Systematic Review
Does Gracilis Preservation Matter in Anterior Cruciate Ligament Reconstruction? A Systematic Review Avijit Sharma, B.S., David C. Flanigan, M.D., Kyle Randall, M.D., and Robert A. Magnussen, M.D., M.P.H.
Purpose: To analyze the effect of gracilis harvest on hamstring strength, patient-reported outcomes, and anterior knee laxity after anterior cruciate ligament (ACL) reconstruction. Methods: A systematic review of the literature was performed to identify studies comparing the results of semitendinosus (ST) versus semitendinosus-gracilis (ST-G) harvest for ACL reconstruction. A meta-analysis using a random effects model was performed to determine overall pooled estimates of effect for the influence of additional gracilis harvest on hamstring strength, patient-reported outcomes, and anterior knee laxity after ACL reconstruction. Results: Twelve studies were identified and included in the review. ST-G harvest was noted to decrease hamstring isokinetic strength at 60 per second by 3.85% relative to isolated ST harvest (P ¼ .01). Decreased isometric strength was also noted in the ST-G harvest group at both 90 of flexion (mean difference: 5.55%; P ¼ .03) and 105 to 110 of flexion (mean difference: 13.68%; P ¼ .003). Active knee flexion angle loss was also noted to be greater in the ST-G harvest group (mean difference: 3.91 ; P ¼ .006). No differences were found in isokinetic strength at 180 to 240 per second (mean difference: 3.20%; P ¼ .08), patient-reported outcome scores (mean difference: 1.87 points; P ¼ .06), or anterior knee laxity (mean difference: 0.03 mm; P ¼ .78) based on gracilis harvest. Conclusions: The addition of gracilis harvest to an isolated ST harvest for ACL reconstruction results in statistically significant, but likely not clinically relevant differences in isokinetic and isometric hamstring strength as well as patient-reported outcomes. Hamstring strength deficits may be larger at higher flexion angles. Level of Evidence: Level III, systematic review of level I-III studies.
A
nterior cruciate ligament (ACL) is commonly injured and reconstruction is often performed to allow return to active participation in cutting and pivoting sports. Although numerous potential grafts for ACL reconstruction have been described, hamstring autografts are among the most commonly used.1 Numerous hamstring preparation techniques have been described, the most common of which are a quadrupled semitendinosus tendon (ST) graft and a
From the Department of Orthopaedic Surgery, The Ohio State University (A.S., D.C.F., R.A.M.); Sports Health and Performance Institute, The Ohio State University (D.C.F., R.A.M.), Columbus; and Avita Health System (K.R.), Ontario, Ohio, U.S.A. The authors report the following potential conflict of interest or source of funding: D.C.F. receives support from Sanofi, Smith & Nephew. Received November 4, 2014; accepted November 5, 2015. Address correspondence to Robert A. Magnussen, M.D., M.P.H., 2050 Kenny Rd., Columbus, Ohio 43214, U.S.A. E-mail: robert.magnussen@ gmail.com Ó 2016 by the Arthroscopy Association of North America 0749-8063/14933/$36.00 http://dx.doi.org/10.1016/j.arthro.2015.11.027
graft consisting of doubled semitendinosus and gracilis tendons (ST-G). Each hamstring graft configuration has advantages and disadvantages. The ST-G graft technique is familiar to many surgeons and generally yields sufficient length for ACL reconstruction without concern about having enough graft for secure fixation at both ends. The disadvantages of this technique are the need to sacrifice the gracilis and the possibility of relatively small graft diameter. The gracilis may act to reinforce the action of the hamstrings in deep knee flexion and potentially help compensate for the loss of the semitendinosus.2 The relatively small graft diameter that this technique yields in some patients is concerning given increased risk of revision surgery associated with smaller graft diameter.3-5 The ST graft technique has the primary advantage of preservation of the gracilis muscle, potentially improving postoperative hamstring strength. Further, the quadrupled ST graft generally has larger diameter than the ST-G graft. Disadvantages of this technique include the potential for a short graft when the ST is
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2016: pp 1-9
1
2
A. SHARMA ET AL.
Table 1. Inclusion and Exclusion Criteria Inclusion Criteria Published studies comparing outcomes after harvest of isolated semitendinosus or semitendinosus and gracilis Studies reporting patient reported outcomes Studies reporting knee laxity Studies reporting hamstring strength after harvest
Exclusion Criteria Noncomparative studies Technique papers not reporting outcomes Review papers Basic science or anatomical studies Commentaries or other papers not presenting data
quadrupled. New techniques have been described to facilitate the preparation and implantation of such grafts,6 but careful planning is required to avoid length mismatch. A recent review by Ardern et al.2 in 2009 evaluated the effect of gracilis harvest on hamstring strength after ACL reconstruction and noted a trend toward improved strength in those in whom only the semitendinosus was harvested. The effect of gracilis harvest on patientreported outcomes7-9 and anterior knee laxity2,7-14 after ACL reconstruction has been reported recently by several authors. This paper will synthesize these data to inform surgeon decision-making regarding harvest of the gracilis tendon. The purpose of this systematic
review was to analyze the effect of gracilis harvest on hamstring strength, patient-reported outcomes, and anterior knee laxity after ACL reconstruction. We hypothesized that ST-G harvest would result in decreased hamstring strength compared with isolated ST harvest, but no effect would be seen on patient-reported outcomes or anterior knee laxity.
Methods Literature Review A search of MEDLINE and Scopus was preformed to identify all publications from January 1, 1966, through August 21, 2014, comparing outcomes of isolated ST harvest versus ST-G harvest for ACL reconstruction. Searches including the terms “semitendinosus” or “gracilis” and “harvest*,” and “ACL” or “anterior cruciate” yielded 872 articles after removal of duplicate citations. The titles and abstracts of all articles were reviewed (augmented by full text review when needed) and the 862 studies that failed to meet inclusion and exclusion criteria outlined in Table 1 were eliminated. Full text review of the remaining 10 studies confirmed that they met all inclusion and exclusion criteria. A review of all references of the included studies as well as a previous systematic review on the subject15 identified 2 additional studies for inclusion in the review.
Fig 1. Summary of literature search. (ACL, anterior cruciate ligament; ST, semitendinosus; ST-G, semitendinosus-gracilis.)
3
GRACILIS PRESERVATION IN ACL RECONSTRUCTION Table 2. Study Demographics
Author Adachi et al.7
Study Design Level of Year Evidence 2003 Prospective cohort level 2 Arthroscopy 2010 Retrospective cohort level 3 Int Orthop 2013 Retrospective cohort level 3 Arthroscopy 1999 Randomized controlled trial level 1 Arthroscopy 2005 Randomized controlled trial level 1 J Orthop Sci 2013 Prospective cohort level 2 KSSTA 2015 Randomized controlled trial level 1 AJSM 1982 Retrospective cohort level 3 Arthroscopy 2002 Retrospective cohort level 2 Arthroscopy 2002 Prospective cohort level 2 AJSM 2003 Randomized controlled trial level 1 KSSTA 2011 Prospective cohort level 2 Journal AOTS
Ardern et al.15 Barenius et al.8 Carter and Edinger16
Gobbi et al.17
Inagaki et al.10 Karimi-Mobarakeh et al.9 Lipscomb et al.18 Nakamura et al.11 Segawa et al.12 Tashiro et al.13
Yosmaoglu et al.14
Modified Coleman Methodology Score 76
Initial Cohort 44
Final Cohort 44 (100%)
Followup (yr) 1.0
ST Harvest 26
ST-G Harvest 18
60
198
50 (25%)
2.7
20
30
46
NR
20
3
10
10
79
76
68 (89%)
0.5
33
35
84
115
97 (84%)
3
50
47
74
143
120 (84%)
2.0
61
59
90
129
119 (92%)
1
58
61
51
482
51 (10.6%)
2.2
26
25
64
74
72 (93.7%)
2.0
49
25
77
62
62 (100%)
1.0
32
30
84
90
85 (94%)
1.5
49
36
56
46
46 (100%)
1.0
23
23
AJSM, American Journal of Sports Medicine; AOTS, Archives of Orthopaedic and Trauma Surgery; Int Orthop, International Orthopedics; J Orthop Sci, Journal of Orthopaedic Science; KSSTA, Knee Surgery, Sports Traumatology, and Arthroscopy; ST, semitendionsis; ST-G, simitendinosus and gracilis.
