This study was p

from the Minis

and also by a g

Cooperative Insu

J Shoulder Elbow Surg (2014) -, 1-6

1058-2746/$ - s

http://dx.doi.org

www.elsevier.com/locate/ymse

The anatomic relationship between themorphology of the greater tubercle of thehumerus and the insertion of the infraspinatustendon

Taiki Nozaki, MDa,b, Akimoto Nimura, MD, PhDa,*, Hitomi Fujishiro, MTa,Tomoyuki Mochizuki, MD, PhDc, Kumiko Yamaguchi, MD, PhDa,Ryuichi Kato, MD, PhDd, Hiroyuki Sugaya, MD, PhDe, Keiichi Akita, MD, PhDa

aDepartment of Clinical Anatomy, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental University,Tokyo, JapanbDepartment of Radiology, St. Luke’s International Hospital, Tokyo, JapancDepartment of Joint Reconstruction, Graduate School, Tokyo Medical and Dental University, Tokyo, JapandNokyo Kyosai Research Institute, Tokyo, JapaneShoulder and Elbow Service, Funabashi Orthopaedic Sports Medicine Center, Funabashi, Japan

Background: The objective of this study was to evaluate the topographic relationship between themorphology of the greater tubercle and the insertion of the tendon of the infraspinatus.Materials and methods: First, we defined an impression of the greater tubercle, which has not been recog-nized in classic textbooks, as the ‘‘lateral impression’’ and then measured the dimensions of the ‘‘lateralimpression’’ of the greater tubercle in 71 samples of dry bone of humeri. Next, we examined 16 cadaverichumeri with rotator cuff tendons by micro–computed tomography to analyze the positional relationshipbetween the lateral impression and the infraspinatus tendon.Results: In all samples of dry bones, the lateral impression could be identified as a triangle shape. Thelateral impression was composed of the border with the highest impression (mean, 6.3 mm), the borderwith the middle impression (mean, 5.0 mm), and the border with the lateral wall of the greater tubercle(mean, 8.5 mm). In all samples of humeri with rotator cuffs, we could confirm the lateral impression,and the border between the highest impression and the lateral impression corresponded to the anteriorborder of the insertion of the infraspinatus tendon.Conclusion: We propose a new anatomic concept of the lateral impression that could enable the precisediagnosis of and facilitate repair techniques for infraspinatus tear, according to specific anatomic charac-teristics, by applying 3-dimensional computed tomography assessment preoperatively.

artly supported by a Grant-in-Aid for Scientific Research

try of Education, Culture, Sports (C) (No. 24791525)

rant from NokyoKyosai Research Institute (Agricultural

rance Research Institute).

*Reprint requests: Akimoto Nimura, MD, PhD, Department of Clinical

Anatomy, Graduate School of Medical and Dental Sciences, Tokyo

Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo

113-8519, Japan.

E-mail address: a-nim@sol.dti.ne.jp (A. Nimura).

ee front matter � 2014 Journal of Shoulder and Elbow Surgery Board of Trustees.

/10.1016/j.jse.2014.09.038

2 T. Nozaki et al.

Level of evidence: Anatomy Study, Imaging.� 2014 Journal of Shoulder and Elbow Surgery Board of Trustees.

Keywords: Greater tubercle; lateral impression; highest impression; middle impression; infraspinatus;

micro-CT

In anatomic textbooks, the greater tubercle is marked by3 flat impressions: the highest impression gives insertion tothe supraspinatus muscle; the middle, to the infraspinatus;and the lowest, to the teres minor.3-5 In these descriptions,the shapes of impressions of the greater tubercle have beensimply described as adjacent squares (Fig. 1).1

However, in dry bone samples for anatomy practice or3-dimensional (3D) computed tomography (CT) images, wesometimes encounter specimens that have a different shapeof the greater tubercle; the tubercle has another impression,in comparison with the conventional description in text-books. In addition, Mochizuki et al7 previously reported thatthe infraspinatus inserted into the anterior edge of the greatertubercle and occupied a substantial area of the greatertubercle, which was in complete contrast to traditionalanatomic concepts of the insertions of the supraspinatus andinfraspinatus to the greater tubercle. On the basis of thestudy of Mochizuki et al,7 strangely enough, the borderbetween the supraspinatus and the infraspinatus also delin-eated the area of the highest impression. We hypothesizedthe consistent existence of this additional impression of thegreater tubercle of the humerus and that this impressionmight be related to the insertion of the infraspinatus tendonthat was newly described in Mochizuki’s report.7

The first aim of this study was to define the additionalimpression of the greater tubercle as the ‘‘lateral impres-sion’’ and to evaluate the existence of this impression. Thesecond aim was to identify the topographic relationshipbetween the morphology of the lateral impression of thegreater tubercle and the insertion of the infraspinatustendon.

