| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Department of Radiology, University of Washington Medical Center, Roosevelt
Clinic, Seattle, WA 98195.
2 Present address: Department of Radiology, University of California Davis
Medical Center, 4860 Y St., Ste. 3100, Sacramento, CA 95817.
3 Present address: Desert Radiologists, Las Vegas, NV 89106.
4 Departments of Orthopaedics and Sports Medicine, University of Washington
Medical Center, Seattle, WA 98195.
5 University of Washington Intercollegiate Athletics, Seattle, WA 98195.
Received February 17, 2005;
accepted after revision August 12, 2005.
Address correspondence to E. M. Escobedo.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. We conclude that posterior glenoid labrum tears are more prevalent in football players than in non-football players.
Keywords: glenoid labrum • MR arthrography • MRI • musculoskeletal imaging • shoulder • sports medicine • trauma
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Classically, posterior shoulder instability has been considered to be less common than its anterior counterpart in the general population [1-3]. There has been a relative paucity of clinical research regarding this injury in contact-sports athletes and, specifically, in football players. However, two series have described posterior glenoid labrum injuries in athletes [4, 5], predominately football players. In both of these studies the patient cohort was small with no control group used.
Mair et al. [4] suggested that these posterior labral injuries might be a "type of occupational hazard for contact athletes." The mechanism they proposed is that this injury is the result of posteriorly directed forces on the shoulder during blocking, leading to excessive shear forces directed against the posterior labrum. They further postulated that repeated episodes of this "microtrauma" eventually lead to posterior labral tears.
We sought to evaluate our population of patients presenting with shoulder pain or instability to test our hypothesis that football players are at increased risk relative to non-football players for development of a tear of the posterior glenoid labrum.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The patients ranged in age from 15 to 74 years (mean age, 32 years) and included 56 females and 101 males. Twenty patients (all male; age range, 17-22 years; mean age, 20 years) were competitive football players. One player was at the high school level and the others were at the collegiate level. The football players included three defensive linemen, one offensive lineman, three running backs, four defensive backs, four linebackers, three receivers, one center, and one quarterback. Only two of the 20 players reported a clear history of a single trauma to the shoulder.
One hundred thirty-seven patients (age range, 15-74 years; mean age, 34 years) did not play football (male-female ratio, 81:56). Of these, 13 patients were known on the basis of their clinical history to be competitive athletes. This included three softball players, three baseball pitchers, three volleyball players, two swimmers, one gymnast, and one hockey player. All were at the collegiate level except two: a high school softball player and baseball pitcher.
|
|
|
All of the football players in our study engaged in weight training, including the bench press. The number of non-football players engaging in weight training is not known.
Imaging Evaluation
MR arthrographic imaging was performed on a 1.5-T unit (Signa, GE
Healthcare). Before imaging, we injected each shoulder under fluoroscopic
guidance with 15-20 mL of gadolinium solution containing a 50:50 mixture of
3.25 mmol/L gadodiamide (Omniscan, Amersham Health) and 30.5% iopamidol
(Isovue-M 300, Bracco Diagnostics). Each patient was imaged using the
following protocol: fast spin-echo T1-weighted fat-saturated sequences (TR
range/TE, 450-700/12; matrix, 320 x 224; echo train, 2) in the coronal
oblique and sagittal oblique planes; fast spin-echo T2-weighted fat-saturated
sequences (3,450-4,300/61; matrix, 320 x 224; echo train, 12) in the
coronal oblique, sagittal oblique, and axial planes; and axial T1-weighted
gradient-echo sequences (TR/TE, 52/10; matrix, 512 x 256; slice
thickness, 2.0-2.2 mm).
All shoulder MR arthrograms were interpreted and dictated by a fellowship-trained musculoskeletal radiologist at the time they were obtained. These radiology reports were reviewed for a mention of "posterior labral tear." Six reports were considered equivocal for posterior labral tear. These MR arthrography studies were reviewed again in consensus by two fellowship-trained musculoskeletal radiologists, and a decision was made about whether a posterior labral tear was present or absent. Fraying was not considered to be a tear.