Table 3. Comparison of ST and ST-G Patients Patient Age Author Adachi et al.7 Ardern et al.15 Barenius et al.8 Carter and Edinger16 Gobbi et al.17 Inagaki et al.10 KarimiMobarakeh et al.9 Lipscomb et al.18 Nakamura et al.11 Segawa et al.12 Tashiro et al.13 Yosmaoglu et al.14
ST ST-G Harvest Harvest 27.7 10.5 25.6 8.9 24.4 4.9 26.1 6.9 26 9 26 7 NR NR 31 28.8 28.2 11.9 26.2 10.3 28.8 8.2 29.7 7.9
Sex (% Male)
Graft Diameter (mm)
ST ST-G ST Significance Harvest Harvest Significance Harvest P ¼ .49 15/26 (57.7%) 12/18 (66.7%) P ¼ .81 NR P ¼ .35 15/20 (75%) 20/30 (66.7%) P ¼ .83 NR P ¼ .9 8/10 (80%) 8/10 (80%) P ¼ 1.0 8.7 1 NR NR NR NR NR P > .05 P ¼ .32 P ¼ .54
31/50 (62%) 26/47 (55.3%) 35/61 (57.1%) 33/59 (55.9%) 48/58 (82.8%) 50/61 (82%)
P ¼ .74 P ¼ .87 P ¼ 1.0
ST-G Harvest NR NR 7.5 1 NR
NR NR NR NR 7.2 0.8 7.9 0.7
Significance NR NR P ¼ .01 NR NR NR P < .001
20.3
19.5
NR
22/26 (84.6%) 23/25 (92%)
P ¼ 1.0
NR
NR
NR
24.3
25.7
NR
28/49 (57.1%)
6/25 (24%)
P ¼ .11
NR
NR
NR
NR 24.5 7.7 29 7
NR 24.8 6.4 28 9
NR P ¼ .85 P ¼ .68
15/30 (50%) 19/36 (53%) NR
P ¼ .83 P ¼ .72 NR
19/32 (59%) 30/49 (61%) NR
NR, not reported; ST, semitendinosus; ST-G, semitendinosus and gracilis.
NR NR 8.8 0.7 8.9 0.7 NR NR
NR P ¼ .51 NR
4
A. SHARMA ET AL.
Table 4. Hamstring Isokinetic Strength and Anterior Laxity Isokinetic Strength 60 /s (Percentage of Contralateral Side) Author Adachi et al.7 Ardern et al.15 Barenius et al.8 Carter and Edinger16 Gobbi et al.17 Inagaki et al.10 KarimiMobarakeh et al.9 Lipscomb et al.18 Nakamura et al.11 Segawa et al.12 Tashiro et al.13 Yosmaoglu et al.14
ST Harvest 93.0 16.4 102.9 23.2 85.8 30 NR
ST-G Harvest 92.8 10.9 95.0 18.4 78.0 36 NR
Specific data Specific data not reported not reported 93.3 17.8 92.6 15.1 NR NR
Isokinetic Strength 180 -240 /s (Percentage of Contralateral Side)
AP Laxity Side-Side Difference (mm)
ST ST-G ST ST-G Significance Harvest Harvest Significance Harvest Harvest Significance P ¼ .96 101.2 20.9 107.1 23.2 P ¼ .38 1.5 1.2 1.9 1.1 P ¼ .36 P ¼ .54 111.7 36.7 97.4 23.6 P ¼ .54 1.4 2.9 1.0 2.8 P ¼ .63 P ¼ .60 NR NR NR 1.8 3.5 1.3 2.5 P ¼ .72 NR 80.6 28.3 81.7 18.5 P ¼ .85 NR NR NR P ¼ NS
Specific data not reported NR NR
P ¼ NS
P ¼ .84 NR
Specific data not reported NR NR
NR NR
1.3 1.4 1.3 1.5 1.2 1.1 1.4 1.2
P ¼ .82 P ¼ .35
1.2
1.6
P ¼ .78
103.5 93.7 12.8
97.5 91.3 16.1
NR P ¼ .49
101.3 89.4 14.3
100.5 86.1 17.6
NR P ¼ .39
NR NR 1.3 1.3 1.3 1.5
NR P ¼ 98
93.8 25.0 91 12 91.0 13.3
93.6 22.0 87 14 78.7 12.1
P ¼ .32 P ¼ .16 P ¼ .002
NR 93 14 93.3 13.7
NR 88 13 90.4 11.3
NR P ¼ .10 P ¼ .43
1.9 1.9 1.7 1.3 1.5 1.1 2.3 0.9 2.2 1.3
P > .05 P ¼ .46 P ¼ .76
NR, not reported; NS, not significant; ST, semitendinosus; ST-G, semitendinosus and gracilis.
These 12 studies provide the dataset for this review.7-18 The literature search is summarized in Figure 1. Data Extraction Extracted data included study characteristics, patient demographics, surgical technique, graft choice and size, length of follow-up, and outcome variables for each graft type including patient-reported outcome scores, anterior laxity, isokinetic and isometric hamstring strength, and active knee flexion angle. Data were extracted by 2 authors (A.S. and K.R.) independently and discrepancies were resolved by consensus.19,20 Assessment of Study Quality A modified Coleman methodology score was used to assess the quality of included studies.21,22 This scoring system awards points for study design and size, patient
selection, length and completeness of follow-up, and outcomes assessment. Points were totaled to yield a maximum of 100 points. Statistics Collected data were tabulated and summarized. Patient factors and results were compared between the ST and ST-G harvest groups in each individual study. P values reported by the original authors were used when available, whereas P values were calculated when possible based on reported means, standard deviations, and patient numbers when not reported. Unpaired t-tests were used to compare continuous variables and Fisher’s exact test was used to compare categorical variables. All statistical tests were performed with STATACorp version 13.1 (Stata, College Station, TX).
Fig 2. Meta-analysis of 8 studies reporting isokinetic strength at 60 per second. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
5
GRACILIS PRESERVATION IN ACL RECONSTRUCTION
Fig 3. Meta-analysis of 6 studies reporting isokinetic strength at 180 to 240 per second. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
To compare overall patient-reported outcomes, anterior laxity, and isokinetic hamstring strength for ST harvest versus ST-G harvest, a random effects model was developed in Review Manager 5.23 The method of DerSimonian-Laird was used to calculate an overall pooled estimate of effect for each outcome variable.23 Studies were weighted by the inverse variance method. Study heterogeneity was assessed qualitatively by comparing the populations and designs of individual studies, as well as quantitatively using a c-squared test. For outcome variables in which a measurement of dispersion (standard deviation or confidence interval) was lacking, formal metaanalysis was not performed. Because of the heterogeneity of study design and follow-up of the included studies, 2 sensitivity analyses were performed using the isokinetic strength data. These analyses were performed by repeating the metaanalysis first using only data from randomized controlled trials, and again using only data from studies with at least 80% follow-up at minimum 2 years after surgery.
