Surg Radiol Anat DOI 10.1007/s00276-015-1486-8
ORIGINAL ARTICLE
Mapping the quadriceps tendon: an anatomic and morphometric study to guide tendon harvesting Damien Potage1 • Fabrice Duparc1 • Amaury D’Utruy1 • Olivier Courage1 Xavier Roussignol1
•
Received: 5 March 2015 / Accepted: 28 April 2015 Ó Springer-Verlag France 2015
Abstract Introduction The quadriceps tendon (QT) may be used in first-line knee ligament surgery (Anterior and Posterior Cruciate Ligaments), surgical revision and multiligament surgery. There are few published anatomic guides to QT harvesting. The present anatomic study sought to determine the ideal harvesting site and examined possible correlation between patellar and graft sizes. Materials and methods A descriptive morphometric anatomic study was performed on 12 cadaveric knees. The reference anatomical landmark was the center of the superior edge of the patella. The QT was dissected and sliced longitudinally into five 5-mm strips. The central strip corresponded to the anatomic center of the patella. QT thickness was measured every 10 mm over a length of 100 mm. Data were analyzed on Pearson correlation test and Student, Bartlett and Fisher tests (a risk = 0.05). Results QT thickness ranged from 0.7 to 9.78 mm, for a mean 4.94 mm. Mean thickness in the lateral, central and medial strips was, respectively, 3.464, 6.040 and 3.899 mm. Central and centromedial strips were thicker than medial, centrolateral and lateral strips; central and centromedial strips were similar at, respectively, 6.040 and 6.041 mm (non-significant: p = 0.95), and significantly thicker than lateral strips. QT thickness showed significant correlation with patellar length (r = 0.75; p = 0.0048; 95 % CI [?0.31; ?0.93]). Conclusion The present anatomical study confirmed that QT should be harvested from the central and centromedial & Fabrice Duparc [email protected] 1
Faculte´ de Me´decine Pharmacie, Rouen University, Rouen Cedex 1, France
regions. Mean thickness was 7.84 mm at the patellar insertion, 7.37 mm at 20 mm from the insertion, 6.41 at 40 mm, 5.61 at 60 mm and 4.33 at 100 mm. Keywords Quadriceps tendon Knee surgery Ligamentoplasty Cruciate ligaments repair
Introduction Ligament injuries of the knee are common. Their management justifies the use of single or multiple transplants. The most commonly used grafts are those of the patellar ligament (PL) and the hamstrings (semitendinosus and/or gracilis), to a lesser extent the fascia lata and quadriceps tendon (QT). There are few anatomical studies that describe the ideal sampling site for quadriceps Tendon. The main objective of the study was to perform a descriptive and morphometric study with high clinical correlation of QT to optimize and standardize the sufficient autograft harvesting technique in knee ligament surgery and reduce the risk of unsatisfactory grafts. The second objective was to try to establish a correlation between general morphological characteristics such as patella and QT.
Materials and methods A descriptive anatomical and morphometric study was performed at the Anatomy Laboratory of the Faculty of Medicine of Rouen University. Twelve freshly preserved adult knees from donated bodies to science were examined for this study. There were 7 male and 5 female knees used. Known demographics including age and gender were
123
Surg Radiol Anat
Fig. 1 a Determining the center of the quadriceps tendon (mean value from 3 measurements). Section of the quadriceps tendon (QT) into 5 bands: 1 medial; 2 centromedial; 3 central; 4 contralateral; 5 lateral. VL vastus lateralis, VM vastus medialis, PL patellar ligament;
b section of the quadriceps tendon with the 6-blade device. The central band is centered on the central landmark. c The quadriceps tendon is divided in 5 bands (5 mm width)
recorded. The mean age of the 12 dissected knees was 80 years. All knees were without prior surgery, free of scars, and had a complete range of normal mobility (0° extension, greater than 120° bending). Skin and subcutaneous tissues were removed to expose the extensor system, the anterior surface of the patella, the QT and the anterior face with muscle fibers of the quadriceps femoris muscle (QFM). The QT was removed in a single block with the attached patella, subsequently the patellar ligament, medial and lateral patellar retinacula, meniscopatellar ligaments, capsule and different parts of the QFM were detached in muscle zones away from the musculotendinous junction. The patellar insertion of the QFM on the patella was identified. The center of the quadriceps tendon was defined using three successive measurements of the width of the patella. The average of the 3 measurements allowed us to determine the center of the QT. Five bands of 5 mm wide in the axis of the tendon were removed using a specific ancillary with 6 blades. The central band was centered on the center mark, which was taken from the center of the quadriceps tendon. Each band was identified from the medial border of the patella; the medial band, centromedial band, central band, centrolateral band and lateral band. (Figure 1) We measured the thickness of the 5 QT bands every 10 mm over the 100 mm lengths. All measurements were performed using a calibrated rigid ruler every 10 mm to maintain and fix the tendon. The thickness of the ruler was to 2.80 mm. If muscle fibers were present on the bands
of QT, they were previously excised to allow accurate measurement of single tendon fibers. For measurements of the patella, the entire soft tissues were resected to allow accurate measurements of height and width (Fig. 2). All measurements were taken in the same manner by the same investigator for all 12 knees. All measurements were made to the nearest 0.01 mm with a digital caliper. Data were collected on Microsoft Excel 2007 (12.0.4518.1014) Office spreadsheet and were analyzed using Pearson correlation tests, Student’s test, Bartlett test, Fisher test and Mann–Whitney test (a = 0.05 risked) using the R software 2.15.
