- Email: [email protected]
Characterization of Symptomatic Hip Impingement in Butterfly Ice Hockey Goalies
Characterization of Symptomatic Hip Impingement in Butterfly Ice Hockey Goalies James R. Ross, M.D., Asheesh Bedi, M.D., Rebecca M. Stone, M.S., A.T.C., Elizabeth Sibilsky Enselman, M.Ed., A.T.C., Bryan T. Kelly, M.D., and Christopher M. Larson, M.D.
Purpose: This study aimed to characterize the radiographic deformity observed in a consecutive series of butterfly goalies with symptomatic mechanical hip pain and to use computer-based software analysis to identify the location of impingement and terminal range of motion. We also compared these analyses to a matched group of positional hockey players with symptomatic femoroacetabular impingement (FAI). Methods: A consecutive series of 68 hips in 44 butterfly-style hockey goalies and a matched group of 34 hips in 26 positional hockey players who underwent arthroscopic correction for symptomatic FAI were retrospectively analyzed. Each patient underwent preoperative anteroposterior (AP) and modified Dunn lateral radiographs and computed tomography (CT) of the affected hips. Common FAI measurements were assessed on plain radiographs. Patient-specific, CT-based 3-dimensional (3D) models of the hip joint were developed, and the femoral version, alpha angles at each radial clock face position, and femoral head coverage were calculated. Maximum hip flexion, abduction, internal rotation in 90 flexion (IRF), flexion/adduction/internal rotation (FADIR), and butterfly position were determined, and the areas of bony collision were defined. Results: Butterfly goalies had an elevated mean alpha angle on both AP (61.3 ) and lateral radiographs (63.4 ) and a diminished beta angle (26.0 ). The mean lateral center-edge angle (LCEA) measured 27.3 and acetabular inclination was 6.1 . A crossover sign was present in 59% of the hips. The maximum alpha angle on the radial reformatted computed tomographic scan was significantly higher among the butterfly goalies (80.9 v 68.6 ; P < .0001) and was located in a more lateral position (1:00 o’clock v. 1:45 o’clock; P < .0001) compared with positional players. Conclusions: Symptomatic butterfly hockey goalies have a high prevalence of FAI, characterized by a unique femoral cam-type deformity and noted by an elevated alpha angle and loss of offset, which is greater in magnitude and more lateral when compared with that in positional hockey players. Associated acetabular dysplasia is also common among hockey goalies. Level of Evidence: Level IV, prognostic case series.
I
n ice hockey players, hip and groin injuries are among the most common anatomic locations for sport-related injury.1-4 Many have attributed these injuries to overuse, a developmental process, or traumatic contact injuries. Femoroacetabular impingement (FAI)
From Sports Medicine and Shoulder Service, University of Michigan (J.R.R., A.B., E.S.E.), Ann Arbor, Michigan; Minnesota Orthopedic Sports Medicine Institute at Twin Cities Orthopedics (R.M.S., C.M.L.), Edina, Minnesota; and Hospital for Special Surgery (B.T.K.), New York, New York, U.S.A. The authors report the following potential conflict of interest or source of funding: C.M.L. receives support from Smith & Nephew and A3 Surgical. Received March 7, 2014; accepted October 23, 2014. Address correspondence to Asheesh Bedi, M.D., University of Michigan, Sports Medicine and Shoulder Service, MedSport, 24 Frank Lloyd Wright Dr, Ann Arbor, MI 48106, U.S.A. E-mail: [email protected] Ó 2015 by the Arthroscopy Association of North America 0749-8063/14196/$36.00 http://dx.doi.org/10.1016/j.arthro.2014.10.010
is increasingly recognized and diagnosed as a potential cause of hip and groin pain in hockey players at the youth, collegiate, and professional levels.5-8 The demands for terminal hip range of motion associated with skating and hockey mechanics may uniquely predispose these athletes to symptomatic FAI.9 Although positional hockey players’ hip mechanics have recently been examined,9 no study has specifically investigated symptomatic FAI in hockey goalies. Historically, ice hockey goaltenders used a “stand-up” style in which a majority of saves were made while standing up. However, with the creation of the hockey goalie mask and improvement in the protection of chest and arm pad equipment, goaltenders adopted a style of play in which they drop to their knees to guard the lower part of the net from scoring attempts. Popularized by Glen Hall, the “butterfly style” is a technique distinguished when the goaltender drops to the knees and internally rotates the hips to 90 to allow the lower
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 31, No 4 (April), 2015: pp 635-642
635
636
J. R. ROSS ET AL.
extremity padding to position parallel to the ice (Fig 1A). the unique demands of this technique may predispose athletes to an increased risk of symptomatic impingement, given the repetitive internal rotation of the hip with axial loading as the knees make contact with the ice. Even more recently, the butterfly style has been modified to include “profly,” “hybrid,” and “V-H” techniques, which also require extreme hip flexion, abduction, and internal rotation, and may thus predispose the athlete to symptomatic impingement in the setting of cam- or pincer-type morphologic conditions (or both) (Fig 1B). Whether an open or arthroscopic surgical technique is used to address symptomatic FAI, favorable clinical outcomes are predicated on a comprehensive correction of the pathologic cam- or pincer-type deformity, or both, as well as eliminating mechanical conflict in the terminal range of motion, because symptom recurrence and need for revision hip arthroscopy is most commonly a result of incomplete resection.10-12 Computed tomography (CT)-based computer modeling can precisely and reliably characterize the location and topography of cam-type deformity in patients with symptomatic FAI.13 This advanced imaging modality may therefore allow surgeons to more accurately and reliably recognize and eliminate these impingement lesions. The purpose of this study was to analyze the radiographic deformity observed in a consecutive series of butterfly-style hockey goalies with symptomatic mechanical hip pain. Using computer-based software analysis, the location and topography of the deformity were defined and the terminal range of motion with various hip maneuvers assessed with dynamic simulation. In addition, these analyses were compared with a matched group of nongoalie hockey “position players” with symptomatic FAI. We hypothesized that hockey goalies would have a higher prevalence of FAI and a larger deformity than would position players, resulting in a secondary decrease in terminal hip range of motion.
Methods This study was performed under an institutional review boardeapproved protocol obtained at all participating sites. We performed a retrospective review of a consecutive series of 68 hips in 44 high school, collegiate, and professional butterfly-style hockey goalies. Inclusion criteria included any ice hockey goalie who presented to 2 of the authors’ (B.T.K., C.M.L.) clinic with symptomatic hip pain from 2010 to 2013. Additionally, the patients had to be diagnosed with symptomatic FAI by the treating surgeon and subsequently undergo a corrective arthroscopic hip procedure during this period. Additionally, to be included in the study, all patients were required to have well-aligned preoperative anteroposterior (AP) pelvis and modified Dunn lateral radiographs (45 hip flexion, 20 external rotation, and 20 abduction)14 in addition to highresolution preoperative computed tomographic scans of the involved hip and lower extremity. A modified CT protocol using a decreased radiation exposure of 2.85 mSV was used to maximize patient safety as described by Milone et al.13 Positioning of the patient in the scanner was standardized, with the legs in neutral abduction/adduction and the patellae pointing directly anterior. Exclusion criteria included any patient with pediatric hip pathologic conditions (e.g. Perthes disease or slipped capital femoral epiphysis) or the lack of any of the preceding radiographic requirements. Evaluation of plain radiographs was performed by an independent fellowship-trained hip preservation surgeon (J.R.R.), who was blinded from the hockey player’s position and other demographic information. Measurements of the lateral center-edge angle (LCEA),15 alpha angle,16 acetabular inclination,17 and the presence or absence of a “crossover sign”18 were obtained on an AP pelvis view. Measurements of alpha angle,16 femoral head-neck offset ratio,19 and beta angle20 were obtained on the modified Dunn lateral radiograph.
Fig 1. (A) The butterfly-style technique is distinguished by the goaltender dropping to the knees and internally rotating the hips to allow the lower extremity padding to position parallel to the ice. (B) Profly, hybrid, and V-H techniques also require extreme hip flexion, abduction, and internal rotation.
