- Email: [email protected]
Using Magnetic Resonance Imaging to Predict Adequate Graft Diameters for Autologous Hamstring Double-Bundle Anterior Cruciate Ligament Reconstruction
Using Magnetic Resonance Imaging to Predict Adequate Graft Diameters for Autologous Hamstring Double-Bundle Anterior Cruciate Ligament Reconstruction Gregory Wernecke, M.B.B.S., Ian A. Harris, M.B.B.S., F.R.A.C.P.(Ortho), Michael T.W. Houang, M.B.B.S., Bradley G. Seeto, M.B.B.S., F.R.A.C.P.(Ortho), Darren B. Chen, M.B.B.S., F.R.A.C.P.(Ortho), and Samuel J. MacDessi, M.B.B.S., F.R.A.C.P.(Ortho)
Purpose: To determine whether the preoperative magnetic resonance imaging (MRI) cross-sectional area (CSA) of the hamstring tendons can predict intraoperative bundle diameters during doublebundle anterior cruciate ligament reconstruction. Methods: A prospective study of 34 patients undergoing anterior cruciate ligament reconstruction with hamstring autografts was performed. CSAs of independent and combined hamstring tendon diameters were correlated to preoperative magnetic resonance images. Results: Intraoperative tendon diameter measurement positively correlated with preoperative MRI tendon CSA measurement for gracilis (P ⫽ .0006), semitendinosus (P ⫽ .001), and final graft size (P ⫽ .001). Double-stranded gracilis grafts greater than or equal to 5 mm in diameter had a mean preoperative MRI gracilis CSA of 9.98 mm2 compared with a mean of 7.76 mm2 for grafts less than 5 mm (P ⫽ .002). Double-stranded semitendinosus grafts greater than or equal to 6 mm had a mean preoperative MRI tendon CSA of 17.33 mm2 compared with 14.80 mm2 for grafts less than 6 mm (P ⫽ .02). Final grafts of diameter greater than or equal to 7 mm had a mean preoperative MRI total tendon CSA of 26.54 mm2 compared with 22.22 mm2 for grafts under 7 mm (P ⫽ .06). Conclusions: Preoperative MRI is a clinically useful tool to assess hamstring tendon graft diameter. We recommend preoperative CSA threshold values of 10 mm2 and 17 mm2 for the gracilis and semitendinosus tendons, respectively, to reliably predict the potential for a double-bundle anterior cruciate ligament reconstruction. Level of Evidence: Level IV, therapeutic case series.
A
nterior cruciate ligament (ACL) rupture is a common sporting injury and is one of the most commonly reconstructed ligaments because of its importance in knee stability. There has been much debate in
From Sydney Knee Specialists (G.W., B.G.S., D.B.C., S.J.M.), Edgecliff; the South Western Sydney Clinical School, University of New South Wales (I.A.H.), Liverpool; and Castlereagh Imaging (M.T.W.H.), Edgecliff, Australia. The authors report no conflict of interest. Received June 18, 2010; accepted February 24, 2011. Address correspondence to Gregory Wernecke, Sydney Knee Specialists, Level 2, Suite 211 Edgecliff Centre, 203-233 New South Head Road, Edgecliff, New South Wales 2027, Australia. E-mail: [email protected] © 2011 by the Arthroscopy Association of North America 0749-8063/10359/$36.00 doi:10.1016/j.arthro.2011.02.035
the literature over the use of single- or double-bundle reconstruction techniques as a means to provide the greatest benefit to the patient. The addition of a separate posterolateral bundle may contribute to improved rotatory stability of the knee joint1-5 when compared with single-bundle techniques. When hamstring autografts are being used for double-bundle reconstruction, the larger semitendinosus tendon is most often used for the anteromedial bundle, with the smaller gracilis tendon being used for the posterolateral bundle. It is unclear what rotational forces these separate bundles must withstand during normal and sporting activity. As such, there is little evidence to guide the orthopaedic surgeon on what is an acceptable graft diameter for the anteromedial and posterolateral portions of a double-bundle hamstring reconstruction.
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 27, No 8 (August), 2011: pp 1055-1059
1055
1056
G. WERNECKE ET AL.
It has been shown that for single-bundle ACL reconstruction, a minimum graft diameter of 7 mm offers improved outcomes6,7 and re-creates the average native ACL diameter.8 It has been documented that there is a wide variation in graft size from patient to patient.9 It has also been shown that preoperative magnetic resonance imaging (MRI) can accurately predict the combined intraoperative diameter of hamstring tendons for ACL reconstruction in adolescents whose grafts may be under 7 mm.10,11 It has been our experience that the diameter of doubled gracilis tendon in isolation is often too small to consider its use for reconstruction of the posterolateral bundle. It would be of benefit if the surgeon, using preoperative MRI, could predict individual graft diameters before performing a double-bundle hamstring ACL reconstruction. This would assist in surgical planning and allow for adequate informed consent regarding the intervention that would be taking place. The aim of this study was to test for a correlation between preoperative MRI measurements of hamstring tendon cross-sectional areas (CSAs) and intraoperative tendon graft diameters. We hypothesized that we could identify threshold CSAs on preoperative MRI that would reliably predict intraoperative diameters of the gracilis tendon of at least 5 mm, semitendinosus tendon of at least 6 mm, and quadrupled single-bundle graft of at least 7 mm because bundle diameters smaller than these would be undersized.
