When dealing with muscle injuries, making an accurate diagnosis and determining injury severity needs of as much information as possible related with injury mechanism, affected muscles, anatomical location of the injury inside the muscle (myofascial, intramuscular, muscle-tendon junction, intra tendinous), size of the injury, tendon involvement and finally history of previous injuries in the lower limbs, groyne or low back. This information together with some other more subjective variables, such as pain, may be useful in guiding us through the rehabilitation programme.

The injury mechanism (sprinting, changing direction, kicking, dominant/non-dominant, stance/swing leg and trauma) may be one of the first aspects to consider. Muscle contusions usually need a shorter recovery period than the non-contact injuries, thought to be caused by lengthening beyond the optimal length of the activated muscles [12]. In a sample of 2003 thigh muscle injuries of the UEFA Elite League injury study, the mean lay-off time for indirect (strain) injuries was of 18.5 days, significantly longer than the 7 days of direct (contusion) injuries [13].

Lower limb muscle injuries have a high rate of recurrence, and reinjuries seem to cause up to 30% longer absence time than first-episode muscle injuries [1]. The positive correlations found between lower limb muscle injury incidence and previous lower limb muscle, knee or groyne injuries [2, 9, 14] could support the hypothesis of biomechanical changes and related post-injury inadequate compensations as the cause of further injuries. This highlights the importance for clinicians to investigate the player’s injury history and regularly monitor those biomechanical alterations that could be related with previous injuries, before allowing players to RTP [2].

The purpose of the first physical examination is to determine the location and severity of the injury. More than the length of the painful area, it has been suggested that it is the distance from the point of maximum pain to the proximal insertion that is associated with the absence period [15]. The more proximal the site of maximum pain, the greater the time needed to return to preinjury level [16].

Related with the lower limb affected muscle, quadriceps injuries are usually the ones requiring longer rehabilitation periods for RTP, while groyne muscle injuries cause the shorter absence [1, 3]. Injuries involving damage to the proximal free tendon [17, 18] or the central tendon of the muscle [19–22] are usually more severe and take longer for RTP. On the contrary, injuries located within the muscle belly, not involving the tendon or muscle-tendon junction, usually have a more benign prognosis and a shorter recovery period until RTP [23].

More than 1 day to walk pain-free has been shown to be associated with a longer (> 3 weeks) recovery period to RTP in hamstring injuries [24]. Symptoms that persist for more than 5 days might need considering more extensive tissue damage or intramuscular hematoma and most probably require special attention and a longer recovery [10].

Verrall et al. [25] found that patient reported pain and clinician’s estimate of injury severity correlated with the RTP in 83 Australian Rules football players with posterior thigh injuries. A few years later, the same group of researchers [26] found that swelling, bruising, tenderness and pain on hamstring contraction in the initial clinical examination had no value in predicting the likelihood of reinjury in 30 hamstring injuries of Australian Rules footballers. Besides the above-mentioned conflicting findings, looking at early clinical signs and symptoms and continuous monitoring during the rehabilitation period is highly recommended. We still need further studies to confirm the validity of clinical findings for predicting the severity and prognosis of lower limb muscle injuries.

A lack of core stability has been widely considered as a risk factor for hamstring injuries [10]. It has been proposed that any sacroiliac joint and/or pelvic dysfunction can affect hamstring mechanical behaviour, alter the load transfer from the spine to the legs and thereby increase the injury risk [10]. Correcting these alterations early in the acute phase may help in restoring the lumbopelvic function and lead to a more successful RTP in thigh muscle injuries.

Sherry and Best [27] found that a rehabilitation programme consisting of progressive agility and trunk stabilization exercises is more effective in promoting return to sports and preventing injury recurrence in athletes suffering an acute hamstring strain than a more traditional isolated hamstring stretching and strengthening programme. Although the morphological and neuromuscular factors were not measured, these results suggest that there may be a role for lumbopelvic neuromuscular control exercises in the prevention of hamstring injuries and reinjuries [28]. The lumbar muscles provide localized segmental control of the lumbar spine, and it has been shown that a smaller size (cross-sectional area) of the multifidus or quadratus lumborum muscles may be predictive of lower limb injury incidence in Australian Rules football players [29].