123
Results The average tendons thickness was 4.94 mm (0.7–9.78 mm). The average thickness of the lateral, central and medial bands was, respectively, 3.46, 6.04 and 3.90 mm. The thickness of the QT bands decreased in 2 ways depending on their location. The central, centromedial and medial bands thickness decreased linearly unlike the centrolateral and lateral bands. There was an additional thickness corresponding to a tendinous expansion, which was located about 50–60 mm from the upper edge of the patella to the centro-sideband and 60–80 mm for the sideband (Fig. 3). This additional thickness corresponded to a fibrous overlayer of the vastus lateralis. This expansion
Surg Radiol Anat
Fig. 4 Distribution of the mean values of the thickness of the five tendinous bands Fig. 2 a, b Measurements of the thickness of the quadriceps tendon with the electronic caliper (MitutoyoR), precision 0.01 mm
Fig. 3 Linear decreasing of the thickness of the tendinous bands: medial (red), centromedial (Brown), Central (Black), centrolateral (dark blue), lateral (light blue). The thickness is majored due to a tendinous expansion, at 50–60 mm from the upper edge of the patella for the contralateral band (dark blue), and 60–80 mm for the lateral band (light blue)
was gradual and off center moving towards the top and outside. This tendinous thickening was mostly visible on the posterior surface of the QT.
The central band and centromedial band had a linear decreasing profile with similar interindividual variance. The average values of these measurements allowed us to model the decreasing tendon thickness of each band (Fig. 3). The Kruskal–Wallis test had showed that all the bands were not identical (p = 0.0021). The Mann–Whitney test was used to compare interindividual thicknesses confirming that the centromedial band was statistically thicker than the lateral and medial bands. Central and centromedial strips were not significantly different in terms of profile or thickness. They were significantly thicker than the lateral band. For centrolateral and medial bands, there were no significant trends. The centromedial bands were not significantly thicker than the central bands (p = 0.95). The medial bands were significantly thinner than the centromedial bands (p = 0.04). In the box diagram (Fig. 4), we observed that for each band, the interindividual variance was low except for the medial bands that had a high interindividual variance, which was significant (Fisher’s test). This figure confirms that the central and centromedial bands were thicker than the other bands, and measured on average 6.04 mm. Of all 12 analyzed tendons, the central and centromedial bands included in their proximal extension the tendon of the rectus femoris muscle. This individually identifiable muscle became increasingly sinewy the closer it was to patellar insertion. At the distal end, the tendon flattened and formed a progressive fiber blade making it
123
Surg Radiol Anat Table 1 Coefficients of correlation between the mean thickness of the tendon and the anthropometric data Correlation with the thickness (Pearson), q
Level of significance
CI 95 %
Age
q = -0.37
p = 0.28
[- 0.81; ?0.33]
Size of the body
q = 0.53
p = 0.074
[- 0.06; ?0.85]
Height of the patella
q = 0.75
p = 0.0048
[? 0.31; ?0.93]
Width of the patella
q = 0.52
p = 0.08
[- 0.08; ?0.84]
indistinguishable from the fibrous patellar insertion of the 3 vastus muscles. The tendon of the rectus femoris muscle corresponded to the superficial layer of the QT. The Pearson correlation tests (Table 1) allowed us to demonstrate a correlation between the thickness of the tendon and the height of the patella which was statistically significant (r = 0.75, p = 0.0048, 95 % CI [0.31; 0.93]). For the width of the patella and the size of the individual, the degree of significance was between 0.05 and 0.10.
