- Email: [email protected]
Contact geometry at the undersurface of the acromion with and without a rotator cuff tear
Contact Geometry at the Undersurface of the Acromion With and Without a Rotator Cuff Tear Seok-Beom Lee, M.D., Ph.D., Eiji Itoi, M.D., Shawn W. O’Driscoll, M.D., Ph.D., and Kai-Nan An, Ph.D.
Purpose: The purpose of this study was to investigate the difference in contact geometry at the undersurface of acromion in shoulders with and without a rotator cuff (RC) tear. Type of Study: Case-control study. Methods: Forty fresh cadaveric shoulders (average age at death, 61 years) without gross osteoarthritic changes were divided into the intact RC group (n ⫽ 20) and the RC tear group (n ⫽ 20). Clinical impingement was simulated by compressing the humeral head and the intact portion of the RC against the coracoacromial arch with an axial compressive force of 25 kg while the humerus was held abducted 20° in the scapular plane. The contact pattern between the acromion and the RC was measured with Fuji Prescale super low-pressure–sensitive film (Fuji Photo Film Co, Ltd, Tokyo, Japan). The imprint image was analyzed using Global Lab image software (Automatix, Marlboro, MA). Results: The percentage of the maximum anteroposterior dimension of the imprint on Fuji film to the anteroposterior diameter of the acromial undersurface was 29% ⫾ 9% in intact RC shoulders, and 39% ⫾ 13% in shoulders with an RC tear (P ⬎ .05). The percentage of the maximum mediolateral dimension of the imprint to the mediolateral diameter of the corresponding part of the acromial undersurface was 27% ⫾ 12% in intact RC shoulders, and 48% ⫾ 11% in shoulders with an RC tear. This difference was statistically significant (P ⬍ .005). Conclusions: The contact geometry of the acromial undersurface with the underlying RC in the anteroposterior dimension, which might be related to the appearance in supraspinatus outlet view, was not significantly different between shoulders with and without an RC tear. These findings suggest that factors other than acromial shape play a significant role in the pathogenesis of RC tears. The implication regarding the role of acromioplasty remains to be clarified. Key Words: Acromion—Rotator cuff—Contact geometry.
I
mpingement syndrome has been considered the primary cause of rotator cuff (RC) tears.1 Neer2 and others3-6 have implicated the undersurface of the acromion in contributing to impingement syndrome. Bigliani and Morrison7 and Bigliani et al.8 used ca-
From the Biomechanics Laboratory, Division of Orthopedic Research, Mayo Clinic and Mayo Foundation, Rochester, Minnesota,U.S.A. Supported by Grant No. AR41171 from the National Insitutes of Health. Address correspondence and reprint requests to Kai-Nan An, Ph.D., Department of Orthopaedic Surgery, Hallym University, Sacred Heart Hospital, 896 Pyungchon-dong, Dongan-ku, Anyang, Kyunggi-do, Korea. © 2001 by the Arthroscopy Association of North America 0749-8063/01/1704-2517$35.00/0 doi:10.1053/jars.2001.19974
daver dissections to determine the morphology of the acromion and classified acromial morphology into type I (flat), type II (curved), and type III (hooked). They described a correlation between the prevalence of type III acromion and RC tears. Although differences in acromial morphology have been proposed to play a significant role in the etiology of RC tears, significant disagreement exists in the literature regarding the reported incidence of each acromial type.8-10 An alternative explanation has been made by other investigators11-14 who believe that degenerative changes are the primary cause of RC tears and that osteophytes under the anterior one third of the acromion are secondary to the RC tears. A vicious cycle subsequently develops, so that the irregularity of the acromial undersurface abrades the cuff, causing further acromial changes.
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 17, No 4 (April), 2001: pp 365–372
365
366
S-B. LEE ET AL.
Regardless of the cause-and-effect relationship between acromial shape and RC tearing, an understanding of the relationship between acromial shape and RC tears is valuable. However, bony morphology alone may not reflect fully the contact patterns with the cuff beneath. If superior migration of the humeral head occurs for any reason, the RC will be in contact with a certain part of the acromial undersurface, regardless of the acromial type. However, to date, little has been reported regarding differences in contact patterns at the acromial undersurface with the underlying cuff in shoulders with and without cuff tears.15,16 We hypothesized that a description of the anterior acromial morphology according to its contact geometry with the underlying RC could be more valuable than a simple description of the bony acromial morphology alone without considering the underlying soft tissue. The presence of offending sharp contact area at the acromial undersurface in shoulders with an RC tear, as could be imagined by the term “hooked acromion,” would support the theory that acromial morphology plays a significant role in the etiology of RC tears. But, if there is no significant difference in the contact geometry in shoulders with and without an RC tear, it could be concluded that cuff tears are related to tendon degeneration. The purpose of this study was to determine the difference in contact geometry, if any, at the undersurface of the acromion to the underlying RC in shoulders with and without an RC tear.
the medial half of the scapular body were removed to expose both sides of the bony scapula. The coracoacromial ligament was maintained intact throughout the test. Loading Frame and Testing Protocol The scapula was firmly potted in the loading frame, maintaining the plane of the scapular body and the face of the glenoid perpendicular to the horizon. Contact geometry was studied by compressing the humeral head and the intact portion of the RC tendon against the coracoacromial arch in the scapular plane with an axial force of 250 N using the loading frame (Fig 1). The humeral head was maintained in neutral rotation in the scapular plane and was abducted 20° so that the intact portion of the RC in group 2 could be impressed against the acromial and coracoacromial ligament undersurface. To compare the contact geom-
METHODS Preparation of Specimens Twenty fresh-frozen cadaveric shoulders (average age at death, 60 years) with intact RC tendon (group 1) and 20 fresh-frozen cadaveric shoulders (average age at death, 64 years) with isolated supraspinatus tendon tear (group 2) were prepared for the experiment. Shoulders with gross osteoarthritic changes and tears in 2 or more cuff tendons were excluded in this study. In each specimen, the scapulothoracic and acromioclavicular junctions were disarticulated and the humerus was amputated 20-cm distal to the center of the humeral head. The skin and subcutaneous tissues were removed and the deltoid muscle was released from its origin to expose the subacromial space. The subacromial space was freed by releasing a part of the subacromial bursa and the coracohumeral ligament to allow placement of the Fuji Prescale super low-pressure–sensitive film (Fuji Photo Film Co Ltd, Tokyo, Japan) to be described. All soft tissues and muscles on
FIGURE 1. A loading frame to compress the coracoacromial arch with the humeral head and its intact part of the RC. The scapula was potted into a frame with the scapular plane and glenoid face perpendicular to the horizon. An axial force was applied to the humeral shaft to get the imprint of contact pressure on Fuji pressure-sensitive film over the acromial undersurface.