Results Twelve studies that compared the outcomes of isolated ST harvest with ST-G harvest were identified in
the literature, including 4 randomized controlled trials (level I evidence),9,13,16,17 4 prospective comparative studies (level II evidence),7,10,12,14 and 4 retrospective comparative studies (level III evidence).8,11,15,18 The modified Coleman methodology scores were highest for the randomized controlled trials (mean: 84.25), followed by the prospective cohort studies (mean: 70.75) and the retrospective cohort studies (mean: 55.25) (Table 2). Follow-up ranged from 6 months to 3 years. Mean patient age for each study ranged from 20 to 29 years, and most of the studies reported more male than female patients. No significant differences in patient age or sex were noted between the ST and ST-G groups in any of the studies (Table 3). Only 3 studies reported graft size for single bundle ACL reconstructions with ST and ST-G grafts. Barenius et al.8 noted larger grafts with ST harvest, whereas KarimiMobarakeh et al.9 noted larger grafts with the ST-G technique. Tashiro et al.13 noted similar graft size with both techniques. Isokinetic strength at 60 per second was reported in 10 studies (Table 4).7,8,10-15,17,18 Only 1 study reported a statistically significant decrease in strength in the ST-G group,14 although a trend in that direction was noted in most studies. Meta-analysis of the 8 studies from which data could be extracted showed a
Table 5. Hamstring Isometric Strength Isometric Strength 90 (Percentage of Contralateral Side) Author Adachi et al.7 Ardern et al.15 Barenius et al.8 KarimiMobarakeh et al.9 Nakamura et al.11 Tashiro et al.13
Isometric Strength 105 -110 (Percentage of Contralateral Side)
Active Knee Flexion Angle Loss (Relative to the Contralateral Side)
ST Harvest NR 81.5 26.4 82.7 36 68.3 9.1
ST-G Harvest NR 76.2 23.4 82.6 27 65.5 8.6
Significance NR P ¼ .46 P ¼ 1.0 P ¼ NS
ST Harvest NR 85.2 42.5 NR NR
ST-G Harvest NR 72.5 48.8 NR NR
Significance NR P ¼ .08 NR NR
ST Harvest 8.9 7.0 3.3 6.5 4.0 5.0 NR
ST-G Harvest 16.7 10.5 4.6 8.6 7.0 11.0 NR
NR
NR
NR
NR
NR
NR
5.0 8.0
9.0 11.0
84 12
73 18
P < .05
82 15
68 22
P < .05
NR
NR
NR, not reported; NS, not significant; ST, semitendinosus; ST-G, semitendinosus and gracilis.
Significance P < .05 P ¼ NS P ¼ .40 NR
P ¼ .014 NR
6
A. SHARMA ET AL.
Fig 4. Meta-analysis of 4 studies reporting isometric strength at 90 of knee flexion. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
difference in relative hamstring strength of 3.85% (95% confidence interval: 0.77% to 6.92%), with the ST-G group showing significantly more weakness relative to the ST group (P ¼ .01) (Fig 2). Sensitivity analyses showed no qualitative differences in isokinetic strength at 60 per second when including only data from randomized controlled trials or studies with greater than 80% follow-up at minimum 2 years after surgery. Isokinetic strength at 180 to 300 per second was reported in 8 studies (Table 4),7,11,13-18 with none showing any significant difference between the ST and ST-G groups. Meta-analysis of the 6 studies from which data could be extracted showed a difference in relative hamstring strength of 2.39% (95% confidence interval: 1.12% to 5.89%), with no significant difference noted between the 2 groups (P ¼ .18) in isokinetic strength between 180 and 240 per second (Fig 3). Sensitivity analyses showed no qualitative differences in isokinetic strength at 180 to 240 per second when including only data from randomized controlled trials or studies with greater than 80% follow-up at minimum 2 years after surgery. Isometric strength was reported in 4 studies (Table 5).8,9,13,15 Two studies reported results only at 90 of flexion,8,9 and 2 reported testing at 90 as well as higher flexion angles (105 to 110 ).13,15 One of 4 studies showed decreased strength in the ST-G group at 90 of flexion and 1 of 2 showed decreased strength in the ST-G group at high flexion angles. Meta-analysis of the 4 studies reporting data
at 90 of flexion showed a difference in relative hamstring strength of 5.55% (95% confidence interval: 0.44% to 10.65%), with a significant difference noted between the 2 groups (P ¼ .03) (Fig 4). Metaanalysis of the 2 studies reporting data at 105 to 110 of flexion showed a difference in relative hamstring strength of 13.68% (95% confidence interval: 4.57% to 22.80%), with a significant difference noted between the 2 groups (P ¼ .003) (Fig 5). Active knee flexion loss was reported in 4 studies (Table 5), 2 of which showed more active flexion loss in the ST-G group. Meta-analysis of the 4 studies showed a difference in loss of active knee flexion of 3.91 (95% confidence interval: 1.15 to 6.67 ), with a significant difference noted between the 2 groups (P ¼ .006) (Fig 6). Patient-reported outcomes (subjective International Knee Documentation Committee, Lysholm, and Cincinnati knee scores) were described in 6 studies (Table 6),8-10,12,15,17 with none showing any significant difference between the ST and ST-G groups. Meta-analysis of the 3 studies reporting subjective International Knee Documentation Committee scores showed no significant differences between the 2 groups (Fig 7). Instrumented anterior laxity was reported in 10 studies (Table 6),7-15,17 with none showing any significant difference between the ST and ST-G groups. Meta-analysis of the 8 studies from which data could be extracted showed no significant differences between the 2 groups (Fig 8).
Fig 5. Meta-analysis of 2 studies reporting isometric strength at 105 to 110 of knee flexion. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
7
GRACILIS PRESERVATION IN ACL RECONSTRUCTION
Fig 6. Meta-analysis of 4 studies reporting active knee flexion loss. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
Discussion
The most important finding of this meta-analysis is that although additional gracilis harvest leads to a statistically significantly more hamstring weakness in isokinetic testing at 60 per second as well as isometric strength testing at 90 of flexion, the overall differences in the 3% to 5% range likely do not reach clinical significance. The difference between groups did not reach statistical significance when isokinetic testing was done at 180 to 240 per second and the 2 groups demonstrated similar patient-reported outcome scores in the 2 groups. The larger differences noted in isometric strength at high flexion angles (13.68%) likely approach clinical significance, although the activities performed at such high knee flexion angles are limited. However, specific situations may exist in which the loss of the gracilis is more significant. Activities that require deep knee flexion may be more affected by gracilis harvest as noted in the meta-analysis above. Similarly, Segawa et al.12 and Gobbi et al.17 both noted decreased lower limb isokinetic internal rotation torque after ST-G harvest relative to ST harvest. These specific deficits could be relevant to specific patient populations and should be considered by clinicians. Prior studies evaluating the effect of ST versus ST-G harvest have focused on hamstring strength and also noted mixed findings. Ardern et al.2 in 2009 noted no differences in isokinetic strength between the 2 groups, but did comment on the decreased hamstring strength in deep flexion that was observed by Tashiro et al.13 after ST-G harvest. Petersen et al.24 in 2014
similarly noted specific cases such as deep knee flexion and internal rotation that may be more affected by gracilis harvest. The prior review did not address the effects of ST versus ST-G harvest on knee laxity or patient-reported outcomes, but the current study finds no evidence of an effect of gracilis harvest on anterior laxity and any difference in patient-reported outcomes is too small to be clinically relevant. Limitations Weaknesses include the variation within the individual studies in regard to a surgical technique, outcome measures, length of follow-up, and study design. These differences are likely reflected in the differences in standard deviations noted among the studies, which were modeled using a random effects statistical model. The inclusion of nonrandomized comparative studies also introduces bias, but the data from the 4 randomized controlled trials included in the analysis are qualitatively similar to those from the nonrandomized comparative studies, minimizing this risk.
