Humeral Insertion of the Supraspinatus and Infraspinatus: New Anatomical Findings Regarding the Footprint of the Rotator Cuff

Mochizuki, Tomoyuki MD; Sugaya, Hiroyuki MD; Uomizu, Mari MD; Maeda, Kazuhiko MD; Matsuki, Keisuke MD; Sekiya, Ichiro MD; Muneta, Takeshi MD; Akita, Keiichi MD

Journal of Bone & Joint Surgery - American Volume: 01 May 2008 - Volume 90 - Issue 5 - p 962–969
doi: 10.2106/JBJS.G.00427
Scientific Articles

Background: It is generally believed that the supraspinatus is the most commonly involved tendon in rotator cuff tears. Clinically, however, atrophy of the infraspinatus muscle is frequently observed in patients with even small to medium-size rotator cuff tears. This fact cannot be fully explained by our current understanding of the anatomical insertions of the supraspinatus and infraspinatus. The purpose of this study was to reinvestigate the humeral insertions of these tendons.

Methods: The study included 113 shoulders from sixty-four cadavers. The humeral insertion areas of the supraspinatus and infraspinatus were investigated in ninety-seven specimens. In sixteen specimens, all muscular portions of the supraspinatus and infraspinatus were removed, leaving the tendinous portions intact, in order to define the specific characteristics of the tendinous portion of the muscles. Another twenty-six shoulders were used to obtain precise measurements of the footprints of the supraspinatus and infraspinatus.

Results: The supraspinatus had a long tendinous portion in the anterior half of the muscle, which always inserted into the anteriormost area of the highest impression on the greater tuberosity and which inserted into the superiormost area of the lesser tuberosity in 21% of the specimens. The footprint of the supraspinatus was triangular in shape, with an average maximum medial-to-lateral length of 6.9 mm and an average maximum anteroposterior width of 12.6 mm. The infraspinatus had a long tendinous portion in the superior half of the muscle, which curved anteriorly and extended to the anterolateral area of the highest impression of the greater tuberosity. The footprint of the infraspinatus was trapezoidal in shape, with an average maximum medial-to-lateral length of 10.2 mm and an average maximum anteroposterior width of 32.7 mm.

Conclusions: The footprint of the supraspinatus on the greater tuberosity is much smaller than previously believed, and this area of the greater tuberosity is actually occupied by a substantial amount of the infraspinatus.

Clinical Relevance: The present study suggests that rotator cuff tears that were previously thought to involve only the supraspinatus tendon may in fact have had a substantial infraspinatus component as well.

1Section of Orthopaedic Surgery, Division of Cartilage Regeneration, Graduate School, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan

2Funabashi Orthopaedic Sports Medicine Center, 1-833 Hazama, Funabashi, Chiba 274-0822, Japan

3Unit of Clinical Anatomy, Graduate School, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan. E-mail address for K. Akita: akita.fana@tmd.ac.jp

4Department of Orthopedic Surgery, Teikyo University Chiba Medical Center, 3426-3 Anegasaki Ichihara, Chiba 299-0111, Japan

5Section of Orthopaedic Surgery, Graduate School, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan

Article Outline

Tears of the rotator cuff, which frequently occur in elderly patients as a result of age-related degenerative changes and/or injury, can lead to dysfunction of the shoulder. Therefore, identification of the precise location and extent of the tear may be important to facilitate proper surgical treatment to restore function. Tears are assessed preoperatively with the use of ultrasonography and magnetic resonance imaging and are diagnosed on the basis of intraoperative findings. In both examinations, the exposed osseous surface of the greater and lesser tuberosities is the critical finding generally used to determine which tendon is damaged.

Most anatomy textbooks state that the supraspinatus inserts into the highest impression of the greater tuberosity and the infraspinatus inserts into the middle impression of the greater tuberosity1,2. The authors of several anatomical studies have delineated the footprints of the supraspinatus and infraspinatus individually and reported numeric values for their maximum length (medial to lateral) and width (anterior to posterior)3-6. Few reports differ from the textbook descriptions. However, Clark and Harryman pointed out the difficulty of separating these tendons and delineating their footprints because of their interdigitated fibers7. Minagawa et al. reported the overlapping areas of these two tendons on the footprint and found the footprint of the supraspinatus to have a wider area than had been previously reported6.