637
HIP IMPINGEMENT IN BUTTERFLY GOALIES
The preoperative computed tomographic scans were uploaded into a CT-based computer modeling software program (DYONICS PLAN Hip Impingement Planning System Smith & Nephew Endoscopy, Andover, MA) to generate patient-specific, 3-dimensional (3D) models of the hip joint. This software system was used to measure the femoral neck-shaft angle and femoral neck version relative to the posterior condylar axis of the knees. In addition, the alpha angles of the various clock face positions at 15-minute increments were measured circumferentially around the entire femoral head on radial sequences. The acetabular version was also measured between the 1:00 and 4:00 o’clock positions in 15-minute increments, as was the presence or absence of a crossover sign on a virtual AP hip radiograph that was constructed from the computed tomographic images and corrected for any errant pelvic obliquity and tilt to minimize the risk of false-positive results that could be observed with the actual plain radiograph. The LCEA and the percent coverage of the femoral head by the acetabulum were also calculated. The 3D model of the pelvis was also used to characterize the morphologic features of the anterior inferior iliac spine by using a “head on view” and “ischium view” as previously described by Hetsroni et al.21 Simulated range of motion was performed with the 3D-generated model as previously described by Bedi et al.22,23 In the simulation, the proximal femur and acetabulum were set to collide. The pelvis was fixed in space, and the femur was free to translate in all directions but was constrained to rotate around the proscribed rotation axis, against the congruous acetabular surface. A small posteriorly and superiorly directed force was applied to the femur to maintain reduction of the femur during simulation. During the simulated range of motion maneuvers, the femur was translated until contact between the femur and acetabulum occurred, causing a translation of the femoral head. This point was defined as the first point of mechanical impingement. Specifically, we measured maximum direct hip flexion, abduction, internal rotation in 90 hip flexion (IRF), internal rotation in 90 hip flexion with 15 of adduction (FADIR), and internal rotation in the “butterfly” position (70 of hip flexion and 10 of adduction). The areas of contact on both the proximal femur and the acetabular rim were also defined using standard clock face nomenclature. The impingement location on the femur was also measured as the distance distal to the location of the alpha angle measurement along a line parallel to the femoral neck axis. The radiographs and computed tomographic scans of 34 hips in 26 positional hockey players with symptomatic FAI, who also underwent corrective arthroscopic surgery during the study period, were also analyzed in an identical manner and provided a matched control group of nongoalie hockey athletes for comparison.
To assess inter-rater reliability of the various computed tomographic measurements, 10 patients’ computed tomographic scans from a previous study24 were reuploaded, and measurements were made by a second observer. A 2-way analysis of variance using SPSS version 22 (SPSS, Chicago, IL) was performed to determine the absolute interobserver reliability (intraclass correlation coefficient for numerical and kappa for categorical variables). Femoral measurements showed excellent interobserver reliability (version, 0.98; maximum alpha angle, 0.88; alpha angle range for the 12:00 to 3:00 o’clock vector, 0.840.95). All computed tomographic acetabular measurements also showed excellent interobserver reliability (1:30 o’clock version, 0.99; 3:00 o’clock version, 0.99; LCEA, 0.98; acetabular inclination, 0.97; crossover sign, 1.00; posterior wall sign, 0.80; prominent ischial spine sign, 0.62). Finally, range of motion also showed excellent interobserver reliability (flexion, 0.96; IRF, 0.96; FADIR, 0.96). Statistics Statistical analysis was performed with Microsoft Excel software (Microsoft, Redmond, WA). The c-square test was used for comparison of categorical variables and the t test was used for continuous variables, with P < .05 considered significant. Demographics, as well as the plain radiographic and CT-derived measurements were compared between hockey butterfly goalies and positional players. A 2-tailed Pearson correlation coefficient was also used to determine any correlation between the age of the player during the time of surgery and the maximum alpha angle that was observed.
Results Demographics Sixty-eight hips (44 hockey goalies) were identified during the study period. Thirty-five (51.5%) of the surgical procedures involved the left hip. The average age of the butterfly hockey goalies was 21.1 5.7 years (range, 14 to 43 years). The mean body mass index was 24.7 3.0 kg/m2 (range, 18.5 to 32.1 kg/m2). The demographic data for the comparison positional hockey players were not significantly different when compared with hockey goalies (Table 1). Plain Film Measurements Ninety percent (61 of 68) of the hockey goalies had a femoral cam-type deformity with an elevated alpha angle on either the AP or modified Dunn lateral view (> 50 ) (Table 2). Mean LCEA was significantly lower among the hockey goalies (27.3 v 29.6; P ¼ .03). Nine hockey (13.2%) goalies had acetabular dysplasia with an LCEA less than 20 or acetabular inclination
638
J. R. ROSS ET AL.
Table 1. Patient Demographics of the Butterfly Goalies and Position Hockey Players Butterfly Goalies (n ¼ 68) Variable Age at surgery, yr Body mass index, kg/m2 Side Left Right
Average 21.1 5.7 24.7 3.0
Positional Players (n ¼ 10)
Range 14-43 18.5-32.1
Average 21.0 7.2 24.3 1.6
51.5% 48.5%
Range 14-49 22.3-26.5
58.8% 41.2%
greater than 10 , whereas another 11 (16.2%) goalies had borderline acetabular dysplasia with an LCEA between 20 and 25 and a normal acetabular inclination (0 to 10 ). Conversely, only 3% of hockey positional players had acetabular dysplasia, whereas another 12% had borderline acetabular dysplasia. Fifty-nine percent of the hockey goalies had anterosuperior overcoverage, shown by the presence of a crossover sign on a well-positioned AP pelvis radiograph, whereas 4 (5.9%) had evidence of pincer morphologic features with an LCEA greater than 40 or an acetabular inclination less than 0 . Although the alpha angle measured on the AP pelvic radiograph showed a trend for a greater measurement among the hockey goalies (61.3 v 55.1 ; P ¼ .06), the positional players’ plain film measurements were otherwise not significantly different (Table 2). CT Measurements The mean femoral version and femoral neck-shaft angle were not significantly different between hockey goalies and positional players (P ¼ .43 and .66, respectively) (Table 3). Nine hockey goalies’ hips (13.2%) had relative femoral retroversion (< 5 ), whereas another 16 hips (23.5%) had increased femoral anteversion (> 20 ). Among the hockey goalies, there was no significant difference in the maximum alpha angle or maximum alpha angle location when compared by version (< 0 , 0 to 20 , and > 20 ). There was no significant correlation between the age of the player at the time of surgery and the maximum alpha angle that was observed (P ¼ .08). Among the femoral version groups, relative femoral
P Value .95 .63 .48
retroversion showed a trend toward having a lower incidence of a crossover sign when compared with normal femoral version (22.2% v 55.8%; P ¼ .07). The butterfly hockey goalies, however, did have significantly greater maximum alpha angles (80.9 v 68.6 ; P < .0001) and a more lateral location of the maximum alpha angle (1:00 o’clock v 1:45 o’clock; P < .0001) (Fig 2). The average CT-measured LCEA was 27.6 5.3 (range, 17 to 47 ) among the hockey goalies, which was significantly lower than that in positional players (30.0 5.4 ; range 19.0 to 39.0 ; P ¼ .04). Cranial (1:30 o’clock position) acetabular version and the presence of a crossover sign (52.9% v 44.1%; P ¼ .40) were not significantly different between goalies and positional players. Central (3:00 o’clock position) acetabular version, however, was significantly less among the hockey goalies (11.5 v 15.5 ; P ¼ .002) (Fig 3). Additionally, positional players had significantly greater acetabular version at 2:00 o’clock through 4:00 o’clock. Thirty-seven hockey goalies (54.4%) were classified as having a type I anterior inferior iliac spine, whereas another 26 (38.2%) were classified as having type II and 5 (7.4%) as having type III.21 Range of Motion Simulation Butterfly hockey goalies had significantly higher flexion angles (121.3 v 115.6 ; P ¼ .03); however, there were no significant differences in abduction (P ¼ .47), IRF (P ¼ .31), or FADIR (P ¼ .36) when compared with positional players. Mean internal rotation in the butterfly position was 33.5 12.8 (range, 2 to 65 ), which was significantly less than that in positional players (40.0 14.1 ; P ¼ .03). With straight
Table 2. Radiographic Measurements of Preoperative Anteroposterior Pelvis and Modified Dunn Lateral Radiographs Butterfly Goalies (n ¼ 68) Measurements AP pelvis Lateral center-edge angle Acetabular inclination Alpha angle Positive crossover sign, n Modified Dunn lateral Alpha angle Head-neck offset ratio Beta angle
Positional Players (n ¼ 34)
Mean SD
Range
Mean SD
Range
P Value
27.3 5.5 6.1 3.7 61.3 16.7 40 (58.8%)
15.3 -46.1 3.9 -14.0 38.2 -91.0
29.6 4.5 4.7 3.1 54.0 13.5 15 (44.1%)
21.5 -38.9 0.4 -10.4 40.8 -84.0
.03 .04 .02 .16
63.4 11.1 0.16 0.03 26.0 13.2
38.2 -84.4 0.03-0.21 2.1 -56.6
60.4 10.6 0.15 0.02 33.5 16.5
42.1 -80.4 0.11 -0.18 5.3 -62.8
.20 .10 .03
639
HIP IMPINGEMENT IN BUTTERFLY GOALIES Table 3. Preoperative Computed Tomographic Measurements of the Proximal Femur and Acetabulum Butterfly Goalies (n ¼ 68) Variable Femur Femoral anteversion Femoral neck-shaft angle Maximum alpha angle* Maximum alpha angle position* Acetabulum Lateral center-edge angle, n* < 20 20 -25 25 -40 > 40 Femoral head coverage Positive crossover sign
Position Players (n ¼ 34)
Mean
Range
Mean
Range
P Value
14.2 9.2 132.5 4.5 80.9 10.2 1:00
8 -43 116 -142 53 -103
12.9 6.8 132.9 4.4 68.6 11.3 1:45
1 -26 121 -141 49 -94
.43 .66 < .0001 < .0001
27.6 5.3 2 (2.9%) 17 (25.0%) 47 (69.2%) 2 (2.9%) 78.0% 52.9%
17 -47
30.0 5.4 1 (2.9%) 4 (11.8%) 29 (85.3%) 0 80.0% 44.1%
19 -39
.04 1.00 .12 .08 .31 .08 .40
69%-95%
70 -89%
*Significant difference.