Conclusions The addition of gracilis harvest to an isolated semitendinosus harvest for ACL reconstruction results in statistically significant, but likely not clinically relevant differences in isokinetic and isometric hamstring strength as well as patient-reported outcomes. Hamstring strength deficits may be larger at higher flexion angles.
Table 6. Patient-Reported Outcomes IKDC Subjective Author Ardern et al.15 Barenius et al.8 Gobbi et al.17 Inagaki et al.10 Karimi-Mobarakeh et al.9 Segawa et al.12
ST Harvest 94.4 3.6 94.1 3.7 NR NR 62.6 9.5 NR
ST-G Harvest 91.6 8.1 94.5 4.0 NR NR 61.4 8.6 NR
Significance P ¼ .16 P ¼ .80 NR NR P ¼ .47 NR
Lysholm ST Harvest NR NR 92.2 97.0 4.6 55.1 8.9 96.9
ST-G Harvest NR NR 93.6 96.7 5.4 57.4 9.3 96.3
Cincinnati Significance NR NR P ¼ .72 P ¼ .78 P ¼ .26 P > .05
ST Harvest NR NR 84.5 NR NR NR
ST-G Harvest NR NR 82.3 NR NR NR
IKDC, International Knee Documentation Committee; NR, not reported; ST, semitendinusis; ST-G, semitendinosus and gracilis.
Significance NR NR P ¼ .72 NR NR NR
8
A. SHARMA ET AL.
Fig 7. Meta-analysis of the 3 studies reporting subjective International Knee Documentation Committee scores. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
Fig 8. Meta-analysis of the 8 studies reporting instrumented anterior laxity. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
References 1. Samuelsson K, Andersson D, Ahlden M, Fu FH, Musahl V, Karlsson J. Trends in surgeon preferences on anterior cruciate ligament reconstructive techniques. Clin Sports Med 2013;32:111-126. 2. Ardern CL, Webster KE. Knee flexor strength recovery following hamstring tendon harvest for anterior cruciate ligament reconstruction: A systematic review. Orthop Rev (Pavia) 2009;1:e12. 3. Magnussen RA, Lawrence JT, West RL, Toth AP, Taylor DC, Garrett WE. Graft size and patient age are predictors of early revision after anterior cruciate ligament reconstruction with hamstring autograft. Arthroscopy 2012;28:526-531. 4. Mariscalco MW, Flanigan DC, Mitchell J, et al. The influence of hamstring autograft size on patient-reported outcomes and risk of revision after anterior cruciate ligament reconstruction: A Multicenter Orthopaedic Outcomes Network (MOON) cohort study. Arthroscopy 2013;29:1948-1953. 5. Park SY, Oh H, Park S, Lee JH, Lee SH, Yoon KH. Factors predicting hamstring tendon autograft diameters and resulting failure rates after anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc 2013;21:1111-1118. 6. Lubowitz JH. All-inside anterior cruciate ligament graft link: Graft preparation technique. Arthrosc Tech 2012;1: e165-e168. 7. Adachi N, Ochi M, Uchio Y, Sakai Y, Kuriwaka M, Fujihara A. Harvesting hamstring tendons for ACL reconstruction influences postoperative hamstring muscle performance. Arch Orthop Trauma Surg 2003;123: 460-465.
8. Barenius B, Webster WK, McClelland J, Feller J. Hamstring tendon anterior cruciate ligament reconstruction: Does gracilis tendon harvest matter? Int Orthop 2013;37:207-212. 9. Karimi-Mobarakeh M, Mardani-Kivi M, Mortazavi A, Saheb-Ekhtiari K, Hashemi-Motlagh K. Role of gracilis harvesting in four-strand hamstring tendon anterior cruciate ligament reconstruction: A double-blinded prospective randomized clinical trial. Knee Surg Sports Traumatol Arthrosc 2015;23:1086-1091. 10. Inagaki Y, Kondo E, Kitamura N, et al. Prospective clinical comparisons of semitendinosus versus semitendinosus and gracilis tendon autografts for anatomic double-bundle anterior cruciate ligament reconstruction. J Orthop Sci 2013;18:754-761. 11. Nakamura N, Horibe S, Sasaki S, et al. Evaluation of active knee flexion and hamstring strength after anterior cruciate ligament reconstruction using hamstring tendons. Arthroscopy 2002;18:598-602. 12. Segawa H, Omori G, Koga Y, Kameo T, Iida S, Tanaka M. Rotational muscle strength of the limb after anterior cruciate ligament reconstruction using semitendinosus and gracilis tendon. Arthroscopy 2002;18:177-182. 13. Tashiro T, Kurosawa H, Kawakami A, Hikita A, Fukui N. Influence of medial hamstring tendon harvest on knee flexor strength after anterior cruciate ligament reconstruction. A detailed evaluation with comparison of single- and double-tendon harvest. Am J Sports Med 2003;31: 522-529. 14. Yosmaoglu HB, Baltaci G, Ozer H, Atay A. Effects of additional gracilis tendon harvest on muscle torque, motor coordination, and knee laxity in ACL reconstruction. Knee Surg Sports Traumatol Arthrosc 2011;19:1287-1292.
GRACILIS PRESERVATION IN ACL RECONSTRUCTION 15. Ardern CL, Webster KE, Taylor NF, Feller JA. Hamstring strength recovery after hamstring tendon harvest for anterior cruciate ligament reconstruction: A comparison between graft types. Arthroscopy 2010;26:462-469. 16. Carter TR, Edinger S. Isokinetic evaluation of anterior cruciate ligament reconstruction: Hamstring versus patellar tendon. Arthroscopy 1999;15:169-172. 17. Gobbi A, Domzalski M, Pascual J, Zanazzo M. Hamstring anterior cruciate ligament reconstruction: Is it necessary to sacrifice the gracilis? Arthroscopy 2005;21: 275-280. 18. Lipscomb AB, Johnston RK, Snyder RB, Warburton MJ, Gilbert PP. Evaluation of hamstring strength following use of semitendinosus and gracilis tendons to reconstruct the anterior cruciate ligament. Am J Sports Med 1982;10: 340-342. 19. Greenhalgh T. How to read a paper: The basics of evidence based medicine, Ed 2. London: BMJ Publishing Group, 2001.
9
20. Spindler KP, Kuhn JE, Dunn W, Matthews CE, Harrell FE, Dittus RS. Reading and reviewing the orthopaedic literature: A systematic, evidence-based medicine approach. J Am Acad Orthop Surg 2005;13:220-229. 21. Coleman BD, Khan KM, Maffulli N, Cook JL, Wark JD. Studies of surgical outcome after patellar tendinopathy: Clinical significance of methodological deficiencies and guidelines for future studies. Victorian Institute of Sport Tendon Study Group. Scand J Med Sci Sports 2000;10:2-11. 22. Magnussen RA, Carey JL, Spindler KP. Does autograft choice determine intermediate-term outcome of ACL reconstruction? Knee Surg Sports Traumatol Arthrosc 2011;19:462-472. 23. Collaboration TC. Review manager (RevMan), Ed 5.0. Copenhagen: The Nordic Cochrane Center, 2008. 24. Petersen W, Taheri P, Forkel P, Zantop T. Return to play following ACL reconstruction: A systematic review about strength deficits. Arch Orthop Trauma Surg 2014;134: 1417-1428.