Previous reports of the specific extents of tendon involvement in rotator cuff tears have been based on the conventional understanding of the footprint and have stated that most rotator cuff tears mainly involve the supraspinatus, and not the infraspinatus8-14. However, muscle atrophy of the infraspinatus is often observed in patients with rotator cuff tears. This atrophy has been believed to be due to a tear of the infraspinatus itself and/or traction on the suprascapular nerve following a tear of the rotator cuff15-17. In daily practice, even shoulders in which a localized supraspinatus tendon tear has been diagnosed with imaging studies can often have muscle atrophy of the infraspinatus18. For these reasons, we reinvestigated the humeral insertions of the supraspinatus and infraspinatus. The purpose of this study was to reevaluate macroscopically the humeral insertions and tendinous structures of the supraspinatus and infraspinatus in cadaver shoulders.

Back to Top | Article Outline

Materials and Methods

One hundred and twenty-eight cadaver shoulders from sixty-four Japanese donors (twenty-five males and thirty-nine females; average age at the time of death, 77.3 years old) were used in this study. All cadavers were fixed in 8% formalin and preserved in 30% ethanol. The entire scapula and the proximal one-third of the humerus and clavicle, with soft tissues, were obtained by cutting the humerus and clavicle. The skin, the subcutaneous tissues, and the deltoid muscle were removed from the shoulder. After resection of the acromion, the coracohumeral ligament and the connective tissues overlying the supraspinatus and infraspinatus were removed (Fig. 1). At this stage, fifteen shoulders with severe degenerative changes and/or torn tendons were excluded because of difficulty in precisely delineating the insertions of the rotator cuff.

In ninety-seven specimens, after observation of the macroscopic characteristics of the muscular and tendinous fibers in the myotendinous cuff, the supraspinatus and infraspinatus were peeled away from their origins on the scapula toward their insertion on the tuberosities of the humerus to delineate the humeral insertion of each tendon (Fig. 2).

In sixteen randomized specimens, the muscular portion was removed from the myotendinous units of the supraspinatus and infraspinatus and the tendons were left attached to the humerus in order to evaluate their specific characteristics. The border between the supraspinatus and infraspinatus was examined by tracing the course of the remaining tendinous portion of both muscles (Fig. 3).

We measured the maximum length (medial to lateral) and width (anterior to posterior) of the footprint of both the supraspinatus and the infraspinatus in twenty-six shoulders of another fourteen cadavers (eight males and six females; average age at the time of death, 73.3 years old). Two (of the original twenty-eight) shoulders with a torn tendon were excluded. We also measured the medial-to-lateral length of the attachment of the joint capsule at the posterior edge of the footprint of the supraspinatus.

Back to Top | Article Outline

Results

Humeral Insertions of the Supraspinatus and Infraspinatus

There was a consistent pattern in all specimens in terms of the anatomy of the supraspinatus and infraspinatus. The supraspinatus muscle originated from the supraspinatus fossa and the superior surface of the spine of the scapula, and it ran laterally. The infraspinatus muscle originated from both the infraspinatus fossa and the inferior surface of the spine of the scapula, and it ran superolaterally. The supraspinatus and infraspinatus appeared to fuse into one structure at their insertions on the humerus. However, in all specimens, after removal of the coracohumeral ligament and the loose connective tissues overlying the supraspinatus and infraspinatus near their insertions, the anterior border of the infraspinatus could be clearly traced and the border between the two muscles became more apparent (Fig. 1). The anterior margin of the infraspinatus was slightly protuberant compared with the posterior, adjacent margin of the supraspinatus. The anterior part of the infraspinatus partially covered the posterolateral part of the supraspinatus (Fig. 1, B).

