What's wrong with my hamstring? How hamstring harvest can influence muscular support of the knee.

The Anterior Cruciate Ligament (ACL) is one of the major ligaments of the knee that act to provide stability and constrain anterior tibial translation. General population estimates show 32 to 52 per 100,000 people per year rupture their ACL, with the majority occurring during sport (Gianotti et al., 2008; Granan et al., 2009; Janssen et al., 2012). Of even greater concern, is the frequency of ACL re-injury, with 27% of patients suffering a second ACL injury within ten years of the initial reconstruction (Pinczewski et al., 2007; Pinczewski et al., 2006). Across various sports, non-contact injuries have been found to make up 50-80% of ACL injuries (Arendt et al., 1999; Boden et al., 2000; Cochrane et al., 2007), with video analysis showing manoeuvres such as side-stepping and single leg landing to be a common cause of ACL injury (Cochrane et al., 2007; Olsen et al., 2004).

ACL injuries are reported to occur early after initial ground contact with knee flexion angles less than 30-degrees, and the knee typically giving way in valgus and internal rotation (Cochrane et al., 2007; Krosshaug et al., 2007). If activated appropriately, knee musculature may have the ability to help stabilise the valgus and internal rotation moments that occur during sidestepping (Besier et al. 2003). This may act to reduce loading of the ligament, however, when external loads are large enough and muscular stabilisation is insufficient, this can lead to ACL rupture (Lloyd, 2001).

Surgical intervention is generally required to restore stability to the tibiofemoral joint, however no perfect graft choice exists for reconstruction of the ACL, with all graft choices having potential advantages and disadvantages. Graft options that are currently used include autografts from the quadriceps tendon, patella tendon or hamstring tendon, a variety of allograft options, and synthetic graft choices (Romanini et al., 2010). The quadrupled hamstring autograft taken from the semitendinosus (ST) and gracilis (GR) muscles is the most common choice for orthopaedic surgeons. However, harvest of the ST and GR tendons leads to post-operative donor muscle atrophy, as well as proximal retraction of the musculotendinous junction (Figure 1)(Konrath et al., 2016a). This in turn, has implications for tibiofemoral joint function, stability and loading.

Previous studies have revealed the morphology of these donor muscles to be substantially altered following tendon harvest, with either no tendon, or abnormal tendon regeneration resulting in knee flexion and internal tibial rotation weakness (Konrath et al. 2016a). Moreover, there is emerging evidence of this harvest to significantly reduce knee muscular protection during sidestep cutting.

Work from my PhD using computational neuro-musculoskeletal models showed the contribution of donor muscles to muscular support of the tibiofemoral joint (TFJ) during sidestep cutting to be decreased following ACL reconstruction (Konrath et al. 2016b). An EMG-driven neuro-musculoskeletal model was used to determine the muscle force estimates of 34 musculotendinous units of the lower limb and to determine the contribution of donor muscles to muscular support (Figure 2) based on normal muscle-tendon parameters (Standard model), and with subject-specific changes to muscle-tendon parameters for the donor muscles (Adjusted model).

The combined contribution of the donor muscles to muscular support about the medial and lateral compartments were reduced by 52% and 42% respectively in the adjusted compared to standard model. Figure 3 shows the contribution of the ST, GR and semimembranosus (SM) muscles toward support about the medial and lateral compartments across stance phase plotted against the knee flexion angle. As we observe, for the standard model, the ST contributes on average 35% of muscular support about the lateral compartment at early stance, when knee flexion angles are at 20 degrees (Figure 3). In the adjusted model, representing donor muscle morbidity, the contribution to muscular support is reduced to roughly 20% at early stance. 

 Figure 3: The mean contribution to muscular support about the medial compartment (MC) and lateral compartment (LC) with the knee flexion angle during stance phase for the (A) semitendinosus (B) gracilis and (C) semimembranosus. Shaded regions indicate ± one standard error. The red shows the standard model, while the blue shows the adjusted model.

As mentioned previously, video analysis of sports-related ACL injuries report the knee to collapse into valgus and/or internal tibial rotation with knee flexion angles of less than 30 degrees shortly after initial foot strike (Cochrane et al. 2007). In addition to this, cadaveric work has shown the ACL to be more susceptible to strain under combined flexion and valgus moments with small knee flexion angles (Markolf et al., 1995). Given the ST and GR have the mechanical potential to contribute to support of external valgus moments at smaller knee flexion angles, their reduced function may infer higher loads to the ACL during early stance if they are not compensated by other non-donor knee muscles (Konrath et al. 2016). These results suggest that hamstring harvest may reduce muscular support of the knee during evasive sidestepping movements, and may increase the risk of sustaining a re-rupture of the ACL.