Does Gracilis Preservation Matter in Anterior Cruciate Ligament Reconstruction? A Systematic Review Avijit Sharma, B.S., David C. Flanigan, M.D., Kyle Randall, M.D., and Robert A. Magnussen, M.D., M.P.H.
Purpose: To analyze the effect of gracilis harvest on hamstring strength, patient-reported outcomes, and anterior knee laxity after anterior cruciate ligament (ACL) reconstruction. Methods: A systematic review of the literature was performed to identify studies comparing the results of semitendinosus (ST) versus semitendinosus-gracilis (ST-G) harvest for ACL reconstruction. A meta-analysis using a random effects model was performed to determine overall pooled estimates of effect for the influence of additional gracilis harvest on hamstring strength, patient-reported outcomes, and anterior knee laxity after ACL reconstruction. Results: Twelve studies were identified and included in the review. ST-G harvest was noted to decrease hamstring isokinetic strength at 60 per second by 3.85% relative to isolated ST harvest (P ¼ .01). Decreased isometric strength was also noted in the ST-G harvest group at both 90 of flexion (mean difference: 5.55%; P ¼ .03) and 105 to 110 of flexion (mean difference: 13.68%; P ¼ .003). Active knee flexion angle loss was also noted to be greater in the ST-G harvest group (mean difference: 3.91 ; P ¼ .006). No differences were found in isokinetic strength at 180 to 240 per second (mean difference: 3.20%; P ¼ .08), patient-reported outcome scores (mean difference: 1.87 points; P ¼ .06), or anterior knee laxity (mean difference: 0.03 mm; P ¼ .78) based on gracilis harvest. Conclusions: The addition of gracilis harvest to an isolated ST harvest for ACL reconstruction results in statistically significant, but likely not clinically relevant differences in isokinetic and isometric hamstring strength as well as patient-reported outcomes. Hamstring strength deficits may be larger at higher flexion angles. Level of Evidence: Level III, systematic review of level I-III studies.
A
nterior cruciate ligament (ACL) is commonly injured and reconstruction is often performed to allow return to active participation in cutting and pivoting sports. Although numerous potential grafts for ACL reconstruction have been described, hamstring autografts are among the most commonly used.1 Numerous hamstring preparation techniques have been described, the most common of which are a quadrupled semitendinosus tendon (ST) graft and a
From the Department of Orthopaedic Surgery, The Ohio State University (A.S., D.C.F., R.A.M.); Sports Health and Performance Institute, The Ohio State University (D.C.F., R.A.M.), Columbus; and Avita Health System (K.R.), Ontario, Ohio, U.S.A. The authors report the following potential conflict of interest or source of funding: D.C.F. receives support from Sanofi, Smith & Nephew. Received November 4, 2014; accepted November 5, 2015. Address correspondence to Robert A. Magnussen, M.D., M.P.H., 2050 Kenny Rd., Columbus, Ohio 43214, U.S.A. E-mail: robert.magnussen@ gmail.com Ó 2016 by the Arthroscopy Association of North America 0749-8063/14933/$36.00 http://dx.doi.org/10.1016/j.arthro.2015.11.027
graft consisting of doubled semitendinosus and gracilis tendons (ST-G). Each hamstring graft configuration has advantages and disadvantages. The ST-G graft technique is familiar to many surgeons and generally yields sufficient length for ACL reconstruction without concern about having enough graft for secure fixation at both ends. The disadvantages of this technique are the need to sacrifice the gracilis and the possibility of relatively small graft diameter. The gracilis may act to reinforce the action of the hamstrings in deep knee flexion and potentially help compensate for the loss of the semitendinosus.2 The relatively small graft diameter that this technique yields in some patients is concerning given increased risk of revision surgery associated with smaller graft diameter.3-5 The ST graft technique has the primary advantage of preservation of the gracilis muscle, potentially improving postoperative hamstring strength. Further, the quadrupled ST graft generally has larger diameter than the ST-G graft. Disadvantages of this technique include the potential for a short graft when the ST is
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2016: pp 1-9
1
2
A. SHARMA ET AL.
Table 1. Inclusion and Exclusion Criteria Inclusion Criteria Published studies comparing outcomes after harvest of isolated semitendinosus or semitendinosus and gracilis Studies reporting patient reported outcomes Studies reporting knee laxity Studies reporting hamstring strength after harvest
Exclusion Criteria Noncomparative studies Technique papers not reporting outcomes Review papers Basic science or anatomical studies Commentaries or other papers not presenting data
quadrupled. New techniques have been described to facilitate the preparation and implantation of such grafts,6 but careful planning is required to avoid length mismatch. A recent review by Ardern et al.2 in 2009 evaluated the effect of gracilis harvest on hamstring strength after ACL reconstruction and noted a trend toward improved strength in those in whom only the semitendinosus was harvested. The effect of gracilis harvest on patientreported outcomes7-9 and anterior knee laxity2,7-14 after ACL reconstruction has been reported recently by several authors. This paper will synthesize these data to inform surgeon decision-making regarding harvest of the gracilis tendon. The purpose of this systematic
review was to analyze the effect of gracilis harvest on hamstring strength, patient-reported outcomes, and anterior knee laxity after ACL reconstruction. We hypothesized that ST-G harvest would result in decreased hamstring strength compared with isolated ST harvest, but no effect would be seen on patient-reported outcomes or anterior knee laxity.
Methods Literature Review A search of MEDLINE and Scopus was preformed to identify all publications from January 1, 1966, through August 21, 2014, comparing outcomes of isolated ST harvest versus ST-G harvest for ACL reconstruction. Searches including the terms “semitendinosus” or “gracilis” and “harvest*,” and “ACL” or “anterior cruciate” yielded 872 articles after removal of duplicate citations. The titles and abstracts of all articles were reviewed (augmented by full text review when needed) and the 862 studies that failed to meet inclusion and exclusion criteria outlined in Table 1 were eliminated. Full text review of the remaining 10 studies confirmed that they met all inclusion and exclusion criteria. A review of all references of the included studies as well as a previous systematic review on the subject15 identified 2 additional studies for inclusion in the review.
Fig 1. Summary of literature search. (ACL, anterior cruciate ligament; ST, semitendinosus; ST-G, semitendinosus-gracilis.)
3
GRACILIS PRESERVATION IN ACL RECONSTRUCTION Table 2. Study Demographics
Author Adachi et al.7
Study Design Level of Year Evidence 2003 Prospective cohort level 2 Arthroscopy 2010 Retrospective cohort level 3 Int Orthop 2013 Retrospective cohort level 3 Arthroscopy 1999 Randomized controlled trial level 1 Arthroscopy 2005 Randomized controlled trial level 1 J Orthop Sci 2013 Prospective cohort level 2 KSSTA 2015 Randomized controlled trial level 1 AJSM 1982 Retrospective cohort level 3 Arthroscopy 2002 Retrospective cohort level 2 Arthroscopy 2002 Prospective cohort level 2 AJSM 2003 Randomized controlled trial level 1 KSSTA 2011 Prospective cohort level 2 Journal AOTS
Ardern et al.15 Barenius et al.8 Carter and Edinger16
Gobbi et al.17
Inagaki et al.10 Karimi-Mobarakeh et al.9 Lipscomb et al.18 Nakamura et al.11 Segawa et al.12 Tashiro et al.13
Yosmaoglu et al.14
Modified Coleman Methodology Score 76
Initial Cohort 44
Final Cohort 44 (100%)
Followup (yr) 1.0
ST Harvest 26
ST-G Harvest 18
60
198
50 (25%)
2.7
20
30
46
NR
20
3
10
10
79
76
68 (89%)
0.5
33
35
84
115
97 (84%)
3
50
47
74
143
120 (84%)
2.0
61
59
90
129
119 (92%)
1
58
61
51
482
51 (10.6%)
2.2
26
25
64
74
72 (93.7%)
2.0
49
25
77
62
62 (100%)
1.0
32
30
84
90
85 (94%)
1.5
49
36
56
46
46 (100%)
1.0
23
23
AJSM, American Journal of Sports Medicine; AOTS, Archives of Orthopaedic and Trauma Surgery; Int Orthop, International Orthopedics; J Orthop Sci, Journal of Orthopaedic Science; KSSTA, Knee Surgery, Sports Traumatology, and Arthroscopy; ST, semitendionsis; ST-G, simitendinosus and gracilis.