MRI Technique Proton density fat-suppressed MR images (GE Hdx 1.5 T MRI system; General Electric, Waukesha, WI) were taken with the patient’s leg in full extension. A General Electric ADW 4.2 software-viewing platform was used to measure the CSA of the gracilis and semitendinosus in the axial plane (Fig 1). The chosen level of CSA measurement was at the widest point of the medial femoral epicondyle, which is easily viewed first in the coronal plane and then translated into the axial plane. This is similar to the method described by Bickel et al.11 The described level is where the tendon has a rounded or oval shape and, in our surgical experience, in the region of tendon graft that is passed into the femoral tunnel after being doubled over through an EndoButton Closed Loop device (Smith & Nephew, Andover, MA). A musculoskeletal MR radiologist and orthopaedic surgeon were involved in the measurements of the tendon CSAs. General agreement was reached on what represented the circumference of each tendon. By use of a freehand region-of-interest tool on the viewing platform software, a line was drawn around each tendon. The software then automatically calculated the CSA in square millimeters for each tendon. Any tendon slips or vincula that were seen to be emanating from a tendon were not included in the area of interest, because intraoperatively, these are debrided at the time of graft preparation (Fig 2). Occa-
METHODS This was a prospective study determining the ability of preoperative magnetic resonance (MR) images to predict the independent diameters of the gracilis and semitendinosus tendons in patients undergoing ACL reconstruction. Local ethics committee approval was granted. The preoperative MR images of 37 patients were reviewed. Inclusion criteria were patients who were considered to be skeletally mature with confirmed ACL tears on MR scans. Patients were excluded if they had previous knee surgery other than arthroscopic meniscectomy. We excluded 3 patients: 2 because the MR axial views were inadequate and 1 because a revision ACL reconstruction was performed with contralateral hamstring tendons and an MR scan was not performed on the donor leg. The mean age of the 34 patients included in this analysis was 30 years (range, 15 to 54 years). Male patients outnumbered female patients 25 to 9.
FIGURE 1. Axial MR slice at level of widest point of distal femur with both gracilis tendon and semitendinosus tendon outlined and CSA measured.
USE OF MRI TO PREDICT GRAFT DIAMETER
1057
don diameter when doubled over and the preoperative tendon MR CSA. Correlation was also performed between final intraoperative graft diameter (gracilis and semitendinosus combined and doubled over) and the combined MR CSAs of the gracilis and semitendinosus. We used t tests to describe the relation between MR CSA and whether the doubled gracilis diameter was greater than or equal to 5 mm, whether the doubled semitendinosus diameter was greater than or equal to 6 mm, and whether the total graft diameter was greater than or equal to 7 mm. A P value of .05 was considered statistically significant. FIGURE 2. Axial MR slice of distal femur showing vincula extending from semitendinosus tendon. It should be noted that this is not included in the measurement of CSA.
sionally, notches in a tendon were also seen. These areas were directly crossed and not outlined with the region-of-interest tool (Fig 3). Surgical Procedure Three surgeons who all routinely perform hamstring autograft reconstructions were involved in the analysis. A longitudinal incision was performed over the pes anserinus bursa. The gracilis and semitendinosus tendons were retrieved with Lahey forceps after an incision was made in the investing layer of the sartorial fascia. The tendons were left attached at their distal insertion, and vincula were identified and released. The individual tendons were harvested with a closed-end tendon harvester (Linvatec, Largo, FL), and all muscle fibers were removed. Each tendon was then doubled over around a suture and passed through a closed-hole sizing block. The sizing block measured 0.5-mm incremental diameters from 3 to 12 mm. The diameter of the graft was determined by the narrowest hole the tendon could fit through at its widest point, which was generally the portion wrapped over the suture. It is this portion of the tendon that will be passed into the femoral tunnel to create a snug fit. We then recorded the combined diameter when the gracilis and semitendinosus were placed together to create a standard single-bundle 4-strand graft. All patients in this series then underwent a single-bundle reconstruction.
