The two most popular tendons used for anterior cruciate ligament (ACL) reconstruction are bone-patellar tendon bone (BPTB) and hamstring tendon . However, there are some donor-site morbidities when using these grafts, e.g., anterior knee pain and risk of patella fracture after harvesting the BPTB autograft  and graft laxity and hamstring weakness after harvesting hamstring autograft .
The quadriceps tendon is an alternative graft choice that provides promising outcomes for ACL reconstruction reported from several studies [4, 5]. Advantages of using this type of graft are lower risk of anterior knee pain  and more preserved hamstring function. Quadriceps tendon can be used with or without patella bone plug. Most of the proposed techniques require at least 7–9 cm of tendon portion [2, 6, 7, 8].
The biomechanical properties of quadriceps tendon have been studied [6, 9]. However, there are few studies evaluating the anatomical structures of quadriceps tendon as graft in ACL reconstruction [9, 10, 11, 12]. Most studies had small number of specimens. A study by Iriuchishima et al.  reported the morphology of quadriceps tendon using Asian cadaveric knees, but the study focused on the rectus femoris tendon which is only a part of quadriceps tendon. Lippe et al.  reported that quadriceps tendon is asymmetrical. Findings from our data may change the current harvesting techniques of quadriceps tendon that is mostly harvested at the central 10 mm of the tendon [13, 14].
The purpose of this study was to evaluate the anatomical structures of quadriceps tendon and the suitability to use quadriceps tendon graft as an alternative for ACL reconstruction.
This study was approved by the Institutional Review Board, Faculty of Medicine, Chulalongkorn University (certificate of approval no. 005/2014).
Forty-nine formalin-fixed cadaveric knees from 33 cadavers were used. Our inclusion criteria were specimens that were used for medical students. Our exclusion criteria were specimens that were not in good condition. Skin and subcutaneous tissue were removed to expose the superficial quadriceps tendon and patella. We used electronic caliper (Baker model EC-10, Mumbai, India) for all measurements.
The most proximal point of superficial quadriceps tendon was identified and marked with a surgical marking pen. After that, we created a transverse line connecting medial and lateral edges of the patella tendon insertion. Then, we drew a line from the most proximal point of the tendon down and perpendicular to the transverse line passing proximal edge of patella. The line from the most proximal point of the tendon to the proximal edge of the patella represents tendon maximum length (
Next, we measured the length of quadriceps tendon from the musculotendinous junction to the patella tendon insertion, starting from the line of the maximum length of the tendon and every 5-mm increments medially and laterally to the edge of the tendon at patella insertion.
After all of the surface measurements were done, we dissected the quadriceps tendon to measure the tendon thickness (
Two investigators were assigned to carry out each measurement. The interobserver reliability was evaluated using the intraclass correlation coefficients (ICCs; two-way mixed average measures). Descriptive statistics were presented as mean and standard deviation (SD). Paired
The ICCs for intra- and inter-rater agreements were >0.9 for all measurements. The ICCs value of >0.8 defined an excellent agreement. Mean age of our population was 75.6 years (range 38–96) and mean height was 165.4 cm (range 149–184).
The maximum length of superficial quadriceps tendon was 63.2 ± 12.5 mm (range 37.6–83.4). The tendon length was decreased to 53.0 ± 14.5 mm and 38.2 ± 15.5 mm at a distance of 5 mm and 10 mm medial to the maximum length of the tendon, respectively. The tendon length was decreased to 49.5 ± 12.8 mm and 38.0 ± 13.9 mm at a distance of 5 mm and 10 mm lateral to the maximum length of the tendon, respectively (
By stratifying the height of the cadavers into three groups (140–155 cm, 156–170 cm, and 171–185 cm). We found positive correlation between the height of cadaver and the maximum length of the superficial quadriceps tendon (
The maximum length of quadriceps tendon at the deepest part was 57.2 ± 11.5 mm (range 30.9–83.4). We found that musculotendinous junction of the deepest part of the tendon was located more distal than the superficial part in 32 (65.31%) of 49 of the specimens.