Next, the infraspinatus was dissected free of the scapula and the humerus. During this process, the supraspinatus and the joint capsule underlying the infraspinatus were not damaged (Fig. 2, A). The upper surface of the greater tuberosity has been generally described as being marked by three impressions: the highest, the middle, and the lowest1. We found that the humeral insertion of the infraspinatus occupied about half of the highest impression and all of the middle impression. The anteriormost region of the humeral insertion of the infraspinatus almost reached the anterior margin of the highest impression of the greater tuberosity (Fig. 2, A). Next, we removed the supraspinatus from the scapula and the humerus. The supraspinatus could be separated from the underlying joint capsule, which was kept intact during this process. The supraspinatus was found to insert into the anteromedial area of the highest impression of the greater tuberosity. The footprint of the supraspinatus was in the shape of a right triangle, with the base lying along the articular surface. It was widest anteriorly and became narrow posteriorly (Fig. 2, B). On the greater tuberosity, the footprints of the supraspinatus and infraspinatus showed a consistent pattern in all specimens.

The supraspinatus also inserted into the lesser tuberosity (in addition to the greater tuberosity) in twenty-four (21%) of the 113 shoulders (Fig. 4). In these specimens, the anteriormost portion of the supraspinatus tendon covered the superior part of the bicipital groove. In the specimens in which the supraspinatus tendon also inserted into the lesser tuberosity, the insertion area of the infraspinatus did not extend to the lesser tuberosity.

Back to Top | Article Outline

Muscular and Tendinous Portions of the Supraspinatus and Infraspinatus

Most of the muscle fibers of the supraspinatus, especially those of its superficial layer, ran anterolaterally toward the anterior tendinous portion, while the rest of the fibers from the deep layer ran laterally toward the medial margin of the highest impression on the greater tuberosity (Figs. 1; 2, A; and 3, A). The muscle fibers of the infraspinatus, which originated from the infraspinatus fossa, ran diagonally to the greater tuberosity and the fibers from the spine ran horizontally to the greater tuberosity (Fig. 3, B). The muscle fibers from the spine attached to the dorsal surface of the fibers from the infraspinatus fossa.

We removed the muscular portion from the myotendinous unit of the supraspinatus and infraspinatus and investigated the remaining tendinous portions (Fig. 3, C and D). The supraspinatus tendon was composed of two portions: the anterior half was long and thick, and the posterior half was short and thin (Fig. 3, C). Similarly, the superior half of the infraspinatus tendon was long and thick, while the inferior half was short and thin (Fig. 3, D). The interval between the tendinous portion of the supraspinatus and that of the infraspinatus was readily observed, as was the interval between the tendinous portions of the infraspinatus and the teres minor (Fig. 3, C and D).

Back to Top | Article Outline

Measurement of the Footprint of the Supraspinatus and Infraspinatus

We measured the maximum length (medial to lateral) and the width (anterior to posterior) on both the medial and the lateral sides of the footprints of the supraspinatus and infraspinatus. The footprint of the supraspinatus had a triangular shape, which tapered away from the joint capsule. The average maximum length of the footprint (and standard deviation) was 6.9 ± 1.4 mm. The average width was 12.6 ± 2.0 mm on the medial margin and 1.3 ± 1.4 mm on the lateral margin. The footprint of the infraspinatus was shaped like a trapezoid, which was wider laterally compared with the more medial insertion along the joint capsule. The average maximum length of the infraspinatus footprint was 10.2 ± 1.6 mm. The average width was 20.2 ± 6.2 mm on the medial margin and 32.7 ± 3.4 mm on the lateral margin. The average medial-to-lateral length of the attachment of the articular capsule was 4.5 ± 0.5 mm at the posterior edge of the footprint of the supraspinatus (Fig. 5 and Table I).

Back to Top | Article Outline

Discussion

Most anatomy textbooks describe the supraspinatus, infraspinatus, and teres minor as running almost parallel to each other and as inserting separately into discrete and flattened impressions on the greater tuberosity, with the supraspinatus inserting into the highest impression, the infraspinatus inserting into the middle impression, and the teres minor inserting into the lowest impression1,2. The authors of most studies in which humeral insertions have been assessed have reported footprints similar to those described in the anatomy textbooks3-5. In the present study, however, the supraspinatus was found to be localized to the anteromedial area of the highest impression on the greater tuberosity, while the infraspinatus occupied the anterolateral area of the highest impression and all of the middle impression. We believe that differences in the dissection methods brought about these discrepancies. Clark and Harryman reported that fibers from the supraspinatus and infraspinatus fuse together and that it is difficult to separate them7. Dugas et al. selected a sharp dissection along with a discrete muscular interval between the supraspinatus and infraspinatus4. Minagawa et al. removed the overlying thick and parallel fiber bundles to expose the rotator cuff; however, they did not state how they had separated these muscles6. In the present study, we removed the overlying coracohumeral ligament and the loose connective tissues, which enabled us to detect a distinct border between the supraspinatus and infraspinatus and to separate them by precisely tracing the anterior margin of the superior tendinous portion of the infraspinatus. The overlying coracohumeral ligament comprised loose connective tissue as described by Edelson et al.19; therefore, it was easy to distinguish and remove the overlying coracohumeral ligament from the underlying supraspinatus and infraspinatus tendons.