CONTACT GEOMETRY etry of the acromion to the underlying cuff with group 2 (cuff tear), identical glenohumeral position was maintained in group 1 (intact cuff) during the test. The contact geometry of an acromial undersurface with the bursal surface of the RC during loading was measured with Fuji Prescale super low-pressure–sensitive film. This film registers pressure between 0.5 and 2.5 MPa. The unit area of measurement has a resolution of 0.1 mm.7 The Prescale film consists of 2 thin sheets, between which lies a layer of microcapsules. As the pressure increases, a greater number of microcapsules rupture creating a stronger image. The film was cut to approximately 6 ⫻ 3 cm to fit the prepared subacromial space. The film was sealed within thin polyethylene sheets for use within each saline-lubricated subacromial space. The total thickness of the film and polyethylene sheets was approximately 0.4 mm. The film was fixed to the acromial undersurface by sutures on the coracoid process and posterolateral corner of the acromion. Before load application, the anterior tip of the acromion was identified on the film with a marking pen. Three measurements were obtained with each specimen to ensure reproducibility. Films with any evidence of crinkling artifact were discarded and repeated. Analysis of the Imprint Pattern The Fuji film images were scanned on a HewlettPackard (Palo Alto, CA) Desk Scan II color scanner. Global Lab image program (version 2.0, Data translation and Automatix, Marlboro, MA) converted the Fuji film image into a scaled image with 256 levels of gray. The contact pressure was quantified with a standard curve and corrected for relative humidity and temperature, which allowed conversion of color density to numerical values. Although super low-pressure–sensitive film has a useful pressure range of 0.5 to 2.5 MPa, the spectrum used for computer analysis of the imprint pattern was 0.8 to 2.5 MPa, as this range was effective in eliminating artifact caused by handling and cutting of the film. The imprint pattern on the Fuji film was analyzed from the digitized image by the Global Lab image program (Fig 2). The anteroposterior diameter of the imprint (A2) on the pressure-sensitive film was defined as the maximum distance between the most anterior and posterior parts of the imprint parallel to the face of the glenoid. The mediolateral diameter of the imprint (M2) on the pressure-sensitive film was defined as a maximum distance between the most medial and lateral parts of the imprint perpendicular to
367
FIGURE 2. Schematic drawings of the acromial undersurface and imprint on the pressure-sensitive film. (A) Direction to observe the acromial undersurface. (B) Gross appearance of the acromial undersurface. A, acromion; C, coracoid process; G, glenoid; CAL, coracoacromial ligament. (C) A1, anteroposterior diameter of the acromion; M1, mediolateral diameter of the acromion; A2, anteroposterior diameter of the imprint; M2, mediolateral diameter of the imprint. Percent anteroposterior dimension of the imprint to acromion ⫽ A2/A1 ⫻ 100; percent mediolateral dimension of the imprint to acromion ⫽ M2/M1 ⫻ 100.
the face of the glenoid. Dimensions of the acromion in each specimen were measured with a caliper for normalization. The anteroposterior diameter of the acromion (A1) was defined as a distance parallel to the face of the glenoid between the most anterior and posterior parts of the acromion. The mediolateral diameter of the acromion (M1) was defined as a distance from the anteromedial corner to the lateral margin of the acromial undersurface perpendicular to the direction of the anteroposterior diameter. To compare quantitatively the contact patterns in group 1 and 2, the percent anteroposterior and percent mediolateral dimensions were calculated by dividing the anteroposterior and mediolateral diameters of the imprint by the anteroposterior and mediolateral diameters of the acromion in each specimen. Supraspinatus outlet radiographs were obtained to identify the acromial shape of specimens in both groups. The radiographs were reviewed by 4 board-certified orthopaedic surgeons. Data were analyzed using a multivariate analysis of variance with a significance level of P ⬍ .05. RESULTS Gross Finding The undersurface of the acromion in group 1 (intact cuff) was smooth, but the area of the coracoacromial ligament insertion on the acromion formed a promi-
368
S-B. LEE ET AL.
nence anterolaterally. In group 2 (cuff tear), characteristic excrescences at the undersurface of anterior acromion were frequently seen. The excrescence always included the area underneath the insertion of the coracoacromial ligament, which was frayed and hypertrophied and, therefore, broader than the anterolateral prominence in group 1. The edge of the elevated area was relatively rough, but the contour at the middle of the area of the excrescence looked congruent with the underlying cuff. The excrescence was sometimes smooth and extended posteriorly into the middle third in some specimens (Fig 3). The undersurface of the spur at the anterior corner was always directed to the coracoid as a coracoacromial ligament spanned between the acromion and the coracoid. The length of the complete tear in group 2 averaged 1.6 cm (range, 0.8 to 2.4 cm). All the tears were confined to the supraspinatus tendon in the attachment to the greater tuberosity of the humerus. The average width of the tear was 2.2 cm (range, 0.5 to 3.0 cm).
The tear was L-shaped in 2 specimens, longitudinally split in 1, and retracted to leave an ovoid defect in the tendon in 17 specimens. Contact Geometry The imprints on the Fuji pressure-sensitive film were located consistently in the anterior one third of the undersurface of the acromion in both groups. In group 1 (intact cuff), the imprint on the Fuji pressuresensitive film was intense at the anterolateral part of the acromial undersurface, particularly at insertion of the coracoacromial ligament into the acromion (Fig 4). The anterior tips of the acromions in 16 of 20 specimens were posterior to the 12 o’clock direction of the glenoid. Only 4 of 20 specimens in group 1 had an anterior acromion extending anteriorly to reach the scapular plane. In group 2 (cuff tear), more than half of the acromions (12 of 20) had the anterior tip anteriorly on or beyond the 12 o’clock direction of the superior glenoid (scapular plane). The film showed the intense imprint area over the excrescence on the undersurface of anterior acromion (Fig 5). The maximum anteroposterior dimension of the imprint on the film as a percentage of the anteroposterior diameter of the acromion, was 29% ⫾ 9% in group 1 and 39% ⫾ 13% in group 2, which was not significantly different (P ⬎ .05). The percent maximum mediolateral diameter of the imprint to that of the acromion was 27% ⫾ 12% in group 1 and 48% ⫾ 11% in group 2. This difference was statistically significant (P ⬍ .005) (Table 1 and Fig 6). On the supraspinatus outlet view, there were 27% type I, 48% type II, and 25% type III acromion in group I, and there were 8% type I, 25% type II, 67% type III acromion in group II. DISCUSSION
FIGURE 3. Photographs of an acromion with the RC tear (group 2). The excrescence at the undersurface of the acromion has a posterior extension that is following the contour of the underlying cuff.