RESULTS The mean intraoperative gracilis tendon diameter when doubled over was 4.65 mm (SD, 0.57 mm; minimum, 3.5 mm; maximum 6.0 mm), and the mean semitendinosus tendon diameter when doubled was 5.96 mm (SD, 0.64 mm; minimum, 4.5 mm; maximum 7.0 mm). The mean final graft size (doubled gracilis and doubled semitendinosus) was 7.50 mm (SD, 0.59 mm; minimum, 6.5 mm; maximum 8.5 mm). Intraoperative doubled tendon diameter measurement positively correlated with preoperative MR tendon CSA measurement for gracilis (P ⫽ .0006), semitendinosus (P ⫽ .001), and final graft size (P ⫽ .001) (Pearson correlation coefficients in Table 1). We performed t tests comparing gracilis, semitendinosus, and final graft size versus preoperative MR CSA (Table 2). Doubled gracilis grafts greater than or equal to 5 mm in diameter had a mean preoperative MR gracilis CSA of 9.98 mm2 compared with a mean of 7.76 mm2
Statistics Pearson correlation coefficients were calculated to determine the relation between the intraoperative ten-
FIGURE 3. Axial MR slice of distal femur showing notched appearance of semitendinosus tendon.
1058 TABLE 1.
G. WERNECKE ET AL. Correlation of Tendon Size With Preoperative MR CSA
Correlation between intraoperative gracilis size and preoperative gracilis CSA Correlation between intraoperative semitendinosus size and preoperative semitendinosus CSA Correlation between final graft size and preoperative total CSA (gracilis ⫹ semitendinosus)
Pearson Coefficient
P Value
0.56
.0006
0.53
.001
0.53
.001
for grafts less than 5 mm (P ⫽ .002). Looking at gracilis CSA as a predictor, we found that only 1 of 9 grafts in patients with preoperative CSAs of less than 9 mm2 reached 5 mm in diameter whereas 8 of 12 grafts with CSAs of 10 mm2 or higher were 5 mm or greater in diameter. Regarding the predictive value of preoperative gracilis CSA, using a cutoff of at least 10 mm2 resulted in a graft diameter of at least 5 mm in 67% of cases and a diameter of at least 4.5 mm in 100% of cases. A gracilis CSA of less than 10 mm2 resulted in a graft diameter of at least 5 mm in only 23% of cases. Doubled semitendinosus grafts greater than or equal to 6 mm had a mean preoperative MR tendon CSA of 17.33 mm2 compared with 14.8 mm2 for grafts less than 6 mm (P ⫽ .02). Regarding the predictive value of preoperative semitendinosus CSA, using a cutoff of at least 17 mm2 resulted in a graft diameter of at least 6 mm in 87% of cases and a diameter of at least 5 mm in 100% of cases. In tendons less than 17 mm2, only 47% of cases had intraoperative diameters of at least 6 mm. Final graft diameters (doubled gracilis and semitendinosus) greater than or equal to 7 mm had a mean preoperative MR total tendon CSA of 26.54 mm2 compared with 22.22 mm2 for grafts under 7 mm (P ⫽ .06). Regarding the predictive value of total CSA, using a cutoff of at least 22 mm2 resulted in a (4strand) graft diameter of at least 7 mm in 93% of cases and a diameter of at least 6.5 mm in 100% of cases. DISCUSSION The use of preoperative MR scans to estimate hamstring autograft size has been previously reviewed, with mixed results.2,11,12 We found no article in the literature exploring the use of individual gracilis and semitendinosus tendon sizes for the purpose of deter-
mining whether a double-bundle ACL reconstruction was feasible. Hamada et al.2 used preoperative T1-weighted MR images at the level of the joint to evaluate semitendinosus CSAs in 78 patients. They showed a close statistical relation between the MR CSA (6.3 to 15.1 mm2) and the intraoperative CSA (7.1 to 17.0 mm2). The mean MR CSA of the semitendinosus in that cohort was 10.1 mm2 (mean age, 23 years). Similarly, Yasumoto et al.12 used 3-dimensional computed tomography reconstructions to estimate semitendinosus CSA at the level of the joint in 28 patients and calculated a mean area of 11.9 ⫾ 4.6 mm2. Our mean semitendinosus CSA was 16.5 mm2, but this likely represents increased musculotendinous bulk because our measurement was taken at the level of the medial femoral epicondyle. Regarding the use of MRI tendon assessment in single-bundle ACL reconstructions, our results did not show that the combination of gracilis and semitendinosus was unsatisfactory as a quadrupled singlebundle construct. The issue of adequate single-bundle size is more relevant in the adolescent population, where tendons have not fully developed. In an adolescent cohort Bickel et al.11 used T1-weighted MRI to analyze hamstring tendons at the level of the femoral epiphysis. A close approximation was found between preoperative CSA and intraoperative tendon diameter of the 2 tendons doubled together. The authors reported a mean preoperative semitendinosus CSA of 13.3 mm2 in 26 adolescent patients (mean age, 15 years). The mean gracilis CSA was 6.97 mm2; that is nearly 3 mm2 less than our result (9.52 mm2). However, our mean age was double that of the adolescent cohort. We are unaware of any study that correlates individual graft diameters and outcomes in doublebundle reconstruction. We believe that a minimum tunnel diameter of 5 mm is required for the posterolateral bundle and 6 mm for the anteromedial bundle. TABLE 2.
