Anatomy of the anterior cruciate ligament: Three-bundle concept with an apparent twist – A cadaveric study

INTRODUCTION

Despite advancements in our comprehension of anterior cruciate ligament (ACL) anatomy, biomechanics, biology, and the progression of techniques and instruments, we still lack reliability in accurately identifying the anatomical ACL footprint and positioning the femoral tunnel correctly.^[1] ACL reconstruction (ACLR) has significantly progressed over the past 50 years. Initially, this was marked by the recognition of the limitations of extra-articular procedures in the 1970s and the necessity to reconstruct the ACL. In the 1980s, the conventional method evolved into anteromedial (AM) bundle reconstruction, which employed the middle third of the patellar ligament. However, inadequate control of rotational laxity prompted the introduction of double-bundle ACL reconstruction in 2000.^[2] This approach, along with an increasing focus on preserving the ACL remnant, subsequently led to selective reconstruction in cases of partial tears, as well as biological reconstruction that maintains the ACL remnant. Double bundle reconstruction has since lost popularity due to the limited evidence supporting its superiority in objective assessments of knee stability and its effectiveness in preventing repeat ACL injuries or new meniscal damage. Despite enhancements in surgical methods and rehabilitation, the rates of return to sport and re-tear incidents remain unsatisfactory.^[3]

To address this issue in the latter half of the second decade of this century, lateral extra-articular tenodesis (LET) was reintroduced alongside ACL reconstruction (Jesani and Getgood).^[4] The advantages of incorporating a LET procedure into ACL surgery may encompass improved rotational stability, reduced re-tear rates, and a better return to play by restoring normal knee kinematics, rectifying a high-grade pivot shift, and correcting excessive internal tibial rotation.^[5] Literature also suggests the reconstruction of (Anterolateral ligament (ALL) in conjunction with ACLR for addressing rotational instability.^[6] However, the contemporary techniques for single-bundle ACL reconstruction, whether or not accompanied by a lateral extra-articular procedure, are contributing to the onset of early osteoarthritis and a delay in the recovery of quadriceps strength (with rotary stability being addressed by the lateral extra-articular procedure).

Among the tunnels in the femur and tibia, the misplacement of the femoral tunnel is the primary reason for the failure of the initial ACL reconstruction, leading to the necessity for revision ACL surgery. In addition, during the surgical procedure, the femoral tunnel is often positioned excessively anteriorly in most cases due to the challenges in visualizing the “over the top” position of the femur during endoscopic ACL surgeries, as the distinction between an anatomical tunnel and an anterior tunnel is merely 3 mm. The task of identifying and reconstructing the ACL femoral anatomy is intricate. A systematic review conducted by Brown et al.^[7] in 2020 indicated that individuals who underwent ACLR exhibited lower quadriceps strength in the injured limb compared to the limbs of controls matched for age, sex, and activity level. The research objectives involve a thorough examination of the ACL anatomy, with a focus on identifying the anatomical footprint of the ACL, particularly on the femur, as well as analyzing the course and measurements of the ACL. The comprehensive anatomical description of the natural attachment of the ACL may provide surgeons with guidance on the appropriate locations for creating tunnels, especially femoral tunnels during ACL reconstruction.

MATERIALS AND METHODS

A cadaveric study was carried out at our institute from May 2023 to February 2025. Cadavers with a post-embalming period of <1 month and age at death <40 years were included in the study. The cadavers not satisfying the inclusion criteria were excluded. We conducted a comprehensive macroscopic evaluation of all specimens that met the inclusion criteria, focusing on the anatomy of the ACL. The parameters of the study encompassed the origin and insertion points of the ACL, its bundles, their orientation, and measurements of thickness, length, and footprints. For an in-depth analysis, we performed a stepwise dissection on 20 cadaver knees in the Anatomy Department of the local Government Medical College.

Initially, the bone shafts were meticulously cleaned of soft tissue, followed by the excision of collateral ligaments, the patella, the patellar ligament, and the quadriceps to ensure optimal access to the ACL, while preserving the menisci. Dissection of the synovial tissue revealed the bundles of the ACL, which were then subjected to careful blunt dissection to elucidate the individual footprints. The medial femoral condyle was removed using a saw to provide an unobstructed view of the ACL. By tracing each bundle to its tibial and femoral attachments, to facilitate the identification of the footprints. Measurements of the footprints and bundles were taken using a vernier scale. For a detailed anatomical understanding, cross-sections of the ACL bundles were obtained. Photographs and video [Video 1] were captured throughout the study for future reference.