Bigliani and Morrison7 and other investigators8 have studied the shape of the anterior acromion in anatomic specimens and in patients. They were able to identify 3 types of acromions: type I (flat), type II (curved), and type III (hooked). In their anatomic specimen studies, 70% of the RC tears were associated with type III acromions (on a supraspinatus outlet radiograph); in their patients, 80% of the cuff tears were associated with these hooked acromions. The remaining patients with RC ruptures all had type II acromions; none was type I. Although acromial morphology has been most commonly classified radiologically,10 significant disagree-
CONTACT GEOMETRY
369
Regardless of the cause-and-effect relationship between acromial shape and the prevalence of an RC tear, an understanding of the variations in acromial shape and their association with RC tears is valuable. Flatow et al.15 and Nasca et al.16 reported the area of surface contact and proximity of areas between the subacromial surface and the superior surface of the supraspinatus tendon as an indication of the regions where contact may occur, using 9 human cadaveric shoulders. They measured the difference in contact patterns at various positions and stated that the type III acromion had increased contact area on its surface. Wuelker et al.19 reported that pressures at the coracoacromial arch were centered at the anterolateral border of the acromion during active glenohumeral
FIGURE 4. (A) Photograph of acromial undersurface in group 1 (shoulders with intact RC). (B) Imprint pattern obtained using the Fuji super low-pressure–sensitive film at the undersurface of the acromion in group 1.
ment exists in the literature regarding the reported incidence of each acromial type.7,8,17 The difference in the incidence was most marked between the type II and the type III acromion. The prevalence of a type III acromion has been reported in the range of 0%17 to 40%.7,8 Jacobson et al.18 showed that there are high degrees of interobserver and intraobserver variations in determining the 3 types of acromion. They concluded that classifying acromial shape on the basis of the Bigliani system is not reliable.
FIGURE 5. (A) Photograph of acromial undersurface in group 2 (shoulders with an RC tear). (B) Imprint pattern obtained using the Fuji super low-pressure–sensitive film at the undersurface of the acromion in group 2.
370
S-B. LEE ET AL.
TABLE 1. Percent Dimension of the Fuji PressureSensitive Film Imprint at the Acromial Undersurface
Group 1 Group 2
Anteroposterior Dimension
Mediolateral Dimension
29% ⫾ 9% 39% ⫾ 13%
27% ⫾ 12% 48% ⫾ 11%*
* The difference between groups 1 and 2 is significant, P ⬍ .05.
motion. Although contact areas on the cuff were measured in these studies, the data did not show the differences in shoulders with and without cuff tears in quantitative measures. However, they did not measure the difference in pressure distribution with respect to acromial morphology between shoulders with and without an RC tear. If superior migration of the humeral head occurs for any reason, the RC will impinge at a certain part of the acromial undersurface, regardless of the acromial type. Contact geometry of the anterior acromial undersurface to the underlying RC might, therefore, be important in the pathogenesis of RC tears. A hooked acromion, in contradiction to what could be imagined by the term hooked, could not offend the bursal surface of the cuff if the hook locates further anteriorly over the underlying cuff. We hypothesized that the morphology of the anterior acromion, as it pertains to subacromial impingement, could be described according to its contact pattern with the underlying RC. In the present study, we fixed the Prescale pressure-sensitive film to the undersurface of the acromion to show the contact geometry of the acromion to the underlying cuff tendon. Because we were interested in the relative difference in the contact pattern of the acromial undersurface in both groups, not in the absolute geometry that occurs in the impingement position, we have thus only evaluated at a given glenohumeral position. To provide more information on the contact pattern, an impression of the acromial undersurface with the corresponding cuff and humeral head was tried. The degree of glenohumeral abduction was limited to 20° to avoid the defect in the cuff in group 2 during the impression. More congruent surfaces would give imprints of larger size on pressure-sensitive film when identical compressive forces are applied on them. Our study revealed that the imprint area was located at the undersurface of the anterior third of the acromion as described by the other investigators for impingement syndrome.1,2,6,20 However, the percent maximum anteroposterior dimension of the contact
area at the acromial undersurface was not significantly different between groups 1 and 2. The percent maximum anteroposterior dimension of the imprint over the defined contact pressure might be related to the radiographic appearance in supraspinatus outlet view, i.e., the lower the value, the more hooked the acromion. Thus, the result of the present study suggests that different acromial shapes in the supraspinatus outlet view are not necessarily related either to the congruence of the acromial undersurface with the cuff or to an RC tear. In the mediolateral direction, on the other hand, the acromial undersurface with an RC tear was more congruent to the bursal surface of the RC than the one with an intact cuff. This finding corresponded to our gross observation that the anterolateral spurs underlying the insertion of the coracoacromial ligament were developed more medially in shoulders with an RC tear and looked broader than those in the intact specimens. If the acromial morphology had been primarily responsible for the development of the RC tear in these specimens, those with an RC tear should have shown evidence of an offending spur under the anterior third. The contact area of the shoulders with an RC tear did not show any focal region of contact, but the width of contact area was increased. Gross observation of the anterior acromion in group 2 also revealed that the prominence was congruent to the bursal surface of the cuff, although a more prominent structure, even with equal contact geometry, may cause more damage to the tendon. These findings would be consistent with the theory that RC tendinopathy may be the primary pathology and lead to the secondary changes seen on the undersurface of the acromion. In fact, Neer1 described, after inspection of 100 dissected scapulae,
FIGURE 6. The percent dimensions of the imprint in the anteroposterior and mediolateral directions. The percent anteroposterior dimensions of both groups did not reveal significant difference (P ⬎ .05). The percent mediolateral dimensions in group 2 (shoulder with an RC tear, gray bar) was significantly higher than that in group 1 (shoulders with intact RC tear, white bar) (P ⬍ .005).