In the additional dissection, the removal of the muscular portion from the supraspinatus and infraspinatus demonstrated that both muscles had a long, thick tendinous portion anteriorly and a short, thin tendinous portion posteriorly. These findings ensured the accuracy of our dissection method with regard to separating the infraspinatus from the supraspinatus along with the superior tendinous portion of the infraspinatus.

The footprint of the supraspinatus was found to have a triangular shape that tapered away from the joint capsule. The footprint of the infraspinatus was trapezoidal in shape and became wider laterally. The average maximum medial-to-lateral length of the footprint of the supraspinatus was 6.9 mm, and that of the infraspinatus was 10.2 mm. Both were smaller than previously reported3,4. We believe that this finding is primarily due to the fact that, in the previous reports, the footprint contained the insertion area of the joint capsule. The total medial-to-lateral length of the footprint plus the capsular attachment in our study was almost the same as that reported by Dugas et al.4. The maximum anteroposterior width of the footprint of the supraspinatus in our study averaged 12.6 mm, which was shorter than previously reported3,4,6. On the other hand, the maximum anteroposterior width of the footprint of the infraspinatus averaged 32.7 mm, which was longer than previously reported3,4,6.

In summary, the supraspinatus insertion at the level of the articular side was found to be almost as broad as previously reported3,4,6. Only laterally, where its insertion was anteriorward, did it become very narrow. Additionally, the infraspinatus insertion was located posteriorward at the level of the anatomical neck, just as previously described in anatomy textbooks1,2. The difference that we found was that the infraspinatus then curved anteriorly as it extended laterally (Fig. 6).

The frequency of rotator cuff tears and the classification of the involved tendons have been described in several anatomical8-11 and clinical studies12-14. Jerosch et al. reported, in their cadaver study, that the supraspinatus was involved in all fifty-six specimens with rotator cuff tears, whereas the infraspinatus was involved in only eleven shoulders (20%)8. Several clinical studies have demonstrated the frequency of supraspinatus involvement in rotator cuff tears to range from 86% to 100% and the frequency of infraspinatus involvement to range from 14% to 48%12-14. However, in light of the finding in the present study that the infraspinatus had a substantially wider footprint than had been previously believed, the infraspinatus tendon may be involved in a much higher proportion of rotator cuff tears.

Muscle atrophy of the infraspinatus is sometimes observed in patients with an apparently isolated supraspinatus tear; this atrophy has been investigated in several studies15-17. In a cadaver study, Albritton et al. observed that retraction of the supraspinatus muscle following rotator cuff tears increased tension on the suprascapular nerve15. In clinical studies, Vad et al. reported that two of seven patients with an abnormal electromyographic finding following a full-thickness rotator cuff tear were diagnosed as having a suprascapular neuropathy16 and Warner reported that seven of twenty-six patients with a massive rotator cuff tear and fatty replacement of the muscles showed dysfunction of the suprascapular nerve17. The authors of these two clinical studies proposed that suprascapular nerve damage could result in the infraspinatus atrophy in association with some rotator cuff tears, although most of their patients with muscle atrophy did not show dysfunction of the suprascapular nerve16,17. Our anatomical findings suggest that there may be a higher frequency of involvement of the infraspinatus in rotator cuff tears, which can explain the development of infraspinatus atrophy without suprascapular nerve damage.

The supraspinatus has traditionally been considered to be an important abductor among the rotator cuff muscles20,21. However, several researchers have reported that the infraspinatus contributes as much to abduction as does the supraspinatus22-25. Although the analysis of the footprints in our cadaver study did not allow us to draw conclusions regarding the function of the rotator cuff, our findings that the infraspinatus occupied about half of the highest impression on the greater tuberosity, which has been believed to be the footprint of the supraspinatus, support the concept that the infraspinatus may contribute more to shoulder abduction than previously believed.