Comparison of Tendon Diameter and CSA
Tendon Diameter Gracilis doubled ⱖ5 mm Gracilis doubled ⬍5 mm Semitendinosus doubled ⱖ6 mm Semitendinosus doubled ⬍6 mm Final ACL graft ⱖ7 mm Final ACL graft ⬍7 mm
Mean CSA (mm2)
Confidence Interval
P Value (t Test)
10.90 8.66 17.33
9.88-11.92 7.74-9.58 16.18-18.49
.002
14.80
12.85-16.75
26.54 22.22
24.89-28.19 19.84-24.60
.02
.06
USE OF MRI TO PREDICT GRAFT DIAMETER Although it is possible to triple or even quadruple hamstring tendons to achieve a minimum graft diameter when performing a double-bundle reconstruction, this not only significantly shortens the tendons, impacting distal aperture fixation, but also is technically more demanding and time-consuming. The other option to maximize individual bundle diameters is to use soft-tissue allografts. However, in many regions of the world, allograft is not readily available and orthopaedic surgeons need to rely on autograft tissue. Our results show that preoperative MR CSA of individual tendon diameters positively correlates with intraoperative tendon measurements. In addition, the mean MR CSA values presented allow reasonably accurate prediction of whether the doubled gracilis tendon diameter would be greater than or equal to 5 mm and, similarly, whether the doubled semitendinosus tendon diameter would be greater than or equal to 6 mm. We recommend preoperative CSA threshold values of 10 mm2 and 17 mm2 for the gracilis and semitendinosus tendons, respectively, to reliably predict the potential for a double-bundle ACL reconstruction. For surgeons who have concerns that inadequate individual graft diameters may compromise the result of a double-bundle hamstring reconstruction, this tool may assist in preoperative planning regarding whether to proceed with this technique or a single-bundle construct.
CONCLUSIONS Preoperative MRI is a clinically useful tool to assess hamstring graft diameter. We recommend preoperative CSA threshold values of 10 mm2 and 17 mm2 for the gracilis and semitendinosus tendons, respectively, to reliably predict the potential for a double-bundle ACL reconstruction.
1059
REFERENCES 1. Markolf KL, Park S, Jackson SR, McAllister DR. Anteriorposterior and rotatory stability of single and double-bundle anterior cruciate ligament reconstructions. J Bone Joint Surg Am 2009;91:107-118. 2. Hamada M, Shino K, Horibe S, et al. Single- versus bi-socket anterior cruciate ligament reconstruction using autogenous multiple-stranded hamstring tendons with EndoButton femoral fixation: A prospective study. Arthroscopy 2001;17:801-807. 3. Yagi M, Kuroda R, Nagamune K, Yoshiya S, Kurosaka M. Double-bundle ACL reconstruction can improve rotational stability. Clin Orthop Relat Res 2007;454:100-107. 4. Jarvela T. Double-bundle versus single-bundle anterior cruciate ligament reconstruction: A prospective, randomized clinical study. Knee Surg Sports Traumatol Arthrosc 2007;15:500507. 5. Yasuda K, Kondo E, Ichiyama H, Tanabe Y, Tohyama H. Clinical evaluation of anatomic double-bundle anterior cruciate ligament reconstruction procedure using hamstring tendon grafts: Comparisons among 3 different procedures. Arthroscopy 2006;22:240-251. 6. Noyes FR, Grood ES. The strength of the anterior cruciate ligament in humans and rhesus monkeys. J Bone Joint Surg Am 1976;58:1074-1082. 7. Grood ES, Walz-Hasselfeld KA, Holden JP, et al. The correlation between anterior-posterior translation and cross-sectional area of anterior cruciate ligament reconstructions. J Orthop Res 1992;10:878-885. 8. Maeda A, Shino K, Horibe S, Nakata K, Buccafusca G. Anterior cruciate ligament reconstruction with multistranded autogenous semitendinosus tendon. Am J Sports Med 1996;24: 504-509. 9. Pagnani MJ, Warner JJ, O’Brien SJ, Warren RF. Anatomic considerations in harvesting the semitendinosus and gracilis tendons and a technique of harvest. Am J Sports Med 1993; 21:565-571. 10. Hamada M, Shino K, Mituoka T, Abe N, Horibe S. Crosssectional area measurement of the semitendinosus tendon for anterior cruciate ligament reconstruction. Arthroscopy 1998; 14:696-701. 11. Bickel BA, Fowler TT, Mowbray JG, et al. Preoperative magnetic resonance imaging cross-sectional area for the measurement of hamstring autograft diameter for reconstruction of the adolescent anterior cruciate ligament. Arthroscopy 2008; 24:1336-1341. 12. Yasumoto M, Deie M, Sunagawa T, et al. Predictive value of preoperative 3-dimensional computer tomography measurement of semitendinosus tendon harvested for anterior cruciate ligament reconstruction. Arthroscopy 2006;22:259-264.
