Literature search

We performed a comprehensive search of PubMed, Medline, Cochrane, CINAHL and Embase databases using various combinations of keywords “sport”, “sports”, “youth sports”, “young athletes”, “former athletes”, “football players”, “football”, “soccer players”, “soccer”, “tennis players”, “tennis”, “swimmers”, “swimming”, “divers”, “wrestlers”, “wrestling”, “cricket”, “gymnastics”, “skiers”, “baseball players”, “baseball”, “basketball players”, “basketball”, “osteoarthritis”, “former athletes”, “strain”, “contusion”, “distortion”, “injuries”, “trauma”, “injury”, “drop out”, “dropping out”, “attrition”, “young”, “youth”, “adolescent”, “long term”, “follow up” and “athlete”, over the years 1966–2009. All articles relevant to the subject were retrieved, and their bibliographies hand searched for further references in the context of long-term health outcomes of youth sports participation. Given the linguistic capabilities of the research team, we considered publications in English, Italian, French, Spanish and Portuguese. The search was limited to articles published in peer-reviewed journals. Studies on catastrophic injuries and related outcomes were excluded from the review, as these are addressed elsewhere in this special journal issue.

Dropping out due to injury

It is estimated that 8% of Australian adolescents drop out of recreational sporting activities annually because of injury occurrence.9 Some young athletes discontinue sports completely, whereas other children may drop out of one sport and still participate in other sports.10 The most common reasons that children drop out include: “no longer interested in the sport” (highest for both boys and girls), “no longer fun”, “the coach played favourites/was a poor teacher” and “wanting to participate in other activities”.11

Data from several case series provide some insight into a relationship between injury and continued sport participation. Jackson et al12 reviewed 10 cases of osteochondritis dissecans of the humeral capitellum in seven high-performance female gymnasts, ranging in age from 10 to 17 years. The average length of follow-up was 2.9 years. Only one of the athletes, at the time of follow-up, was still in competitive gymnastics. Each has a minimal limitation in range of motion, with crepitus and often catching noted by history and on examination.

Singer and Roy13 reported seven cases of osteochondrosis of the capitellum occurring in five high performance female gymnasts between the ages of 11 and 13 years. Two of the gymnasts were managed by surgical excision of the loose osteochondral fragment in three elbows. Four of the five gymnasts, including the two undergoing surgical management, were able to return to full workouts without recurrence of symptoms within the 3-year follow-up.

Katz and Scerpella14 reported on seven female gymnasts aged 11–17 years with a diagnosis of injury to the anterior or middle column of the thoracolumbar spine. All subjects had either ceased or markedly reduced gymnastic participation because of back pain at the final follow-up. Injury was career-ending for three of the gymnasts. One additional subject continued to compete for 2 years but was forced to quit when she advanced to a higher competitive level and her back pain returned. Two other gymnasts returned to their sport but were forced to end their careers because of other injuries 2 and 10 months later. One returned to gymnastics after treatment.

In a 10-year cohort study of Australian elite gymnasts, seven women and one man (8/116) retired as a result of injury.15 Injuries that resulted in retirement included chronic rotator cuff injury, navicular stress fracture, loose bodies in the ankle joint, medial and lateral meniscus lesions, ACL rupture and osteochondritis dissecans of the elbow joint. In three cohort studies,16 17 18 16.3–52.4% of dropouts were injured when they withdrew from participation, thus implicating injury as a likely contributing factor in the decision to retire from gymnastics.

Baseball is one of the safest US high school team sports;19 however, participation in this sport may represent a significant cause of long-term disability among participants, particularly among pitchers.20 A one-season study of 130 US high school pitchers identified injuries such as self-reported elbow pain resulting from pitching during a preseason or regular season game.21 Six of these pitchers experienced an acute episode of pain that inhibited performance, and four of the six missed at least one pitching turn. Ten pitchers had a previous history of elbow injury and had residual pain during the season studied, and 26 pitchers reported pain in the elbow without a previous history. No pitcher missed game time during the season as a result of elbow discomfort.

Two studies reported the frequency of season or tournament-ending injuries among young wrestlers. Of the 418 high school wrestlers studied by Pasque and Hewitt,22 23 (5.5%) sustained season-ending injuries. The most common of these injuries involved the knee (44%) and shoulder (22%). During the 2006 US junior freestyle and Greco-Roman wrestling tournament, 56% of injured wrestlers and 50% of injured Greco-Roman wrestlers discontinued participation in the tournament.23

Physeal injuries and growth disturbance

A concern regarding children’s participation in youth sports is that the tolerance limits of the physis may be exceeded by the mechanical stresses of sports such as football and hockey or by the repetitive physical loading required in sports such as baseball, gymnastics and distance running.24 Unfortunately, what is known about the frequency of acute sport-related physeal injuries is derived primarily from case reports and case series data. In a previous systematic review on the frequency and characteristics of sports-related growth plate injuries affecting children and youth, we found that 38.3% of 2157 acute cases were sport related and among these 14.9% were associated with growth disturbance.24 These injuries were incurred in a variety of sports, although football is the sport most often reported.

