thomee et al., 2011). In the literature there is a vast amount of studies comparing muscle function and functional capacity of the ACL reconstructed leg to the non-injured leg at different time-points after ACL reconstruction. At 9–12 months after ACL reconstruction studies have shown knee extensor (KE) (Burks, Crim, Fink, Boylan, & Greis, 2005; de Jong, van Caspel, van Haeff, & Saris, 2007; Kobayashi et al., 2004; Palmieri-Smith, Thomas, & Wojtys, 2008; Yosmaoglu, Baltaci, Kaya, & Ozer, 2011) and knee flexor (KF) (Burks et al., 2005; de Jong et al., 2007; Kobayashi et al., 2004; Yosmaoglu et al., 2011) muscle strength deficits of 5–27% and +1–21%, respectively, when comparing the ACL reconstructed leg to the non-injured leg. Generally subjects with a Bone Patellar Tendon Bone (BPTB) graft exhibits a larger deficit in KE muscle strength, whereas subjects with a semitendinosus-gracilis (STG) reveals larger deficits in KF muscle strength (Thomee et al., 2011). Also, functional capacity, in terms of one-legged hop tests such as Counter Movement Jump (CMJ) and Single Hop for Distance (SHD), have shown deficits varying from 5-11%, when compared to the non-injured leg 9–12 months post ACL reconstruction (Augustsson, Thomee, & Karlsson, 2004; de Jong et al., 2007; Hopper et al., 2002; Yosmaoglu et al., 2011). It appears that some studies exhibit LSI scores within the acceptable range of >90% and others reveal larger deficits at the time of recommended return to sport. These results could be biased because the non-injured leg, which is used as reference, seems to be influenced by the change in activity level that occurs when injured. Several authors have reported that a unilateral ACL injury results in bilateral muscle strength deficits, making the non-injured leg sub-optimal as a ‘‘healthy’’ reference (Hiemstra, Webber, MacDonald, & Kriellaars, 2000, 2007; Krishnan & Williams, 2011; PalmieriSmith et al., 2008; Urbach, Nebelung, Becker, & Awiszus, 2001). Several studies exhibit somewhat larger deficit in the range of 16–27% when comparing the ACL reconstructed leg to a healthy control group (Hiemstra et al., 2000, 2007; Mattacola et al., 2002; Xergia, Pappas, Zampeli, Georgiou, & Georgoulis, 2013). Yet, none of the existing studies made their examination within the timeframe for recommended return to sports and therefore, it is unknown to what degree deficits exists regarding muscle strength and functional capacity at this timeframe. Hence, knowledge is warranted in order to evaluate if the timeframe 9–12 months is appropriate or to promote the need for objective measurements to evaluate readiness for return to sport. Another important aspect of muscle function is the rate of force development (RFD), which quantifies the ability to rapidly generate muscle force.
75
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
RFD is related to performance in sports that require the ability to generate muscle strength quickly and furthermore, hamstring RFD is important for protection against ACL injuries (Angelozzi et al., 2012; Zebis, Andersen, Ellingsgaard, & Aagaard, 2011). However, RFD has only been sparsely investigated after ACL reconstruction. Angelozzi and colleagues (Angelozzi et al., 2012) showed a KE RFD deficit in professional soccer players 6 months post-surgery compared to a routine evaluation performed before ACL injury. Twelve months post-surgery the RFD level of the ACL reconstructed leg was recovered. Although RFD remains an important factor in sports performance and an aspect in injury prevention, studies’ examining this area in recreational athletes is non-existing and therefore warranted. Thus, the purpose of the present study was to examine muscle strength, including RFD, and functional capacity when comparing an ACL reconstructed group to a matched healthy control group at the time of recommended sport return, 9–12 months post-surgery. A secondary purpose was to compare muscle strength and functional capacity between the ACL reconstructed leg and the non-injured leg. It was hypothesized that muscle strength and functional capacity would be reduced (LSI < 90%) in (1) the ACL group compared to the matched healthy control group, (2) the ACL reconstructed leg compared to the ACL patients non-injured leg and (3) the non-injured leg compared to the matched healthy control group. 2. Methods 2.1. Study design and approvals A cross-sectional study was conducted at Department of Public Health, Section of Sports Science, Aarhus University, Denmark. Data on muscle strength and functional capacity was evaluated from ACL reconstructed subjects 9–12 months post-surgery and from healthy matched controls. The study was approved by the local scientific ethics committee (Videnskabsetisk Komite, Region Midtjylland) and performed according to the Helsinki Declaration. Written informed consent was obtained from all participants. 2.2. Study population All patients undergoing ACL reconstruction between November 2010 and March 2011 at the Division of Sportstrauma, Orthopedic Department, Aarhus University Hospital, were invited by mail Table 1 Subjects characteristics.