RESULTS

In a study involving 20 cadaver knees, degeneration of the ACL was observed in one, which hindered the examination of its footprints and fibers. The age of cadavers at the time of death was <40; it was noted that 12 were male, and 7 were female. Our study demonstrated that ACL is a three-bundle structure with each bundle having individual footprints on both the femoral and tibial sides. The three bundles are AM bundle, the intermediate bundle (we termed it as anterolateral [AL] bundle by considering its tibial footprint), and posterolateral (PL) bundle [Figure 1a and b]. The femoral footprint of all three bundles of ACL is on the resident’s ridge (RR)/lateral intercondylar ridge. There is only a synovial extension outside the RR, which appears as a fanning of ACL fibers. The RR is crescentic in shape. This crescentic footprint appears as a continuation of the posterior cortex of the femur. The footprints of individual bundles on the femoral side are that AM bundle is at the posterosuperior part of the femoral lateral intercondylar ridge (RR), AL bundle is at the apex of the crescent of the RR – anteroinferior to the attachment of the AM bundle, and PL bundle is at the distal part of RR – anteroinferior to the attachment of the AL bundle [Figure 2a-d]. This special configuration of the footprints prevents the impingement of AL on the PL bundle.

Close

Figure 1:

(a) Cadaver specimen of tibia plateau showing mid-substance cross-section of anterior cruciate ligament (ACL) showing three bundles (AM, AL, and PL) (Yellow box indicate the area shown in Figure 1b). (b) Cadaver specimen of tibia plateau showing mid-substance cross-section of ACL showing three bundles (AM, AL, and PL) -marked) - Zoom view. AM: Anteromedial, AL: Anterolateral, PL: Posterolateral.

Export to PPT

Close

Figure 2:

(a) Cadaver specimen sagittal view, showing a medial aspect of the lateral femoral condyle and anterior cruciate ligament (ACL) remnants after removing medial femoral condyle, (b) Cadaver specimen Sagittal view, showing a medial aspect of the lateral femoral condyle and residents ridge after removing medial femoral condyle, (Foot print of AM, AL and PL bundle, represented by green, blue and red line respectively) (c) Cadaver specimen sagittal view, showing a medial aspect of the lateral femoral condyle and Residents’ Ridge (ACL footprint of each bundle) after removing the medial femoral condyle. (d) 3-dimensional model picture of the medial aspect of the lateral femoral condyle showing the ACL footprint of each bundle. Femoral footprints of ACL: (1) Anteromedial (AM) Bundle (Green) - posterosuperior part of the femoral lateral intercondylar ridge (Resident’s Ridge) in continuation with posterior femoral cortex. (2) Anterolateral (AL)/Intermediate bundle (Blue) - the middle of the lateral intercondylar ridge, anteroinferior to the attachment of the AM bundle. (3) Posterolateral (PL) bundle (Red) attaches anteroinferior to the attachment of AL bundle on the lateral intercondylar ridge.

Export to PPT

We recorded that the mean width of the femoral footprint is 3.5 mm (3–4 mm) and the mean length of the femoral footprint is 15 mm (13 mm–17 mm). From the posterior articular cartilage to the center of ACL femoral attachment (residents ridge), there is an 8–10 mm free space [Figure 3a and b]. In the intra-articular area, the mean width of ACL is 9 mm (8–10 mm), thickness is 3.5 mm (3–4 mm), length of AM bundle is 29 mm (28–30 mm), length of AL bundle is 23 mm (22–24 mm), and length of PL bundle is 15 mm (13–17 mm). Even though the course of ACL appears to be twisted 90° from the tibia to the femur during flexion, while considering each bundle separately, they traverse in a straight path. In extension, the bundles are parallel to each other [Figure 4]. Even though the posterior margin of tibial attachment of ACL reaches the anterior margin of PCL, there is no chance for PCL impingement because of this apparent twisted arrangement of bundles. The ACL unwinds when the femoral condyle is rotated anticlockwise to 90° in the left knee and clockwise in the right knee in a 90° flexed knee.