CONTACT GEOMETRY that a characteristic ridge of proliferative spurs and excrescences on the undersurface of the anterior process were frequently seen and apparently caused by repeated impingement of the RC and humeral head, with traction on the coracoacromial ligament. Because the bony spur develops in the coracoacromial ligament, the undersurface of the traction spur might conform in shape of the ligament, which itself is usually congruent with the cuff. Based on the contact pattern of the acromion to the cuff in the present study, we believe the different radiological types seen on the supraspinatus outlet view may result from the different sizes of the acromion and the spur (Fig 7).
FIGURE 7. Different radiological types of the acromion on the supraspinatus outlet view can result from the difference in acromial length in the anteroposterior direction. The longer the acromion anteriorly, it can look to have a sharp beak offending the path of the RC as a consequence of the oblong shape of the acromial undersurface and its constant radius of curvature.
371
The spurs at the anterior acromion were mostly directed to the coracoid process as were those of the coracoacromial ligament. With increasing length of the spur in the anteroposterior direction defined in this study, acromions can look flat, curved, or even hooked on radiographs with different angles of projection, although they all follow the previous contour of the coracoacromial arch. For most of the acromions looking hooked (type III) in our study, the anterior tips extended beyond the 12 o’clock position of the glenoid, whereas seemingly flat or curved acromions remained posterior to this reference point. Description of the acromial morphology in this manner corresponds to the study reported by Zucherman et al.,21 in which the apex of the coracoacromial arch was skewed toward the coracoid process in shoulders with RC tears. In this study, we were unable to compare contact patterns in the 3 radiological types of acromion because the interobserver reliability among the 5 orthopaedic surgeons was so poor. This probably stems from the lack of truly objective criteria in the literature for radiologic classification of the acromial shape.7,8 The causes of RC tears are not well understood. What is not yet clear is the causal relationship between the various findings commonly observed in patients with an RC tear. Furthermore, recent investigations point to the importance of contact and load transfer between the RC and the coracoacromial arch in the function of normal shoulders. This is particularly true for superior stability. Acromioplasty, with coracoacromial ligament release, may facilitate anterosuperior subluxation of the humeral head, which is a very difficult problem in patients with massive cuff tears. In fact, it is a problem without a correct solution. The present study shows that the congruence of the acromial undersurface to the bursal surface of the RC is not significantly different between shoulders with and without an RC tear in the anteroposterior direction. The results of this study suggest that factors other than acromial shape may play an important role in the pathogenesis of RC tears. Secondary osteophytes can be hypertrophic, rough, and offending to underlying soft tissues if the height of the coracoacromial arch is decreased. The implication regarding the role of acromioplasty remains to be clarified. REFERENCES 1. Neer CS II. Anterior acromioplasty for the chronic impingement syndrome of the shoulder. J Bone Joint Surg Am 1972; 54:41-50.
372
S-B. LEE ET AL.
2. Neer CS II. Impingement lesions. Clin Orthop 1983;173:7077. 3. Peterson CJ, Gentz CF. Ruptures of the supraspinatus tendon: The significance of distally pointing acromioclavicular osteophytes. Clin Orthop 1983;174:43-48. 4. Post M, Cohen J. Impingement syndrome: A review of late stage II and early stage II lesions. Clin Orthop 1986;207:126132. 5. Thorling J, Bjerneld H, Hallin G, Hovelius L, Hagg O. Acromioplasty for impingement syndrome. Acta Orthop Scand 1985;56:147-148. 6. Watson M. The refractory painful arc syndrome. J Bone Joint Surg Br 1978;60:544-546. 7. Bigliani LU, Morrison DS. The clinical significance of variation in acromial morphology. Orthop Trans 1987;11:234 (abstr). 8. Bigliani LU, Morrison DS, April EW. The morphology of the acromion and its relationship to rotator cuff tears. Orthop Trans 1986;10:228 (abstr). 9. Browne AO, Bigliani LU. The shoulder. Impingement and rotator cuff tears. Contemp Orthop 1987;14:53-55. 10. Mallon WJ, Brown HR, Vogler JB III, Matinez SM. Radiographic and geometric anatomy of the scapula. Clin Orthop 1992;277:142-154. 11. Ogata S, Uhthoff HK. Histological postmortem investigation of the coracoacromial arch. Clin Orthop 1990;254:39-48. 12. Ozaki J, Fujimoto S, Nakagawa Y, Masuhara K, Itami S. Tears of the rotator cuff of the shoulder associated with pathological changes in the acromion. J Bone Joint Surg Am 1988;70:12241230. 13. Sarkar K, Taine W, Uhthoff HK. The ultrastructure of the
14.
15.
16. 17.
18.
19. 20. 21.
coracoacromial ligament in patients with chronic impingement syndrome. Clin Orthop 1990;254:49-54. Uhthoff HK, Hammond DI, Sarkar K, Hooper GJ, Papoff WJ. The role of the coracoacromial ligament in the impingement syndrome: A clinical, bursographic and histologic study. Int Orthop 1988;12:97-105. Flatow EL, Soslowsky LJ, Ticker JB, Pawluk RJ, Hepler M, Ark J, Mow VC, Bigliani LU. Excursion of the rotator cuff under the acromion. Patterns of subacromial contact. Am J Sports Med 1994;22:779-788. Nasca RJ, Salter EG, Weil CE. Contact area of the “subacromial” joint. In: BatemanJE, Welsh RP, eds. Surgery of the shoulder. New York: Decker 1984;134-139. Aoki M, Ishii S, Usui M. Clinical application for measuring the slope of the acromion. In: Post M, Morrey B, Hawkins R, eds. Surgery of the shoulder. St. Louis: Mosby-Year Book, 1990;200-203. Jacobson SR, Speer KP, Moor JT, Janda DH, Saddemi SR, MacDonald PB, Mallon WJ. Reliability of radiographic assessment of acromial morphology. J Shoulder Elbow Surg 1995;4:449-453. Wuelker N, Roetman B, Roessig S. Coracoacromial pressure recording in a cadaveric model. J Shoulder Elbow Surg 1995; 4:462-467. Codman EA. The shoulder, rupture of the supraspinatus tendon and other lesions in or about the subacromial bursa. Boston: Thomas Todd, 1934. Zuckerman JD, Kummer FJ, Cuomo F, Simon J, Rosenblum S. The influence of coracoacromial arch anatomy on rotator cuff tears. J Shoulder Elbow Surg 1992;1:4-14.