a
Age (years) Weight (kg)a Height (cm)a Body Mass Indexa Male-female ratio, n (%) Dominant leg, right, n (%) Occupational physical activitya Spare-time physical activitya Surgical leg, right, n (%) Graft type, BPTB, n, (%) Graft type, STG, n, (%) Time from injury to ACL reconstruction (months)a Time from ACL reconstruction to test (months)a Isolated ACL reconstruction, n Partial meniscectomy, n Cartilage debridement, n
ACL (n = 16)
CON (n = 16)
p-Value
29.6 (9.7) 78.8 (15) 176 (8.8) 22.3 (3.9) 9:7 (56%/44%) 14 (88%) 1.8 (1.0) 2.7 (1.0) 9 (56%) 2 (13%) 14 (87%) 8.4 (6.3) 11.1 (0.8) 7 8 3
29.9 (9) 72.2 (11.1) 179 (8) 20.1 (2.3) 9:7 (56%/44%) 15 (94%) 2.3 (0.9) 3.4 (0.6)
.94 .17 .36 .06
BPTB: Bone Patella Tendon Bone graft, STG: Semitendinosus/gracilis graft. a Mean ± standard deviation. * p 6 .05.
.14 .03*
76
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
to participate in the study (Table 1). Patients were ACL reconstructed with a BPTB autograft or STG autograft and the surgeries were performed by six different surgeons. After surgery, standardized rehabilitation was initiated by the hospital physiotherapy department before patients were discharged on the same day. During rehabilitation, weight bearing was allowed and encouraged and initiation of range of motion exercises, as well as neuromuscular exercises, was introduced. After discharge, the patients continued their rehabilitation under supervision in their local community for 3 months, before continuing training unsupervised. Generally, rehabilitation included progressive strength training, neuromuscular exercises and balance training in accordance with current rehabilitation concepts (Adams, Logerstedt, Hunter-Giordano, Axe, & Snyder-Mackler, 2012) (Rehabilitation milestones are outlined in Appendix A). Included patients were recreational athletes between 18–50 years old and had to be 9–12 months post-surgery. Exclusion criteria were: (1) additional surgical procedures than ACL reconstruction except meniscal resections or cartilage debridement, (2) previous ACL reconstruction, (3) other musculoskeletal injuries that could interfere with the tests performed and (4) participation in other clinical trials. Details on the inclusion process can be seen in the flowchart depicted in Fig. 1. The healthy control subjects were sex- and age matched (within ±5 years) in a 1:1 ratio to the ACL reconstructed subjects, and were recruited among teammates, if possible and otherwise among friends, colleagues and relatives of the ACL group or of the authors. Testing was performed between November 2011 and April 2012.
2.3. Sample size calculation and primary outcome measure The recruited number of participants was based on a sample size calculation where the primary outcome was KE muscle strength. The sample size estimate revealed that 10 subjects per group would be sufficient. In the calculation an expected difference in mean eccentric KE muscle strength of 0.88 N m kg 1 (alpha level = 0.05, statistical power of 80%) was applied, based on data from Hiemstra et al. (2007). Hence, 58 patients were invited to participate due to an expected participation rate of approximately 20%.
Contacted: n=58
Unwillingly to parcipate: n=2 No response: n=34
Exclusions: n=6: Previous ACL injury: n=2 Unavailable during the tesng period n=2 Parcipang in other clinical trial: n=1 Mulligament injury: n=1
Willingly to parcipate: n=22
Inclusions: n=16
Fig. 1. Study flowchart.