Close

Figure 3:

(a) Cadaver specimen showing anterior cruciate ligament (ACL) and medial aspect of lateral femoral condyle after removing the medial femoral condyle. (b) Cadaver specimen showing remnants ACL and the medial aspect of the lateral femoral condyle after removing the medial femoral condyle. We can see the posterior part of the lateral femoral condyle behind the resident’s ridge, which is free of any ACL attachment. The attachment appears as a continuation of the posterior cortex of the femur. (LFC: Lateral femoral condyle).

Export to PPT

Close

Figure 4:

Cadaver specimen front view anterior cruciate ligament showing the three bundles (AM, AL, and PL), after removing the medial condyle, the anteromedial (marked as Green Color) and posterolateral bundles (marked as Red Color), the anterolateral or intermediate bundle (marked as Blue Color) is seen in between. AM: Anteromedial, AL: Anterolateral, PL: Posterolateral.

Export to PPT

In the tibial footprint, the AM bundle is posterior to the anterior horn of the medial meniscus (without any attachment to the meniscus), at the lateral aspect of the medial tibial spine. The AL (intermediate) bundles are at the medial part of the lateral tibial spine with few fibers arising from the anterior root of the lateral meniscus. The PL bundle is posterior to AM and AL bundles, more toward the lateral side, and 3 mm AM to the lateral meniscal posterior root [Figure 5a and b]. The tibial footprint appears to be kite-shaped, considering all three bundles as a single unit, but we do not consider that to be important, rather than the individual origin of the three bundles as discussed above.

Close

Figure 5:

(a) Cadaver specimen oblique view of tibia plateau showing three bundles of anterior cruciate ligament (ACL). (b) Three-dimensional model picture representing tibia footprints of each bundle of ACL. Tibial footprints of ACL: (1) Anteromedial (AM) Bundle - Posterior to the anterior horn of the medial meniscus (without any attachment to the meniscus), from the lateral aspect of the medial tibial spine. (2) Anterolateral (AL)/Intermediate bundle - medial aspect of the lateral tibial spine (a few fibers arising from the anterior root of lateral meniscus also). (3) Posterolateral (PL) bundle - posterior to AM and AL bundles more toward the lateral side and 3–4 mm anterior to the lateral meniscal posterior root.

Export to PPT

DISCUSSION

In our analysis of 19 cadaveric knee specimens, we found that the ACL is predominantly a three-bundle structure, except for a bifid AM bundle noted in seven knees. A review of previous literature indicates that the three-bundle concept was initially introduced by Norwood and Cross^[8] in 1979. Despite numerous researchers discussing this concept over the years, it has not achieved widespread acceptance in the field. Some academics describe the ACL as a single continuous structure, while others recognize two separate bundles. Importantly, Mommersteeg et al.^[9-12] documented the existence of 6–10 individual bundles within each ACL. Our findings also suggested the presence of more than three bundles in degenerated knees; however, this was not observed in younger patients under the age of 40. Therefore, we conclude that the ACL consistently consists of three bundles, with any additional bundles that are seen being linked to the aging process. Each bundle of the ACL has unique attachment sites on both the femoral and tibial sides. However, Tan et al.^[13] have reported that they did not find any distinct anatomical bundles in the ACL and concluded that these bundles are more functional than anatomical. Śmigielski et al.^[14] and Śmigielski et al.^[15] have proposed a ribbon-like flat ACL concept devoid of any bundles. Lalwani et al.^[16] which is an Indian subcontinental study, characterized the three-bundle structure of the ACL, noting a twist in its trajectory, with each bundle having different lengths. Further studies will be required to identify any racial variations in the anatomy of ACL.

Femoral tunnel malposition is the main cause of primary ACL failure, resulting in revision ACL reconstruction. The femoral footprint of all three bundles of ACL is on the RR/lateral intercondylar ridge [Figure 2]. There is only a synovial extension outside the RR, which appears as a fanning of ACL fibers. Norwood and Cross^[8] reported that the AM bundle attaches to the femur posteriorly and superiorly on the medial surface of the lateral femoral condyle (LFC). The PL bundle attaches anteriorly and inferiorly to the LFC. The intermediate bundle attaches between the PL and AL bundles on the femur and completes a crescentic femoral attachment proximal to the articular cartilage of the LFC and posterior to the intercondylar shelf. The biggest difference from our study is with regard to the femoral footprint; they characterized it as parallel to Blumensaat’s line in a linear fashion and not on the lateral intercondylar ridge. Otsubo et al.^[17] reported the 3-bundle concept of ACL in 2011. They stated that the femoral footprint was posterior to the lateral intercondylar ridge.