Discussion The QT is a site of tendon harvesting which is not frequently exploited but can be used to reconstruct the ACL [3, 4, 8, 9], the PCL, the medial patellofemoral ligament (MPFL), and the lateral collateral ligament [1]. The use of QT was originally described by Marshall et al. [2]. This autograft shares biological and mechanical properties with the LP, sometimes with superiority. Its harvest has a lower morbidity than using PL [4] and hamstrings. Patients relate no loss of flexion, less anterior pain, decrease or absence of patella fracturing in the case of using grafts with patellar bone, and the absence of paresthesia on the outer edge of the knee, the sensitive area of the medial saphenous nerve. The QT may include a patellar bone fragment [4] or be free [9]. It can be used in single bundle [4] or double bundle [3]. Few articles discussed the tendinous transplant and its preparation with fixed and reproducible anatomical landmarks; they did not specify the exact size and morphometric characteristics of autografts. Several authors [1, 2, 4] described a technique within the sample seemed to be harvested with only one landmark, the patella center. Our study showed that the autografts of QT should be harvested at the central and centromedial bands to maintain a maximum thickness. In several studies, the graft transplant was measured 10 mm in width, with or without patellar bone plug. An incision of just a few centimeters long could be made in relation to the patellar insertion of the QT. This was sufficient to harvest the graft due to the elasticity of the skin [3]. The insertion of the QT was
123
located; the medial and lateral borders of the patella were palpated to determine the center of the QT. Once the center of the patella was located, we could harvest the central and centromedial strips. Geib et al. [3] and Harris et al. [5] have described a fat plane in the QT during the dissection. This fat plane corresponded to the limit between the common tendon of the large lateral and medial muscles, vastus intermedius and vastus medialis, and the tendon of the rectus femoris muscle. We observed this fat plan in our 12 dissected cadaveric knees. This fat plan inside the QT could not be individually identified at the patellar insertion. Harris et al. [5] have measured the division between the two tendons at 6 cm proximal to the insertion of the patella. As the QT is divided into two layers, the graft required a thorough preparation with proximal sewing of the two strands to allow its implementation in tunnels. Vastus medialis and vastus lateralis tendons behave like two fiber blades that reinforce the thickness of QT. The centrolateral and lateral bands showed a non-linear decrease for all the 12 tendons with a tendon thickening which was measured at 50–60 mm from the proximal edge of the patella for the centrolateral band and 60–80 mm for the lateral band. Central and centromedial bands were at mean thicker than all other bands. The analyses of the 12 tendons showed that these two bands included the distal tendon part of the rectus femoris muscle. The quadriceps tendon is considered as trilaminar, with a ventral part coming from the rectus femoris, a dorsal part coming from the vastus intermedius, and a thin fatty layer placed between this two tendinous layers. This aspect was not confirmed in the comprehensive review by Slone et al. [10]. The rectus femoris muscle at the junction between middle and distal thirds became tendinous with a large tendinous insertion on the upper pole of the patella. Lippe et al. [6] confirmed in their anatomical study the observations by Harris et al. [5] who had described an asymmetric surface of the QT with a lower insertion of the vastus medialis compared to the vastus lateralis. Lippe et al. [6] proposed a reference point for guiding the graft harvesting of QT. This point corresponded to the greatest length of the
Surg Radiol Anat
tendon visible at the surface of QT where the largest thickness is measured. This point was at mean at 61.6 % of the width of the insertion of the QT from the medial edge of the patella. Lippe et al. [6] recommended taking this reference point as the lateral edge of the graft to obtain a sufficient thickness. In their MRI study, Xerogeanes et al. [11] showed that the QT could provide a significantly bigger graft than the PT which measured 60 or 80 mm. Moreover, they showed a thickness uniformity along the first 60 mm of QT above the insertion, and explained that the 2.5 cm of graft which is intra-articular is much thicker. In our study, we observed a gradual decrease on the same portion. Xerogeanes et al. [11] and Harris et al. [5] have both measured QT, respectively, at 88 % and 1.8 times bigger than the PT. Xerogeanes et al. [11] found a correlation between the size of the patient and the length of the graft. Our study showed a correlation between the thickness of the tendon and the height of the patella. The width of the patella and the size of the individual had a degree of significance between 0.05 and 0.10. The risk of rupture of the quadriceps tendon after harvesting of the median band is not clear. No true rupture was described, but rectus femoris retraction when the tendon harvesting is extended to the myotendinous junction [10]. According to Lund et al, the quadriceps tendon has been considered as choice graft, the risk of rupture of the extensor system has not been cited [7]. Our study confirmed that if the surgeon wants a thicker graft, he must harvest only the central and centromedial bands of the QT. Our results were similar to the recommendations of Lippe et al. [6] who advocated harvesting the medial part of the quadriceps tendon peak which is 61.6 % from the medial border of the QT to have the longest graft. Xerogeanes et al. [11] suggested a more lateral harvesting of the autograft, by placing it in the middle of the patella and taking 5 mm on either side.
