Anatomical Complexity of the T-Junction
Though still an emerging focus, the distal T-junction of the biceps femoris has become recognised as a problematic site of hamstring injury. Proposed high recurrence rates and extended return-to-play timelines underline its clinical importance (Shamji et al., 2021) though not all research to date has been in agreement (Kerin et al., 2024; Pollock et al., 2022).
Entwisle et al. (2017) was the first to characterise this region in detail, describing how the confluence of the long and short heads of biceps femoris creates an interface with distinct structural vulnerabilities (Figure 1). The long head is biarticular (crossing hip and knee joints) and innervated by the tibial division of the sciatic nerve, whereas the short head is uniarticular (crossing knee only) and supplied by the common peroneal nerve. Coordinating these differently controlled muscle components places high demands on neuromuscular function (Nakao et al., 2025a, 2025b).
Figure 1: Axial anatomy of the distal biceps femoris (Entwisle et al., 2017). Proximal images show long head preponderance, the mid-thigh forms the T-junction, and distally the short head predominates.
Anatomical variation adds further complexity: the point at which the long and short heads meet, and their relative contributions to the junction, differ across individuals. This variability likely contributes to the unpredictable behaviour of injuries in this region. Entwisle also described how healing at the T-junction tends to lag behind more proximal sites, with scar formation and remodelling delayed. This results in prolonged structural weakness, a possible factor in reported recurrence rates exceeding 50% at two years (Entwisle et al., 2017). A further challenge is that these injuries do not fit neatly within existing grading systems (Paton et al., 2023). They straddle both musculotendinous and tendinous zones, with the junction effectively acting as an aponeurosis—a flattened epimyseal sheet (deep component in Figure 2). As such, they fall outside current classification criteria and generate variability in terminology. Many are also underestimated on MRI, with true instability often only revealed intra-operatively (Kayani et al., 2020).
In this piece we will discuss how we can apply aspects of the evolving yet still limited literature to practically influence rehabilitation, specifically the anatomy and pathology of T-Junction injuries, the influence of intermuscular connections and activation patterns, and the role of rotational mechanisms in both injury and rehabilitation.
Diagnostic Limitations and Emerging Tools
The T-junction poses some of the greatest diagnostic challenges in hamstring injury. The variable contributions of the long and short heads to the deep and superficial limbs mean these injuries often do not fit neatly into existing classification systems (Figure 2). The British Athletics Muscle Injury Classification (BAMIC), although widely used, was not designed with this region in mind. Lesions frequently span tendon limbs and connective tissue planes, making them difficult to categorise and, in practice, often under-called or misclassified. This leaves a central question unanswered: which injuries are relatively minor and low risk, and which demand a more cautious approach?
Figure 2: Cross-sectional anatomy of the distal thigh highlighting the biceps femoris T-junction. Arrows indicate the components of the T-Junction, comprised of the long and short heads.
These limitations have driven growing interest in dynamic ultrasound as an adjunct or potential replacement of MRI for diagnostic evaluation where T-junction injuries are suspected. In addition to MRI, it provides superior spatial resolution for superficial structures and, crucially, enables dynamic assessment of a dynamic junction (Cronin et al., 2024). Features such as gapping or asynchronous tendon limb motion during resisted contraction are impossible to visualise on static imaging. Figure 3 shows an apparently stable injury above, yet a gap opens during an isometric prone curl below). Cronin et al. (2024) recently reported the use of ultrasound to track resolution of such features in an elite footballer. Building on this, the lead author has proposed a five-type ultrasound classification tailored to the T-junction. This classification system does require testing in vivo to establish validity. However, given that Kerin et al. (2024) reported an interquartile range of 10–44 days for return to full training in these injuries; better sub-classification may clarify where on this spectrum each case sits.
Figure 3: (A). Transverse image showing a clear defect between the distal Biceps Femoris short head (BS) and the distal Biceps Femoris long head at rest. (B). Transverse image showing an enlargement and separation of the defect between the distal Biceps Femoris short head (BS) and the distal Biceps Femoris long head with a resisted isometric hamstring contraction (courtesy of Kevin Cronin).
A limitation however of this framework is that the terminology does not align neatly with MRI-based systems such as BAMIC (Pollock et al., 2014), which limits cross-comparison. It is also a dynamic classification that depends on sonographer expertise and interpretation. Yet, when integrated with MRI, it offers a way to capture the distinct anatomy and healing response of the T-Junction (Figure 4), with potential to separate low-risk injuries from those more prone to recurrence, and to provide a valuable tool for monitoring remodelling.
Figure 4: Ultrasound and corresponding schematic of a hamstring “T-Junction” injury.