The multi-fascicular structure of the ACL was described by Amis and Dawkins^[18] in 1991 in terms of three functional fiber bundles, which is in line with our observation. Ferretti et al.^[19] discovered two distinct osseous markers. All arthroscopic patients and cadaveric knees had an osseous ridge running from the proximal to the distal ends. It was called the lateral intercondylar ridge. Another bony landmark was found between the femoral attachments of the AM and PL bundles, which run from anterior to posterior. However, we did not discover a bifurcate ridge posterior to the lateral intercondylar ridge in all patients, and all three ACL bundles were connected to the lateral intercondylar ridge (RR). In reviewing the literature, most of them described the ACL attachment as between the residents’ ridge and posterior articular cartilage (Arnoczky, 1983),^[10] which was a contradiction in our study. Only a few literature such as Sasaki et al.^[20] and Norman et al.^[21] suggested that the ACL footprint is on the lateral intercondylar ridge. Sasaki et al. (2018)^[20] histologically, the ACL fibers of midsubstance form a thin direct insertion posterior and along the lateral intercondylar ridge, followed by a fan-like indirect insertion toward the posterior femoral cartilage. Mochizuki et al.^[22] distinguished between the primary attachment of ACL mid-substance fibers and the attachment of thin fibrous tissue. Later, they extended from the mid-substance fibers and spread out like a fan on the posterior condyle. The authors named these fibers as fan-like extension fibers. According to Duthon et al.,^[23] the bony attachment of ACL in the femur is located at the posterior part of the inner surface of the LFC and not, as sometimes believed, at the roof of the intercondylar notch. We also found, from the posterior articular cartilage to the center of ACL femoral attachment (residents ridge), there is an 8–10 mm free space.

The concept of the flat ribbon ACL graft was introduced by Śmigielski et al. in 2012.^[14] He proposed that the femoral attachment of the direct fibers of ACL is on the RR, which is correct as per the findings of our study. Smigielski could not find the bundles at the femoral attachment area, maybe because he used a fresh frozen cadaver. However, we could differentiate all three bundles at the femoral and tibial attachment sites. A computed tomography investigation by Norman et al.^[21] discovered a design of cortical thickening compatible with the functional imprint of the ACL. On average, this section exhibited a thickness that was 3 times greater than that of the surrounding bone and significantly thicker than the remaining lateral wall. This thickening was roughly elliptical in shape. Shino et al.^[24,25] presented the first anatomic triple bundle reconstruction with twin femoral and triple tibial tunnels. The tunnel apertures they described corresponded to the attachment regions of the three fiber bundles in studies by Norwood and Cross^[8] and Amis and Dawkins,^[18] however, femoral tunnels were drilled posterior to the intercondylar ridge, which is not the anatomic position according to our findings.

The landmarks of the tibial footprint from our study: the AM bundle is posterior to the anterior horn of the medial meniscus (without any attachment to the meniscus), at the lateral aspect of the medial tibial spine. The AL (intermediate) bundle is at the medial part of the lateral tibial spine, with few fibers arising from the anterior root of the lateral meniscus. The PL is posterior to AM and AL bundles, more toward the lateral side, and 3 mm AM to the lateral meniscal posterior root [Figure 5]. The tibial footprint appears to be kite-shaped, considering all three bundles as a single unit, but we do not consider that to be important, rather than the individual origin of the three bundles as discussed above. According to the study conducted by Norwood and Cross,^[8] the tibial foot is shaped like a triangle that points toward the posterior. The AM bundle is anchored on the medial side of the intercondylar eminence of the tibia, constituting the medial corner of this triangle. Meanwhile, the intermediate bundle is connected at the midline and the lateral side of the eminence, positioned laterally to the AM bundle, thereby forming the lateral corner of the triangular attachment. The PL bundle serves as the apex of the triangle, directed posteriorly, with its attachment located just lateral to the midline of the intercondylar eminence and slightly to the side of the most lateral attachment of the intermediate bundle. Tibial footprints are almost identical, except for a kite-shaped appearance of the footprint formed by fibers from three bundles. However, the shape of the footprint, whether it is a triangle or kite, is not significant; rather, the origin of the individual bundles is more important for the normal functioning of a reconstructed ACL.