Materials and methods

This study is an anatomic research using dry bone samples andembalmed cadavers at Tokyo Medical and Dental University. Allof the donors voluntarily declared before their death that theirremains would be donated as materials for education and study.This voluntary donor system of cadavers is universally spreadthroughout Japan, and our study completely complies with thecurrent laws of Japan. History of any shoulder problem was notavailable. Of 78 samples of dry bone of humeri, we excluded7 samples with severe deformities or destruction of the greatertubercle. In 71 samples including 34 right and 37 left samples, weobserved another triangular impression that was located postero-lateral to the highest impression, anterolateral to the middleimpression, and medial to the lateral wall of the greater tubercle;we defined it as the lateral impression of the greater tubercle

(Fig. 2). Then, we measured the dimensions of the lateralimpression according to each boundary with the highest impres-sion (star in Fig. 3), the middle impression (square in Fig. 3), andthe lateral wall of the greater tubercle (circle in Fig. 3) with avernier caliper. One observer repeated these measurements twiceand calculated the intraclass correlation coefficient to evaluate thevalidity of the measurement within each group. The scores ofintraclass correlation coefficient for star, square, and circle were0.810, 0.792, and 0.821, respectively.

Next, a 3D image of the greater tubercle in 28 cadaveric hu-meri with rotator cuff tendons was taken with a micro-CT scanner(inspeXio SMX-100 CT; SHIMADZU, Kyoto, Japan) withapplication software (VGStudio Max 2.0, Heidelberg, Germany).Of the 28 samples, we excluded 12 specimens with an unclearbone surface (10 specimens) or marked osteophytes on the greatertubercle (2 specimens) on 3D CT images (Supplemental file).After imaging, we carefully dissected the remaining 16 samples soas not to damage the bone surface and identified the anteromedialborder of the infraspinatus tendon. Radiopaque markers (0.8 mmin width) were placed along the identified border of the infra-spinatus at the base of the tendon. We performed micro-CT scanand analyzed the positional relationship between the lateralimpression and the identified border of the infraspinatus.

Finally, the spatial relationship between the lateral impressionand the anterior edge of the infraspinatus with the radiopaquemarker was analyzed by the consensus of a board-certifiedorthopedic surgeon and a radiologist.

Results

We could identify a remarkable impression delineated asa triangle that was located posterolateral to the highestimpression, anterolateral to the middle impression, andmedial to the lateral wall of the greater tubercle. We termedthis triangular impression the lateral impression. In allsamples of dry bones, the lateral impression could beidentified, although there was some variability in the size(Fig. 4). To analyze the size variance of the lateral impres-sion, we measured each side of the triangle in all samples(Fig. 3). The lateral impression was composed of the borderwith the highest impression (mean, 6.3 mm; star in Fig. 3)(Table I ), the border with the middle impression (mean,5.0 mm; square in Fig. 3), and the border with the lateralwall of the greater tubercle (mean, 8.5 mm; circle in Fig. 3).

To understand the significance of the lateral impressionof the greater tubercle in reference to the insertion of ro-tator cuff tendons, we took 3D images of cadaveric humeriwith rotator cuff tendons by micro-CT. In all 16 samplesthat showed the clear surface of the greater tubercle, we

Figure 2 Definition of the lateral impression of the greater tu-bercle. Right proximal humerus, lateral view. Another impression(asterisk) could be observed posterolateral to the highest impres-sion, anterolateral to the middle impression, and medial to thelateral wall of the greater tubercle. Post, posterior; Sup, superior.

Figure 3 Measurement sites of the lateral impression on thegreater tubercle. The shape of the lateral impression wasapproximately that of a triangle. We measured each side of thetriangle. Star, the border with the highest impression; square, theborder with the middle impression; circle, the border with thelateral wall of the greater tubercle. Data are listed in Table I. Post,posterior; Sup, superior.

Figure 1 Illustration of the conventionally described greatertubercle of humerus. The shapes of impressions of the greatertubercle (here illustrated as the highest and middle impressions)have been simply described as adjacent squares.

‘‘Lateral impression’’ of the greater tubercle 3

could confirm the lateral impression as the triangle shape(Fig. 5, A and B). In addition, we marked the anterior edgeof the infraspinatus tendon with a radiopaque marker andtook 3D images of the same specimens again. The anterior

side of the lateral impression, in other words, the borderbetween the highest impression and the lateral impression,was shown to correspond to the insertion of the anterioredge of the infraspinatus tendon in all 16 samples (Fig. 5,C and D).