Intermuscular Control and Adaptation
The hamstrings operate as an integrated system rather than isolated linear units. The epimysial convergence of the long and short heads forms a structural pathway for force transmission and redistribution (Nakao et al., 2025a, 2025b). As Schuermans et al. (2014) noted, shared connective tissues can be up- or down-regulated after injury, with the ‘teammates or competitors’ concept underscoring the group’s interdependence on each other. This supports viewing hamstring mechanics systemically rather than in isolation. T-junction injury is often a sequela after semitendinosus harvest for ACL reconstruction. Whether this reflects load transfer from semitendinosus to biceps femoris, or a global reduction in knee flexor capacity leading to greater short head reliance, the pattern emphasise that intermuscular coordination sits at the heart of T-junction pathology.
Recent work by Nakao et al. (2025a, 2025b) has clarified the mechanical interplay between the long and short heads. Using ultrasonic shear wave elastography (SWE), they mapped tension distribution across the biceps femoris complex in cadavers, with shear modulus (resistance to deformation) as a proxy for stiffness and load. The limbs of the T-junction were shown to interact through their shared tendon and fascia rather than functioning independently. Detachment of short head increased long head shear modulus likely forcing it to absorb disproportionate strain (Nakao et al., 2025a). Conversely, increasing tension of the short head reduced long head tension by up to 19% (Nakao et al., 2025b).
Figure 5: Reduced left biceps femoris long head muscle mass in a previously injured athlete, visualised using a Springbok Analytics protocol.
Schuermans et al. (2014) provide a framework where hamstring muscles with shared connective tissues may act as ‘teammates’ or ‘competitors’ to each other. This applies to the long head–semitendinosus connection via their fibrous membrane, and now too to the long head–short head link through the shared tendon. Ideally, the short head would buffer tension within the long head. But when absent or over-developed, this balance may invert, leaving the long head overloaded or under-conditioned. Clinically, this raises a key question: should rehabilitation enhance short head contribution, or could excessive reliance shield the long head and promote injury at their confluence (Figure 6)?
This hypothesis aligns with Fyfe et al. (2013) and Silder et al. (2008). Fyfe showed that hamstring strain injury often triggers neuromuscular inhibition of the long head, reducing voluntary activation and leading to long head atrophy and short head hypertrophy. The frequent distal injuries that follow more proximal injury can therefore be seen as a downstream effect of this altered interaction through this sequence: proximal long head injury → compensatory short head reliance → T-junction lesion (Figure 6).
Figure 6: Recent left-sided T-junction injury showing inhibition of the biceps femoris long head proximal to the lesion. On the contralateral (uninjured) side, clear activation and definition are visible in both medial and lateral hamstrings. On the injured side, although medial activation is evident, there is a notable lack of definition and proximal engagement above the T-junction. Without targeted rehabilitation, this maladaptive pattern may persist, with the athlete continuing to offload the long head—leaving it underprepared and increasing strain on adjacent structures.
While athletes may restore function quickly, range specific knee flexor weakness is still a frequent finding after T-junction injury, particularly at the extremes of flexion and extension. This pattern reflects the differing contributions of various muscles across the range of motion and highlights the need for considered range-specific rehabilitation. Programmes should emphasise heaviest loading at outer range, where the length–tension relationship is most favourable and demands high load, moderate resistance through mid-range, and lighter or even assisted work in inner range to provide an appropriate stimulus. This approach also serves as a diagnostic tool, helping clinicians identify the range in which the athlete is most deficient and target their rehabilitation accordingly.
However, there is a need to restore balance between the hamstring heads rather than rely on non-specific strengthening. Clinically, two patterns can be observed.
- A stiff, overloaded long head may indicate an underactive short head and respond well to short head–biased work, such as knee flexion exercises (e.g. curl variations in external tibial rotation or Nordics with a band around the feet).
- Conversely, a long or under-tensioned long head could point to short head dominance and benefit from long head–focused strategies: hip-extension exercises with hip external rotation.
It is important to note, however, that Nakao and colleagues work is limited to the measurement of passive stiffness using SWE. While this provides valuable insight into the behaviour of the muscle-tendon unit under unloaded conditions, it does not capture dynamic function in vivo. Clinically, the manifestation of stiffness or excessive range following injury is not determined solely by passive tissue mechanics, but also reflects neuromuscular inhibition, scar tissue formation, altered recruitment strategies, and impaired coordination.