Purpose: To determine whether the preoperative magnetic resonance imaging (MRI) cross-sectional area (CSA) of the hamstring tendons can predict intraoperative bundle diameters during doublebundle anterior cruciate ligament reconstruction. Methods: A prospective study of 34 patients undergoing anterior cruciate ligament reconstruction with hamstring autografts was performed. CSAs of independent and combined hamstring tendon diameters were correlated to preoperative magnetic resonance images. Results: Intraoperative tendon diameter measurement positively correlated with preoperative MRI tendon CSA measurement for gracilis (P ⫽ .0006), semitendinosus (P ⫽ .001), and final graft size (P ⫽ .001). Double-stranded gracilis grafts greater than or equal to 5 mm in diameter had a mean preoperative MRI gracilis CSA of 9.98 mm2 compared with a mean of 7.76 mm2 for grafts less than 5 mm (P ⫽ .002). Double-stranded semitendinosus grafts greater than or equal to 6 mm had a mean preoperative MRI tendon CSA of 17.33 mm2 compared with 14.80 mm2 for grafts less than 6 mm (P ⫽ .02). Final grafts of diameter greater than or equal to 7 mm had a mean preoperative MRI total tendon CSA of 26.54 mm2 compared with 22.22 mm2 for grafts under 7 mm (P ⫽ .06). Conclusions: Preoperative MRI is a clinically useful tool to assess hamstring tendon graft diameter. We recommend preoperative CSA threshold values of 10 mm2 and 17 mm2 for the gracilis and semitendinosus tendons, respectively, to reliably predict the potential for a double-bundle anterior cruciate ligament reconstruction. Level of Evidence: Level IV, therapeutic case series.
A
nterior cruciate ligament (ACL) rupture is a common sporting injury and is one of the most commonly reconstructed ligaments because of its importance in knee stability. There has been much debate in
From Sydney Knee Specialists (G.W., B.G.S., D.B.C., S.J.M.), Edgecliff; the South Western Sydney Clinical School, University of New South Wales (I.A.H.), Liverpool; and Castlereagh Imaging (M.T.W.H.), Edgecliff, Australia. The authors report no conflict of interest. Received June 18, 2010; accepted February 24, 2011. Address correspondence to Gregory Wernecke, Sydney Knee Specialists, Level 2, Suite 211 Edgecliff Centre, 203-233 New South Head Road, Edgecliff, New South Wales 2027, Australia. E-mail: [email protected] © 2011 by the Arthroscopy Association of North America 0749-8063/10359/$36.00 doi:10.1016/j.arthro.2011.02.035
the literature over the use of single- or double-bundle reconstruction techniques as a means to provide the greatest benefit to the patient. The addition of a separate posterolateral bundle may contribute to improved rotatory stability of the knee joint1-5 when compared with single-bundle techniques. When hamstring autografts are being used for double-bundle reconstruction, the larger semitendinosus tendon is most often used for the anteromedial bundle, with the smaller gracilis tendon being used for the posterolateral bundle. It is unclear what rotational forces these separate bundles must withstand during normal and sporting activity. As such, there is little evidence to guide the orthopaedic surgeon on what is an acceptable graft diameter for the anteromedial and posterolateral portions of a double-bundle hamstring reconstruction.
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 27, No 8 (August), 2011: pp 1055-1059
1055
1056
G. WERNECKE ET AL.
It has been shown that for single-bundle ACL reconstruction, a minimum graft diameter of 7 mm offers improved outcomes6,7 and re-creates the average native ACL diameter.8 It has been documented that there is a wide variation in graft size from patient to patient.9 It has also been shown that preoperative magnetic resonance imaging (MRI) can accurately predict the combined intraoperative diameter of hamstring tendons for ACL reconstruction in adolescents whose grafts may be under 7 mm.10,11 It has been our experience that the diameter of doubled gracilis tendon in isolation is often too small to consider its use for reconstruction of the posterolateral bundle. It would be of benefit if the surgeon, using preoperative MRI, could predict individual graft diameters before performing a double-bundle hamstring ACL reconstruction. This would assist in surgical planning and allow for adequate informed consent regarding the intervention that would be taking place. The aim of this study was to test for a correlation between preoperative MRI measurements of hamstring tendon cross-sectional areas (CSAs) and intraoperative tendon graft diameters. We hypothesized that we could identify threshold CSAs on preoperative MRI that would reliably predict intraoperative diameters of the gracilis tendon of at least 5 mm, semitendinosus tendon of at least 6 mm, and quadrupled single-bundle graft of at least 7 mm because bundle diameters smaller than these would be undersized.