The quadriceps tendon width at patella insertion was 46.2 ± 4.7 mm (range 36.7–56.7). The maximum length of quadriceps tendon was located at 61.4% ± 13% (range 31.8–83.6) from the medial edge of the patella insertion. It is located at approximately 5 mm laterally from the midline of patella insertion of the quadriceps tendon. The quadriceps tendon thickness at its maximum length of the tendon on patella insertion was 6.9 ± 2.0 mm (range 1.5–9.5).
There were no differences in the maximum length, the location of the maximum length, the width at patella insertion, and the thickness at patella insertion of the quadriceps tendon between the left and right sides of the same cadaver.
ACL injuries can be managed with surgical or nonsurgical treatments. Surgical treatment is indicated to a patient who has knee instability, young age, high-functional demand, or failed to nonsurgical treatment. The ACL reconstruction became a common orthopedic surgery, and the number is increasing every year [15, 16]. Available graft options include autografts, allografts, and synthetic grafts. The numbers of allografts and synthetic grafts used are increasing but limited by their cost and availability [17, 18]. Hamstring tendon, patellar tendon, and quadriceps tendon are commonly used due to biological advantage of autografts . The quadriceps tendon grafts have been proposed as a promising alternative due to its lower donor site morbidity. It also has comparable clinical outcomes, patient satisfaction, range of motion, and complications to other autografts [20, 21, 22].
The majority of techniques for harvesting quadriceps tendon either with or without patella bone plug require at least 7–9 cm of the tendon portion [2, 6, 7, 8]. According to the anatomical study of the quadriceps tendon reported by Lippe et al. , the maximum (peak) length of quadriceps tendon was 88.3 mm (range 78.3–99.7 mm). However, in our study, we found that the maximum length of the quadriceps tendon was only 63.23 ± 12.49 (range 37.61–83.41) at the superficial quadriceps tendon and 57.23 ± 11.53 (range 30.94–83.41) at the deepest part. If the minimal length of quadriceps tendon required for ACL reconstruction was 7 cm, we found that 61.22% of our quadriceps tendons (30 of 49) were too short. These results might warn surgeons that quadriceps tendon autograft might not be adequate for ACL reconstruction, especially in short statured patients. Our study is the first study to investigate the correlation between the height of cadaver and the maximum length of the quadriceps tendon. We found a positive correlation between these two variables.
Similar to previous study , we found that the maximum length of quadriceps tendon was not located at the center of the tendon width at the patella insertion but located about 61.4% from the medial border of the width (approximately 5 mm lateral from the center of tendon insertion). In addition, the length of quadriceps tendon decreased dramatically even 5 mm from either medial or lateral to the area of maximum quadriceps tendon length. Thus, harvesting quadriceps graft using the original technique (10 mm midline) might not achieve its maximum graft length. The graft will get significantly shorter if surgeons move the harvesting area only 5 mm medially. We recommend marking midline of patella as the medial border of the graft and harvesting the graft laterally from this line (
Another important finding, which is different from the previous study , that we found is the musculotendinous junction of the deepest part of some quadriceps tendon was located more distal compared to the superficial part. From our study, surgeons should not rely only on superficial anatomy of quadriceps tendon when harvesting the graft, otherwise graft shortening might occur.
The strengths of this study were that we evaluated the anatomical structures of quadriceps tendon in Asian cadavers. The sample size of this study was relatively large. However, our limitation was that we performed the study in embalmed cadaveric knees which might not actually reflect the in vivo situation. However, a previous study reported by Cutts  showed that graft shrinkage of embalmed specimen was minimal if it was fixed to bone. We believe that our measured data similarly resembled in vivo conditions.
The maximum length of quadriceps tendon was located 5 mm lateral to the center of quadriceps tendon insertion. Thus, we recommend marking midline of the quadriceps tendon insertion as a medial border of the harvested graft to get maximum length of the graft. Quadriceps tendon might not be suitable to use as an autograft in short statured patients because of graft length inadequacy. Graft length measurement using superficial anatomy should be cautioned because graft shortening might occur.