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
77
2.4. Outcome measures The test battery consisted of 3 muscle strength tests for the KE and the KF, 2 hop tests and two questionnaires. The order of testing was (1) completion of questionnaires, (2) CMJ, (3) SHD, (4) isokinetic muscle strength and (5) maximal voluntary isometric contraction (MVIC). 2.4.1. Questionnaires A subjective measurement of knee performance was obtained through the Lysholm Score (Briggs, Steadman et al., 2009). The Lysholm score provides an overall score of 0–100 based on 8 domains: limp, locking, pain, stair climbing, support, instability, swelling and squatting. Previous studies (Ahlen & Liden, 2011; Briggs, Steadman, Hay, & Hines, 2009; Risberg, Holm, Steen, & Beynnon, 1999) have interpreted a score of 95–100 as ‘‘Excellent’’, 84–94 as ‘‘Good’’, 65–83 as ‘‘Fair’’ and 90% was deemed acceptable. 2.6. Statistical analysis Data were checked for normal distribution using Bland–Altman plots, scatterplot, histogram and QQ-plots. Data is presented as mean ± standard deviation (SD) unless otherwise stated. For comparisons of demographic data between the ACL group and the control group Student´s unpaired t-tests or the Mann Whitney test were applied. The study´s primary outcome measure was comparison of muscle strength, and functional capacity between the ACL reconstructed leg and the matched control group (ACLOP vs. CONLeg1). A mixed linear model was used for the comparisons between the ACL group and the control group. The mixed linear model used ‘‘subject’’ as random effect and ‘‘leg’’ (ACL reconstructed leg, non-injured leg or control leg) ‘‘side’’ (dominant or non-dominant leg), ‘‘weight’’, ‘‘sex’’ and ‘‘age’’ as fixed effects (with weight and age applied as continues covariates). Side, weight, age and sex were added to the model to account for possible confounding. A Mann Whitney test was used to compare the Lysholm Score between the ACL group and the control group. A Pearson’s correlation analysis was performed to evaluate possible linear relationships between muscle strength and functional capacity in the ACL group. Furthermore, a Spearman rank correlation analysis was performed to evaluate possible linear relationships between Lysholm score, muscle strength and functional capacity. The correlation analysis included Lysholm score, CMJ, SHD, concentric and eccentric KE and KF strength at 60°/s and isometric KE and KF strength. A correlation above 0.90 was interpreted as very high, 0.70–0.89 as high, 0.50–0.69 as moderate, 0.30–0.49 as low, and less than 0.29 as little, if any, relation (Gijbels et al., 2010). Stata software, version 11.2 (StataCorp) was used for all statistical analyses, with the significance level set at p 6 .05. 3. Results The ACL reconstructed group and the control group did not differ in terms of age (p = .94), weight (p = .17), height (p = .36) or occupational physical activity (p = .14). There was a difference between the groups concerning spare-time physical activity (p = .03) (Table 1). 3.1. Muscle strength The ACLOP group had significant lower muscle strength in all types of measurements compared to the control group (Fig. 2A and B). Differences in LSI varied between 16% and 39%, and were, therefore, not within the acceptable LSI limit (Table 2). Regarding the two ACL legs there were significant lower muscle strength in all types of muscle measurements in the ACL reconstructed leg. Differences in LSI varied between 8% and 23%, and were not within the acceptable LSI limit, except for the KF MVIC measurement (Table 2). For the comparison between ACLHealthy and the control group all types of muscle strength measurements were in favor of the control group, although not all differences were significant. Differences in LSI varied between 3% and 16%, and were not within the acceptable LSI limit, except for the KF MVIC measurement (Table 2). 3.2. Rate of force development Significant differences were observed when comparing ACLOP and CON for KE RFD at 30 ms, 50 ms, 100 ms and 200 ms and KF RFD at 200 ms (Fig. 3A and B). In all other RFD measurements ACLOP revealed lower scores, although not significant (Table 3).
80
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
Fig. 2. Comparison of the force velocity curves of the maximal (A) knee extensor muscle strength between groups and (B) knee flexor muscle strength between groups in the operated leg of an ACL reconstructed group (ACLOP) 9–12 months post-surgery to the leg of a of sex and age matched group of healthy subjects (CONLeg1). ⁄p 6 .05.