Purpose: The purpose of this study was to investigate the difference in contact geometry at the undersurface of acromion in shoulders with and without a rotator cuff (RC) tear. Type of Study: Case-control study. Methods: Forty fresh cadaveric shoulders (average age at death, 61 years) without gross osteoarthritic changes were divided into the intact RC group (n ⫽ 20) and the RC tear group (n ⫽ 20). Clinical impingement was simulated by compressing the humeral head and the intact portion of the RC against the coracoacromial arch with an axial compressive force of 25 kg while the humerus was held abducted 20° in the scapular plane. The contact pattern between the acromion and the RC was measured with Fuji Prescale super low-pressure–sensitive film (Fuji Photo Film Co, Ltd, Tokyo, Japan). The imprint image was analyzed using Global Lab image software (Automatix, Marlboro, MA). Results: The percentage of the maximum anteroposterior dimension of the imprint on Fuji film to the anteroposterior diameter of the acromial undersurface was 29% ⫾ 9% in intact RC shoulders, and 39% ⫾ 13% in shoulders with an RC tear (P ⬎ .05). The percentage of the maximum mediolateral dimension of the imprint to the mediolateral diameter of the corresponding part of the acromial undersurface was 27% ⫾ 12% in intact RC shoulders, and 48% ⫾ 11% in shoulders with an RC tear. This difference was statistically significant (P ⬍ .005). Conclusions: The contact geometry of the acromial undersurface with the underlying RC in the anteroposterior dimension, which might be related to the appearance in supraspinatus outlet view, was not significantly different between shoulders with and without an RC tear. These findings suggest that factors other than acromial shape play a significant role in the pathogenesis of RC tears. The implication regarding the role of acromioplasty remains to be clarified. Key Words: Acromion—Rotator cuff—Contact geometry.
I
mpingement syndrome has been considered the primary cause of rotator cuff (RC) tears.1 Neer2 and others3-6 have implicated the undersurface of the acromion in contributing to impingement syndrome. Bigliani and Morrison7 and Bigliani et al.8 used ca-
From the Biomechanics Laboratory, Division of Orthopedic Research, Mayo Clinic and Mayo Foundation, Rochester, Minnesota,U.S.A. Supported by Grant No. AR41171 from the National Insitutes of Health. Address correspondence and reprint requests to Kai-Nan An, Ph.D., Department of Orthopaedic Surgery, Hallym University, Sacred Heart Hospital, 896 Pyungchon-dong, Dongan-ku, Anyang, Kyunggi-do, Korea. © 2001 by the Arthroscopy Association of North America 0749-8063/01/1704-2517$35.00/0 doi:10.1053/jars.2001.19974
daver dissections to determine the morphology of the acromion and classified acromial morphology into type I (flat), type II (curved), and type III (hooked). They described a correlation between the prevalence of type III acromion and RC tears. Although differences in acromial morphology have been proposed to play a significant role in the etiology of RC tears, significant disagreement exists in the literature regarding the reported incidence of each acromial type.8-10 An alternative explanation has been made by other investigators11-14 who believe that degenerative changes are the primary cause of RC tears and that osteophytes under the anterior one third of the acromion are secondary to the RC tears. A vicious cycle subsequently develops, so that the irregularity of the acromial undersurface abrades the cuff, causing further acromial changes.
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 17, No 4 (April), 2001: pp 365–372
365
366
S-B. LEE ET AL.
Regardless of the cause-and-effect relationship between acromial shape and RC tearing, an understanding of the relationship between acromial shape and RC tears is valuable. However, bony morphology alone may not reflect fully the contact patterns with the cuff beneath. If superior migration of the humeral head occurs for any reason, the RC will be in contact with a certain part of the acromial undersurface, regardless of the acromial type. However, to date, little has been reported regarding differences in contact patterns at the acromial undersurface with the underlying cuff in shoulders with and without cuff tears.15,16 We hypothesized that a description of the anterior acromial morphology according to its contact geometry with the underlying RC could be more valuable than a simple description of the bony acromial morphology alone without considering the underlying soft tissue. The presence of offending sharp contact area at the acromial undersurface in shoulders with an RC tear, as could be imagined by the term “hooked acromion,” would support the theory that acromial morphology plays a significant role in the etiology of RC tears. But, if there is no significant difference in the contact geometry in shoulders with and without an RC tear, it could be concluded that cuff tears are related to tendon degeneration. The purpose of this study was to determine the difference in contact geometry, if any, at the undersurface of the acromion to the underlying RC in shoulders with and without an RC tear.
the medial half of the scapular body were removed to expose both sides of the bony scapula. The coracoacromial ligament was maintained intact throughout the test. Loading Frame and Testing Protocol The scapula was firmly potted in the loading frame, maintaining the plane of the scapular body and the face of the glenoid perpendicular to the horizon. Contact geometry was studied by compressing the humeral head and the intact portion of the RC tendon against the coracoacromial arch in the scapular plane with an axial force of 250 N using the loading frame (Fig 1). The humeral head was maintained in neutral rotation in the scapular plane and was abducted 20° so that the intact portion of the RC in group 2 could be impressed against the acromial and coracoacromial ligament undersurface. To compare the contact geom-
METHODS Preparation of Specimens Twenty fresh-frozen cadaveric shoulders (average age at death, 60 years) with intact RC tendon (group 1) and 20 fresh-frozen cadaveric shoulders (average age at death, 64 years) with isolated supraspinatus tendon tear (group 2) were prepared for the experiment. Shoulders with gross osteoarthritic changes and tears in 2 or more cuff tendons were excluded in this study. In each specimen, the scapulothoracic and acromioclavicular junctions were disarticulated and the humerus was amputated 20-cm distal to the center of the humeral head. The skin and subcutaneous tissues were removed and the deltoid muscle was released from its origin to expose the subacromial space. The subacromial space was freed by releasing a part of the subacromial bursa and the coracohumeral ligament to allow placement of the Fuji Prescale super low-pressure–sensitive film (Fuji Photo Film Co Ltd, Tokyo, Japan) to be described. All soft tissues and muscles on
FIGURE 1. A loading frame to compress the coracoacromial arch with the humeral head and its intact part of the RC. The scapula was potted into a frame with the scapular plane and glenoid face perpendicular to the horizon. An axial force was applied to the humeral shaft to get the imprint of contact pressure on Fuji pressure-sensitive film over the acromial undersurface.