Otsubo et al.^[17] findings were similar to Norwood and Cross^[8] regarding the tibial footprint. According to Śmigielski et al.,^[14] the tibial attachment was found to be C-shaped, which is different from our findings. In our study, the ACL was a three-bundle structure that appears to be sheet-like in the intraarticular area, with a tibial footprint that appears kite-shaped. Each bundle has a separate attachment in the footprint area. He missed the kite shape because he had not taken into consideration, the fibers of the AL bundle that arise from the lateral meniscus. Norwood, Smigielski, and Otsubo have missed the fibers of the AL bundle arising from the lateral meniscus.

From our study, the course of ACL appears to be twisted 90° to the right in the right knee and 90° to the left in the left knee at 90° of flexion of the knee. But while considering each bundle separately, they traverse in a straight path. This is achieved by the peculiar footprints of each bundle as discussed above. In extension, the bundles are parallel to each other. Even though the posterior margin of the tibial attachment of ACL reaches the anterior margin of PCL, there is no chance for PCL impingement because of this apparent twisted arrangement of bundles during flexion. Otsubo et al.^[17] reported the 3-bundle concept of ACL in 2011. According to them, the AM and intermediate bundles were distributed parallel to the PL bundle in the extended-knee position, whereas they were twisted in the 90° knee flexion. This result agrees with our finding, but we consider the twist in ACL at 90° flexion to be an apparent twist because of the attachment of individual bundles on the femur and tibia without actual twist of the individual bundles. If we take individual bundles even at 90°, they travel in a straight line. Smigielski^[14] also proposed the twisting of ACL up to 90° in the ribbon concept. Lalwani et al.^[16] also characterized the three-bundle structure of the ACL, noting a twist in its trajectory, with each bundle having different lengths. This agrees with our findings, as the AM bundle is the longest and the PL bundle is the smallest due to their separate attachments in the tibial and femoral footprint. Although they employed different terminology for each bundle, their findings were consistent with ours.

The results from our study can be used to alter further the way an ACLR is done. The primary challenges associated with triple bundle ACLR involve the establishment of three distinct tibial and femoral tunnels for each bundle of the ACL. From our study, the femoral footprint is on the RR, the AM bundle tunnel position can be identified at posterosuperior part of the RR close to the posterior cortex of femur, AL bundle tunnel can be identified anteroinferior to AM bundle on the RR, and the PL bundle tunnel can be identified anteroinferior to the AL bundle on the RR. For the identification of the tibial footprint, the meniscus roots and tibial spine can be used as references. The PL bundle position is 3 mm AM to the lateral meniscal posterior root, the AL bundle position is close to the anterior horn of the lateral meniscus, and the AM bundle position is the same as the normal single bundle ACLR done nowadays. Once these tunnel positions are attained, the new graft bundles will have a 90° twist to the right in the right knee and the left in the left knee in 90° flexion of the knee, and each bundle will be parallel to the other in full extension. Authors are of the opinion that there is a scope for exploring surgical techniques utilizing these anatomical facts for effective restoration of rotational stability (to avoid extra-articular procedures such as LET and ALL reconstruction), achieving early recovery of quadriceps strength, and preventing progression of osteoarthritis in ACLR. However, further studies are required to establish these hypotheses.

While unable to provide commentary on the kinematics of each bundle, the model is capable of accurately explaining the ACL footprints and detailing each bundle comprehensively.

CONCLUSION

ACL is a three-bundle structure. Each bundle of ACL has different lengths and separate footprints in the femoral and tibial sides. The femoral footprint of ACL is on the RR, which appears as a continuation of the posterior cortex of the femur. Each bundle travels in a straight path from the tibia to the femur, and there appears to be a twist in the ACL during flexion, although no actual twist occurs due to the unique attachments on the femur and tibia. The AL bundle/intermediate bundle of ACL has an attachment to the lateral meniscus anterior horn. The comprehensive anatomical description of the natural attachment of the ACL will guide surgeons in creating three different femoral and tibial tunnels during anatomical triple-bundle ACL reconstruction.