Rotational Mechanisms and Rehabilitation Pathways
Severe T-Junction injuries can appear benign in the clinic, with pain often resolving quickly. Indeed, it is common for athletes to report minimal symptoms after a matter of days, despite substantial damage evident on subsequent scans (Entwisle et al., 2017). The result is a diagnostic trap that fosters either premature or unnecessarily delayed return to play.
Part of this challenge lies in our reliance on sagittal-plane assessment. Clinical provocation tests such as straight leg raise, 90° hip flexion stretch, bridges, or prone curls, and objective measures such as NordBord, hip-iso push, or isokinetics, predominantly capture sagittal performance. While useful for evaluating sagittal plane function—knee flexion and hip extension—these tests overlook the biceps femoris’ role in transverse plane control as an external rotator of the hip (long head) and knee (short head), which is critical given the multiplanar demands characteristic of T-junction injury mechanisms (Kerin et al., 2022). The dominant mechanisms of injury—open chain long kicking with ipsilateral trunk rotation or closed chain reaching while deceleration—are poorly challenged through such testing protocols. This limitation explains why athletes can pass standard benchmarks while still remaining vulnerable.
While tools such as the SMAS (Sprint Movement Assessment Scale) (Bramah et al., 2024) and CMAS (Cutting Movement Assessment Scale) (Dos’Santos et al., 2019) help identify aspects of trunk and lumbopelvic control that influence transverse plane stress, no single objective test can reliably predict a successful rehabilitation outcome. Periodising and gradually integrating rotational demands within rehab helps bridge this gap. By systematically layering anti-rotation, rotation-generating, and high-speed decelerative tasks, clinicians can recreate the stresses associated with injury while promoting progressive tissue remodelling (Figure 8).
Figure 8: Elevate × Kerin Performance rotational curriculum, integrating rotational loading streams with rotational running progressions.
Embedding progressive, rotationally biased tasks serves three key purposes: 1) it reflects the mechanism of injury, 2) stimulates robust scar formation, and 3) rebuilds both athlete and clinician confidence. It also offers a reproducible framework for progression that can provide a thread through rehabilitation in addition to more standardised and established strength tests (Figure 9).
Figure 9: Example of the Elevate × Kerin Performance rotational curriculum. Indoor rehabilitation tasks target knee, hip, and trunk streams, while outdoor rotational running progresses from linear to multi-planar and transitional drills.
Key Takeaways and Reflections
- The T-junction is distinct
Dual innervation, variable anatomy, connective tissue overlap, and delayed scarring make it structurally and biologically different and potentially more vulnerable than proximal strains. - Current grading systems fall short
T-junction injuries may be underestimated on MRI because they do not fit existing classification systems, making prognosis challenging. Ultrasound, particularly when used dynamically, can add value by identifying the injured tissue, showing gapping or asynchronous motion, and tracking healing over time. - Intermuscular control > strength
Recent work from Mendiguchia and colleagues highlights the role of the short head in buffering load from the long head. Inefficiencies in this process can create imbalances that increase vulnerability at the T-junction. Effective rehabilitation therefore depends on restoring coordinated interaction between heads and targeting the specific muscle involved, rather than defaulting to protocol-driven, general loading, and improving strength. - Rotation is both mechanism and a potential assessment tool
T-Junction injuries often occur during rotational lengthening tasks in both the open- and closed-chain. Embedding rotation in rehab can have a multifactorial role: acting as a progressive competency marker, stimulating collagen alignment and neotendon stiffening, and developing athlete and clinician confidence in the position of greatest T-Junction strain.
