Summary of findings
This appears to be the first systematic review on OIE in adults comparing the occurrence in various sports. We attempted to identify any differences between sports in relation to diagnoses and anatomical areas most likely to be injured. We were able to retrieve 10 studies on three different sports: soccer (N = 8), triathlon (N = 1) and beach volleyball (N = 1). Methodological differences between studies and a limited number of studies and sports studied made it difficult to provide clear answers. However, in relation to the proportion of OIE it varied between 0.2 to 13.3 per 1000 h of exposure, with soccer not having the highest estimates. This proportion is generally more important in adults than in youngsters, where results around 0.5 were found [5].
Injury site was, as for the youngsters [5], mainly the lower limb. However, in adults this was reported rather for the knee, tibia, thigh and pelvis/hip/groin whereas in youngsters it was the knee and the lower leg.
As for the diagnoses, they were most frequently (when at all provided) reported to be tendinitis/bursitis, and strain, whereas in children and adolescents the most commonly reported diagnoses were tendinitis/bursitis and periostitis [5].
Methodological aspects of the articles reviewed
A large body of literature on sports injuries of adults, as well as of children, is written by a group of researchers that uses the same methodological approach when surveying injuries in different sports. Typically, they study injuries in single sport clubs or during specific sports events with the ultimate goal to compare risk estimates for various sport activities. To record a sufficiently large number of injuries of specific sports in the general population is of course difficult, hence this approach. However, when choosing such a tactic, it would be relevant to collect similar data from several clubs/events, in order to even out any bias associated with single convenience samples of such type.
After having reviewed this literature on both children/adolescents and adults, it is clear that even when multiple studies are found for similar sports, data are often collected at different intervals, in different ways, using different definitions for injury, and for different specific types of injuries. Authors do not clearly report diagnosis and anatomical areas of injury, and if they do, they often leave out the one or the other. This, also, makes it difficult to make comparisons and to establish risk estimates. A simple example is the difference in estimates expected when the presence of an “injury” is reported as “complaint”, as “sought care”, or “time loss”. Further, in the case of “overuse”, absence of a traumatic etiology seems often automatically to result in a diagnosis of an “overuse” injury, merely because the person with the complaint was involved in a sporting activity. It is not logical that people involved in studies on sport injuries only have these two possible diagnoses, traumatic or overuse injury. Surely patients from the general population are diagnosed from a larger spectrum of possibilities. Clear criteria for this diagnostic label have been proposed [5] and discussed in the literature [20], but seem to be largely ignored, at least when reports are written up.
As for the definitions of “incidence” and “prevalence”, true incidence and prevalence estimates are usually not distinguished in studies within this area. The incidence is defined as number of injuries based on 1000 h of session (training, competition or both), in general without regards concerning the previous injury. In fact, this should not really be called incidence but prevalence. This issue has been previously discussed by Bahr [7]. Further, the numbers of potential and included study subjects are often not reported. Clearly, an injury rate (per 1000 h) would be more credible when obtained from many study subjects than from a few. It would therefore be useful for the reader to have access to both these denominators.
Admittedly, the objectives of our review were not the same as the objectives of the studies under review, which makes difficult the extraction of information in our review. Nevertheless, as we have already discussed in our previous review on children/adolescents [5] in our opinion, this research area would benefit from a well-reasoned consensus approach to the various definitions.
Methodological aspects of our review
Our review followed the current guidelines, using a transparent approach, searched several databases, and data were extracted blindly by two reviewers. However, it is possible that some articles could have been missed, as only texts written in English, French and Scandinavian languages were acceptable for inclusion. Checklists for data extraction have been previously tested and used in a previous review and were therefore known to be user-friendly and relevant.
Sometimes we had to make assumptions regarding the nature of injuries, when exact information regarding the site of injury was missing. Thus two diagnoses, tendinopathy and periostitis, were systematically considered as extremity injuries, whereas some diagnoses such as strain was not, because it could affect the spine.
Discussion of findings regarding the incidence of OIE
We did not find any information in the literature on OIE in the general population of adults. However, the incidence of OIE in general population of schoolchildren has been reported to be 2.3(1.6–3.0 95% CI) for the upper extremity and 3.7(3.5–4.0) for the lower extremity [4].
Discussion of findings regarding the anatomical site of OIE
As observed in the previous review on children and adolescents [5], the lower extremity is more often affected than the upper extremity in the sports studied. Only three sports could be considered in this review, so it is difficult to compare the localisation of OIE between sports. However, we noted that in soccer, in youngsters and in adults, the pelvis/hip/groin are more often affected than in the other sports. We assumed that this is due to the shearing force often imposed on the pelvis in soccer.
Discussion of findings regarding the diagnosis of OIE
Only four articles provided good information on the diagnosis of OIE and they all studied soccer, making it impossible to compare this finding with other sports. In childhood, 8 articles reported the diagnosis making a comparison relevant. However, for all sports covered, it was always the two same diagnoses that were reported.
Tendinis/bursitis is the most common diagnosis both in childhood and adulthood, followed in adults by synovitis, and in youngsters by periostitis. Probably because of the difference in bone skeletal maturity, osteochondral disorders, present in youngsters, did not appear in adults.