Table 3. Comparison of ST and ST-G Patients Patient Age Author Adachi et al.7 Ardern et al.15 Barenius et al.8 Carter and Edinger16 Gobbi et al.17 Inagaki et al.10 KarimiMobarakeh et al.9 Lipscomb et al.18 Nakamura et al.11 Segawa et al.12 Tashiro et al.13 Yosmaoglu et al.14
ST ST-G Harvest Harvest 27.7 10.5 25.6 8.9 24.4 4.9 26.1 6.9 26 9 26 7 NR NR 31 28.8 28.2 11.9 26.2 10.3 28.8 8.2 29.7 7.9
Sex (% Male)
Graft Diameter (mm)
ST ST-G ST Significance Harvest Harvest Significance Harvest P ¼ .49 15/26 (57.7%) 12/18 (66.7%) P ¼ .81 NR P ¼ .35 15/20 (75%) 20/30 (66.7%) P ¼ .83 NR P ¼ .9 8/10 (80%) 8/10 (80%) P ¼ 1.0 8.7 1 NR NR NR NR NR P > .05 P ¼ .32 P ¼ .54
31/50 (62%) 26/47 (55.3%) 35/61 (57.1%) 33/59 (55.9%) 48/58 (82.8%) 50/61 (82%)
P ¼ .74 P ¼ .87 P ¼ 1.0
ST-G Harvest NR NR 7.5 1 NR
NR NR NR NR 7.2 0.8 7.9 0.7
Significance NR NR P ¼ .01 NR NR NR P < .001
20.3
19.5
NR
22/26 (84.6%) 23/25 (92%)
P ¼ 1.0
NR
NR
NR
24.3
25.7
NR
28/49 (57.1%)
6/25 (24%)
P ¼ .11
NR
NR
NR
NR 24.5 7.7 29 7
NR 24.8 6.4 28 9
NR P ¼ .85 P ¼ .68
15/30 (50%) 19/36 (53%) NR
P ¼ .83 P ¼ .72 NR
19/32 (59%) 30/49 (61%) NR
NR, not reported; ST, semitendinosus; ST-G, semitendinosus and gracilis.
NR NR 8.8 0.7 8.9 0.7 NR NR
NR P ¼ .51 NR
4
A. SHARMA ET AL.
Table 4. Hamstring Isokinetic Strength and Anterior Laxity Isokinetic Strength 60 /s (Percentage of Contralateral Side) Author Adachi et al.7 Ardern et al.15 Barenius et al.8 Carter and Edinger16 Gobbi et al.17 Inagaki et al.10 KarimiMobarakeh et al.9 Lipscomb et al.18 Nakamura et al.11 Segawa et al.12 Tashiro et al.13 Yosmaoglu et al.14
ST Harvest 93.0 16.4 102.9 23.2 85.8 30 NR
ST-G Harvest 92.8 10.9 95.0 18.4 78.0 36 NR
Specific data Specific data not reported not reported 93.3 17.8 92.6 15.1 NR NR
Isokinetic Strength 180 -240 /s (Percentage of Contralateral Side)
AP Laxity Side-Side Difference (mm)
ST ST-G ST ST-G Significance Harvest Harvest Significance Harvest Harvest Significance P ¼ .96 101.2 20.9 107.1 23.2 P ¼ .38 1.5 1.2 1.9 1.1 P ¼ .36 P ¼ .54 111.7 36.7 97.4 23.6 P ¼ .54 1.4 2.9 1.0 2.8 P ¼ .63 P ¼ .60 NR NR NR 1.8 3.5 1.3 2.5 P ¼ .72 NR 80.6 28.3 81.7 18.5 P ¼ .85 NR NR NR P ¼ NS
Specific data not reported NR NR
P ¼ NS
P ¼ .84 NR
Specific data not reported NR NR
NR NR
1.3 1.4 1.3 1.5 1.2 1.1 1.4 1.2
P ¼ .82 P ¼ .35
1.2
1.6
P ¼ .78
103.5 93.7 12.8
97.5 91.3 16.1
NR P ¼ .49
101.3 89.4 14.3
100.5 86.1 17.6
NR P ¼ .39
NR NR 1.3 1.3 1.3 1.5
NR P ¼ 98
93.8 25.0 91 12 91.0 13.3
93.6 22.0 87 14 78.7 12.1
P ¼ .32 P ¼ .16 P ¼ .002
NR 93 14 93.3 13.7
NR 88 13 90.4 11.3
NR P ¼ .10 P ¼ .43
1.9 1.9 1.7 1.3 1.5 1.1 2.3 0.9 2.2 1.3
P > .05 P ¼ .46 P ¼ .76
NR, not reported; NS, not significant; ST, semitendinosus; ST-G, semitendinosus and gracilis.
These 12 studies provide the dataset for this review.7-18 The literature search is summarized in Figure 1. Data Extraction Extracted data included study characteristics, patient demographics, surgical technique, graft choice and size, length of follow-up, and outcome variables for each graft type including patient-reported outcome scores, anterior laxity, isokinetic and isometric hamstring strength, and active knee flexion angle. Data were extracted by 2 authors (A.S. and K.R.) independently and discrepancies were resolved by consensus.19,20 Assessment of Study Quality A modified Coleman methodology score was used to assess the quality of included studies.21,22 This scoring system awards points for study design and size, patient
selection, length and completeness of follow-up, and outcomes assessment. Points were totaled to yield a maximum of 100 points. Statistics Collected data were tabulated and summarized. Patient factors and results were compared between the ST and ST-G harvest groups in each individual study. P values reported by the original authors were used when available, whereas P values were calculated when possible based on reported means, standard deviations, and patient numbers when not reported. Unpaired t-tests were used to compare continuous variables and Fisher’s exact test was used to compare categorical variables. All statistical tests were performed with STATACorp version 13.1 (Stata, College Station, TX).
Fig 2. Meta-analysis of 8 studies reporting isokinetic strength at 60 per second. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
5
GRACILIS PRESERVATION IN ACL RECONSTRUCTION
Fig 3. Meta-analysis of 6 studies reporting isokinetic strength at 180 to 240 per second. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
To compare overall patient-reported outcomes, anterior laxity, and isokinetic hamstring strength for ST harvest versus ST-G harvest, a random effects model was developed in Review Manager 5.23 The method of DerSimonian-Laird was used to calculate an overall pooled estimate of effect for each outcome variable.23 Studies were weighted by the inverse variance method. Study heterogeneity was assessed qualitatively by comparing the populations and designs of individual studies, as well as quantitatively using a c-squared test. For outcome variables in which a measurement of dispersion (standard deviation or confidence interval) was lacking, formal metaanalysis was not performed. Because of the heterogeneity of study design and follow-up of the included studies, 2 sensitivity analyses were performed using the isokinetic strength data. These analyses were performed by repeating the metaanalysis first using only data from randomized controlled trials, and again using only data from studies with at least 80% follow-up at minimum 2 years after surgery.