CONTACT GEOMETRY etry of the acromion to the underlying cuff with group 2 (cuff tear), identical glenohumeral position was maintained in group 1 (intact cuff) during the test. The contact geometry of an acromial undersurface with the bursal surface of the RC during loading was measured with Fuji Prescale super low-pressure–sensitive film. This film registers pressure between 0.5 and 2.5 MPa. The unit area of measurement has a resolution of 0.1 mm.7 The Prescale film consists of 2 thin sheets, between which lies a layer of microcapsules. As the pressure increases, a greater number of microcapsules rupture creating a stronger image. The film was cut to approximately 6 ⫻ 3 cm to fit the prepared subacromial space. The film was sealed within thin polyethylene sheets for use within each saline-lubricated subacromial space. The total thickness of the film and polyethylene sheets was approximately 0.4 mm. The film was fixed to the acromial undersurface by sutures on the coracoid process and posterolateral corner of the acromion. Before load application, the anterior tip of the acromion was identified on the film with a marking pen. Three measurements were obtained with each specimen to ensure reproducibility. Films with any evidence of crinkling artifact were discarded and repeated. Analysis of the Imprint Pattern The Fuji film images were scanned on a HewlettPackard (Palo Alto, CA) Desk Scan II color scanner. Global Lab image program (version 2.0, Data translation and Automatix, Marlboro, MA) converted the Fuji film image into a scaled image with 256 levels of gray. The contact pressure was quantified with a standard curve and corrected for relative humidity and temperature, which allowed conversion of color density to numerical values. Although super low-pressure–sensitive film has a useful pressure range of 0.5 to 2.5 MPa, the spectrum used for computer analysis of the imprint pattern was 0.8 to 2.5 MPa, as this range was effective in eliminating artifact caused by handling and cutting of the film. The imprint pattern on the Fuji film was analyzed from the digitized image by the Global Lab image program (Fig 2). The anteroposterior diameter of the imprint (A2) on the pressure-sensitive film was defined as the maximum distance between the most anterior and posterior parts of the imprint parallel to the face of the glenoid. The mediolateral diameter of the imprint (M2) on the pressure-sensitive film was defined as a maximum distance between the most medial and lateral parts of the imprint perpendicular to
367
FIGURE 2. Schematic drawings of the acromial undersurface and imprint on the pressure-sensitive film. (A) Direction to observe the acromial undersurface. (B) Gross appearance of the acromial undersurface. A, acromion; C, coracoid process; G, glenoid; CAL, coracoacromial ligament. (C) A1, anteroposterior diameter of the acromion; M1, mediolateral diameter of the acromion; A2, anteroposterior diameter of the imprint; M2, mediolateral diameter of the imprint. Percent anteroposterior dimension of the imprint to acromion ⫽ A2/A1 ⫻ 100; percent mediolateral dimension of the imprint to acromion ⫽ M2/M1 ⫻ 100.
the face of the glenoid. Dimensions of the acromion in each specimen were measured with a caliper for normalization. The anteroposterior diameter of the acromion (A1) was defined as a distance parallel to the face of the glenoid between the most anterior and posterior parts of the acromion. The mediolateral diameter of the acromion (M1) was defined as a distance from the anteromedial corner to the lateral margin of the acromial undersurface perpendicular to the direction of the anteroposterior diameter. To compare quantitatively the contact patterns in group 1 and 2, the percent anteroposterior and percent mediolateral dimensions were calculated by dividing the anteroposterior and mediolateral diameters of the imprint by the anteroposterior and mediolateral diameters of the acromion in each specimen. Supraspinatus outlet radiographs were obtained to identify the acromial shape of specimens in both groups. The radiographs were reviewed by 4 board-certified orthopaedic surgeons. Data were analyzed using a multivariate analysis of variance with a significance level of P ⬍ .05. RESULTS Gross Finding The undersurface of the acromion in group 1 (intact cuff) was smooth, but the area of the coracoacromial ligament insertion on the acromion formed a promi-
368
S-B. LEE ET AL.
nence anterolaterally. In group 2 (cuff tear), characteristic excrescences at the undersurface of anterior acromion were frequently seen. The excrescence always included the area underneath the insertion of the coracoacromial ligament, which was frayed and hypertrophied and, therefore, broader than the anterolateral prominence in group 1. The edge of the elevated area was relatively rough, but the contour at the middle of the area of the excrescence looked congruent with the underlying cuff. The excrescence was sometimes smooth and extended posteriorly into the middle third in some specimens (Fig 3). The undersurface of the spur at the anterior corner was always directed to the coracoid as a coracoacromial ligament spanned between the acromion and the coracoid. The length of the complete tear in group 2 averaged 1.6 cm (range, 0.8 to 2.4 cm). All the tears were confined to the supraspinatus tendon in the attachment to the greater tuberosity of the humerus. The average width of the tear was 2.2 cm (range, 0.5 to 3.0 cm).
The tear was L-shaped in 2 specimens, longitudinally split in 1, and retracted to leave an ovoid defect in the tendon in 17 specimens. Contact Geometry The imprints on the Fuji pressure-sensitive film were located consistently in the anterior one third of the undersurface of the acromion in both groups. In group 1 (intact cuff), the imprint on the Fuji pressuresensitive film was intense at the anterolateral part of the acromial undersurface, particularly at insertion of the coracoacromial ligament into the acromion (Fig 4). The anterior tips of the acromions in 16 of 20 specimens were posterior to the 12 o’clock direction of the glenoid. Only 4 of 20 specimens in group 1 had an anterior acromion extending anteriorly to reach the scapular plane. In group 2 (cuff tear), more than half of the acromions (12 of 20) had the anterior tip anteriorly on or beyond the 12 o’clock direction of the superior glenoid (scapular plane). The film showed the intense imprint area over the excrescence on the undersurface of anterior acromion (Fig 5). The maximum anteroposterior dimension of the imprint on the film as a percentage of the anteroposterior diameter of the acromion, was 29% ⫾ 9% in group 1 and 39% ⫾ 13% in group 2, which was not significantly different (P ⬎ .05). The percent maximum mediolateral diameter of the imprint to that of the acromion was 27% ⫾ 12% in group 1 and 48% ⫾ 11% in group 2. This difference was statistically significant (P ⬍ .005) (Table 1 and Fig 6). On the supraspinatus outlet view, there were 27% type I, 48% type II, and 25% type III acromion in group I, and there were 8% type I, 25% type II, 67% type III acromion in group II. DISCUSSION
FIGURE 3. Photographs of an acromion with the RC tear (group 2). The excrescence at the undersurface of the acromion has a posterior extension that is following the contour of the underlying cuff.