Discussion
abduction, butterfly goalies’ acetabular contact position was significantly more lateral (12:15 o’clock v 12:30 o’clock; P ¼ .04). The mean alpha angle at this location was also greater among the goalies (71.5 v 56.7 ; P < .001). The femoral contact position with butterfly positioning was also significantly more anterior (2:45 v 2:15; P < .001) among the hockey goalies (Table 4). The alpha angle was also greater among the hockey goalies (63.1 v 56.7; P ¼ .002). Relative femoral retroversion (< 5 ) was associated with significantly lower IRF (18.8 v 27.3 ; P ¼ .02), FADIR (11.1 v 18.6 ; P ¼ .01), and internal rotation in the butterfly position (23.1 v 34.0 ; P ¼ .0004). Hockey goalies with evidence of acetabular dysplasia or borderline dysplasia (LCEA < 25 ) had significantly greater mean flexion (132.9 v 118.4 ; P < .0001), mean abduction angles (74.3 v 68.9 ; P ¼ .003), and FADIR (21.9 v 16.9 ; P ¼ .048).
Ice hockey players are at increased risk for lower abdominal and groin injuries, which are a common and major cause of disability.2 A National Hockey League study found a cumulative incidence of 20 groin/ abdominal injuries per 100 players per year during the 1996 to 1997 season, which was significantly increased from the 1991 to 1992 season.2 Although Stull et al.9 recently pointed to the sprint start position as a potential cause for impingement in positional hockey players, no study has previously investigated symptomatic FAI in hockey goalies. According to a surveillance study in the NHL from 2006 to 2010,6 hockey goalies had a significantly higher rate of hip injuries per 1000 player game appearances (1.84) when compared with positional players (0.34 to 0.47). This discrepancy may reflect the significant terminal hip range of motion
Fig 2. Comparison of the average femoral head-neck junction deformities between hockey butterfly goalies and positional players. The hockey goalies had significantly greater values of average alpha angles, which were also situated in a more superolateral position.
Fig 3. Comparison of the mean acetabular version between hockey butterfly goalies and positional players. Asterisk (*) denotes significant difference, with P < .05.
640
J. R. ROSS ET AL.
Table 4. Comparison of the Dynamic Analysis of Motion Between Butterfly Goalies and Position Players Butterfly Goalies Variable 1 Flexion Acetabulum Femur
Abduction Acetabulum Femur
IRF Acetabulum Femur
FADIR Acetabulum Femur
Butterfly Acetabulum Femur
Variable 2 Motion Clock face position Clock face position Alpha angle Distance Motion Clock face position Clock face position Alpha angle Distance Motion Clock face position Clock face position Alpha angle Distance Motion Clock face position Clock face position Alpha angle Distance Motion Clock face position Clock face position Alpha angle Distance
Mean 121.3 1:45 5:15 43.1 6.3 mm 69.7 12:30 12:45 71.5 13.6 mm 26.4 1:30 3:00 58.8 8.0 mm 17.8 1:45 3:15 58.1 7.8 mm 33.5 2:00 2:45 63.1 8.7 mm
Range 77 -149 1:00-2:45 3:00-6:15 36 -71 2-12 mm 41 -82 11:45-1:45 12:00-2:00 40 -99 3-29 mm 0 -54 12:45-2:30 2:15-4:30 34 -86 2-17 mm 0 -45 1:15-2:30 2:30-4:30 34 -90 2-17 mm 2 -65 1:30-2:30 1:45-3:30 39 -86 3-19 mm
Position Players Mean 115.6 1:45 5:00 43.6 6.2 mm 70.8 12:15 12:30 56.7 9.2 mm 29.0 1:30 3:00 50.2 7.5 mm 20.1 2:00 3:15 47.9 6.8 mm 40.0 1:45 2:15 56.7 8.5 mm
Range 93 -138 1:00-2:30 3:00-5:45 37 -62 3-13 mm 58 -84 11:30-1:00 12:00-1:15 42 -81 4-17 mm 3 -51 12:45-2:15 1:45-4:00 37 -78 2-17 mm 0 -41 1:15-2:30 2:15-4:15 37 -67 2-16 mm 15 -67 12:45-2:30 1:30-3:15 40 -74 3-16 mm
P Value .03* .32 .04* .63 .89 .47 .04* .10 < .001* < .001* .31 .70 .02* .0001* .53 .36 .62 .39 < .001* .17 .03* .002* < .001* .002* .75
FADIR, flexion/adduction/internal rotation; IRF, internal rotation in 90 flexion. *Significant difference.