Results Twelve studies that compared the outcomes of isolated ST harvest with ST-G harvest were identified in
the literature, including 4 randomized controlled trials (level I evidence),9,13,16,17 4 prospective comparative studies (level II evidence),7,10,12,14 and 4 retrospective comparative studies (level III evidence).8,11,15,18 The modified Coleman methodology scores were highest for the randomized controlled trials (mean: 84.25), followed by the prospective cohort studies (mean: 70.75) and the retrospective cohort studies (mean: 55.25) (Table 2). Follow-up ranged from 6 months to 3 years. Mean patient age for each study ranged from 20 to 29 years, and most of the studies reported more male than female patients. No significant differences in patient age or sex were noted between the ST and ST-G groups in any of the studies (Table 3). Only 3 studies reported graft size for single bundle ACL reconstructions with ST and ST-G grafts. Barenius et al.8 noted larger grafts with ST harvest, whereas KarimiMobarakeh et al.9 noted larger grafts with the ST-G technique. Tashiro et al.13 noted similar graft size with both techniques. Isokinetic strength at 60 per second was reported in 10 studies (Table 4).7,8,10-15,17,18 Only 1 study reported a statistically significant decrease in strength in the ST-G group,14 although a trend in that direction was noted in most studies. Meta-analysis of the 8 studies from which data could be extracted showed a
Table 5. Hamstring Isometric Strength Isometric Strength 90 (Percentage of Contralateral Side) Author Adachi et al.7 Ardern et al.15 Barenius et al.8 KarimiMobarakeh et al.9 Nakamura et al.11 Tashiro et al.13
Isometric Strength 105 -110 (Percentage of Contralateral Side)
Active Knee Flexion Angle Loss (Relative to the Contralateral Side)
ST Harvest NR 81.5 26.4 82.7 36 68.3 9.1
ST-G Harvest NR 76.2 23.4 82.6 27 65.5 8.6
Significance NR P ¼ .46 P ¼ 1.0 P ¼ NS
ST Harvest NR 85.2 42.5 NR NR
ST-G Harvest NR 72.5 48.8 NR NR
Significance NR P ¼ .08 NR NR
ST Harvest 8.9 7.0 3.3 6.5 4.0 5.0 NR
ST-G Harvest 16.7 10.5 4.6 8.6 7.0 11.0 NR
NR
NR
NR
NR
NR
NR
5.0 8.0
9.0 11.0
84 12
73 18
P < .05
82 15
68 22
P < .05
NR
NR
NR, not reported; NS, not significant; ST, semitendinosus; ST-G, semitendinosus and gracilis.
Significance P < .05 P ¼ NS P ¼ .40 NR
P ¼ .014 NR
6
A. SHARMA ET AL.
Fig 4. Meta-analysis of 4 studies reporting isometric strength at 90 of knee flexion. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
difference in relative hamstring strength of 3.85% (95% confidence interval: 0.77% to 6.92%), with the ST-G group showing significantly more weakness relative to the ST group (P ¼ .01) (Fig 2). Sensitivity analyses showed no qualitative differences in isokinetic strength at 60 per second when including only data from randomized controlled trials or studies with greater than 80% follow-up at minimum 2 years after surgery. Isokinetic strength at 180 to 300 per second was reported in 8 studies (Table 4),7,11,13-18 with none showing any significant difference between the ST and ST-G groups. Meta-analysis of the 6 studies from which data could be extracted showed a difference in relative hamstring strength of 2.39% (95% confidence interval: 1.12% to 5.89%), with no significant difference noted between the 2 groups (P ¼ .18) in isokinetic strength between 180 and 240 per second (Fig 3). Sensitivity analyses showed no qualitative differences in isokinetic strength at 180 to 240 per second when including only data from randomized controlled trials or studies with greater than 80% follow-up at minimum 2 years after surgery. Isometric strength was reported in 4 studies (Table 5).8,9,13,15 Two studies reported results only at 90 of flexion,8,9 and 2 reported testing at 90 as well as higher flexion angles (105 to 110 ).13,15 One of 4 studies showed decreased strength in the ST-G group at 90 of flexion and 1 of 2 showed decreased strength in the ST-G group at high flexion angles. Meta-analysis of the 4 studies reporting data
at 90 of flexion showed a difference in relative hamstring strength of 5.55% (95% confidence interval: 0.44% to 10.65%), with a significant difference noted between the 2 groups (P ¼ .03) (Fig 4). Metaanalysis of the 2 studies reporting data at 105 to 110 of flexion showed a difference in relative hamstring strength of 13.68% (95% confidence interval: 4.57% to 22.80%), with a significant difference noted between the 2 groups (P ¼ .003) (Fig 5). Active knee flexion loss was reported in 4 studies (Table 5), 2 of which showed more active flexion loss in the ST-G group. Meta-analysis of the 4 studies showed a difference in loss of active knee flexion of 3.91 (95% confidence interval: 1.15 to 6.67 ), with a significant difference noted between the 2 groups (P ¼ .006) (Fig 6). Patient-reported outcomes (subjective International Knee Documentation Committee, Lysholm, and Cincinnati knee scores) were described in 6 studies (Table 6),8-10,12,15,17 with none showing any significant difference between the ST and ST-G groups. Meta-analysis of the 3 studies reporting subjective International Knee Documentation Committee scores showed no significant differences between the 2 groups (Fig 7). Instrumented anterior laxity was reported in 10 studies (Table 6),7-15,17 with none showing any significant difference between the ST and ST-G groups. Meta-analysis of the 8 studies from which data could be extracted showed no significant differences between the 2 groups (Fig 8).
Fig 5. Meta-analysis of 2 studies reporting isometric strength at 105 to 110 of knee flexion. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
7
GRACILIS PRESERVATION IN ACL RECONSTRUCTION
Fig 6. Meta-analysis of 4 studies reporting active knee flexion loss. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
Discussion
The most important finding of this meta-analysis is that although additional gracilis harvest leads to a statistically significantly more hamstring weakness in isokinetic testing at 60 per second as well as isometric strength testing at 90 of flexion, the overall differences in the 3% to 5% range likely do not reach clinical significance. The difference between groups did not reach statistical significance when isokinetic testing was done at 180 to 240 per second and the 2 groups demonstrated similar patient-reported outcome scores in the 2 groups. The larger differences noted in isometric strength at high flexion angles (13.68%) likely approach clinical significance, although the activities performed at such high knee flexion angles are limited. However, specific situations may exist in which the loss of the gracilis is more significant. Activities that require deep knee flexion may be more affected by gracilis harvest as noted in the meta-analysis above. Similarly, Segawa et al.12 and Gobbi et al.17 both noted decreased lower limb isokinetic internal rotation torque after ST-G harvest relative to ST harvest. These specific deficits could be relevant to specific patient populations and should be considered by clinicians. Prior studies evaluating the effect of ST versus ST-G harvest have focused on hamstring strength and also noted mixed findings. Ardern et al.2 in 2009 noted no differences in isokinetic strength between the 2 groups, but did comment on the decreased hamstring strength in deep flexion that was observed by Tashiro et al.13 after ST-G harvest. Petersen et al.24 in 2014
similarly noted specific cases such as deep knee flexion and internal rotation that may be more affected by gracilis harvest. The prior review did not address the effects of ST versus ST-G harvest on knee laxity or patient-reported outcomes, but the current study finds no evidence of an effect of gracilis harvest on anterior laxity and any difference in patient-reported outcomes is too small to be clinically relevant. Limitations Weaknesses include the variation within the individual studies in regard to a surgical technique, outcome measures, length of follow-up, and study design. These differences are likely reflected in the differences in standard deviations noted among the studies, which were modeled using a random effects statistical model. The inclusion of nonrandomized comparative studies also introduces bias, but the data from the 4 randomized controlled trials included in the analysis are qualitatively similar to those from the nonrandomized comparative studies, minimizing this risk.