Bigliani and Morrison7 and other investigators8 have studied the shape of the anterior acromion in anatomic specimens and in patients. They were able to identify 3 types of acromions: type I (flat), type II (curved), and type III (hooked). In their anatomic specimen studies, 70% of the RC tears were associated with type III acromions (on a supraspinatus outlet radiograph); in their patients, 80% of the cuff tears were associated with these hooked acromions. The remaining patients with RC ruptures all had type II acromions; none was type I. Although acromial morphology has been most commonly classified radiologically,10 significant disagree-
CONTACT GEOMETRY
369
Regardless of the cause-and-effect relationship between acromial shape and the prevalence of an RC tear, an understanding of the variations in acromial shape and their association with RC tears is valuable. Flatow et al.15 and Nasca et al.16 reported the area of surface contact and proximity of areas between the subacromial surface and the superior surface of the supraspinatus tendon as an indication of the regions where contact may occur, using 9 human cadaveric shoulders. They measured the difference in contact patterns at various positions and stated that the type III acromion had increased contact area on its surface. Wuelker et al.19 reported that pressures at the coracoacromial arch were centered at the anterolateral border of the acromion during active glenohumeral
FIGURE 4. (A) Photograph of acromial undersurface in group 1 (shoulders with intact RC). (B) Imprint pattern obtained using the Fuji super low-pressure–sensitive film at the undersurface of the acromion in group 1.
ment exists in the literature regarding the reported incidence of each acromial type.7,8,17 The difference in the incidence was most marked between the type II and the type III acromion. The prevalence of a type III acromion has been reported in the range of 0%17 to 40%.7,8 Jacobson et al.18 showed that there are high degrees of interobserver and intraobserver variations in determining the 3 types of acromion. They concluded that classifying acromial shape on the basis of the Bigliani system is not reliable.
FIGURE 5. (A) Photograph of acromial undersurface in group 2 (shoulders with an RC tear). (B) Imprint pattern obtained using the Fuji super low-pressure–sensitive film at the undersurface of the acromion in group 2.
370
S-B. LEE ET AL.
TABLE 1. Percent Dimension of the Fuji PressureSensitive Film Imprint at the Acromial Undersurface
Group 1 Group 2
Anteroposterior Dimension
Mediolateral Dimension
29% ⫾ 9% 39% ⫾ 13%
27% ⫾ 12% 48% ⫾ 11%*
* The difference between groups 1 and 2 is significant, P ⬍ .05.
motion. Although contact areas on the cuff were measured in these studies, the data did not show the differences in shoulders with and without cuff tears in quantitative measures. However, they did not measure the difference in pressure distribution with respect to acromial morphology between shoulders with and without an RC tear. If superior migration of the humeral head occurs for any reason, the RC will impinge at a certain part of the acromial undersurface, regardless of the acromial type. Contact geometry of the anterior acromial undersurface to the underlying RC might, therefore, be important in the pathogenesis of RC tears. A hooked acromion, in contradiction to what could be imagined by the term hooked, could not offend the bursal surface of the cuff if the hook locates further anteriorly over the underlying cuff. We hypothesized that the morphology of the anterior acromion, as it pertains to subacromial impingement, could be described according to its contact pattern with the underlying RC. In the present study, we fixed the Prescale pressure-sensitive film to the undersurface of the acromion to show the contact geometry of the acromion to the underlying cuff tendon. Because we were interested in the relative difference in the contact pattern of the acromial undersurface in both groups, not in the absolute geometry that occurs in the impingement position, we have thus only evaluated at a given glenohumeral position. To provide more information on the contact pattern, an impression of the acromial undersurface with the corresponding cuff and humeral head was tried. The degree of glenohumeral abduction was limited to 20° to avoid the defect in the cuff in group 2 during the impression. More congruent surfaces would give imprints of larger size on pressure-sensitive film when identical compressive forces are applied on them. Our study revealed that the imprint area was located at the undersurface of the anterior third of the acromion as described by the other investigators for impingement syndrome.1,2,6,20 However, the percent maximum anteroposterior dimension of the contact
area at the acromial undersurface was not significantly different between groups 1 and 2. The percent maximum anteroposterior dimension of the imprint over the defined contact pressure might be related to the radiographic appearance in supraspinatus outlet view, i.e., the lower the value, the more hooked the acromion. Thus, the result of the present study suggests that different acromial shapes in the supraspinatus outlet view are not necessarily related either to the congruence of the acromial undersurface with the cuff or to an RC tear. In the mediolateral direction, on the other hand, the acromial undersurface with an RC tear was more congruent to the bursal surface of the RC than the one with an intact cuff. This finding corresponded to our gross observation that the anterolateral spurs underlying the insertion of the coracoacromial ligament were developed more medially in shoulders with an RC tear and looked broader than those in the intact specimens. If the acromial morphology had been primarily responsible for the development of the RC tear in these specimens, those with an RC tear should have shown evidence of an offending spur under the anterior third. The contact area of the shoulders with an RC tear did not show any focal region of contact, but the width of contact area was increased. Gross observation of the anterior acromion in group 2 also revealed that the prominence was congruent to the bursal surface of the cuff, although a more prominent structure, even with equal contact geometry, may cause more damage to the tendon. These findings would be consistent with the theory that RC tendinopathy may be the primary pathology and lead to the secondary changes seen on the undersurface of the acromion. In fact, Neer1 described, after inspection of 100 dissected scapulae,
FIGURE 6. The percent dimensions of the imprint in the anteroposterior and mediolateral directions. The percent anteroposterior dimensions of both groups did not reveal significant difference (P ⬎ .05). The percent mediolateral dimensions in group 2 (shoulder with an RC tear, gray bar) was significantly higher than that in group 1 (shoulders with intact RC tear, white bar) (P ⬍ .005).