Table 2 Muscle strength and LSI in the operated and non-injured leg of ACL patients and in healthy matched controls. Measured in Newton-meter.
KE Conc. 60°/sa KE Conc 180°/sa KE Ecc. 60°/sa KE Ecc. 180°/sa KE MVICa KF Conc. 60°/sa KF Conc. 180°/sa KF Ecc. 60°/sa KF Ecc. 180°/sa KF MVICa
ACLOP (n = 16)
LSI difference ACLOP vs. CONa
ACLHealthy (n = 16)
LSI difference ACLOP vs. ACLHealthya
CON (n = 16, 32 legs)
LSI difference ACLHealthy vs. CONa
126.1 (48.6) 77.3 (28.7) 163.7 (56.7) 146.6 (53.7) 188.5 (65.1) 89.9 (30.8) 67.6 (23.5) 117.7 (34.1) 110.4 (36.0) 133.7 (31.9)
29 ± 26% 33 ± 21% 39 ± 18% 31 ± 25% 27 ± 25% 35 ± 21% 34 ± 20% 32 ± 16% 34 ± 18% 16 ± 27%
158.3 (52.5) 94.8 (37.3) 213.6 (66.9) 182.3 (63.3) 234.2 (72.2) 112.4 (31.2) 82.2 (25.7) 137.2 (32.7) 135.4 (38.3) 146.8 (33.6)
17 ± 25% 15 ± 26% 23 ± 15% 20 ± 14% 17 ± 19% 20 ± 21% 18 ± 17% 14 ± 15% 19 ± 20% 8 ± 13%
172.5 (54.1)* 106.9 (34.4)*,§ 248.8 (79.6)* 205.4 (74.2)* 249.0 (73.6)* 125.4 (33.2)*,§ 92.4 (25.9)*,§ 157.5 (41.4)*,§ 153.4 (39.3)*,§ 151.0 (43.5)*
13 ± 28% 13 ± 27% 16 ± 21% 13 ± 29% 12 ± 18% 12 ± 21% 12 ± 25% 14 ± 20% 15 ± 22% 3 ± 29%
a Mean ± standard deviation. KE: Knee extensor. KF: Knee flexor. Conc: Concentric. Ecc: Eccentric. MVIC: Maximal Voluntary Isometric Contraction. s: Seconds. * Significant difference between ACLOP vs. CON, p 6 .05. Significant difference between ACLOP vs. ACLHealthy, p 6 .05. § Significant difference between ACLHealthy vs. CON, p 6 .05.
Significant differences between the ACL legs were observed for KE RFD at 30 ms, 50 ms and 100 ms. For all other RFD measurements ACLOP elicited lower scores, although not significant (Table 3). Significant differences between ACLHealthy and CON were observed for KE RFD at 30 ms, 50 ms and 100 ms, as well as for KF RFD at 200 ms. For all other RFD measurements ACLHealthy had the lowest scores, although not significant (Table 3).
3.3. Functional capacity The ACLOP group performed significantly lower in the CMJ and SHD tests compared to the control group. Differences in LSI were 38% for CMJ and 33% for SHD, and were therefore not within the acceptable LSI limit (Table 4). The ACLOP leg showed significant lower scores in the CMJ and SHD tests compared to the ACLHealthy leg. Differences in LSI were 23% for CMJ and 20% for SHD, and therefore not within the acceptable LSI limit (Table 4).
81
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
Fig. 3. Comparison of the rate of force development for maximal (A) knee extensor muscle strength between groups and (B) knee flexor muscle strength between groups in the operated leg of an ACL reconstructed group (ACLOP) 9–12 months postsurgery to the leg of a of sex and age matched group of healthy subjects (CONLeg1). ⁄p 6 .05.
Table 3 Rate of force development in the operated and non-injured leg of ACL patients and in healthy matched controls. Measured in Newton-meter.