Discussion

In classic textbooks of anatomy, the shapes of impressions ofthe greater tubercle have been simply described as adjacentsquares, superior and middle impressions.1,3-5 However, wecould confirm the existence of the lateral impression as atriangle facet that was separated from the highest and middleimpressions in all samples of dry bones and on 3Dmicro-CTimages. The size of the lateral impression varied from smallto large. As a reason for the discrepancy between previousknowledge and actual images of the greater tubercle of thehumeri, it could be speculated that for the smaller lateralimpression, the configuration of the greater tubercle seemedmore similar to that described in classic textbooks. Incontrast, for the larger lateral impression, this impressioncould be clearly identified to be like cases that we havesometimes encountered in clinical images and in anatomypractice dissections.

In the previous literature, the highest impressionwas expressed as the prominence where the supraspinatusinserted and the middle impression where the infraspinatusinserted. The configurations of each impression were

Figure 4 Variable sizes of the lateral impression on the greater tubercle. The lateral impression is composed of boundaries with thehighest impression, the middle impression, and the lateral wall of the greater tubercle. The size of the lateral impression varies from large(A) to small (B).

Table I Dimensions of the lateral impression on the greatertubercle in dry bone samples

Length from the location ofmeasurement

Average and standarddeviation (range)

Border with highest impression(star)

6.3 � 3.2 (2.0-16.3)

Border with middle impression(square)

5.0 � 1.7 (1.8-9.9)

Border with lateral wall of thegreater tuberosity (circle)

8.5 � 3.7 (3.0-18.8)

Locations of measurement sites are indicated in Figure 3.

4 T. Nozaki et al.

approximately delineated as simply adjacent squares. Curtiset al2 and Minagawa et al6 also analyzed the supraspinatusand infraspinatus insertions on the basis of the concept thatthe greater tubercle was simply divided into three ‘‘facets.’’In contrast to these concepts, Mochizuki et al7,8 reportedthat the tendon of the infraspinatus extends to the antero-lateral area of the highest impression of the greater tuber-cle, and the infraspinatus occupies a substantial area of thegreater tubercle, meaning that it was not limited to themiddle impression. However, it was enigmatic that theborder between the supraspinatus and the infraspinatusinsertion also served to separate the area of the highestimpression of the greater tubercle, as Mochizuki et aldescribed. In this study, the border between the highestimpression and the lateral impression corresponded to theanterior border of the insertion of the infraspinatus tendon.In other words, the lateral impression corresponded to theanterior part of the infraspinatus insertion.

On the basis of the current study, the new anatomicconcepts for the infraspinatus insertion described byMochizuki et al could be simply understood by the proof onthe bone surface of the greater tubercle.7 The anteriorborder of the infraspinatus in Mochizuki’s concept wasthought to obliquely cross the highest impression in the

classic definition of the greater tubercle. In contrast, theinfraspinatus is assumed to be inserted into not only themiddle impression but also the lateral impression in thenew definition of this study (Fig. 6).

As a clinical implication, the lateral impression could bea clue to facilitate innovation of the clinical imaging orsurgical techniques for rotator cuff tears. To date, there hasbeen no bone landmark for the identification of the originalinsertion of the infraspinatus, even though the newanatomic concepts regarding the insertion were reported byMochizuki et al.7 However, on the basis of this study, thelocation of the lateral impression could be identified bypreoperative assessment of 3D CT images of the humerus,and this will be useful for the specific diagnosis of infra-spinatus tear and for the planning of anatomic repair of theinfraspinatus tendon.

Limitations of this study should be considered. This studywas limited from the standpoint of pure anatomy, and thepathologic issues were speculative. We excluded 12 speci-mens from the CT imaging series. Most of the excludedspecimens had an unclear bone surface on 3D CT images,which was likely due to osteoporosis (Supplemental file).Although we tried to adjust the ‘‘window level and width’’ toemphasize the bone cortex and to conceal soft tissue, wecould not adjust it to achieve a clear bone surface in theosteoporotic bone. This was an inevitable technical limita-tion and might be a bias for the sample selection. In theclinical setting, cases with an unclear bone surface ordeformity on the greater tubercle will not be applicable.