needed by hockey goalies to meet the unique demands of their position, including kick saves and the butterfly position. We hypothesized that hockey goalies would have a high prevalence of FAI and a larger deformity when compared with a matched control group of positional hockey players, resulting in a secondary decrease in terminal hip range of motion. Although cephalad retroversion was observed in more than half of the butterfly goalies, as found on CT scans (52.9%), the goalie population was universally characterized by a significant cam-type deformity and loss of offset, which was seen maximally at the 1:00 o’clock position and noted to have an elevated mean alpha angle between 11:30 and 3:30 o’clock. This deformity is larger and more lateral than that of a matched group of positional hockey players, which resulted in significantly less internal rotation in the butterfly position. The difference in the position of cam deformity may support the hypothesis that this morphologic defect results from a local induction of the proximal femoral physis with dynamic impingement.25 Hockey goalies place their hips in extreme range of motion positions (in particular abduction) more frequently, which according to our data creates bony contact on the proximal femur at the 12:45 o’clock position, which may stimulate this area to induce a more lateral cam deformity during development. These mechanics would be different from those
of a hockey positional player, who has been shown previously to have demands of hip flexion and internal rotation.9 We also did not find any relationship between the severity of deformity and the age of the athlete at the time of surgery. It may be possible that an increased number of years or level of competition may be associated with a larger deformity; however, these data were not available for analysis in our study population. We noted a significant difference in the proximal femoral deformity, which was also accompanied by significant differences in the flexion- and butterflysimulated range of motion analysis between goalie and positional athletes, with goalies having increased flexion angles and decreased internal rotation in the butterfly position. This may reflect the more anterior contact position on the femur in goalies compared with positional players with direction hip flexion, given that the location of bony impingement on the acetabular rim was similar between the groups. The contact areas of bony impingement on the femur and acetabulum were also notably anterior in hockey goalies with butterfly positioning, which may explain the diminished motion in this position. Although the significant differences such as the location of the maximum cam deformity (1:00 o’clock v 1:45 o’clock) and the degree of maximum alpha angle (80.9 v 68.6 ) were small, we do believe that they are
641
HIP IMPINGEMENT IN BUTTERFLY GOALIES
clinically significant and important to consider. This is especially true during the arthroscopic treatment, because the surgeon must know where in the femoral head-neck junction to focus the correction and ensure that access can be safely achieved. Interestingly, acetabular dysplasia was more common among hockey goalies (29%) than among positional athletes (15%). This may be a structural adaptation while they are placing their hips in extreme motion, thus protecting their hips in the presence of the larger cam deformities. Alternatively, this finding may result from selection bias in that hockey goalies with acetabular dysplasia have more extreme motion and thus are more skillful and able to compete at higher levels. In comparison, the largest population-based investigations of hip joint malformations noted a much lower prevalence of acetabular dysplasia (3.9% of 3,620 hips).26 It is important to recognize those players with acetabular dysplasia who undergo surgical treatment to avoid errant rim resection and the secondary risk of iatrogenic structural instability. The examination and evaluation of these hips with a large cam deformity, however, were more consistent with impingement rather than instability based on hip range of motion. In addition, hockey players, and goalies in particular, spend more time in hip flexion, abduction, and internal rotation (impingement positions) than in extension and external rotation (instability positions). Additionally, there was a small but significant difference in the LCEA between the 2 groupsdhockey goalies, on average, had a smaller LCEA by 2 . This may be clinically significant, however, because it could represent the difference between a normal-coverage acetabulum and one that is dysplastic. Hip preservation surgery in the setting of mixed and combined deformities has been described with favorable outcomes by Clohisy et al.27 but must be performed carefully and with appropriate caution; the increased motion that occurs with acetabular dysplasia may be beneficial in this patient population, and surgical treatment with a reorienting periacetabular osteotomy may actually hinder hip motion and make these patients more symptomatic. Limitations Our study is not without limitations. This study is based on a virtual analysis from reconstructed computed tomographic scans and does not reflect the in vivo range of motion given that it does not account for soft tissue structures that may play a role in range of motion, such as the acetabular labrum, chondral surfaces, hip capsule, and proximal tendinous structures. In addition, the dynamic analysis assumes a stable and fixed pelvis, with terminal range of motion defined with femoral motion only. Dynamic changes in pelvic position may allow for increased compensatory motion and may suggest a role for functional rehabilitation;
however, this concept has not yet been defined in the literature. Finally, the study population is subject to selection bias in that we included only athletes who presented with symptoms. In this regard, we are unable to comment on the applicability of this study to the entire population of hockey goalies.
Conclusions Symptomatic butterfly hockey goalies have a high prevalence of FAI characterized by a unique femoral cam-type deformity and noted by an elevated alpha angle and loss of offset, which is greater in magnitude and more lateral when compared with that in positional hockey players. Associated acetabular dysplasia is also common among hockey goalies.
References 1. Agel J, Dompier TP, Dick R, Marshall SW. Descriptive epidemiology of collegiate men’s ice hockey injuries: National Collegiate Athletic Association Injury Surveillance System, 1988-1989 through 2003-2004. J Athl Train 2007;42:241-248. 2. Emery CA, Meeuwisse WH, Powell JW. Groin and abdominal strain injuries in the National Hockey League. Clin J Sport Med 1999;9:151-156. 3. Irshad K, Feldman LS, Lavoie C, Lacroix VJ, Mulder DS, Brown RA. Operative management of “hockey groin syndrome”: 12 years of experience in National Hockey League players. Surgery 2001;130:759-764. 4. Sim FH, Simonet WT, Melton LJ, Lehn TA. Ice hockey injuries. Am J Sports Med 1987;15:30-40. 5. Bizzini M, Notzli HP, Maffiuletti NA. Femoroacetabular impingement in professional ice hockey players: a case series of 5 athletes after open surgical decompression of the hip. Am J Sports Med 2007;35:1955-1959. 6. Epstein DM, McHugh M, Yorio M, Neri B. Intra-articular hip injuries in National Hockey League players: a descriptive epidemiological study. Am J Sports Med 2013;41:343-348. 7. Philippon MJ, Weiss DR, Kuppersmith DA, Briggs KK, Kay CJ. Arthroscopic labral repair and treatment of femoroacetabular impingement in professional hockey players. Am J Sports Med 2010;38:99-104. 8. Shindle MK, Marx RG, Kelly BT, Bisson L, Burke CJ. Hockey injuries: a pediatric sport update. Curr Opin Pediatr 2010;22:54-60. 9. Stull JD, Philippon MJ, LaPrade RF. “At-risk” positioning and hip biomechanics of the Peewee ice hockey sprint start. Am J Sports Med 2011;39:29S-35S. 10. Bogunovic L, Gottlieb M, Pashos G, Baca G, Clohisy JC. Why do hip arthroscopy procedures fail? Clin Orthop Relat Res 2013;471:2523-2529. 11. Heyworth BE, Shindle MK, Voos JE, Rudzki JR, Kelly BT. Radiologic and intraoperative findings in revision hip arthroscopy. Arthroscopy 2007;23:1295-1302. 12. Philippon MJ, Schenker ML, Briggs KK, Kuppersmith DA, Maxwell RB, Stubbs AJ. Revision hip arthroscopy. Am J Sports Med 2007;35:1918-1921.
642
J. R. ROSS ET AL.
13. Milone MT, Bedi A, Poultsides L, et al. Novel CT-based three-dimensional software improves the characterization of cam morphology. Clin Orthop Rel Res 2013;471: 2484-2491. 14. Clohisy JC, Carlisle JC, Beaule PE, et al. A systematic approach to the plain radiographic evaluation of the young adult hip. J Bone Joint Surg Am 2008;90:47-66 (suppl 4). 15. Wiberg G. Studies on dysplastic acetabula and congenital subluxation of the hip: with special reference to the complication of osteoarthritis. Acta Chir Scand 1939;58: 7-38. 16. Notzli HP, Wyss TF, Stoecklin CH, Schmid MR, Treiber K, Hodler J. The contour of the femoral head-neck junction as a predictor for the risk of anterior impingement. J Bone Joint Surg Br 2002;84:556-560. 17. Tonnis D, Legal H, Graf R, eds. Congenital dysplasia and dislocation of the hip in children and adults. Berlin, Germany: Springer-Verlag, 1987. 18. Reynolds D, Lucas J, Klaue K. Retroversion of the acetabulum. A cause of hip pain. J Bone Joint Surg Br 1999;81: 281-288. 19. Peelle MW, Della Rocca GJ, Maloney WJ, Curry MC, Clohisy JC. Acetabular and femoral radiographic abnormalities associated with labral tears. Clin Orthop Rel Res 2005;441:327-333. 20. Brunner A, Hamers AT, Fitze M, Herzog RF. The plain beta-angle measured on radiographs in the assessment of femoroacetabular impingement. J Bone Joint Surg Br 2010;92:1203-1208.
21. Hetsroni I, Poulsides L, Bedi A, Larson CM, Kelly BT. Anterior inferior iliac spine morphology correlates with hip range of motion: a classification system and dynamic model. Clin Orthop Rel Res 2013;471:2497-2503. 22. Bedi A, Dolan M, Hetsroni I, et al. Surgical treatment of femoroacetabular impingement improves hip kinematics: a computer-assisted model. Am J Sports Med 2011;39: S43-S49 (suppl). 23. Bedi A, Dolan M, Magennis E, Lipman J, Buly R, Kelly BT. Computer-assisted modeling of osseous impingement and resection in femoroacetabular impingement. Arthroscopy 2012;28:204-210. 24. Ross JR, Nepple JJ, Philippon MJ, Kelly BT, Larson CM, Bedi A. Effect of changes in pelvic tilt on range of motion to impingement and radiographic parameters of acetabular morphologic characteristics. Am J Sports Med 2014;42: 2404-2409. 25. Ng VY, Ellis TJ. More than just a bump: Cam-type femoroacetabular impingement and the evolution of the femoral neck. Hip Int 2011;21:1-8. 26. Gosvig KK, Jacobsen S, Sonne-Holm S, Palm H, Troelsen A. Prevalence of malformations of the hip joint and their relationship to sex, groin pain, and risk of osteoarthritis: a populations-based study. J Bone Joint Surg Am 2010;92:1162-1169. 27. Clohisy JC, Nunley RM, Currey MC, Schoenecker PL. Periacetabular osteotomy for the treatment of acetabular dysplasia associated with major aspherical femoral head deformities. J Bone Joint Surg Am 2007;89: 1417-1423.