REFERENCES
Bramah, C., Tawiah-Dodoo, J., Rhodes, S., Elliott, J. D., & Dos’Santos, T. (2024). The Sprint Mechanics Assessment Score: A Qualitative Screening Tool for the In-field Assessment of Sprint Running Mechanics. Am J Sports Med, 52(6), 1608-1616. https://doi.org/10.1177/03635465241235525
Cronin, K., Kerin, F., O’Flanagan, S., & Delahunt, E. (2024). ‘T-junction’ hamstring muscle injury in an elite-level football player. Br J Sports Med, 58(16), 930-931. https://doi.org/10.1136/bjsports-2024-108463
Dos’Santos, T., McBurnie, A., Donelon, T., Thomas, C., Comfort, P., & Jones, P. A. (2019). A qualitative screening tool to identify athletes with ‘high-risk’ movement mechanics during cutting: The cutting movement assessment score (CMAS). Phys Ther Sport, 38, 152-161. https://doi.org/10.1016/j.ptsp.2019.05.004
Entwisle, T., Ling, Y., Splatt, A., Brukner, P., & Connell, D. (2017). Distal Musculotendinous T Junction Injuries of the Biceps Femoris: An MRI Case Review. Orthop J Sports Med, 5(7), 2325967117714998. https://doi.org/10.1177/2325967117714998
Fyfe, J. J., Opar, D. A., Williams, M. D., & Shield, A. J. (2013). The role of neuromuscular inhibition in hamstring strain injury recurrence. J Electromyogr Kinesiol, 23(3), 523-530. https://doi.org/10.1016/j.jelekin.2012.12.006
Kayani, B., Ayuob, A., Begum, F., Singh, S., & Haddad, F. S. (2020). Surgical Repair of Distal Musculotendinous T Junction Injuries of the Biceps Femoris. Am J Sports Med, 48(10), 2456-2464. https://doi.org/10.1177/0363546520938679
Kerin, F., Farrell, G., Tierney, P., McCarthy Persson, U., De Vito, G., & Delahunt, E. (2022). Its not all about sprinting: mechanisms of acute hamstring strain injuries in professional male rugby union-a systematic visual video analysis. Br J Sports Med, 56(11), 608-615. https://doi.org/10.1136/bjsports-2021-104171
Kerin, F., O’Flanagan, S., Coyle, J., Curley, D., Farrell, G., Persson, U. M., De Vito, G., & Delahunt, E. (2024). Are all hamstring injuries equal? A retrospective analysis of time to return to full training following BAMIC type ‘c’ and T-junction injuries in professional men’s rugby union. Scand J Med Sci Sports, 34(2), e14586. https://doi.org/10.1111/sms.14586
Nakao, G., Nara, G., Adachi, R., Ishiyama, K., Kozawa, K., Sekiguchi, K., Nagaishi, K., Shiwaku, K., Hayashi, N., Mendiguchia, J., Kawama, R., Aoki, N., Katayose, M., & Taniguchi, K. (2025a). Mechanical interactions between the biceps femoris long and short heads: Implications for T-junction hamstring injuries. Clin Physiol Funct Imaging, 45(5), e70026. https://doi.org/10.1111/cpf.70026
Nakao, G., Nara, G., Adachi, R., Ishiyama, K., Kozawa, K., Sekiguchi, K., Nagaishi, K., Shiwaku, K., Hayashi, N., Mendiguchia, J., Kawama, R., Aoki, N., Katayose, M., & Taniguchi, K. (2025b). Regional changes in shear modulus of the biceps femoris long head following load application to the biceps femoris short head. J Biomech, 192, 112947. https://doi.org/10.1016/j.jbiomech.2025.112947
Paton, B. M., Court, N., Giakoumis, M., Head, P., Kayani, B., Kelly, S., Kerkhoffs, G., Moore, J., Moriarty, P., Murphy, S., Plastow, R., Pollock, N., Read, P., Stirling, B., Tulloch, L., van Dyk, N., Wilson, M. G., Wood, D., & Haddad, F. (2023). London International Consensus and Delphi study on hamstring injuries part 1: classification. Br J Sports Med, 57(5), 254-265. https://doi.org/10.1136/bjsports-2021-105371
Pollock, N., James, S. L., Lee, J. C., & Chakraverty, R. (2014). British athletics muscle injury classification: a new grading system. Br J Sports Med, 48(18), 1347-1351. https://doi.org/10.1136/bjsports-2013-093302
Pollock, N., Kelly, S., Lee, J., Stone, B., Giakoumis, M., Polglass, G., Brown, J., & MacDonald, B. (2022). A 4-year study of hamstring injury outcomes in elite track and field using the British Athletics rehabilitation approach. Br J Sports Med, 56(5), 257-263. https://doi.org/10.1136/bjsports-2020-103791
Schuermans, J., Van Tiggelen, D., Danneels, L., & Witvrouw, E. (2014). Biceps femoris and semitendinosus–teammates or competitors? New insights into hamstring injury mechanisms in male football players: a muscle functional MRI study. Br J Sports Med, 48(22), 1599-1606. https://doi.org/10.1136/bjsports-2014-094017
Shamji, R., James, S. L. J., Botchu, R., Khurniawan, K. A., Bhogal, G., & Rushton, A. (2021). Association of the British Athletic Muscle Injury Classification and anatomic location with return to full training and reinjury following hamstring injury in elite football. BMJ Open Sport Exerc Med, 7(2), e001010. https://doi.org/10.1136/bmjsem-2020-001010
Silder, A., Heiderscheit, B. C., Thelen, D. G., Enright, T., & Tuite, M. J. (2008). MR observations of long-term musculotendon remodeling following a hamstring strain injury. Skeletal Radiol, 37(12), 1101-1109. https://doi.org/10.1007/s00256-008-0546-0