Conclusion This anatomical study has clarified the morphometric mapping of the thickness of QT based on the medial and lateral edges of the patella. This allows us to harvest the tendinous autograft with reliable anatomical landmarks taken from the patella edges. The central and centromedial bands are thicker. The average thickness of the transplant can be anticipated based on the distance from the upper
edge of the patella and the location relative to the center of the patella. We are then able to manage the dimensions of the tendinous transplant in relation to the ligament surgery required. Acknowledgments None of the authors has any conflict of interest related to the topic of this study. Ethical standard France.
This study complies with the laws in force in
References 1. Chen CH, Chen WJ, Shih CH (2001) Lateral collateral ligament reconstruction using quadriceps tendon–patellar bone autograft with bioscrew fixation. Arthrosc J Arthrosc Relat Surg 17(5):551–554 2. Fulkerson JP, Langeland R (1995) An alternative cruciate reconstruction graft: the central quadriceps tendon. Arthrosc J Arthrosc Relat Surg 11(2):252–254 3. Geib TM, Shelton WR, Phelps RA, Clark L (2009) Anterior cruciate ligament reconstruction using quadriceps tendon autograft: intermediate-term outcome. Arthrosc J Arthrosc Relat Surg 25(12):1408–1414 4. Han HS, Seong SC, Lee S, Lee MC (2008) Anterior cruciate ligament reconstruction: quadriceps versus patellar autograft. Clin Orthop Rel Res 466(1):198–204 5. Harris NL, Smith DAB, Lamoreaux L, Purnell M (1997) Central quadriceps tendon for anterior cruciate ligament reconstruction: Part I: morphometric and biomechanical evaluation. Am J Sports Med 25(1):23–28 6. Lippe J, Armstrong A, Fulkerson JP (2012) Anatomic guidelines for harvesting a quadriceps free tendon autograft for anterior cruciate ligament reconstruction. Arthrosc J Arthrosc Relat Surg Off Publ Arthrosc Assoc N Am Int Arthrosc Assoc 28(7):980–984 7. Lund B, Nielsen T, Phys BA, Faun P, Christiansen SE, Lind M (2014) Is quadriceps tendon a better graft choice than patellar tendon? A prospective randomized study. Arthroscopy 30(5):593–598 8. Marshall JL, Warren RF, Wickiewicz TL, Reider B (1979) The anterior cruciate ligament: a technique of repair and reconstruction. Clin Orthop Rel Res 143:97–106 9. Schulz AP, Lange V, Gille J, Voigt C, Fro¨hlich S, Stuhr M et al (2013) Anterior cruciate ligament reconstruction using bone plugfree quadriceps tendon autograft: intermediate-term clinical outcome after 24-36 months. Open Access J Sports Med 4:243–249 10. Slone HS, Romine SE, Premkumar A, Xerogeanes JW (2015) Quadriceps tendon autograft for anterior cruciate ligament reconstruction: a comprehensive review of current literature and systematic review of clinical results. Arthroscopy 31(3):541–554 11. Xerogeanes JW, Mitchell PM, Karasev PA, Kolesov IA, Romine SE (2013) Anatomic and morphological evaluation of the quadriceps tendon using 3-Dimensional magnetic resonance imaging reconstruction. Applications for anterior cruciate ligament autograft choice and procurement. Am J Sports Med 41(10):2392–2399
123
ORIGINAL ARTICLE
Mapping the quadriceps tendon: an anatomic and morphometric study to guide tendon harvesting Damien Potage1 • Fabrice Duparc1 • Amaury D’Utruy1 • Olivier Courage1 Xavier Roussignol1
•
Received: 5 March 2015 / Accepted: 28 April 2015 Ó Springer-Verlag France 2015
Abstract Introduction The quadriceps tendon (QT) may be used in first-line knee ligament surgery (Anterior and Posterior Cruciate Ligaments), surgical revision and multiligament surgery. There are few published anatomic guides to QT harvesting. The present anatomic study sought to determine the ideal harvesting site and examined possible correlation between patellar and graft sizes. Materials and methods A descriptive morphometric anatomic study was performed on 12 cadaveric knees. The reference anatomical landmark was the center of the superior edge of the patella. The QT was dissected and sliced longitudinally into five 5-mm strips. The central strip corresponded to the anatomic center of the patella. QT thickness was measured every 10 mm over a length of 100 mm. Data were analyzed on Pearson correlation test and Student, Bartlett and Fisher tests (a risk = 0.05). Results QT thickness ranged from 0.7 to 9.78 mm, for a mean 4.94 mm. Mean thickness in the lateral, central and medial strips was, respectively, 3.464, 6.040 and 3.899 mm. Central and centromedial strips were thicker than medial, centrolateral and lateral strips; central and centromedial strips were similar at, respectively, 6.040 and 6.041 mm (non-significant: p = 0.95), and significantly thicker than lateral strips. QT thickness showed significant correlation with patellar length (r = 0.75; p = 0.0048; 95 % CI [?0.31; ?0.93]). Conclusion The present anatomical study confirmed that QT should be harvested from the central and centromedial & Fabrice Duparc [email protected] 1