There are also accumulating reports of stress-related physeal injuries affecting young athletes in a variety of sports, including baseball, basketball, climbing, cricket, distance running, football, soccer, gymnastics, rugby, swimming and tennis.24 Although most of these stress-related conditions resolved without growth complications during short-term follow-up, there are several reports of stress-related premature partial or complete distal radius physeal closure of young gymnasts.25 26 27 28 29 There are also two reports of varus changes subsequent to sports-related stress injury to the distal femoral and/or proximal tibial physes (rugby and tennis players).30 31 These data indicate that sport training, if of sufficient duration and intensity, may precipitate pathological changes of the growth plate and, in extreme cases, produce growth disturbance.24 32

Disturbed physeal growth as a result of injury can result in length discrepancy, angular deformity, or altered joint mechanics, and may cause significant long-term disability.33 However, the incidence of long-term health outcomes of physeal injuries in children’s and youth sports is largely unknown.

Spine pathology

Young athletes involved in sports with great demands on the spine such as gymnastics and diving are at risk of overuse and acute injury to the spine. The radiological findings indicate the potential damaging effects of repetitive microtrauma and/or acute macrotrauma with injuries to the spine during growth, including damage to the pars interarticularis resulting in spondylolysis or spondylolysisthesis, discogenic pathology and abnormalities of the vertebral endplates and vertebral ring apophyses of the thoracolumbar spine.34 35 36 37 38 39 40 41 42 43

Two recent studies provide some insight into the long-term outcome of sport-related spinal conditions experienced by young athletes. Baranto et al35 conducted a 5-year follow-up magnetic resonance imaging (MRI) study of back pain and spinal abnormalities in young elite divers, aged 10–21 years. They found that 65% of the divers had MRI abnormalities in the thoracolumbar spine at baseline and that the total number of abnormalities increased by only 29% at follow-up. Sixteen of 18 divers (89%) had a previous or present history of back pain. Baranto et al34 also conducted a 15-year follow-up of back pain and MRI changes in the thoracolumbar spine of top adult athletes (aged 18–41 years at baseline) representing four different sports. Similar to their earlier findings, the authors reported that most of the spinal abnormalities seen on follow-up were also present at baseline.

Of particular concern is the possible relation between back pain, particularly long-term pain or dysfunction and the spinal abnormalities observed in young athletes. Several reports indicate that back pain is common among young athletes35 40 42 44 45 and MRI studies of athletes have shown a correlation between disc degeneration and low back pain.46 47 48 49 50 It has also been reported that athletes with abnormalities of the vertebrae and disc degenerations have more back pain than other athletes and non-athletes.36 40 43 51 52 53 54 However, Lundin et al55 did not find a higher frequency of back pain in athletes compared with non-athletes despite significantly more radiological abnormalities among the athletes.46

These study results associate intense physical loading and/or acute trauma with injuries to the spine during growth. However, even though sporting activity is regarded as an important predisposing factor in the development of spinal pathologies, there is limited knowledge of the long-term effects of spinal trauma in athletes, both regarding morphological appearance and back pain.35

Knee injury and osteoarthritis

Knee, ankle, hip and foot injuries in adults significantly increase the risk of the early development of osteoarthritis.56 57 58 59 What is the evidence that long-term effects of acute knee injury during adolescence promote osteoarthritis? In long-term follow-up of young athletes with meniscus surgery, more than 50% of patients will have knee osteoarthritis and associated pain and functional impairment.60 61 62 63 Follow-up studies of young non-athletes exposed to meniscectomy during adolescence are consistent with these results.64 65 66 67

ACL injuries are becoming increasingly common among young athletes, particularly young women in the 15–19-year age group.68 Unfortunately, little is known about the effect of age as a risk factor for ACL injury because of a lack of age-related exposure data. A major problem after an ACL injury is that, regardless of treatment, athletes with the injury retire from active participation at a higher rate than athletes without this injury.69 In one study, 80% of soccer players reported reduced activity levels 14 years after ACL injury.59 These outcomes are usually due to residual knee instability, reduced range of motion and/or stiffness and pain.

Long-term follow-up studies provide evidence that 12–20 years after knee injury (meniscus or ACL), more than 50% of those injured will have knee osteoarthritis compared with 5% in the uninjured population.70 71 As such, knee injury and related meniscus or ACL surgery during childhood or adolescence may reduce future involvement in physical activity, leading to less than optimal health in later life.