KE RFD 30 msa KE RFD 50 msa KE RFD 100 msa KE RFD 200 msa KF RFD 30 msa KF RFD 50 msa KF RFD 100 msa KF RFD 200 msa
ACLOP (n = 16)
ACLHealthy (n = 16)
CON (n = 16, 32 legs)
1174 (294) 1043 (332) 989 (358) 870 (322) 992 (218) 843 (248) 758 (273) 615 (199)
1485 (394) 1404 (502) 1377 (644) 1107 (490) 1041 (245) 892 (286) 811 (329) 673 (235)
1613 (416)*,§ 1588 (535)*,§ 1615 (688)*,§ 1277 (523)* 1138 (226) 1019 (277) 978 (357) 773 (279)*,§
a Mean ± standard deviation. KE: Knee extensor. KF: Knee flexor. Conc: Concentric. Ecc: Eccentric. RFD: Rate of Force Development. ms: Milliseconds. * Significant difference between ACLOP vs. CON, p 6 .005. Significant difference between ACLOP vs. ACLHealthy, p 6 .05. § Significant difference between ACLHealthy vs. CON, p 6 .05.
Table 4 Functional capacity and LSI in the reconstructed and non-injured leg of ACL patients and in healthy matched controls.
CMJ (cm)a SHD (cm)a
ACLOP (n = 16)
LSI difference ACLOP vs. CONa
ACLHealthy (n = 16)
LSI difference ACLOP vs. ACLHealthya
CON (n = 16, 32 legs)
LSI difference ACLHealthy vs. CONa
9.4 (5.9) 78.7 (40.0)
38 ± 32% 33 ± 29%
11.9 (5.9) 96.6 (36.9)
23 ± 19% 20 ± 22%
15.7 (6.6)*,§ 120.8 (42.9)*,§
25 ± 23% 18 ± 22%
a
Mean ± standard deviation. CMJ: Counter Movement Jump. SHD: Single Hop for Distance. Significant difference between ACLOP vs. CON, p 6 .0001. Significant difference between ACLOP vs. ACLHealthy, p 6 .0001. Significant difference between ACLHealthy vs. CON, p 6 .001.
* §
The ACLHealthy leg showed significant lower scores in the CMJ and SHD tests compared to the control group. Differences in LSI were 25% for CMJ and 18% for SHD, and thus not within the acceptable LSI limit (Table 4).
82
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
3.4. Questionnaires As shown in Table 5 a significant difference was observed in the Lysholm Score when ACLOP was compared to the control group. Concerning return to sports, 2(13%) ACL reconstructed participants had returned fully to sport, 2(13%) had returned partly to sport and 12(74%) had not returned to sport. 3.5. Correlations Results from the Spearman rank correlation and Pearson correlation analysis can be seen in Table 6. Significant moderate to high correlations between all measures of muscle strength and both CMJ and SHD were found ranging from r = .53–.80 and r = .63–.75, respectively. Furthermore, high correlations between the Lysholm score and both CMJ (r = .78) and SHD (r = .68) were observed. For the Lysholm score and muscle strength significant high to moderate correlations between all variables, except KF MVIC, were found ranging from r = .51–.78. 4. Discussion To our knowledge this is the first study to compare muscle strength and functional capacity of ACL reconstructed patients 9–12 months post-surgery to a healthy age and sex matched control group. The present study observed significant deficits of maximal muscle strength, RFD and functional capacity between the ACL reconstructed group and the control group. This is of interest since 9–12 months post-surgery is typically the time-point recommended for sport return after ACL reconstruction for recreational athletes. Furthermore, significant differences between maximal muscle strength, RFD and functional capacity of the ACL reconstructed leg and the non-injured leg of the ACL patients, were found. The overall pattern is that the ACL reconstructed leg is weaker than the healthy leg, which itself is weaker than that seen in matched healthy controls. Consequently, one should be cautious when evaluating readiness for sport return by comparing the ACL reconstructed leg to the healthy leg. This recommendation has been pointed out in previous studies (Hiemstra et al., 2000, 2007; Krishnan & Williams, 2011; Palmieri-Smith et al., 2008; Urbach et al., 2001) and is further confirmed by the present study. The present study expands previous research by assessing muscle strength at the recommended time-point for sport return and by including measurements of all types of muscle contraction as well as RFD. Our results are consistent with other studies comparing an ACL reconstructed group to a control group, exhibiting a deficit markedly higher than within the accepted LSI (>90%) (Hiemstra et al., 2000, 2007; Mattacola et al., 2002). The long lasting muscle strength deficit after ACL reconstruction may be due to inadequate rehabilitation, but other factors, such as graft type, may also affect longterm recovery (Ageberg, Roos, Silbernagel, Thomee, & Roos, 2009; Hiemstra et al., 2000, 2007; Hopper et al., 2002; Landes, Nyland, Elmlinger, Tillett, & Caborn, 2010; Moisala, Jarvela, Kannus, & Jarvinen, 2007). Impaired quadriceps muscle strength has been shown to correlate with a poor outcome after ACL reconstruction (Wilk, Romaniello, Soscia, Arrigo, & Andrews, 1994). This further
Table 5 Lysholm score for ACL patients and matched healthy controls.