MRI Technique Proton density fat-suppressed MR images (GE Hdx 1.5 T MRI system; General Electric, Waukesha, WI) were taken with the patient’s leg in full extension. A General Electric ADW 4.2 software-viewing platform was used to measure the CSA of the gracilis and semitendinosus in the axial plane (Fig 1). The chosen level of CSA measurement was at the widest point of the medial femoral epicondyle, which is easily viewed first in the coronal plane and then translated into the axial plane. This is similar to the method described by Bickel et al.11 The described level is where the tendon has a rounded or oval shape and, in our surgical experience, in the region of tendon graft that is passed into the femoral tunnel after being doubled over through an EndoButton Closed Loop device (Smith & Nephew, Andover, MA). A musculoskeletal MR radiologist and orthopaedic surgeon were involved in the measurements of the tendon CSAs. General agreement was reached on what represented the circumference of each tendon. By use of a freehand region-of-interest tool on the viewing platform software, a line was drawn around each tendon. The software then automatically calculated the CSA in square millimeters for each tendon. Any tendon slips or vincula that were seen to be emanating from a tendon were not included in the area of interest, because intraoperatively, these are debrided at the time of graft preparation (Fig 2). Occa-
METHODS This was a prospective study determining the ability of preoperative magnetic resonance (MR) images to predict the independent diameters of the gracilis and semitendinosus tendons in patients undergoing ACL reconstruction. Local ethics committee approval was granted. The preoperative MR images of 37 patients were reviewed. Inclusion criteria were patients who were considered to be skeletally mature with confirmed ACL tears on MR scans. Patients were excluded if they had previous knee surgery other than arthroscopic meniscectomy. We excluded 3 patients: 2 because the MR axial views were inadequate and 1 because a revision ACL reconstruction was performed with contralateral hamstring tendons and an MR scan was not performed on the donor leg. The mean age of the 34 patients included in this analysis was 30 years (range, 15 to 54 years). Male patients outnumbered female patients 25 to 9.
FIGURE 1. Axial MR slice at level of widest point of distal femur with both gracilis tendon and semitendinosus tendon outlined and CSA measured.
USE OF MRI TO PREDICT GRAFT DIAMETER
1057
don diameter when doubled over and the preoperative tendon MR CSA. Correlation was also performed between final intraoperative graft diameter (gracilis and semitendinosus combined and doubled over) and the combined MR CSAs of the gracilis and semitendinosus. We used t tests to describe the relation between MR CSA and whether the doubled gracilis diameter was greater than or equal to 5 mm, whether the doubled semitendinosus diameter was greater than or equal to 6 mm, and whether the total graft diameter was greater than or equal to 7 mm. A P value of .05 was considered statistically significant. FIGURE 2. Axial MR slice of distal femur showing vincula extending from semitendinosus tendon. It should be noted that this is not included in the measurement of CSA.
sionally, notches in a tendon were also seen. These areas were directly crossed and not outlined with the region-of-interest tool (Fig 3). Surgical Procedure Three surgeons who all routinely perform hamstring autograft reconstructions were involved in the analysis. A longitudinal incision was performed over the pes anserinus bursa. The gracilis and semitendinosus tendons were retrieved with Lahey forceps after an incision was made in the investing layer of the sartorial fascia. The tendons were left attached at their distal insertion, and vincula were identified and released. The individual tendons were harvested with a closed-end tendon harvester (Linvatec, Largo, FL), and all muscle fibers were removed. Each tendon was then doubled over around a suture and passed through a closed-hole sizing block. The sizing block measured 0.5-mm incremental diameters from 3 to 12 mm. The diameter of the graft was determined by the narrowest hole the tendon could fit through at its widest point, which was generally the portion wrapped over the suture. It is this portion of the tendon that will be passed into the femoral tunnel to create a snug fit. We then recorded the combined diameter when the gracilis and semitendinosus were placed together to create a standard single-bundle 4-strand graft. All patients in this series then underwent a single-bundle reconstruction.
RESULTS The mean intraoperative gracilis tendon diameter when doubled over was 4.65 mm (SD, 0.57 mm; minimum, 3.5 mm; maximum 6.0 mm), and the mean semitendinosus tendon diameter when doubled was 5.96 mm (SD, 0.64 mm; minimum, 4.5 mm; maximum 7.0 mm). The mean final graft size (doubled gracilis and doubled semitendinosus) was 7.50 mm (SD, 0.59 mm; minimum, 6.5 mm; maximum 8.5 mm). Intraoperative doubled tendon diameter measurement positively correlated with preoperative MR tendon CSA measurement for gracilis (P ⫽ .0006), semitendinosus (P ⫽ .001), and final graft size (P ⫽ .001) (Pearson correlation coefficients in Table 1). We performed t tests comparing gracilis, semitendinosus, and final graft size versus preoperative MR CSA (Table 2). Doubled gracilis grafts greater than or equal to 5 mm in diameter had a mean preoperative MR gracilis CSA of 9.98 mm2 compared with a mean of 7.76 mm2
Statistics Pearson correlation coefficients were calculated to determine the relation between the intraoperative ten-
FIGURE 3. Axial MR slice of distal femur showing notched appearance of semitendinosus tendon.