Faculte´ de Me´decine Pharmacie, Rouen University, Rouen Cedex 1, France
regions. Mean thickness was 7.84 mm at the patellar insertion, 7.37 mm at 20 mm from the insertion, 6.41 at 40 mm, 5.61 at 60 mm and 4.33 at 100 mm. Keywords Quadriceps tendon Knee surgery Ligamentoplasty Cruciate ligaments repair
Introduction Ligament injuries of the knee are common. Their management justifies the use of single or multiple transplants. The most commonly used grafts are those of the patellar ligament (PL) and the hamstrings (semitendinosus and/or gracilis), to a lesser extent the fascia lata and quadriceps tendon (QT). There are few anatomical studies that describe the ideal sampling site for quadriceps Tendon. The main objective of the study was to perform a descriptive and morphometric study with high clinical correlation of QT to optimize and standardize the sufficient autograft harvesting technique in knee ligament surgery and reduce the risk of unsatisfactory grafts. The second objective was to try to establish a correlation between general morphological characteristics such as patella and QT.
Materials and methods A descriptive anatomical and morphometric study was performed at the Anatomy Laboratory of the Faculty of Medicine of Rouen University. Twelve freshly preserved adult knees from donated bodies to science were examined for this study. There were 7 male and 5 female knees used. Known demographics including age and gender were
123
Surg Radiol Anat
Fig. 1 a Determining the center of the quadriceps tendon (mean value from 3 measurements). Section of the quadriceps tendon (QT) into 5 bands: 1 medial; 2 centromedial; 3 central; 4 contralateral; 5 lateral. VL vastus lateralis, VM vastus medialis, PL patellar ligament;
b section of the quadriceps tendon with the 6-blade device. The central band is centered on the central landmark. c The quadriceps tendon is divided in 5 bands (5 mm width)
recorded. The mean age of the 12 dissected knees was 80 years. All knees were without prior surgery, free of scars, and had a complete range of normal mobility (0° extension, greater than 120° bending). Skin and subcutaneous tissues were removed to expose the extensor system, the anterior surface of the patella, the QT and the anterior face with muscle fibers of the quadriceps femoris muscle (QFM). The QT was removed in a single block with the attached patella, subsequently the patellar ligament, medial and lateral patellar retinacula, meniscopatellar ligaments, capsule and different parts of the QFM were detached in muscle zones away from the musculotendinous junction. The patellar insertion of the QFM on the patella was identified. The center of the quadriceps tendon was defined using three successive measurements of the width of the patella. The average of the 3 measurements allowed us to determine the center of the QT. Five bands of 5 mm wide in the axis of the tendon were removed using a specific ancillary with 6 blades. The central band was centered on the center mark, which was taken from the center of the quadriceps tendon. Each band was identified from the medial border of the patella; the medial band, centromedial band, central band, centrolateral band and lateral band. (Figure 1) We measured the thickness of the 5 QT bands every 10 mm over the 100 mm lengths. All measurements were performed using a calibrated rigid ruler every 10 mm to maintain and fix the tendon. The thickness of the ruler was to 2.80 mm. If muscle fibers were present on the bands
of QT, they were previously excised to allow accurate measurement of single tendon fibers. For measurements of the patella, the entire soft tissues were resected to allow accurate measurements of height and width (Fig. 2). All measurements were taken in the same manner by the same investigator for all 12 knees. All measurements were made to the nearest 0.01 mm with a digital caliper. Data were collected on Microsoft Excel 2007 (12.0.4518.1014) Office spreadsheet and were analyzed using Pearson correlation tests, Student’s test, Bartlett test, Fisher test and Mann–Whitney test (a = 0.05 risked) using the R software 2.15.