There would be concern for measurement bias regardingthe order of the definition of the lateral impression and theidentification of the infraspinatus tendon. As we were waryof the harmful effects on the actual configuration of thegreater tuberosity of wet bone samples, we first definedthe lateral impression before the identification of theinfraspinatus tendon. In addition, the anatomic method ofthe infraspinatus identification by Mochizuki et al7 is afirmly established method. Hence, we believed that timing

Figure 5 The topographic relationship between the lateral impression and the anterior border of the infraspinatus tendon. (A) Cadavericsample of the humeral head with rotator cuff tendons. (B) The lateral impression could be identified in the 3D CT image by micro-CT(dotted area). (C) We dissected and identified the anterior border of the infraspinatus tendon (ISP) and placed a radiopaque marker(arrowheads) along this border. SSP, supraspinatus tendon. (D) The anterior boundary of the lateral impression corresponded to the anteriorborder of the infraspinatus tendon (arrowheads). Dotted area, the identified lateral impression; arrowheads, the anterior border of theinsertion of the infraspinatus tendon.

Figure 6 Illustrations of the superior aspect of the right humerus showing relationships between insertions of rotator cuffs and theconfiguration of the greater tubercle. (A) An illustration based on the previous concept of the insertion of the supraspinatus (pink area)and infraspinatus (purple area) and the classic description of the greater tubercle. The supraspinatus is shown to insert into the highestimpression. The infraspinatus is shown to insert into the middle impression. (B) An illustration based on Mochizuki’s concept of theinsertion and the classic description of the greater tubercle. The infraspinatus is shown to attach to the middle impression and theposterolateral part of the highest impression. (C) An illustration based on Mochizuki’s concept of the insertion and the definition ofthe lateral impression in this study. The infraspinatus is shown to insert into the middle impression and the lateral impression(asterisk).

‘‘Lateral impression’’ of the greater tubercle 5

6 T. Nozaki et al.

of the definition of the lateral impression was relativelyinsignificant.

Conclusion

The lateral impression of the greater tuberosity could beidentified in dry bone samples and on 3D micro-CTimages of the humerus. The border between the highestimpression and the lateral impression corresponded tothe anterior border of the insertion of the infraspinatustendon. We propose a new anatomic concept of thelateral impression that could enable the precise diag-nosis of and facilitate repair techniques for infraspinatustear, according to specific anatomic characteristics, byapplying 3D CT assessment.

Disclaimer

The authors, their immediate families, and any researchfoundation with which they are affiliated have not re-ceived any financial payments or other benefits from anycommercial entity related to the subject of this article.

Acknowledgments

We thank Hitoshi Shibuya, MD, PhD, for his valuableassistance in this study.

Supplementary data

Supplementary data related to this article can be found athttp://dx.doi.org/10.1016/j.jse.2014.09.038

References

1. Clemente C. Osteology, and muscles and fasciae of the upper limb. In:

Gray’s anatomy. 30th American ed. Philadelphia: Lea & Febiger; 1985.

p. 233-4.

2. Curtis AS, Burbank KM, Tierney JJ, Scheller AD, Curran AR.

The insertional footprint of the rotator cuff: an anatomic study.

Arthroscopy 2006;22:609.e1. http://dx.doi.org/10.1016/j.arthro.2006.

04.001

3. Fick R. Obere Extremit€at. In: Handbuch der Anatomie und Mechanik

der Gelenke unter Ber€ucksichtigung der bewegenden Muskeln. Jena:

Gustav Fisher; 1904. p. 141-285.

4. Gray H. Osteology. In: Pick T, Howden R, editors. Gray’s anatomy.

15th ed. Philadelphia: Lea Bros; 1901. p. 1-187.

5. Kopsch F. Die Lehre von den Muskeln. Myologia. In: Lehrbuch und

Atlas der Anatomie des Menschen Abteilung 3: Muskeln, Gef€aße.Leipzig: Georg Thieme; 1933. p. 1-153.

6. Minagawa H, Itoi E, Konno N, Kido T, Sano A, Urayama M, et al.

Humeral attachment of the supraspinatus and infraspinatus tendons: an

anatomic study. Arthroscopy 1998;14:302-6.

7. Mochizuki T, Sugaya H, Uomizu M, Maeda K, Matsuki K, Sekiya I,

et al. Humeral insertion of the supraspinatus and infraspinatus. New

anatomical findings regarding the footprint of the rotator cuff.

J Bone Joint Surg Am 2008;90:962-9. http://dx.doi.org/10.2106/jbjs.

g.00427

8. Mochizuki T, Sugaya H, Uomizu M, Maeda K, Matsuki K, Sekiya I,

et al. Humeral insertion of the supraspinatus and infraspinatus. New

anatomical findings regarding the footprint of the rotator cuff. Surgical

technique. J Bone Joint Surg Am 2009;91(Suppl 2 Pt 1):1-7. http://dx.

doi.org/10.2106/jbjs.h.01426