Abdominal and lumbar muscles may also be important in preventing quadriceps injuries. During kicking, abdominal muscles are necessary to reduce the quadriceps overload and help in controlling the lateral displacement of the trunk to the non-kicking side at foot-to-ball contact [30]. Kicking may also demand high levels of activation of the quadratus lumborum of the non-kicking side, to counteract torsion and side flexion moments [31]. Core stability exercises have therefore become an important part of most thigh injuries prevention and rehabilitation programmes and should be incorporated early in the subacute phase.

Despite conflicting data in the literature, it has traditionally been suggested that greater flexibility may reduce the risk of muscle injuries due to greater compliance of the passive components of the muscle-tendon unit, at least in sports involving bouncing and jumping activities with a high intensity of stretch-shortening cycles [32]. A few prospective studies have identified relationships between reduced hamstring flexibility and hamstring injuries [33–35] as well as between reduced quadriceps flexibility [35] or quadriceps asymmetries [36] and quadriceps injuries, in professional football players. Malliaropoulos et al. [37] found that knee active ROM deficit was correlated with recovery time in 165 track and field athletes with acute, first-time, unilateral posterior thigh muscle injuries. Recently, Moen et al. [38] reported that passive straight leg raise deficit was associated with time to RTP in 80 non-professional athletes with magnetic resonance imaging (MRI)-positive hamstring injuries. By contrast, Askling et al. [17] found no correlation between measures of hip flexion ROM and time to RTP of hamstring injuries, in 18 elite sprinters and 15 professional dancers. Part of the contradictory results might be related with the absence of gold standard methods for flexibility measurements and with difficulties in stabilizing the hip and lumbar spine in some of the most widely used tests, such as the sit-and-reach, straight leg raise and toe-touch test [39]. Hunter and Speed [40] recommend the active knee extension (AKE) test that measures hamstring flexibility at 90 degrees of hip flexion, but more dynamic tests such as the active hamstring flexibility test proposed by Askling et al. [41] might be more sensitive to detect differences not only in flexibility but also in insecurity before allowing the player to return to full training and competition.

Reduced hip flexors flexibility has also been identified as a risk factor for hamstring injuries. An increase in anterior pelvic tilt, due to tight hip flexors and limited hip extension, could cause excessive hamstring stretch in the opposite limb and thus increase the risk of hamstring injury [10, 42].

Hamstring and hip flexors flexibility exercises should be initiated soon after injury, but always trying to avoid an increase in neural tension and including dynamic and functional exercises that also involve hip stability and neuromuscular control [10].

There are a few reasons to support that achieving optimal levels of both quadriceps and hip flexors flexibility should be a cornerstone of any quadriceps prevention and rehabilitation programme in football players [31]. In the kicking action, the hip flexors generate greater hip flexion moment by using the stretch-shortening cycle. A lack of hip extension during the early swing phase due to a tight iliopsoas may require higher force generation from the rectus femoris and lead to overload and early fatigue during repetitive kicking sessions. Mechanical irritation of the femoral nerve due to a restricted psoas has also been suggested as a possible cause of rectus femoris injuries [31]. The modified Thomas test is recommended for quadriceps and hip flexors flexibility assessment during any thigh injury rehabilitation programme [43].

Decreased ROM of the hip has been suggested to be a risk factor for sports-related chronic groin pain in athletes [44]. A decrease in hip abduction [45] and internal-external rotation [46] ROM of the hip have both been reported to be correlated with occurrence of groyne strains in professional football players. Although more prospective studies are needed, interventions directed at players with limited hip ROM could lead to reduce adductors injury and reinjury incidence and should therefore be considered in any groyne muscle injury rehabilitation programme.

Lower limb muscle injuries are sure related with different joints and groups of muscles, so an analytic approach to flexibility assessment in such a complex system should be avoided. Testing should be ideally performed with dynamic tests and in all the involved muscles. We should be very careful when interpreting results from static measures and avoid generalizing to dynamic actions [5].

Decreased muscle strength and strength imbalances have long been proposed as related to increased risk of thigh muscle injuries. Reduced hamstring strength is commonly perceived to be a risk factor for hamstring injuries, most of which occur during the late swing [47] or early stance phase [48] of sprinting. Hamstring eccentric strength training has been shown to increase the optimum length of tension development [49] and decrease hamstring injury risk in professional and amateur football players [50–52].