Purpose: This study aimed to characterize the radiographic deformity observed in a consecutive series of butterfly goalies with symptomatic mechanical hip pain and to use computer-based software analysis to identify the location of impingement and terminal range of motion. We also compared these analyses to a matched group of positional hockey players with symptomatic femoroacetabular impingement (FAI). Methods: A consecutive series of 68 hips in 44 butterfly-style hockey goalies and a matched group of 34 hips in 26 positional hockey players who underwent arthroscopic correction for symptomatic FAI were retrospectively analyzed. Each patient underwent preoperative anteroposterior (AP) and modified Dunn lateral radiographs and computed tomography (CT) of the affected hips. Common FAI measurements were assessed on plain radiographs. Patient-specific, CT-based 3-dimensional (3D) models of the hip joint were developed, and the femoral version, alpha angles at each radial clock face position, and femoral head coverage were calculated. Maximum hip flexion, abduction, internal rotation in 90 flexion (IRF), flexion/adduction/internal rotation (FADIR), and butterfly position were determined, and the areas of bony collision were defined. Results: Butterfly goalies had an elevated mean alpha angle on both AP (61.3 ) and lateral radiographs (63.4 ) and a diminished beta angle (26.0 ). The mean lateral center-edge angle (LCEA) measured 27.3 and acetabular inclination was 6.1 . A crossover sign was present in 59% of the hips. The maximum alpha angle on the radial reformatted computed tomographic scan was significantly higher among the butterfly goalies (80.9 v 68.6 ; P < .0001) and was located in a more lateral position (1:00 o’clock v. 1:45 o’clock; P < .0001) compared with positional players. Conclusions: Symptomatic butterfly hockey goalies have a high prevalence of FAI, characterized by a unique femoral cam-type deformity and noted by an elevated alpha angle and loss of offset, which is greater in magnitude and more lateral when compared with that in positional hockey players. Associated acetabular dysplasia is also common among hockey goalies. Level of Evidence: Level IV, prognostic case series.
I
n ice hockey players, hip and groin injuries are among the most common anatomic locations for sport-related injury.1-4 Many have attributed these injuries to overuse, a developmental process, or traumatic contact injuries. Femoroacetabular impingement (FAI)
From Sports Medicine and Shoulder Service, University of Michigan (J.R.R., A.B., E.S.E.), Ann Arbor, Michigan; Minnesota Orthopedic Sports Medicine Institute at Twin Cities Orthopedics (R.M.S., C.M.L.), Edina, Minnesota; and Hospital for Special Surgery (B.T.K.), New York, New York, U.S.A. The authors report the following potential conflict of interest or source of funding: C.M.L. receives support from Smith & Nephew and A3 Surgical. Received March 7, 2014; accepted October 23, 2014. Address correspondence to Asheesh Bedi, M.D., University of Michigan, Sports Medicine and Shoulder Service, MedSport, 24 Frank Lloyd Wright Dr, Ann Arbor, MI 48106, U.S.A. E-mail: [email protected] Ó 2015 by the Arthroscopy Association of North America 0749-8063/14196/$36.00 http://dx.doi.org/10.1016/j.arthro.2014.10.010
is increasingly recognized and diagnosed as a potential cause of hip and groin pain in hockey players at the youth, collegiate, and professional levels.5-8 The demands for terminal hip range of motion associated with skating and hockey mechanics may uniquely predispose these athletes to symptomatic FAI.9 Although positional hockey players’ hip mechanics have recently been examined,9 no study has specifically investigated symptomatic FAI in hockey goalies. Historically, ice hockey goaltenders used a “stand-up” style in which a majority of saves were made while standing up. However, with the creation of the hockey goalie mask and improvement in the protection of chest and arm pad equipment, goaltenders adopted a style of play in which they drop to their knees to guard the lower part of the net from scoring attempts. Popularized by Glen Hall, the “butterfly style” is a technique distinguished when the goaltender drops to the knees and internally rotates the hips to 90 to allow the lower
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 31, No 4 (April), 2015: pp 635-642
635
636
J. R. ROSS ET AL.
extremity padding to position parallel to the ice (Fig 1A). the unique demands of this technique may predispose athletes to an increased risk of symptomatic impingement, given the repetitive internal rotation of the hip with axial loading as the knees make contact with the ice. Even more recently, the butterfly style has been modified to include “profly,” “hybrid,” and “V-H” techniques, which also require extreme hip flexion, abduction, and internal rotation, and may thus predispose the athlete to symptomatic impingement in the setting of cam- or pincer-type morphologic conditions (or both) (Fig 1B). Whether an open or arthroscopic surgical technique is used to address symptomatic FAI, favorable clinical outcomes are predicated on a comprehensive correction of the pathologic cam- or pincer-type deformity, or both, as well as eliminating mechanical conflict in the terminal range of motion, because symptom recurrence and need for revision hip arthroscopy is most commonly a result of incomplete resection.10-12 Computed tomography (CT)-based computer modeling can precisely and reliably characterize the location and topography of cam-type deformity in patients with symptomatic FAI.13 This advanced imaging modality may therefore allow surgeons to more accurately and reliably recognize and eliminate these impingement lesions. The purpose of this study was to analyze the radiographic deformity observed in a consecutive series of butterfly-style hockey goalies with symptomatic mechanical hip pain. Using computer-based software analysis, the location and topography of the deformity were defined and the terminal range of motion with various hip maneuvers assessed with dynamic simulation. In addition, these analyses were compared with a matched group of nongoalie hockey “position players” with symptomatic FAI. We hypothesized that hockey goalies would have a higher prevalence of FAI and a larger deformity than would position players, resulting in a secondary decrease in terminal hip range of motion.
Methods This study was performed under an institutional review boardeapproved protocol obtained at all participating sites. We performed a retrospective review of a consecutive series of 68 hips in 44 high school, collegiate, and professional butterfly-style hockey goalies. Inclusion criteria included any ice hockey goalie who presented to 2 of the authors’ (B.T.K., C.M.L.) clinic with symptomatic hip pain from 2010 to 2013. Additionally, the patients had to be diagnosed with symptomatic FAI by the treating surgeon and subsequently undergo a corrective arthroscopic hip procedure during this period. Additionally, to be included in the study, all patients were required to have well-aligned preoperative anteroposterior (AP) pelvis and modified Dunn lateral radiographs (45 hip flexion, 20 external rotation, and 20 abduction)14 in addition to highresolution preoperative computed tomographic scans of the involved hip and lower extremity. A modified CT protocol using a decreased radiation exposure of 2.85 mSV was used to maximize patient safety as described by Milone et al.13 Positioning of the patient in the scanner was standardized, with the legs in neutral abduction/adduction and the patellae pointing directly anterior. Exclusion criteria included any patient with pediatric hip pathologic conditions (e.g. Perthes disease or slipped capital femoral epiphysis) or the lack of any of the preceding radiographic requirements. Evaluation of plain radiographs was performed by an independent fellowship-trained hip preservation surgeon (J.R.R.), who was blinded from the hockey player’s position and other demographic information. Measurements of the lateral center-edge angle (LCEA),15 alpha angle,16 acetabular inclination,17 and the presence or absence of a “crossover sign”18 were obtained on an AP pelvis view. Measurements of alpha angle,16 femoral head-neck offset ratio,19 and beta angle20 were obtained on the modified Dunn lateral radiograph.
Fig 1. (A) The butterfly-style technique is distinguished by the goaltender dropping to the knees and internally rotating the hips to allow the lower extremity padding to position parallel to the ice. (B) Profly, hybrid, and V-H techniques also require extreme hip flexion, abduction, and internal rotation.