Mean Lysholm Score (arbitrary unit)a Median (arbitrary unit) Range ’’Excellent’’, n (%) ’’Good’’, n (%) ’’Fair’’, n (%) ’’Poor’’, n (%) a *
Standard Deviation p 6 0.0001
ACL (n = 16)
CON (n = 16)
77 (21) 85 22-100 3 (19%) 7 (43%) 3 (19%) 3 (19%)
97* (4) 97.5 89-100 13 (81%) 3 (19%)
83
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
Table 6 Correlations between muscle strength, functional capacity and Lysholm score in the operated leg of ACL patients.
Lysholm CMJ SHD KE Conc. 60°/s KE Ecc. 60°/s KE MVIC KF Conc. 60°/s KF Ecc. 60°/s KF MVIC
Lysholma
CMJb
SHDb
1.00 0.78* 0.68* 0.67* 0.78* 0.69* 0.77* 0.53* 0.51
1.00 0.83* 0.58* 0.57* 0.67* 0.80* 0.53* 0.56*
1.00 0.63* 0.64* 0.75* 0.71* 0.68* 0.67*
KE konc. 60°/sb
KE Ecc. 60°/sb
KE MVICb
KF Conc. 60°/sb
KF Ecc. 60°/sb
1.00 0.88* 0.83* 0.68* 0.74* 0.63*
1.00 0.92* 0.59* 0.77* 0.61*
1.00 0.53* 0.63* 0.58*
1.00 0.72* 0.74*
1.00 0.80*
KFMVICb
1.00
*
p 6 .05. a Spearman Rank Correlation. b Pearsons Correlation CMJ: Counter Movement Jump. SHD: Single Hop for Distance. KE: Knee extensor. KF: Knee flexor. Conc: Concentric. Ecc: Eccentric. MVIC: Maximal Voluntary Isometric Contraction. s: Seconds.
highlights the need for adequate rehabilitation, especially before commencing sports participation in order to minimize risk for re-injury. Yet, there is no consensus to what extent deficits remains acceptable, as different authors have suggested various deficit rates between 10% and 35% (Kyritsis & Witvrouw, 2014). Schmidt et al. found that deficits of 15% or larger was negatively correlated to function and performance after ACL reconstruction, which could be used as an argument for not accepting deficits larger than 10% and 15% (Schmitt, Paterno, & Hewett, 2012). The reported ranges in deficits usually restricts to comparison between the ACL reconstructed leg and the non-injured leg. The present study has shown that there is a discrepancy, when evaluating muscle strength in the ACL reconstructed leg compared to either the non-injured leg or a control group. It appears that deficits are underestimated when using the non-injured leg as reference. Our results shows that, not only are deficits larger when comparing ACL reconstructed patients to a control group instead of against the non-injured leg, but deficits persist when comparing the non-injured leg to the control group. This highly implies that the non-injured leg is significantly affected by the post-surgery deconditioning phase, causing a biased comparison, when using the non-injured leg as reference for return to sport. Typically the non-injured leg is being used as reference for practical reasons and because there, to the authors knowledge, does not exist normative data for different sports, which could be used as reference values. As a consequence, the authors suggest that marked deficits between the ACL reconstructed leg and the non-injured leg should not be accepted before return to sport. This would minimize the deficits compared to a control group. RFD is only sparsely investigated in patients undergoing ACL reconstruction. The present study demonstrated that RFD of both KE and KF were impaired when compared to matched healthy controls. RFD has been shown to be related to performance in many sports disciplines in which fast movements and muscle contractions are essential (Aagaard, Simonsen, Andersen, Magnusson, & Dyhre-Poulsen, 2002). Angelozzi and colleagues (Angelozzi et al., 2012) targeted a population of professional soccer players who had received an ACL reconstruction. At 6 months post-surgery they observed significant deficits in the subjects KE RFD. After an additional 20 week training period specifically aiming at improving RFD the study found no significant deficits in RFD, when reassessed 12 months postsurgery. The present study observed significant deficits in KE RFD at 30, 50, 100 and 200 ms and in KF RFD at 200 ms 9–12 months post-surgery. The diverging results are probably explained by different study populations (professional athletes vs. recreational athletes) as well as different post-surgery rehabilitation programs (specific RFD program vs. standard rehabilitation program) applied in the two studies. The majority of patients in the present study were reconstructed using a STG autograft and therefore, it is somewhat surprisingly that the KE, and not the KF, exhibits the most significant deficits in RFD as subjects generally exhibits larger deficits from the graft donor site (Thomee et al.,