Isokinetic strength testing is frequently used in football, but there is still lack of consensus regarding its usefulness in RTP decisions [53]. Peak torque at greater knee flexion angle (i.e. shorter optimum length) [12] and eccentric hamstring strength deficits compared to the uninjured side, as well as the ratios of concentric hamstring to concentric quadriceps strength and eccentric hamstring to concentric quadriceps strength, have been proposed as the most useful variables to predict hamstring injury risk [54, 55]. In a recent study in 52 professional male football players with MRI-positive hamstring injuries, Tol et al. [11] found that when compared with the uninjured leg, 67% of the clinically recovered players showed at least one hamstring isokinetic testing deficit of more than 10%. They concluded that normalization of hamstring isokinetic strength variables does not seem to be required for successfully completing a football-specific rehabilitation programme. Because of the low reinjury incidence, it was not possible to conclude if there was any association between isokinetic strength deficits and reinjury incidence. In their study with hamstring injuries in sprinters and dancers, Askling et al. [17] did not find any correlation between knee flexion isometric strength (measured in prone position and with knee extended) and time to RTP. Freckleton et al. [14] demonstrated a significant deficit in preseason single leg hamstring bridge scores on the right thigh of Australian Rules football players that subsequently sustained a right-sided hamstring injury. The lack of correlation between strength and time to RTP in some studies could probably be related with the fact that assessment was performed with non-functional tests that are very different from the football-specific demands. Hamstring strength should be tested and trained at long muscle lengths, with the hip and knee at functional angles and aiming to assess and improve strength, power and endurance.

Hamstring eccentric training may result in greater structural stability at longer muscle lengths and consequently may have interesting implications for injury prevention. Despite the demonstrated benefit of the Nordic hamstring exercise in several studies [50–52], there might be greater benefit by using more functional unilateral eccentric exercises that involve both hip and knee motion, similar to that needed for sprinting and other football activities [28, 56, 57]. An adequate progression would allow the player to include these exercises in the last stages of the rehabilitation programme.

During running, the gluteus maximus has a multifaceted role not only as a powerful hip extensor but also to control trunk flexion of the stance leg and decelerate the swing leg [58]. Gluteus maximus activity has been shown to be much greater during sprinting than during running [59]. Any alteration in gluteus maximus strength, endurance or activation pattern will place greater demand on the hamstrings and may therefore increase the injury risk of this muscle group. Sugiura et al. [60] reported that hamstring injuries in a group of elite sprinters were associated with reduced hip extensors concentric strength. Exercises aimed at teaching how to isolate gluteus maximus activation from hamstrings, and improving gluteus maximus strength and endurance, should be included already in the subacute phase of hamstring injuries rehabilitation programmes [10].

Quadriceps injuries are more common in the kicking leg, most probably related with a greater exposure to high-risk actions [2]. Both the iliopsoas and rectus femoris contribute to hip flexion, and the ability to generate a greater hip flexion moment is critical to achieve a high foot velocity during kicking [31]. A reduction in strength and/or activation of the iliopsoas in football players may result in overload and increased injury risk of the rectus femoris [31]. Achieving optimal and balanced levels of quadriceps and hip flexors strength at long muscle lengths should be considered as a priority in every prevention and rehabilitation programme of rectus femoris injuries.

Eccentric training of the quadriceps has been shown to increase the optimum length of the knee extensors in professional football players [49]. As muscle injuries are thought to occur when muscles are contracted beyond their optimal length [12], rectus femoris prevention and rehabilitation programmes should also include sport-specific eccentric exercises of the knee extensors, with distances (sprints and decelerations), velocities and directional components similar to the football movements [31]. These eccentric exercises aimed to increase the optimal length of the knee extensors should only be incorporated in the last part of the functional phase of the rehabilitation programme.

Weak adductors have been suggested to be a risk factor of groyne injuries in sports requiring side-to-side cuttings, quick accelerations and decelerations and sudden changes in direction. Professional ice hockey players were 17 times more likely to sustain an adductor muscle strain if their adductor strength was less than 80% of the abductor strength [61]. Engebretsen et al. [62] found that the risk for a new groyne injury (22/61 acute injuries) in football players (first, second and third Norwegian divisions) with weak adductors was four times the risk of players with normal strength. An intervention programme aimed at improving adductor’s strength proved to be effective to prevent adductor injuries in professional ice hockey players [63]. Hip adductors/abductor strength assessment with hand-held dynamometer in the supine position has been shown to be a simple and reliable method [64], and athletes with an adduction-to-abduction strength ratio of less than 80% with this test have been considered as at risk [63]. Tyler et al. [63] propose a three-phase adductor muscle strain rehabilitation programme, progressing from submaximal isometric adduction exercises already in the acute phase to eccentric and sport-specific exercises in the last stages.