637
HIP IMPINGEMENT IN BUTTERFLY GOALIES
The preoperative computed tomographic scans were uploaded into a CT-based computer modeling software program (DYONICS PLAN Hip Impingement Planning System Smith & Nephew Endoscopy, Andover, MA) to generate patient-specific, 3-dimensional (3D) models of the hip joint. This software system was used to measure the femoral neck-shaft angle and femoral neck version relative to the posterior condylar axis of the knees. In addition, the alpha angles of the various clock face positions at 15-minute increments were measured circumferentially around the entire femoral head on radial sequences. The acetabular version was also measured between the 1:00 and 4:00 o’clock positions in 15-minute increments, as was the presence or absence of a crossover sign on a virtual AP hip radiograph that was constructed from the computed tomographic images and corrected for any errant pelvic obliquity and tilt to minimize the risk of false-positive results that could be observed with the actual plain radiograph. The LCEA and the percent coverage of the femoral head by the acetabulum were also calculated. The 3D model of the pelvis was also used to characterize the morphologic features of the anterior inferior iliac spine by using a “head on view” and “ischium view” as previously described by Hetsroni et al.21 Simulated range of motion was performed with the 3D-generated model as previously described by Bedi et al.22,23 In the simulation, the proximal femur and acetabulum were set to collide. The pelvis was fixed in space, and the femur was free to translate in all directions but was constrained to rotate around the proscribed rotation axis, against the congruous acetabular surface. A small posteriorly and superiorly directed force was applied to the femur to maintain reduction of the femur during simulation. During the simulated range of motion maneuvers, the femur was translated until contact between the femur and acetabulum occurred, causing a translation of the femoral head. This point was defined as the first point of mechanical impingement. Specifically, we measured maximum direct hip flexion, abduction, internal rotation in 90 hip flexion (IRF), internal rotation in 90 hip flexion with 15 of adduction (FADIR), and internal rotation in the “butterfly” position (70 of hip flexion and 10 of adduction). The areas of contact on both the proximal femur and the acetabular rim were also defined using standard clock face nomenclature. The impingement location on the femur was also measured as the distance distal to the location of the alpha angle measurement along a line parallel to the femoral neck axis. The radiographs and computed tomographic scans of 34 hips in 26 positional hockey players with symptomatic FAI, who also underwent corrective arthroscopic surgery during the study period, were also analyzed in an identical manner and provided a matched control group of nongoalie hockey athletes for comparison.
To assess inter-rater reliability of the various computed tomographic measurements, 10 patients’ computed tomographic scans from a previous study24 were reuploaded, and measurements were made by a second observer. A 2-way analysis of variance using SPSS version 22 (SPSS, Chicago, IL) was performed to determine the absolute interobserver reliability (intraclass correlation coefficient for numerical and kappa for categorical variables). Femoral measurements showed excellent interobserver reliability (version, 0.98; maximum alpha angle, 0.88; alpha angle range for the 12:00 to 3:00 o’clock vector, 0.840.95). All computed tomographic acetabular measurements also showed excellent interobserver reliability (1:30 o’clock version, 0.99; 3:00 o’clock version, 0.99; LCEA, 0.98; acetabular inclination, 0.97; crossover sign, 1.00; posterior wall sign, 0.80; prominent ischial spine sign, 0.62). Finally, range of motion also showed excellent interobserver reliability (flexion, 0.96; IRF, 0.96; FADIR, 0.96). Statistics Statistical analysis was performed with Microsoft Excel software (Microsoft, Redmond, WA). The c-square test was used for comparison of categorical variables and the t test was used for continuous variables, with P < .05 considered significant. Demographics, as well as the plain radiographic and CT-derived measurements were compared between hockey butterfly goalies and positional players. A 2-tailed Pearson correlation coefficient was also used to determine any correlation between the age of the player during the time of surgery and the maximum alpha angle that was observed.
Results Demographics Sixty-eight hips (44 hockey goalies) were identified during the study period. Thirty-five (51.5%) of the surgical procedures involved the left hip. The average age of the butterfly hockey goalies was 21.1 5.7 years (range, 14 to 43 years). The mean body mass index was 24.7 3.0 kg/m2 (range, 18.5 to 32.1 kg/m2). The demographic data for the comparison positional hockey players were not significantly different when compared with hockey goalies (Table 1). Plain Film Measurements Ninety percent (61 of 68) of the hockey goalies had a femoral cam-type deformity with an elevated alpha angle on either the AP or modified Dunn lateral view (> 50 ) (Table 2). Mean LCEA was significantly lower among the hockey goalies (27.3 v 29.6; P ¼ .03). Nine hockey (13.2%) goalies had acetabular dysplasia with an LCEA less than 20 or acetabular inclination
638
J. R. ROSS ET AL.
Table 1. Patient Demographics of the Butterfly Goalies and Position Hockey Players Butterfly Goalies (n ¼ 68) Variable Age at surgery, yr Body mass index, kg/m2 Side Left Right
Average 21.1 5.7 24.7 3.0
Positional Players (n ¼ 10)
Range 14-43 18.5-32.1
Average 21.0 7.2 24.3 1.6
51.5% 48.5%
Range 14-49 22.3-26.5
58.8% 41.2%
greater than 10 , whereas another 11 (16.2%) goalies had borderline acetabular dysplasia with an LCEA between 20 and 25 and a normal acetabular inclination (0 to 10 ). Conversely, only 3% of hockey positional players had acetabular dysplasia, whereas another 12% had borderline acetabular dysplasia. Fifty-nine percent of the hockey goalies had anterosuperior overcoverage, shown by the presence of a crossover sign on a well-positioned AP pelvis radiograph, whereas 4 (5.9%) had evidence of pincer morphologic features with an LCEA greater than 40 or an acetabular inclination less than 0 . Although the alpha angle measured on the AP pelvic radiograph showed a trend for a greater measurement among the hockey goalies (61.3 v 55.1 ; P ¼ .06), the positional players’ plain film measurements were otherwise not significantly different (Table 2). CT Measurements The mean femoral version and femoral neck-shaft angle were not significantly different between hockey goalies and positional players (P ¼ .43 and .66, respectively) (Table 3). Nine hockey goalies’ hips (13.2%) had relative femoral retroversion (< 5 ), whereas another 16 hips (23.5%) had increased femoral anteversion (> 20 ). Among the hockey goalies, there was no significant difference in the maximum alpha angle or maximum alpha angle location when compared by version (< 0 , 0 to 20 , and > 20 ). There was no significant correlation between the age of the player at the time of surgery and the maximum alpha angle that was observed (P ¼ .08). Among the femoral version groups, relative femoral
P Value .95 .63 .48
retroversion showed a trend toward having a lower incidence of a crossover sign when compared with normal femoral version (22.2% v 55.8%; P ¼ .07). The butterfly hockey goalies, however, did have significantly greater maximum alpha angles (80.9 v 68.6 ; P < .0001) and a more lateral location of the maximum alpha angle (1:00 o’clock v 1:45 o’clock; P < .0001) (Fig 2). The average CT-measured LCEA was 27.6 5.3 (range, 17 to 47 ) among the hockey goalies, which was significantly lower than that in positional players (30.0 5.4 ; range 19.0 to 39.0 ; P ¼ .04). Cranial (1:30 o’clock position) acetabular version and the presence of a crossover sign (52.9% v 44.1%; P ¼ .40) were not significantly different between goalies and positional players. Central (3:00 o’clock position) acetabular version, however, was significantly less among the hockey goalies (11.5 v 15.5 ; P ¼ .002) (Fig 3). Additionally, positional players had significantly greater acetabular version at 2:00 o’clock through 4:00 o’clock. Thirty-seven hockey goalies (54.4%) were classified as having a type I anterior inferior iliac spine, whereas another 26 (38.2%) were classified as having type II and 5 (7.4%) as having type III.21 Range of Motion Simulation Butterfly hockey goalies had significantly higher flexion angles (121.3 v 115.6 ; P ¼ .03); however, there were no significant differences in abduction (P ¼ .47), IRF (P ¼ .31), or FADIR (P ¼ .36) when compared with positional players. Mean internal rotation in the butterfly position was 33.5 12.8 (range, 2 to 65 ), which was significantly less than that in positional players (40.0 14.1 ; P ¼ .03). With straight
Table 2. Radiographic Measurements of Preoperative Anteroposterior Pelvis and Modified Dunn Lateral Radiographs Butterfly Goalies (n ¼ 68) Measurements AP pelvis Lateral center-edge angle Acetabular inclination Alpha angle Positive crossover sign, n Modified Dunn lateral Alpha angle Head-neck offset ratio Beta angle
Positional Players (n ¼ 34)
Mean SD
Range
Mean SD
Range
P Value
27.3 5.5 6.1 3.7 61.3 16.7 40 (58.8%)
15.3 -46.1 3.9 -14.0 38.2 -91.0
29.6 4.5 4.7 3.1 54.0 13.5 15 (44.1%)
21.5 -38.9 0.4 -10.4 40.8 -84.0
.03 .04 .02 .16
63.4 11.1 0.16 0.03 26.0 13.2
38.2 -84.4 0.03-0.21 2.1 -56.6
60.4 10.6 0.15 0.02 33.5 16.5
42.1 -80.4 0.11 -0.18 5.3 -62.8
.20 .10 .03
639
HIP IMPINGEMENT IN BUTTERFLY GOALIES Table 3. Preoperative Computed Tomographic Measurements of the Proximal Femur and Acetabulum Butterfly Goalies (n ¼ 68) Variable Femur Femoral anteversion Femoral neck-shaft angle Maximum alpha angle* Maximum alpha angle position* Acetabulum Lateral center-edge angle, n* < 20 20 -25 25 -40 > 40 Femoral head coverage Positive crossover sign
Position Players (n ¼ 34)
Mean
Range
Mean
Range
P Value
14.2 9.2 132.5 4.5 80.9 10.2 1:00
8 -43 116 -142 53 -103
12.9 6.8 132.9 4.4 68.6 11.3 1:45
1 -26 121 -141 49 -94
.43 .66 < .0001 < .0001
27.6 5.3 2 (2.9%) 17 (25.0%) 47 (69.2%) 2 (2.9%) 78.0% 52.9%
17 -47
30.0 5.4 1 (2.9%) 4 (11.8%) 29 (85.3%) 0 80.0% 44.1%
19 -39
.04 1.00 .12 .08 .31 .08 .40
69%-95%
70 -89%
*Significant difference.