123
Results The average tendons thickness was 4.94 mm (0.7–9.78 mm). The average thickness of the lateral, central and medial bands was, respectively, 3.46, 6.04 and 3.90 mm. The thickness of the QT bands decreased in 2 ways depending on their location. The central, centromedial and medial bands thickness decreased linearly unlike the centrolateral and lateral bands. There was an additional thickness corresponding to a tendinous expansion, which was located about 50–60 mm from the upper edge of the patella to the centro-sideband and 60–80 mm for the sideband (Fig. 3). This additional thickness corresponded to a fibrous overlayer of the vastus lateralis. This expansion
Surg Radiol Anat
Fig. 4 Distribution of the mean values of the thickness of the five tendinous bands Fig. 2 a, b Measurements of the thickness of the quadriceps tendon with the electronic caliper (MitutoyoR), precision 0.01 mm
Fig. 3 Linear decreasing of the thickness of the tendinous bands: medial (red), centromedial (Brown), Central (Black), centrolateral (dark blue), lateral (light blue). The thickness is majored due to a tendinous expansion, at 50–60 mm from the upper edge of the patella for the contralateral band (dark blue), and 60–80 mm for the lateral band (light blue)
was gradual and off center moving towards the top and outside. This tendinous thickening was mostly visible on the posterior surface of the QT.
The central band and centromedial band had a linear decreasing profile with similar interindividual variance. The average values of these measurements allowed us to model the decreasing tendon thickness of each band (Fig. 3). The Kruskal–Wallis test had showed that all the bands were not identical (p = 0.0021). The Mann–Whitney test was used to compare interindividual thicknesses confirming that the centromedial band was statistically thicker than the lateral and medial bands. Central and centromedial strips were not significantly different in terms of profile or thickness. They were significantly thicker than the lateral band. For centrolateral and medial bands, there were no significant trends. The centromedial bands were not significantly thicker than the central bands (p = 0.95). The medial bands were significantly thinner than the centromedial bands (p = 0.04). In the box diagram (Fig. 4), we observed that for each band, the interindividual variance was low except for the medial bands that had a high interindividual variance, which was significant (Fisher’s test). This figure confirms that the central and centromedial bands were thicker than the other bands, and measured on average 6.04 mm. Of all 12 analyzed tendons, the central and centromedial bands included in their proximal extension the tendon of the rectus femoris muscle. This individually identifiable muscle became increasingly sinewy the closer it was to patellar insertion. At the distal end, the tendon flattened and formed a progressive fiber blade making it
123
Surg Radiol Anat Table 1 Coefficients of correlation between the mean thickness of the tendon and the anthropometric data Correlation with the thickness (Pearson), q
Level of significance
CI 95 %
Age
q = -0.37
p = 0.28
[- 0.81; ?0.33]
Size of the body
q = 0.53
p = 0.074
[- 0.06; ?0.85]
Height of the patella
q = 0.75
p = 0.0048
[? 0.31; ?0.93]
Width of the patella
q = 0.52
p = 0.08
[- 0.08; ?0.84]
indistinguishable from the fibrous patellar insertion of the 3 vastus muscles. The tendon of the rectus femoris muscle corresponded to the superficial layer of the QT. The Pearson correlation tests (Table 1) allowed us to demonstrate a correlation between the thickness of the tendon and the height of the patella which was statistically significant (r = 0.75, p = 0.0048, 95 % CI [0.31; 0.93]). For the width of the patella and the size of the individual, the degree of significance was between 0.05 and 0.10.