Recently, on the basis of both biomechanical26,27 and clinical28,29 studies, the concept of “footprint reconstruction” with use of a double row of suture anchors has been suggested to be superior to single-row fixation in rotator cuff repair. On the basis of the anatomical findings of our study, it appears that rotator cuff reconstruction to the anatomical footprint requires repair not only of the supraspinatus but also of the infraspinatus to the wider footprint on the highest impression on the greater tuberosity. Although it is difficult to distinguish the infraspinatus from the supraspinatus during surgery, the long, thick tendinous portions of both the supraspinatus and the infraspinatus may help to identify each muscle. Additionally, repair of these thick tendinous portions to the original insertions may be important for the full restoration of shoulder function.

The anatomical findings in the present study suggest that surgeons need to have increased awareness of pathological conditions of the infraspinatus tendon, especially when delamination is observed30,31. It may be important for surgeons to incorporate these new anatomical findings in order to properly restore the geography of the footprint of the torn anterior rotator cuff.

NOTE: The authors thank Akimoto Nimura, MD, for measurements of the footprint; Joji Moriishi, MD, PhD, for proffering advice regarding our studies; and Kumiko Yamaguchi, MD, PhD, for technical assistance.

Disclosure: The authors did not receive any outside funding or grants in support of their research for or preparation of this work. Neither they nor a member of their immediate families received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, division, center, clinical practice, or other charitable or nonprofit organization with which the authors, or a member of their immediate families, are affiliated or associated.

Investigation performed at the Unit of Clinical Anatomy, Graduate School, Tokyo Medical and Dental University, Tokyo, Japan