1058 TABLE 1.
G. WERNECKE ET AL. Correlation of Tendon Size With Preoperative MR CSA
Correlation between intraoperative gracilis size and preoperative gracilis CSA Correlation between intraoperative semitendinosus size and preoperative semitendinosus CSA Correlation between final graft size and preoperative total CSA (gracilis ⫹ semitendinosus)
Pearson Coefficient
P Value
0.56
.0006
0.53
.001
0.53
.001
for grafts less than 5 mm (P ⫽ .002). Looking at gracilis CSA as a predictor, we found that only 1 of 9 grafts in patients with preoperative CSAs of less than 9 mm2 reached 5 mm in diameter whereas 8 of 12 grafts with CSAs of 10 mm2 or higher were 5 mm or greater in diameter. Regarding the predictive value of preoperative gracilis CSA, using a cutoff of at least 10 mm2 resulted in a graft diameter of at least 5 mm in 67% of cases and a diameter of at least 4.5 mm in 100% of cases. A gracilis CSA of less than 10 mm2 resulted in a graft diameter of at least 5 mm in only 23% of cases. Doubled semitendinosus grafts greater than or equal to 6 mm had a mean preoperative MR tendon CSA of 17.33 mm2 compared with 14.8 mm2 for grafts less than 6 mm (P ⫽ .02). Regarding the predictive value of preoperative semitendinosus CSA, using a cutoff of at least 17 mm2 resulted in a graft diameter of at least 6 mm in 87% of cases and a diameter of at least 5 mm in 100% of cases. In tendons less than 17 mm2, only 47% of cases had intraoperative diameters of at least 6 mm. Final graft diameters (doubled gracilis and semitendinosus) greater than or equal to 7 mm had a mean preoperative MR total tendon CSA of 26.54 mm2 compared with 22.22 mm2 for grafts under 7 mm (P ⫽ .06). Regarding the predictive value of total CSA, using a cutoff of at least 22 mm2 resulted in a (4strand) graft diameter of at least 7 mm in 93% of cases and a diameter of at least 6.5 mm in 100% of cases. DISCUSSION The use of preoperative MR scans to estimate hamstring autograft size has been previously reviewed, with mixed results.2,11,12 We found no article in the literature exploring the use of individual gracilis and semitendinosus tendon sizes for the purpose of deter-
mining whether a double-bundle ACL reconstruction was feasible. Hamada et al.2 used preoperative T1-weighted MR images at the level of the joint to evaluate semitendinosus CSAs in 78 patients. They showed a close statistical relation between the MR CSA (6.3 to 15.1 mm2) and the intraoperative CSA (7.1 to 17.0 mm2). The mean MR CSA of the semitendinosus in that cohort was 10.1 mm2 (mean age, 23 years). Similarly, Yasumoto et al.12 used 3-dimensional computed tomography reconstructions to estimate semitendinosus CSA at the level of the joint in 28 patients and calculated a mean area of 11.9 ⫾ 4.6 mm2. Our mean semitendinosus CSA was 16.5 mm2, but this likely represents increased musculotendinous bulk because our measurement was taken at the level of the medial femoral epicondyle. Regarding the use of MRI tendon assessment in single-bundle ACL reconstructions, our results did not show that the combination of gracilis and semitendinosus was unsatisfactory as a quadrupled singlebundle construct. The issue of adequate single-bundle size is more relevant in the adolescent population, where tendons have not fully developed. In an adolescent cohort Bickel et al.11 used T1-weighted MRI to analyze hamstring tendons at the level of the femoral epiphysis. A close approximation was found between preoperative CSA and intraoperative tendon diameter of the 2 tendons doubled together. The authors reported a mean preoperative semitendinosus CSA of 13.3 mm2 in 26 adolescent patients (mean age, 15 years). The mean gracilis CSA was 6.97 mm2; that is nearly 3 mm2 less than our result (9.52 mm2). However, our mean age was double that of the adolescent cohort. We are unaware of any study that correlates individual graft diameters and outcomes in doublebundle reconstruction. We believe that a minimum tunnel diameter of 5 mm is required for the posterolateral bundle and 6 mm for the anteromedial bundle. TABLE 2.