Discussion The QT is a site of tendon harvesting which is not frequently exploited but can be used to reconstruct the ACL [3, 4, 8, 9], the PCL, the medial patellofemoral ligament (MPFL), and the lateral collateral ligament [1]. The use of QT was originally described by Marshall et al. [2]. This autograft shares biological and mechanical properties with the LP, sometimes with superiority. Its harvest has a lower morbidity than using PL [4] and hamstrings. Patients relate no loss of flexion, less anterior pain, decrease or absence of patella fracturing in the case of using grafts with patellar bone, and the absence of paresthesia on the outer edge of the knee, the sensitive area of the medial saphenous nerve. The QT may include a patellar bone fragment [4] or be free [9]. It can be used in single bundle [4] or double bundle [3]. Few articles discussed the tendinous transplant and its preparation with fixed and reproducible anatomical landmarks; they did not specify the exact size and morphometric characteristics of autografts. Several authors [1, 2, 4] described a technique within the sample seemed to be harvested with only one landmark, the patella center. Our study showed that the autografts of QT should be harvested at the central and centromedial bands to maintain a maximum thickness. In several studies, the graft transplant was measured 10 mm in width, with or without patellar bone plug. An incision of just a few centimeters long could be made in relation to the patellar insertion of the QT. This was sufficient to harvest the graft due to the elasticity of the skin [3]. The insertion of the QT was
123
located; the medial and lateral borders of the patella were palpated to determine the center of the QT. Once the center of the patella was located, we could harvest the central and centromedial strips. Geib et al. [3] and Harris et al. [5] have described a fat plane in the QT during the dissection. This fat plane corresponded to the limit between the common tendon of the large lateral and medial muscles, vastus intermedius and vastus medialis, and the tendon of the rectus femoris muscle. We observed this fat plan in our 12 dissected cadaveric knees. This fat plan inside the QT could not be individually identified at the patellar insertion. Harris et al. [5] have measured the division between the two tendons at 6 cm proximal to the insertion of the patella. As the QT is divided into two layers, the graft required a thorough preparation with proximal sewing of the two strands to allow its implementation in tunnels. Vastus medialis and vastus lateralis tendons behave like two fiber blades that reinforce the thickness of QT. The centrolateral and lateral bands showed a non-linear decrease for all the 12 tendons with a tendon thickening which was measured at 50–60 mm from the proximal edge of the patella for the centrolateral band and 60–80 mm for the lateral band. Central and centromedial bands were at mean thicker than all other bands. The analyses of the 12 tendons showed that these two bands included the distal tendon part of the rectus femoris muscle. The quadriceps tendon is considered as trilaminar, with a ventral part coming from the rectus femoris, a dorsal part coming from the vastus intermedius, and a thin fatty layer placed between this two tendinous layers. This aspect was not confirmed in the comprehensive review by Slone et al. [10]. The rectus femoris muscle at the junction between middle and distal thirds became tendinous with a large tendinous insertion on the upper pole of the patella. Lippe et al. [6] confirmed in their anatomical study the observations by Harris et al. [5] who had described an asymmetric surface of the QT with a lower insertion of the vastus medialis compared to the vastus lateralis. Lippe et al. [6] proposed a reference point for guiding the graft harvesting of QT. This point corresponded to the greatest length of the
Surg Radiol Anat
tendon visible at the surface of QT where the largest thickness is measured. This point was at mean at 61.6 % of the width of the insertion of the QT from the medial edge of the patella. Lippe et al. [6] recommended taking this reference point as the lateral edge of the graft to obtain a sufficient thickness. In their MRI study, Xerogeanes et al. [11] showed that the QT could provide a significantly bigger graft than the PT which measured 60 or 80 mm. Moreover, they showed a thickness uniformity along the first 60 mm of QT above the insertion, and explained that the 2.5 cm of graft which is intra-articular is much thicker. In our study, we observed a gradual decrease on the same portion. Xerogeanes et al. [11] and Harris et al. [5] have both measured QT, respectively, at 88 % and 1.8 times bigger than the PT. Xerogeanes et al. [11] found a correlation between the size of the patient and the length of the graft. Our study showed a correlation between the thickness of the tendon and the height of the patella. The width of the patella and the size of the individual had a degree of significance between 0.05 and 0.10. The risk of rupture of the quadriceps tendon after harvesting of the median band is not clear. No true rupture was described, but rectus femoris retraction when the tendon harvesting is extended to the myotendinous junction [10]. According to Lund et al, the quadriceps tendon has been considered as choice graft, the risk of rupture of the extensor system has not been cited [7]. Our study confirmed that if the surgeon wants a thicker graft, he must harvest only the central and centromedial bands of the QT. Our results were similar to the recommendations of Lippe et al. [6] who advocated harvesting the medial part of the quadriceps tendon peak which is 61.6 % from the medial border of the QT to have the longest graft. Xerogeanes et al. [11] suggested a more lateral harvesting of the autograft, by placing it in the middle of the patella and taking 5 mm on either side.