84
J.B. Larsen et al. / Human Movement Science 39 (2015) 73–87
2011). Furthermore, when evaluating the results concerning concentric, eccentric and MVIC between the ACL reconstructed legs and the control group, deficits seems to be of the same magnitude (KE: 27– 39% vs. KF: 16–35%) and all differences were significant, but when evaluating RFD, primarily KE exhibits significant differences. Therefore, it appears that an ACL reconstruction has a marked impact on the KE regarding the ability to rapidly generate muscle force. The findings suggest that ACL rehabilitation programs in recreationally active athletes should also target RFD, especially the KE, to normalize muscle function. Further research on this area is warranted. The present study showed impairments of the functional capacity during both CMJ and SHD in the ACL reconstructed leg when compared to the matched control group. Furthermore, findings revealed a large deficit in functional capacity for the comparison between the non-injured leg and the control group (CMJ: 25% & SHD: 18%). Once again, these results highlight that the non-injured leg is far from optimal as a reference. Outcomes from functional testing can be influenced by the methods used during tests. The present study did not allow any arm involvement to improve the balance during the landing phase in the tests, which is consistent with other studies (Ageberg et al., 2008; Gustavsson et al., 2006; Hopper et al., 2002; Keays, Bullock-Saxton, Newcombe, & Keays, 2003). Yet, no consensus exists since other studies allow involvement of arms during functional tests (de Jong et al., 2007; Mattacola et al., 2002; Yosmaoglu et al., 2011). Therefore, diverging results between studies could be explained by different measurement methods. The current findings underline that the time-since-surgery is less than ideal as a measure for return to sport. Our results show that major deficits still occur regarding muscle strength, RFD and functional capacity at the timeframe 9–12 months postoperatively. Furthermore, only 4 participants (25%) had returned to sport to some extent. It is unknown whether the lack of return to sport in this population is because of physiological, psychological or other reasons. Evaluating, when ready to return to sport, should be made on the background of objective measures, such as LSI scores, concerning muscle strength, RFD and functional capacity, although, no objective guidelines that permits for safe return to sports currently exist. A LSI within acceptable limit does not give any insurance for safe return to sport, but still allows objective measures to be taken into consideration, contrary to the timesince-surgery, which gives no information to what extent muscle strength and functional capacity has recovered. Positive correlations were found between functional capacity and different measures of muscle strength, suggesting that functional capacity can be improved by improving muscle strength, in agreement with previous studies (Petschnig, Baron, & Albrecht, 1998; Wilk et al., 1994). This study has shown that the non-injured leg is less than ideal to act as a control when evaluating muscle strength and functional capacity in the ACL reconstructed leg due to underestimation of deficits. This issue has previously been highlighted by other studies (Hiemstra et al., 2000; Krishnan & Williams, 2011; Palmieri-Smith et al., 2008), but Lautamies, Harilainen, Kettunen, Sandelin, and Kujala, (2008) argues that the non-injured leg is useful as a control when evaluating non-athletes (participating in exercise or sport