Comparison of Tendon Diameter and CSA
Tendon Diameter Gracilis doubled ⱖ5 mm Gracilis doubled ⬍5 mm Semitendinosus doubled ⱖ6 mm Semitendinosus doubled ⬍6 mm Final ACL graft ⱖ7 mm Final ACL graft ⬍7 mm
Mean CSA (mm2)
Confidence Interval
P Value (t Test)
10.90 8.66 17.33
9.88-11.92 7.74-9.58 16.18-18.49
.002
14.80
12.85-16.75
26.54 22.22
24.89-28.19 19.84-24.60
.02
.06
USE OF MRI TO PREDICT GRAFT DIAMETER Although it is possible to triple or even quadruple hamstring tendons to achieve a minimum graft diameter when performing a double-bundle reconstruction, this not only significantly shortens the tendons, impacting distal aperture fixation, but also is technically more demanding and time-consuming. The other option to maximize individual bundle diameters is to use soft-tissue allografts. However, in many regions of the world, allograft is not readily available and orthopaedic surgeons need to rely on autograft tissue. Our results show that preoperative MR CSA of individual tendon diameters positively correlates with intraoperative tendon measurements. In addition, the mean MR CSA values presented allow reasonably accurate prediction of whether the doubled gracilis tendon diameter would be greater than or equal to 5 mm and, similarly, whether the doubled semitendinosus tendon diameter would be greater than or equal to 6 mm. We recommend preoperative CSA threshold values of 10 mm2 and 17 mm2 for the gracilis and semitendinosus tendons, respectively, to reliably predict the potential for a double-bundle ACL reconstruction. For surgeons who have concerns that inadequate individual graft diameters may compromise the result of a double-bundle hamstring reconstruction, this tool may assist in preoperative planning regarding whether to proceed with this technique or a single-bundle construct.
CONCLUSIONS Preoperative MRI is a clinically useful tool to assess hamstring graft diameter. We recommend preoperative CSA threshold values of 10 mm2 and 17 mm2 for the gracilis and semitendinosus tendons, respectively, to reliably predict the potential for a double-bundle ACL reconstruction.
1059
REFERENCES 1. Markolf KL, Park S, Jackson SR, McAllister DR. Anteriorposterior and rotatory stability of single and double-bundle anterior cruciate ligament reconstructions. J Bone Joint Surg Am 2009;91:107-118. 2. Hamada M, Shino K, Horibe S, et al. Single- versus bi-socket anterior cruciate ligament reconstruction using autogenous multiple-stranded hamstring tendons with EndoButton femoral fixation: A prospective study. Arthroscopy 2001;17:801-807. 3. Yagi M, Kuroda R, Nagamune K, Yoshiya S, Kurosaka M. Double-bundle ACL reconstruction can improve rotational stability. Clin Orthop Relat Res 2007;454:100-107. 4. Jarvela T. Double-bundle versus single-bundle anterior cruciate ligament reconstruction: A prospective, randomized clinical study. Knee Surg Sports Traumatol Arthrosc 2007;15:500507. 5. Yasuda K, Kondo E, Ichiyama H, Tanabe Y, Tohyama H. Clinical evaluation of anatomic double-bundle anterior cruciate ligament reconstruction procedure using hamstring tendon grafts: Comparisons among 3 different procedures. Arthroscopy 2006;22:240-251. 6. Noyes FR, Grood ES. The strength of the anterior cruciate ligament in humans and rhesus monkeys. J Bone Joint Surg Am 1976;58:1074-1082. 7. Grood ES, Walz-Hasselfeld KA, Holden JP, et al. The correlation between anterior-posterior translation and cross-sectional area of anterior cruciate ligament reconstructions. J Orthop Res 1992;10:878-885. 8. Maeda A, Shino K, Horibe S, Nakata K, Buccafusca G. Anterior cruciate ligament reconstruction with multistranded autogenous semitendinosus tendon. Am J Sports Med 1996;24: 504-509. 9. Pagnani MJ, Warner JJ, O’Brien SJ, Warren RF. Anatomic considerations in harvesting the semitendinosus and gracilis tendons and a technique of harvest. Am J Sports Med 1993; 21:565-571. 10. Hamada M, Shino K, Mituoka T, Abe N, Horibe S. Crosssectional area measurement of the semitendinosus tendon for anterior cruciate ligament reconstruction. Arthroscopy 1998; 14:696-701. 11. Bickel BA, Fowler TT, Mowbray JG, et al. Preoperative magnetic resonance imaging cross-sectional area for the measurement of hamstring autograft diameter for reconstruction of the adolescent anterior cruciate ligament. Arthroscopy 2008; 24:1336-1341. 12. Yasumoto M, Deie M, Sunagawa T, et al. Predictive value of preoperative 3-dimensional computer tomography measurement of semitendinosus tendon harvested for anterior cruciate ligament reconstruction. Arthroscopy 2006;22:259-264.