CONTACT GEOMETRY that a characteristic ridge of proliferative spurs and excrescences on the undersurface of the anterior process were frequently seen and apparently caused by repeated impingement of the RC and humeral head, with traction on the coracoacromial ligament. Because the bony spur develops in the coracoacromial ligament, the undersurface of the traction spur might conform in shape of the ligament, which itself is usually congruent with the cuff. Based on the contact pattern of the acromion to the cuff in the present study, we believe the different radiological types seen on the supraspinatus outlet view may result from the different sizes of the acromion and the spur (Fig 7).
FIGURE 7. Different radiological types of the acromion on the supraspinatus outlet view can result from the difference in acromial length in the anteroposterior direction. The longer the acromion anteriorly, it can look to have a sharp beak offending the path of the RC as a consequence of the oblong shape of the acromial undersurface and its constant radius of curvature.
371
The spurs at the anterior acromion were mostly directed to the coracoid process as were those of the coracoacromial ligament. With increasing length of the spur in the anteroposterior direction defined in this study, acromions can look flat, curved, or even hooked on radiographs with different angles of projection, although they all follow the previous contour of the coracoacromial arch. For most of the acromions looking hooked (type III) in our study, the anterior tips extended beyond the 12 o’clock position of the glenoid, whereas seemingly flat or curved acromions remained posterior to this reference point. Description of the acromial morphology in this manner corresponds to the study reported by Zucherman et al.,21 in which the apex of the coracoacromial arch was skewed toward the coracoid process in shoulders with RC tears. In this study, we were unable to compare contact patterns in the 3 radiological types of acromion because the interobserver reliability among the 5 orthopaedic surgeons was so poor. This probably stems from the lack of truly objective criteria in the literature for radiologic classification of the acromial shape.7,8 The causes of RC tears are not well understood. What is not yet clear is the causal relationship between the various findings commonly observed in patients with an RC tear. Furthermore, recent investigations point to the importance of contact and load transfer between the RC and the coracoacromial arch in the function of normal shoulders. This is particularly true for superior stability. Acromioplasty, with coracoacromial ligament release, may facilitate anterosuperior subluxation of the humeral head, which is a very difficult problem in patients with massive cuff tears. In fact, it is a problem without a correct solution. The present study shows that the congruence of the acromial undersurface to the bursal surface of the RC is not significantly different between shoulders with and without an RC tear in the anteroposterior direction. The results of this study suggest that factors other than acromial shape may play an important role in the pathogenesis of RC tears. Secondary osteophytes can be hypertrophic, rough, and offending to underlying soft tissues if the height of the coracoacromial arch is decreased. The implication regarding the role of acromioplasty remains to be clarified. REFERENCES 1. Neer CS II. Anterior acromioplasty for the chronic impingement syndrome of the shoulder. J Bone Joint Surg Am 1972; 54:41-50.
372
S-B. LEE ET AL.
2. Neer CS II. Impingement lesions. Clin Orthop 1983;173:7077. 3. Peterson CJ, Gentz CF. Ruptures of the supraspinatus tendon: The significance of distally pointing acromioclavicular osteophytes. Clin Orthop 1983;174:43-48. 4. Post M, Cohen J. Impingement syndrome: A review of late stage II and early stage II lesions. Clin Orthop 1986;207:126132. 5. Thorling J, Bjerneld H, Hallin G, Hovelius L, Hagg O. Acromioplasty for impingement syndrome. Acta Orthop Scand 1985;56:147-148. 6. Watson M. The refractory painful arc syndrome. J Bone Joint Surg Br 1978;60:544-546. 7. Bigliani LU, Morrison DS. The clinical significance of variation in acromial morphology. Orthop Trans 1987;11:234 (abstr). 8. Bigliani LU, Morrison DS, April EW. The morphology of the acromion and its relationship to rotator cuff tears. Orthop Trans 1986;10:228 (abstr). 9. Browne AO, Bigliani LU. The shoulder. Impingement and rotator cuff tears. Contemp Orthop 1987;14:53-55. 10. Mallon WJ, Brown HR, Vogler JB III, Matinez SM. Radiographic and geometric anatomy of the scapula. Clin Orthop 1992;277:142-154. 11. Ogata S, Uhthoff HK. Histological postmortem investigation of the coracoacromial arch. Clin Orthop 1990;254:39-48. 12. Ozaki J, Fujimoto S, Nakagawa Y, Masuhara K, Itami S. Tears of the rotator cuff of the shoulder associated with pathological changes in the acromion. J Bone Joint Surg Am 1988;70:12241230. 13. Sarkar K, Taine W, Uhthoff HK. The ultrastructure of the
14.
15.
16. 17.
18.
19. 20. 21.
coracoacromial ligament in patients with chronic impingement syndrome. Clin Orthop 1990;254:49-54. Uhthoff HK, Hammond DI, Sarkar K, Hooper GJ, Papoff WJ. The role of the coracoacromial ligament in the impingement syndrome: A clinical, bursographic and histologic study. Int Orthop 1988;12:97-105. Flatow EL, Soslowsky LJ, Ticker JB, Pawluk RJ, Hepler M, Ark J, Mow VC, Bigliani LU. Excursion of the rotator cuff under the acromion. Patterns of subacromial contact. Am J Sports Med 1994;22:779-788. Nasca RJ, Salter EG, Weil CE. Contact area of the “subacromial” joint. In: BatemanJE, Welsh RP, eds. Surgery of the shoulder. New York: Decker 1984;134-139. Aoki M, Ishii S, Usui M. Clinical application for measuring the slope of the acromion. In: Post M, Morrey B, Hawkins R, eds. Surgery of the shoulder. St. Louis: Mosby-Year Book, 1990;200-203. Jacobson SR, Speer KP, Moor JT, Janda DH, Saddemi SR, MacDonald PB, Mallon WJ. Reliability of radiographic assessment of acromial morphology. J Shoulder Elbow Surg 1995;4:449-453. Wuelker N, Roetman B, Roessig S. Coracoacromial pressure recording in a cadaveric model. J Shoulder Elbow Surg 1995; 4:462-467. Codman EA. The shoulder, rupture of the supraspinatus tendon and other lesions in or about the subacromial bursa. Boston: Thomas Todd, 1934. Zuckerman JD, Kummer FJ, Cuomo F, Simon J, Rosenblum S. The influence of coracoacromial arch anatomy on rotator cuff tears. J Shoulder Elbow Surg 1992;1:4-14.