The MR arthrograms considered positive for posterior labral tears were reviewed by the same two musculoskeletal radiologists who determined the nature of the tear: Each was categorized as either a labral substance tear or a labral detachment. A substance tear was diagnosed when contrast material on T1-weighted images or fluid on T2-weighted images extended into the substance of the posterior labrum, interrupting its normal contour. An avulsion of the labrum was considered a detachment. Detachments included "reverse Bankart"-type lesions, which were characterized by a labrocapsular avulsion with torn periosteum, and posterior labrocapsular periosteal sleeve avulsions (POLPSAs), in which the posterior labrum is avulsed but remains attached to the glenoid by a sleeve of periosteum [5, 6].
For each study, the presence of anterior and superior labrum anterior-to-posterior (SLAP) tears, cartilaginous lesions, rotator cuff tears, and biceps tendon tears was also noted.
The results of these MR studies were then compared with surgical results when available. Arthroscopic or surgical correlation was available for 21 of the 27 football players' shoulders and in 41 of the 137 non-football players' shoulders.
Statistics
Statistical analysis was performed using 2 x 2 contingency table
analysis and the exact permutation test with StatXact (Cytel Software). The
odds ratio and its CI were also calculated using the exact permutation
test.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ten (67%) of the 15 posterior labral tears in football players were considered to be labral detachments (Fig. 1). The remaining five tears were considered substance tears. Of the 10 detachments, three (30%) were considered POLPSA lesions (Fig. 2). Six (60%) of the 10 detachments were associated with posterior glenoid articular cartilage injuries (Figs. 2 and 3).
Of the 15 shoulders with posterior labral tears, four (27%) also had associated anterior labral tears and four (27%) also had SLAP tears (one extending into a posterior labral tear). Eleven (41%) of the 27 football players' shoulders had other types of labral injuries depicted on MR arthrograms without associated posterior labral tears. These included seven shoulders with isolated anterior labral tears, two with isolated SLAP lesions, and two with both.
Non-Football Players
One hundred thirty-nine MR arthrograms were obtained in 137 non-football
players (two bilateral). Of the 137 non-football players, 10 (7%) showed
posterior labral tears, none of which was bilateral. Six (60%) of 10 posterior
labral tears were detachments. Two of these six were considered to be POLPSA
lesions. The remaining four were substance tears. Of the 10 posterior labral
tears, two were associated with anterior labral tears and two with SLAP
lesions.
Twenty-one (15%) of 139 non-football players' shoulders showed anterior labral tears, SLAP tears, or both with no associated posterior tear. One hundred eight (78%) of 139 had no labral abnormalities.
Two (15%) of the 13 competitive athletes who were not football players had a posterior labral tear; one was considered a detachment.
Thus, of the 25 shoulder MR arthrograms that showed posterior labral tears, 15 were in football players, two were in non-football-playing competitive athletes, and eight were in nonathletes. Inciting activities in the nonathletes included two falls, one motor vehicle accident, one boating accident, one waterskiing accident, an injury while power lifting, and two unknown events.
Surgical correlation was available for 16 of the 25 MR arthrograms that showed posterior labral tears. There was agreement in 14 of the 16 cases. Of the two discordant cases, one was called a "probable healed" tear on MRI and was not seen on arthroscopy, and in the second case, there was no mention of the posterior labrum in the operative report.
Data Analysis
Statistical analysis was performed using 2 x 2 contingency table
analysis and the exact permutation test with StatXact (Cytel Software). The
odds ratio and its CI were also calculated using the exact permutation
test.
When we compared the 20 football players with the other 137 subjects in our study, we found a highly statistically significant association between playing football and the presence of posterior labral injury (p = 0.0000) (Table 1). We also calculated the common odds ratio for this comparison and found that football players are 15.0 times more likely to have a posterior labral injury than non-football players (p = 0.0000). The 95% CI for this odds ratio was 4.52-52.9.
|
Considering that the presence of other associated injuries in addition to posterior labral tears had the potential of skewing the data, we eliminated these studies (eight football players and four non-football players) and recalculated the data for those with isolated posterior labral tears. Repeat statistics continue to indicate a strong association between football playing and posterior labral injuries (p = 0.028).