Discussion
abduction, butterfly goalies’ acetabular contact position was significantly more lateral (12:15 o’clock v 12:30 o’clock; P ¼ .04). The mean alpha angle at this location was also greater among the goalies (71.5 v 56.7 ; P < .001). The femoral contact position with butterfly positioning was also significantly more anterior (2:45 v 2:15; P < .001) among the hockey goalies (Table 4). The alpha angle was also greater among the hockey goalies (63.1 v 56.7; P ¼ .002). Relative femoral retroversion (< 5 ) was associated with significantly lower IRF (18.8 v 27.3 ; P ¼ .02), FADIR (11.1 v 18.6 ; P ¼ .01), and internal rotation in the butterfly position (23.1 v 34.0 ; P ¼ .0004). Hockey goalies with evidence of acetabular dysplasia or borderline dysplasia (LCEA < 25 ) had significantly greater mean flexion (132.9 v 118.4 ; P < .0001), mean abduction angles (74.3 v 68.9 ; P ¼ .003), and FADIR (21.9 v 16.9 ; P ¼ .048).
Ice hockey players are at increased risk for lower abdominal and groin injuries, which are a common and major cause of disability.2 A National Hockey League study found a cumulative incidence of 20 groin/ abdominal injuries per 100 players per year during the 1996 to 1997 season, which was significantly increased from the 1991 to 1992 season.2 Although Stull et al.9 recently pointed to the sprint start position as a potential cause for impingement in positional hockey players, no study has previously investigated symptomatic FAI in hockey goalies. According to a surveillance study in the NHL from 2006 to 2010,6 hockey goalies had a significantly higher rate of hip injuries per 1000 player game appearances (1.84) when compared with positional players (0.34 to 0.47). This discrepancy may reflect the significant terminal hip range of motion
Fig 2. Comparison of the average femoral head-neck junction deformities between hockey butterfly goalies and positional players. The hockey goalies had significantly greater values of average alpha angles, which were also situated in a more superolateral position.
Fig 3. Comparison of the mean acetabular version between hockey butterfly goalies and positional players. Asterisk (*) denotes significant difference, with P < .05.
640
J. R. ROSS ET AL.
Table 4. Comparison of the Dynamic Analysis of Motion Between Butterfly Goalies and Position Players Butterfly Goalies Variable 1 Flexion Acetabulum Femur
Abduction Acetabulum Femur
IRF Acetabulum Femur
FADIR Acetabulum Femur
Butterfly Acetabulum Femur
Variable 2 Motion Clock face position Clock face position Alpha angle Distance Motion Clock face position Clock face position Alpha angle Distance Motion Clock face position Clock face position Alpha angle Distance Motion Clock face position Clock face position Alpha angle Distance Motion Clock face position Clock face position Alpha angle Distance
Mean 121.3 1:45 5:15 43.1 6.3 mm 69.7 12:30 12:45 71.5 13.6 mm 26.4 1:30 3:00 58.8 8.0 mm 17.8 1:45 3:15 58.1 7.8 mm 33.5 2:00 2:45 63.1 8.7 mm
Range 77 -149 1:00-2:45 3:00-6:15 36 -71 2-12 mm 41 -82 11:45-1:45 12:00-2:00 40 -99 3-29 mm 0 -54 12:45-2:30 2:15-4:30 34 -86 2-17 mm 0 -45 1:15-2:30 2:30-4:30 34 -90 2-17 mm 2 -65 1:30-2:30 1:45-3:30 39 -86 3-19 mm
Position Players Mean 115.6 1:45 5:00 43.6 6.2 mm 70.8 12:15 12:30 56.7 9.2 mm 29.0 1:30 3:00 50.2 7.5 mm 20.1 2:00 3:15 47.9 6.8 mm 40.0 1:45 2:15 56.7 8.5 mm
Range 93 -138 1:00-2:30 3:00-5:45 37 -62 3-13 mm 58 -84 11:30-1:00 12:00-1:15 42 -81 4-17 mm 3 -51 12:45-2:15 1:45-4:00 37 -78 2-17 mm 0 -41 1:15-2:30 2:15-4:15 37 -67 2-16 mm 15 -67 12:45-2:30 1:30-3:15 40 -74 3-16 mm
P Value .03* .32 .04* .63 .89 .47 .04* .10 < .001* < .001* .31 .70 .02* .0001* .53 .36 .62 .39 < .001* .17 .03* .002* < .001* .002* .75
FADIR, flexion/adduction/internal rotation; IRF, internal rotation in 90 flexion. *Significant difference.