Conclusion This anatomical study has clarified the morphometric mapping of the thickness of QT based on the medial and lateral edges of the patella. This allows us to harvest the tendinous autograft with reliable anatomical landmarks taken from the patella edges. The central and centromedial bands are thicker. The average thickness of the transplant can be anticipated based on the distance from the upper
edge of the patella and the location relative to the center of the patella. We are then able to manage the dimensions of the tendinous transplant in relation to the ligament surgery required. Acknowledgments None of the authors has any conflict of interest related to the topic of this study. Ethical standard France.
This study complies with the laws in force in
References 1. Chen CH, Chen WJ, Shih CH (2001) Lateral collateral ligament reconstruction using quadriceps tendon–patellar bone autograft with bioscrew fixation. Arthrosc J Arthrosc Relat Surg 17(5):551–554 2. Fulkerson JP, Langeland R (1995) An alternative cruciate reconstruction graft: the central quadriceps tendon. Arthrosc J Arthrosc Relat Surg 11(2):252–254 3. Geib TM, Shelton WR, Phelps RA, Clark L (2009) Anterior cruciate ligament reconstruction using quadriceps tendon autograft: intermediate-term outcome. Arthrosc J Arthrosc Relat Surg 25(12):1408–1414 4. Han HS, Seong SC, Lee S, Lee MC (2008) Anterior cruciate ligament reconstruction: quadriceps versus patellar autograft. Clin Orthop Rel Res 466(1):198–204 5. Harris NL, Smith DAB, Lamoreaux L, Purnell M (1997) Central quadriceps tendon for anterior cruciate ligament reconstruction: Part I: morphometric and biomechanical evaluation. Am J Sports Med 25(1):23–28 6. Lippe J, Armstrong A, Fulkerson JP (2012) Anatomic guidelines for harvesting a quadriceps free tendon autograft for anterior cruciate ligament reconstruction. Arthrosc J Arthrosc Relat Surg Off Publ Arthrosc Assoc N Am Int Arthrosc Assoc 28(7):980–984 7. Lund B, Nielsen T, Phys BA, Faun P, Christiansen SE, Lind M (2014) Is quadriceps tendon a better graft choice than patellar tendon? A prospective randomized study. Arthroscopy 30(5):593–598 8. Marshall JL, Warren RF, Wickiewicz TL, Reider B (1979) The anterior cruciate ligament: a technique of repair and reconstruction. Clin Orthop Rel Res 143:97–106 9. Schulz AP, Lange V, Gille J, Voigt C, Fro¨hlich S, Stuhr M et al (2013) Anterior cruciate ligament reconstruction using bone plugfree quadriceps tendon autograft: intermediate-term clinical outcome after 24-36 months. Open Access J Sports Med 4:243–249 10. Slone HS, Romine SE, Premkumar A, Xerogeanes JW (2015) Quadriceps tendon autograft for anterior cruciate ligament reconstruction: a comprehensive review of current literature and systematic review of clinical results. Arthroscopy 31(3):541–554 11. Xerogeanes JW, Mitchell PM, Karasev PA, Kolesov IA, Romine SE (2013) Anatomic and morphological evaluation of the quadriceps tendon using 3-Dimensional magnetic resonance imaging reconstruction. Applications for anterior cruciate ligament autograft choice and procurement. Am J Sports Med 41(10):2392–2399
123