Comparing the 20 football players with the age- and sex-matched group of non-football players, we again found a highly significant association between playing football and posterior labral tear, with a p value of 0.0057 (Table 2). The common odds ratio was 10.3, with a 95% CI of 1.72-115.
|
Next, we combined the 20 football players with the 13 other competitive athletes and compared these athletes against the remaining 124 subjects in our study (Table 3). Again we found a highly statistically significant association between competitive athletics and the presence of posterior labral injury (p = 0.0000). When we calculated the common odds ratio for this comparison, we found that these athletes are 9.23 times more likely to have a posterior labral injury than nonathletes (p = 0.0000). The 95% CI for this odds ratio was 3.09-29.3.
|
Finally, we compared the 20 football players with the 13 other competitive athletes in our study (Table 4). We found a statistically significant association between the type of athletics and the presence of posterior labral injury (p = 0.033). The common odds ratio for this comparison was 6.3. The p value of 0.052 and the 95% CI of 0.99-73.5 indicate that the odds ratio is marginally statistically significant and suggest that a larger sample size might reveal a more statistically significant odds ratio between the two groups.
|
There was no statistically significant association between playing football and the presence of a posterior detachment (p = 1.0).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In addition to the various causes of posterior labral tears and instability, the findings at surgery in those with posterior instability may be varied and may include posterior labral tears or detachments, glenoid rim cartilage or bone fractures, and posterior capsular injuries [1, 11, 13].
There is also a poor understanding of the mechanism leading to traumatic posterior labral and capsular injuries and posterior instability. A mechanism often described is injury occurring when the arm is flexed to 90°, adducted, and internally rotated at the time of contact [1, 9, 11]. A mechanism for a specific type of posterior labral tear in contact-sports athletes such as football players has been proposed by Mair et al. [4], who described posterior labral detachments in nine athletes: eight football players, seven of whom were offensive linemen, and one lacrosse player. Five also had glenoid chondral surface injuries, but none had capsular injuries or instability. The proposed mechanism of injury was blocking of an opponent when the shoulder is positioned in 90° of flexion with the elbows locked. With contact, the shoulder undergoes a posteriorly directed force leading to a significant shearing force that is transferred to the posterior labrum. The preponderance of labral detachments and articular cartilage injuries as opposed to labral substance tears and capsular injuries in these athletes was thought to be caused by a major compressive force due to protracting the shoulders in anticipation of contact. This results in a shear force directed at the glenoid and labrum. Multiple episodes of this form of microtrauma eventually lead to posterior labral detachment or tears.
Yu et al. [5] described POLPSA lesions in six patients, four of whom were football players. Because POLPSA lesions are considered a form of labral detachment, the findings reported by Yu et al. also support the theory of a shear-type injury directed at the labrum. With POLPSA lesions, even though the capsule remains intact, generally there is instability due to stripping of the periosteum, resulting in a more patulous posterior capsule [5, 6]. Instability was detected in all subjects in that study [5], unlike the study by Mair et al. [4].
In the two studies mentioned [4, 5], subjects were selected on the basis of having a particular posterior labral injury: either a detachment without capsular injury [4] or a POLPSA lesion [5]. Although the labral detachments seen in the former study were not clearly described, it is possible that these were also POLPSA lesions and not reverse Bankart-type lesions, which by definition involve disruption of the capsule [13]. Because other types of posterior labral injuries were not addressed, we chose to look at all posterior labral injuries in our general population of people with shoulder pain or instability to assess the prevalence in football players versus a control group. We found that posterior labral injuries in general are more prevalent in football players than in non-football players. Although we found a preponderance of labral detachments versus labral substance tears, the difference was not statistically significant when compared with non-football players.
We believe that the mechanism leading to posterior labral injury proposed by Mair et al. [4] is at least partially a valid one. The injuries they described were thought to be specific to offensive linemen [4]. However, in our population of football players, we found posterior labral tears in those playing other positions as well. One explanation could be that the defensive players may be countering the offensive technique of blocking with outstretched arms using a similar technique.