ACL reconstruction in children

A systematic review on studies published until 200672 reporting on the prognosis of conservatively managed ACL injuries showed that there was an average reduction of 21% at the level of activities (Tegner score evaluation). ACL reconstruction is therefore a procedure frequently performed in athletic individuals, as they desire to maintain a high level of activities. ACL reconstruction in skeletally immature patients is a relatively new trend.73 The concern with this procedure is intra-operative epiphyseal damage and growth disturbance, a complication that has been avoided in several studies.74 75 76 77 78

One of the early reports showed that there were no growth disturbances at a mean of 3.3 years after surgery in nine children; however, with two re-ruptures. Those children could not return to athletic activities.74

In a series of 57 ACL reconstructions, 15 patients had reached completion of growth when examined at follow-up, none had signs of growth disturbance, whereas clinical scoring was good or excellent in all patients.75

Another study compared the outcomes of two management strategies in 56 children with ACL ruptures, namely ligament reconstruction in the presence of open physis, or delayed reconstruction after skeletal maturity. The “early” reconstruction group had evidence of fewer medial meniscal tears (16% vs 41%) and no evidence of growth disturbance at 27 months mean follow-up.76

After 1.5–7.5 years follow-up of 19 ACL reconstructions in 20 athletic teenagers (age range 11.8–15.6 years), all but one had returned to sports, none had tibiofemoral malalignment or a leg length discrepancy of more than 1 cm and the modified Lysholm score was 93 out of 95.77

Finally, 55 children (aged 8–16 years, mean 13 years) were followed for a mean of 3.2 years (range 1–7.5 years) after ACL reconstruction, with no evidence of growth disturbance. Clinical scores revealed normal or almost normal values (higher than 90 out of a 100 possible points) and 88% of the patients went back to normal or almost normal sports according to the Tegner score.78

Overall, the clinical results are encouraging and iatrogenic epiphysis damage does not appear to present a problem in ACL construction. However, study designs could be improved to answer the question of whether early or delayed ACL reconstruction results in the best possible outcome in skeletally immature patients.79

Moksnes et al80 performed a study evaluating performance-based functional outcome in children aged 12 years or younger affected by ACL injury. The population study consisted of 20 non-operated subjects (21 knees) and six ACL reconstructed subjects (seven knees) who were examined at a minimum of 2 years after ACL injury or reconstruction. The results showed that 65% of the non-operated children (compared with 67% ACL reconstructed subjects) are able to perform sports at their pre-injury activity level without a high risk of meniscus injury; moreover, they demonstrated excellent knee function on performance-based single-legged hop tests.

Aichroth et al81 studied children and adolescents with tears of the ACL, demonstrating that patients treated conservatively develop a severe instability with poor function of the knee at a mean follow-up period of 72 months, whereas patients who underwent reconstruction showed marked improvement in the pivot-shift and Lachman tests at a mean follow-up period of 49 months.

Arbes et al82 evaluated the functional outcome of skeletally immature patient with ACL rupture in a mean follow-up period of 5.4 years. In that study, ACL reconstruction represented the best management. The authors therefore suggest early ACL reconstruction in young patients, but propose delayed ACL reconstruction in patients who have contraindications to or refuse surgery.

Kocher et al83 have proposed a combined intra-articular and extra-articular reconstruction of the ACL in skeletally immature prepubescent children and adolescents. They demonstrated that this technique provides excellent functional outcome with a minimal risk of growth disturbances due to a physeal sparing reconstruction. However, Seon et al84 demonstrated that ACL reconstruction, in skeletally immature adolescents, can be performed using the transphyseal tunnel without significant leg length discrepancy or abnormal angular deformity of the knee joint.

Mizuta et al85 evaluated the conservative management of complete tears of the ACL in 18 skeletally immature patients, with a minimal follow-up of 36 months. Only one patient returned to pre-injury level sports activity, while 11 patients showed radiological evidence of degenerative changes. For this reason, the non-operative treatment for ACL injuries is not acceptable, with fair or poor results in 16 of the 18 treated.

Øiestad et al86 conducted a systematic review of studies on the prevalence of osteoarthritis in the tibiofemoral joint more than 10 years after an ACL injury. Notably, these authors conclude that the prevalence rates of knee osteoarthritis after ACL reconstruction reported by previous reviews have been too high. However, it should be considered that most of the studies included were conducted in skeletally mature patients.

Challenges for future research

Participation in sports offers potential benefits for individuals of all ages, such as combating obesity and enhancing cardiovascular fitness. On the other hand, the negative consequences of musculoskeletal injuries sustained during sports in childhood and adolescence may lead to incomplete recovery and residual symptoms and can cause joint degeneration in the long term, limiting the ability to experience pain-free mobility and engage in fitness-enhancing activity throughout adult life.87 Few well-conducted studies are available on the long-term follow-up of youth athletes and, in general, we are lacking studies reporting on their health-related quality of life compared with the general population.

This review underscores the need for prospective cohort studies to clarify the long-term health outcomes of youth sports injuries. Important to this research is the meticulous documentation of injuries on injury report forms that include age-appropriate designations for injury types that are specific to children and youth. Also important to this research is the determination of injury rates based on time-at-risk exposure data. Reliable descriptive data also provide a basis from which causes of injury can be probed and the effectiveness of preventive measures evaluated.