needed by hockey goalies to meet the unique demands of their position, including kick saves and the butterfly position. We hypothesized that hockey goalies would have a high prevalence of FAI and a larger deformity when compared with a matched control group of positional hockey players, resulting in a secondary decrease in terminal hip range of motion. Although cephalad retroversion was observed in more than half of the butterfly goalies, as found on CT scans (52.9%), the goalie population was universally characterized by a significant cam-type deformity and loss of offset, which was seen maximally at the 1:00 o’clock position and noted to have an elevated mean alpha angle between 11:30 and 3:30 o’clock. This deformity is larger and more lateral than that of a matched group of positional hockey players, which resulted in significantly less internal rotation in the butterfly position. The difference in the position of cam deformity may support the hypothesis that this morphologic defect results from a local induction of the proximal femoral physis with dynamic impingement.25 Hockey goalies place their hips in extreme range of motion positions (in particular abduction) more frequently, which according to our data creates bony contact on the proximal femur at the 12:45 o’clock position, which may stimulate this area to induce a more lateral cam deformity during development. These mechanics would be different from those
of a hockey positional player, who has been shown previously to have demands of hip flexion and internal rotation.9 We also did not find any relationship between the severity of deformity and the age of the athlete at the time of surgery. It may be possible that an increased number of years or level of competition may be associated with a larger deformity; however, these data were not available for analysis in our study population. We noted a significant difference in the proximal femoral deformity, which was also accompanied by significant differences in the flexion- and butterflysimulated range of motion analysis between goalie and positional athletes, with goalies having increased flexion angles and decreased internal rotation in the butterfly position. This may reflect the more anterior contact position on the femur in goalies compared with positional players with direction hip flexion, given that the location of bony impingement on the acetabular rim was similar between the groups. The contact areas of bony impingement on the femur and acetabulum were also notably anterior in hockey goalies with butterfly positioning, which may explain the diminished motion in this position. Although the significant differences such as the location of the maximum cam deformity (1:00 o’clock v 1:45 o’clock) and the degree of maximum alpha angle (80.9 v 68.6 ) were small, we do believe that they are
641
HIP IMPINGEMENT IN BUTTERFLY GOALIES
clinically significant and important to consider. This is especially true during the arthroscopic treatment, because the surgeon must know where in the femoral head-neck junction to focus the correction and ensure that access can be safely achieved. Interestingly, acetabular dysplasia was more common among hockey goalies (29%) than among positional athletes (15%). This may be a structural adaptation while they are placing their hips in extreme motion, thus protecting their hips in the presence of the larger cam deformities. Alternatively, this finding may result from selection bias in that hockey goalies with acetabular dysplasia have more extreme motion and thus are more skillful and able to compete at higher levels. In comparison, the largest population-based investigations of hip joint malformations noted a much lower prevalence of acetabular dysplasia (3.9% of 3,620 hips).26 It is important to recognize those players with acetabular dysplasia who undergo surgical treatment to avoid errant rim resection and the secondary risk of iatrogenic structural instability. The examination and evaluation of these hips with a large cam deformity, however, were more consistent with impingement rather than instability based on hip range of motion. In addition, hockey players, and goalies in particular, spend more time in hip flexion, abduction, and internal rotation (impingement positions) than in extension and external rotation (instability positions). Additionally, there was a small but significant difference in the LCEA between the 2 groupsdhockey goalies, on average, had a smaller LCEA by 2 . This may be clinically significant, however, because it could represent the difference between a normal-coverage acetabulum and one that is dysplastic. Hip preservation surgery in the setting of mixed and combined deformities has been described with favorable outcomes by Clohisy et al.27 but must be performed carefully and with appropriate caution; the increased motion that occurs with acetabular dysplasia may be beneficial in this patient population, and surgical treatment with a reorienting periacetabular osteotomy may actually hinder hip motion and make these patients more symptomatic. Limitations Our study is not without limitations. This study is based on a virtual analysis from reconstructed computed tomographic scans and does not reflect the in vivo range of motion given that it does not account for soft tissue structures that may play a role in range of motion, such as the acetabular labrum, chondral surfaces, hip capsule, and proximal tendinous structures. In addition, the dynamic analysis assumes a stable and fixed pelvis, with terminal range of motion defined with femoral motion only. Dynamic changes in pelvic position may allow for increased compensatory motion and may suggest a role for functional rehabilitation;
however, this concept has not yet been defined in the literature. Finally, the study population is subject to selection bias in that we included only athletes who presented with symptoms. In this regard, we are unable to comment on the applicability of this study to the entire population of hockey goalies.
Conclusions Symptomatic butterfly hockey goalies have a high prevalence of FAI characterized by a unique femoral cam-type deformity and noted by an elevated alpha angle and loss of offset, which is greater in magnitude and more lateral when compared with that in positional hockey players. Associated acetabular dysplasia is also common among hockey goalies.
References 1. Agel J, Dompier TP, Dick R, Marshall SW. Descriptive epidemiology of collegiate men’s ice hockey injuries: National Collegiate Athletic Association Injury Surveillance System, 1988-1989 through 2003-2004. J Athl Train 2007;42:241-248. 2. Emery CA, Meeuwisse WH, Powell JW. Groin and abdominal strain injuries in the National Hockey League. Clin J Sport Med 1999;9:151-156. 3. Irshad K, Feldman LS, Lavoie C, Lacroix VJ, Mulder DS, Brown RA. Operative management of “hockey groin syndrome”: 12 years of experience in National Hockey League players. Surgery 2001;130:759-764. 4. Sim FH, Simonet WT, Melton LJ, Lehn TA. Ice hockey injuries. Am J Sports Med 1987;15:30-40. 5. Bizzini M, Notzli HP, Maffiuletti NA. Femoroacetabular impingement in professional ice hockey players: a case series of 5 athletes after open surgical decompression of the hip. Am J Sports Med 2007;35:1955-1959. 6. Epstein DM, McHugh M, Yorio M, Neri B. Intra-articular hip injuries in National Hockey League players: a descriptive epidemiological study. Am J Sports Med 2013;41:343-348. 7. Philippon MJ, Weiss DR, Kuppersmith DA, Briggs KK, Kay CJ. Arthroscopic labral repair and treatment of femoroacetabular impingement in professional hockey players. Am J Sports Med 2010;38:99-104. 8. Shindle MK, Marx RG, Kelly BT, Bisson L, Burke CJ. Hockey injuries: a pediatric sport update. Curr Opin Pediatr 2010;22:54-60. 9. Stull JD, Philippon MJ, LaPrade RF. “At-risk” positioning and hip biomechanics of the Peewee ice hockey sprint start. Am J Sports Med 2011;39:29S-35S. 10. Bogunovic L, Gottlieb M, Pashos G, Baca G, Clohisy JC. Why do hip arthroscopy procedures fail? Clin Orthop Relat Res 2013;471:2523-2529. 11. Heyworth BE, Shindle MK, Voos JE, Rudzki JR, Kelly BT. Radiologic and intraoperative findings in revision hip arthroscopy. Arthroscopy 2007;23:1295-1302. 12. Philippon MJ, Schenker ML, Briggs KK, Kuppersmith DA, Maxwell RB, Stubbs AJ. Revision hip arthroscopy. Am J Sports Med 2007;35:1918-1921.
642
J. R. ROSS ET AL.
13. Milone MT, Bedi A, Poultsides L, et al. Novel CT-based three-dimensional software improves the characterization of cam morphology. Clin Orthop Rel Res 2013;471: 2484-2491. 14. Clohisy JC, Carlisle JC, Beaule PE, et al. A systematic approach to the plain radiographic evaluation of the young adult hip. J Bone Joint Surg Am 2008;90:47-66 (suppl 4). 15. Wiberg G. Studies on dysplastic acetabula and congenital subluxation of the hip: with special reference to the complication of osteoarthritis. Acta Chir Scand 1939;58: 7-38. 16. Notzli HP, Wyss TF, Stoecklin CH, Schmid MR, Treiber K, Hodler J. The contour of the femoral head-neck junction as a predictor for the risk of anterior impingement. J Bone Joint Surg Br 2002;84:556-560. 17. Tonnis D, Legal H, Graf R, eds. Congenital dysplasia and dislocation of the hip in children and adults. Berlin, Germany: Springer-Verlag, 1987. 18. Reynolds D, Lucas J, Klaue K. Retroversion of the acetabulum. A cause of hip pain. J Bone Joint Surg Br 1999;81: 281-288. 19. Peelle MW, Della Rocca GJ, Maloney WJ, Curry MC, Clohisy JC. Acetabular and femoral radiographic abnormalities associated with labral tears. Clin Orthop Rel Res 2005;441:327-333. 20. Brunner A, Hamers AT, Fitze M, Herzog RF. The plain beta-angle measured on radiographs in the assessment of femoroacetabular impingement. J Bone Joint Surg Br 2010;92:1203-1208.
21. Hetsroni I, Poulsides L, Bedi A, Larson CM, Kelly BT. Anterior inferior iliac spine morphology correlates with hip range of motion: a classification system and dynamic model. Clin Orthop Rel Res 2013;471:2497-2503. 22. Bedi A, Dolan M, Hetsroni I, et al. Surgical treatment of femoroacetabular impingement improves hip kinematics: a computer-assisted model. Am J Sports Med 2011;39: S43-S49 (suppl). 23. Bedi A, Dolan M, Magennis E, Lipman J, Buly R, Kelly BT. Computer-assisted modeling of osseous impingement and resection in femoroacetabular impingement. Arthroscopy 2012;28:204-210. 24. Ross JR, Nepple JJ, Philippon MJ, Kelly BT, Larson CM, Bedi A. Effect of changes in pelvic tilt on range of motion to impingement and radiographic parameters of acetabular morphologic characteristics. Am J Sports Med 2014;42: 2404-2409. 25. Ng VY, Ellis TJ. More than just a bump: Cam-type femoroacetabular impingement and the evolution of the femoral neck. Hip Int 2011;21:1-8. 26. Gosvig KK, Jacobsen S, Sonne-Holm S, Palm H, Troelsen A. Prevalence of malformations of the hip joint and their relationship to sex, groin pain, and risk of osteoarthritis: a populations-based study. J Bone Joint Surg Am 2010;92:1162-1169. 27. Clohisy JC, Nunley RM, Currey MC, Schoenecker PL. Periacetabular osteotomy for the treatment of acetabular dysplasia associated with major aspherical femoral head deformities. J Bone Joint Surg Am 2007;89: 1417-1423.