In our non-football player group, posterior labral tears occurred in a hockey player, a patient injured in an automobile accident, a yard maintenance worker who had fallen off of a ladder, a collegiate softball catcher, and a patient injured in a waterskiing accident. Other studies have shown posterior labral injuries from a variety of activities in addition to football. Injuries in weightlifters, wrestlers, hockey players, and lacrosse players and traumatic injuries from falls, waterskiing, and motor vehicle accidents have been mentioned in more than one other study [2, 4-7, 9-11, 14, 15]. The finding of this injury in non-football players suggests that posterior labral tears could potentially be seen in any patient who undergoes a force that displaces the humeral head posteriorly relative to the glenoid.
For a number of reasons, the presence or absence of instability in our subjects was not specifically addressed. First, we had incomplete information regarding the clinical examinations of our patients. Also, the presence or absence of instability while the patient was under anesthesia was not consistently noted in the operative reports. Finally, the definition and diagnosis of instability can vary among individual orthopedic surgeons [3]. Thus, we believe that this determination is beyond the scope of our study.
One controversial consideration in our study is whether aspects of an athletic training regimen could be contributing to these posterior labral injuries. Weight training, for example, is almost universal in football players, including all of the players in our series. One patient in the series reported by Yu et al. [5] was a non-football playing weightlifter. Mair et al. [4] do not believe weight training contributes significantly to the development of posterior labral injuries because they have not seen this injury in bench-pressing athletes who were not involved in football or in any of the previously mentioned sports. Bench-pressing did, however, exacerbate shoulder symptoms in their subjects. We find it plausible that bench-pressing heavy weights could at least contribute to posterior labral injuries by the previously discussed mechanism of an anticipated posteriorly directed force to the shoulder, and we think that this issue deserves further study.
Limitations
A potential weakness of our study is its relatively small patient cohort.
However, our study population is significantly larger than that of previous
studies [4,
5] and also includes a control
group. Despite this relatively small cohort of players, the association
between playing football and posterior labral tear is highly statistically
significant. Likewise, the odds ratio of the proportion of football players to
non-football players with posterior labral tears is highly significant, both
statistically and clinically.
Admittedly, our patient population is a select subgroup of patients seen by sports medicine physicians and may not truly represent the general population. However, we believe that our patient population is representative of the patient population seen by other sports medicine physicians, orthopedic surgeons, and musculoskeletal radiologists. The comparison of football players to an age- and sex-matched control group lends further credence to our findings.
Because our study was limited to examining the prevalence of posterior labral tears, we did not discuss the high percentage of other shoulder injuries that were present as well. Whether these injuries are associated with the same mechanism that produces posterior labrocapsular injuries or whether they have been caused by a separate mechanism of injury cannot be determined. Perhaps these posterior labral injuries are not a result of a single simple mechanism of injury, but of a more complex mechanism that involves other quadrants of the glenohumeral joint as well. The training regimen of football players is rigorous and could also play a part in the development of these varied injuries.
A final limitation of our study is that we have surgical proof of the MR diagnosis for only 16 of 25 labral tears. However, MR arthrography has been shown to be an accurate technique for evaluating the capsuloligamentous complex and labrum [13, 16], and we are confident diagnosing posterior labral lesions using this technique, especially with the exclusion of equivocal findings.
Conclusions
Our study results suggest that posterior labral injuries are almost 15
times more common in football players than in the general population. In
addition, similar injuries occur in non-football players whose shoulders are
likely subjected to a similar posteriorly directed force leading to posterior
translation of the humeral head relative to the glenoid. We think that most of
the posterior labral tears in our population resulted from multiple episodes
of microtrauma. Given this mechanism, it is plausible that weight-training
regimens may also play a significant role in the development of these
injuries. Further study of these and other etiologic factors is warranted. If
significant associations can be shown between posterior labral tears and
injuries specific to a sport or type of weight training, those findings could
lead to changes in playing practices, athletic gear, and weight-training
regimens. Finally, an increased awareness of posterior labral tears among
interested physicians may lead to a better understanding of the injury.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. H. Brophy, S. Lyman, E. L. Chehab, R. P. Barnes, S. A. Rodeo, and R. F. Warren Predictive Value of Prior Injury on Career in Professional American Football Is Affected by Player Position Am. J. Sports Med., April 1, 2009; 37(4): 768 - 775. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