Conclusions The addition of gracilis harvest to an isolated semitendinosus harvest for ACL reconstruction results in statistically significant, but likely not clinically relevant differences in isokinetic and isometric hamstring strength as well as patient-reported outcomes. Hamstring strength deficits may be larger at higher flexion angles.
Table 6. Patient-Reported Outcomes IKDC Subjective Author Ardern et al.15 Barenius et al.8 Gobbi et al.17 Inagaki et al.10 Karimi-Mobarakeh et al.9 Segawa et al.12
ST Harvest 94.4 3.6 94.1 3.7 NR NR 62.6 9.5 NR
ST-G Harvest 91.6 8.1 94.5 4.0 NR NR 61.4 8.6 NR
Significance P ¼ .16 P ¼ .80 NR NR P ¼ .47 NR
Lysholm ST Harvest NR NR 92.2 97.0 4.6 55.1 8.9 96.9
ST-G Harvest NR NR 93.6 96.7 5.4 57.4 9.3 96.3
Cincinnati Significance NR NR P ¼ .72 P ¼ .78 P ¼ .26 P > .05
ST Harvest NR NR 84.5 NR NR NR
ST-G Harvest NR NR 82.3 NR NR NR
IKDC, International Knee Documentation Committee; NR, not reported; ST, semitendinusis; ST-G, semitendinosus and gracilis.
Significance NR NR P ¼ .72 NR NR NR
8
A. SHARMA ET AL.
Fig 7. Meta-analysis of the 3 studies reporting subjective International Knee Documentation Committee scores. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
Fig 8. Meta-analysis of the 8 studies reporting instrumented anterior laxity. (CI, confidence interval; df, degree of freedom; IV, inverse variance; SD, standard deviation; ST, semitendinosus; ST-G, semitendinosus and gracilis.)
References 1. Samuelsson K, Andersson D, Ahlden M, Fu FH, Musahl V, Karlsson J. Trends in surgeon preferences on anterior cruciate ligament reconstructive techniques. Clin Sports Med 2013;32:111-126. 2. Ardern CL, Webster KE. Knee flexor strength recovery following hamstring tendon harvest for anterior cruciate ligament reconstruction: A systematic review. Orthop Rev (Pavia) 2009;1:e12. 3. Magnussen RA, Lawrence JT, West RL, Toth AP, Taylor DC, Garrett WE. Graft size and patient age are predictors of early revision after anterior cruciate ligament reconstruction with hamstring autograft. Arthroscopy 2012;28:526-531. 4. Mariscalco MW, Flanigan DC, Mitchell J, et al. The influence of hamstring autograft size on patient-reported outcomes and risk of revision after anterior cruciate ligament reconstruction: A Multicenter Orthopaedic Outcomes Network (MOON) cohort study. Arthroscopy 2013;29:1948-1953. 5. Park SY, Oh H, Park S, Lee JH, Lee SH, Yoon KH. Factors predicting hamstring tendon autograft diameters and resulting failure rates after anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc 2013;21:1111-1118. 6. Lubowitz JH. All-inside anterior cruciate ligament graft link: Graft preparation technique. Arthrosc Tech 2012;1: e165-e168. 7. Adachi N, Ochi M, Uchio Y, Sakai Y, Kuriwaka M, Fujihara A. Harvesting hamstring tendons for ACL reconstruction influences postoperative hamstring muscle performance. Arch Orthop Trauma Surg 2003;123: 460-465.
8. Barenius B, Webster WK, McClelland J, Feller J. Hamstring tendon anterior cruciate ligament reconstruction: Does gracilis tendon harvest matter? Int Orthop 2013;37:207-212. 9. Karimi-Mobarakeh M, Mardani-Kivi M, Mortazavi A, Saheb-Ekhtiari K, Hashemi-Motlagh K. Role of gracilis harvesting in four-strand hamstring tendon anterior cruciate ligament reconstruction: A double-blinded prospective randomized clinical trial. Knee Surg Sports Traumatol Arthrosc 2015;23:1086-1091. 10. Inagaki Y, Kondo E, Kitamura N, et al. Prospective clinical comparisons of semitendinosus versus semitendinosus and gracilis tendon autografts for anatomic double-bundle anterior cruciate ligament reconstruction. J Orthop Sci 2013;18:754-761. 11. Nakamura N, Horibe S, Sasaki S, et al. Evaluation of active knee flexion and hamstring strength after anterior cruciate ligament reconstruction using hamstring tendons. Arthroscopy 2002;18:598-602. 12. Segawa H, Omori G, Koga Y, Kameo T, Iida S, Tanaka M. Rotational muscle strength of the limb after anterior cruciate ligament reconstruction using semitendinosus and gracilis tendon. Arthroscopy 2002;18:177-182. 13. Tashiro T, Kurosawa H, Kawakami A, Hikita A, Fukui N. Influence of medial hamstring tendon harvest on knee flexor strength after anterior cruciate ligament reconstruction. A detailed evaluation with comparison of single- and double-tendon harvest. Am J Sports Med 2003;31: 522-529. 14. Yosmaoglu HB, Baltaci G, Ozer H, Atay A. Effects of additional gracilis tendon harvest on muscle torque, motor coordination, and knee laxity in ACL reconstruction. Knee Surg Sports Traumatol Arthrosc 2011;19:1287-1292.
GRACILIS PRESERVATION IN ACL RECONSTRUCTION 15. Ardern CL, Webster KE, Taylor NF, Feller JA. Hamstring strength recovery after hamstring tendon harvest for anterior cruciate ligament reconstruction: A comparison between graft types. Arthroscopy 2010;26:462-469. 16. Carter TR, Edinger S. Isokinetic evaluation of anterior cruciate ligament reconstruction: Hamstring versus patellar tendon. Arthroscopy 1999;15:169-172. 17. Gobbi A, Domzalski M, Pascual J, Zanazzo M. Hamstring anterior cruciate ligament reconstruction: Is it necessary to sacrifice the gracilis? Arthroscopy 2005;21: 275-280. 18. Lipscomb AB, Johnston RK, Snyder RB, Warburton MJ, Gilbert PP. Evaluation of hamstring strength following use of semitendinosus and gracilis tendons to reconstruct the anterior cruciate ligament. Am J Sports Med 1982;10: 340-342. 19. Greenhalgh T. How to read a paper: The basics of evidence based medicine, Ed 2. London: BMJ Publishing Group, 2001.
9
20. Spindler KP, Kuhn JE, Dunn W, Matthews CE, Harrell FE, Dittus RS. Reading and reviewing the orthopaedic literature: A systematic, evidence-based medicine approach. J Am Acad Orthop Surg 2005;13:220-229. 21. Coleman BD, Khan KM, Maffulli N, Cook JL, Wark JD. Studies of surgical outcome after patellar tendinopathy: Clinical significance of methodological deficiencies and guidelines for future studies. Victorian Institute of Sport Tendon Study Group. Scand J Med Sci Sports 2000;10:2-11. 22. Magnussen RA, Carey JL, Spindler KP. Does autograft choice determine intermediate-term outcome of ACL reconstruction? Knee Surg Sports Traumatol Arthrosc 2011;19:462-472. 23. Collaboration TC. Review manager (RevMan), Ed 5.0. Copenhagen: The Nordic Cochrane Center, 2008. 24. Petersen W, Taheri P, Forkel P, Zantop T. Return to play following ACL reconstruction: A systematic review about strength deficits. Arch Orthop Trauma Surg 2014;134: 1417-1428.