1. Clemente CD, editor. Gray's anatomy of the human body. 30th ed. Philadelphia: Lea & Febiger; 1985. Osteology, and muscles and fasciae of the upper limb. p 233-4.
2. Johnson D, Ellis H, editors. Pectoral girdle, shoulder region, and axilla. In: Standring S, Ellis H, Healy JC, Johnson D, Williams A, Collins P, Wigley C, editors. Gray's anatomy: the anatomical basis of clinical practice. 39th ed. Edinburgh: Elsevier Churchill Livingstone; 2005. p 817-49.
3. Curtis AS, Burbank KM, Tierney JJ, Scheller AD, Curran AR. The insertional footprint of the rotator cuff: an anatomic study. Arthroscopy. 2006;22:603-9.
4. Dugas JR, Campbell DA, Warren RF, Robie BH, Millett PJ. Anatomy and dimensions of rotator cuff insertions. J Shoulder Elbow Surg. 2002;11:498-503.
5. Ruotolo C, Fow JE, Nottage WM. The supraspinatus footprint: an anatomic study of the supraspinatus insertion. Arthroscopy. 2004;20:246-9.
6. Minagawa H, Itoi E, Konno N, Kido T, Sano A, Urayama M, Sato K. Humeral attachment of the supraspinatus and infraspinatus tendons: an anatomic study. Arthroscopy. 1998;14:302-6.
7. Clark JM, Harryman DT 2nd. Tendons, ligaments, and capsule of the rotator cuff. Gross and microscopic anatomy. J Bone Joint Surg Am. 1992;74:713-25.
8. Jerosch J, Müller T, Castro WH. The incidence of rotator cuff rupture. An anatomic study. Acta Orthop Belg. 1991;57:124-9.
9. Lehman C, Cuomo F, Kummer FJ, Zuckerman JD. The incidence of full thickness rotator cuff tears in a large cadaveric population. Bull Hosp Jt Dis. 1995;54:30-1.
10. DePalma AF. Surgery of the shoulder. 3rd ed. Philadelphia: JB Lippincott; 1983. p 220-3.
11. Ogata S, Uhthoff HK. The early development and ossification of the human clavicle—an embryologic study. Acta Orthop Scand. 1990;61:330-4.
12. Harryman DT 2nd, Mack LA, Wang KY, Jackins SE, Richardson ML, Matsen FA 3rd. Repairs of the rotator cuff. Correlation of functional results with integrity of the cuff. J Bone Joint Surg Am. 1991;73:982-9.
13. Blevins FT, Warren RF, Cavo C, Altchek DW, Dines D, Palletta G, Wickiewicz TL. Arthroscopic assisted rotator cuff repair: results using a mini-open deltoid splitting approach. Arthroscopy. 1996;12:50-9.
14. Gartsman GM, Khan M, Hammerman SM. Arthroscopic repair of full-thickness tears of the rotator cuff. J Bone Joint Surg Am. 1998;80:832-40.
15. Albritton MJ, Graham RD, Richards RS 2nd, Basamania CJ. An anatomic study of the effects on the suprascapular nerve due to retraction of the supraspinatus muscle after a rotator cuff tear. J Shoulder Elbow Surg. 2003;12:497-500.
16. Vad VB, Southern D, Warren RF, Altchek DW, Dines D. Prevalence of peripheral neurologic injuries in rotator cuff tears with atrophy. J Shoulder Elbow Surg. 2003;12:333-6.
17. Warner JP. Management of massive “irreparable” rotator cuff tears. Read at the Annual Meeting of the American Academy of Orthopaedic Surgeons; 2007 Feb 14-18; San Diego, CA.
18. Goutallier D, Postel JM, Bernageau J, Lavau L, Voisin MC. Fatty muscle degeneration in cuff ruptures. Pre- and postoperative evaluation by CT scan. Clin Orthop Relat Res. 1994;304:78-83.
19. Edelson JG, Taitz C, Grishkan A. The coracohumeral ligament. Anatomy of a substantial but neglected structure. J Bone Joint Surg Br. 1991;73:150-3.
20. Perry J. Muscle control of the shoulder. In: Rowe CR, editor. The shoulder. New York: Churchill Livingstone; 1988. New York. p 17-34.
21. Inman VT, Saunders JB, Abbott LC. Observations on the function of the shoulder joint. J Bone Joint Surg Am. 1944;26:1-30.
22. Kronberg M, Németh G, Broström LA. Muscle activity and coordination in the normal shoulder. An electromyographic study. Clin Orthop Relat Res. 1990;257:76-85.
23. Itoi E, Minagawa H, Sato T, Tabata S. Isokinetic strength after tears of the supraspinatus tendon. J Bone Joint Surg Br. 1997;79:77-82.
24. Otis JC, Jiang CC, Wickiewicz TL, Peterson MG, Warren RF, Santner TJ. Changes in the moment arms of the rotator cuff and deltoid muscles with abduction and rotation. J Bone Joint Surg Am. 1994;76:667-76.
25. Sharkey NA, Marder RA, Hanson PB. The entire rotator cuff contributes to elevation of the arm. J Orthop Res. 1994;12:699-708.
26. Waltrip RL, Zheng N, Dugas JR, Andrews JR. Rotator cuff repair. A biomechanical comparison of three techniques. Am J Sports Med. 2003;31:493-7.
27. Ma CB, Comerford L, Wilson J, Puttlitz CM. Biomechanical evaluation of arthroscopic rotator cuff repairs: double-row compared with single-row fixation. J Bone Joint Surg Am. 2006;88:403-10.
28. Sugaya H, Maeda K, Matsuki K, Moriishi J. Repair integrity and function outcome after arthroscopic double-row rotator cuff repair. A prospective outcome study. J Bone Joint Surg Am. 2007;89:953-60.
29. Lafosse L, Brozska R, Toussaint B, Gobezie R. The outcome and structural integrity of arthroscopic rotator cuff repair with use of the double-row suture anchor technique. J Bone Joint Surg Am. 2007;89:1533-41.
30. Sugaya H, Maeda K, Matsuki K, Moriishi J. Functional and structural outcome after arthroscopic full-thickness rotator cuff repair: single-row versus dual-row fixation. Arthroscopy. 2005;21:1307-16.
31. Zilber S, Carillon Y, Lapner PC, Walch G, Nové-Josserand L. Infraspinatus delamination does not affect supraspinatus tear repair. Clin Orthop Relat Res. 2007;458:63-9.
Copyright 2008 by The Journal of Bone and Joint Surgery, Incorporated