Skip to main content
NCBI home page
Clinical Orthopaedics and Related Research logoLink to Clinical Orthopaedics and Related Research
. 2021 Apr 30;479(10):2323–2331. doi: 10.1097/CORR.0000000000001779

What Are Practical Surgical Anatomic Landmarks and Distances from Relevant Neurologic Landmarks in Cadavers for the Posterior Approach in Shoulder Arthroplasty?

Michael S Bahk 1,✉, R Michael Greiwe 1
PMCID: PMC8445576  PMID: 33938480

Abstract

Background

Traditional total shoulder arthroplasty is performed through the deltopectoral approach and includes subscapularis release and repair. Subscapularis nonhealing or dysfunction may leave patients with persistent pain, impairment, and instability. Alternative approaches that spare the subscapularis include rotator interval and posterior shoulder approaches; however, to our knowledge, a cadaveric study describing pertinent surgical anatomy for a posterior shoulder approach regarding shoulder arthroplasty has not been performed.

Questions/purposes

(1) What are the distances from important neurologic structures of the shoulder for arthroplasty through a posterior approach? (2) What surgical landmarks can help identify the internervous interval between the infraspinatus and teres minor?

Methods

Twelve hemitorso cadaver specimens with intact rotator cuffs were dissected to study posterior shoulder anatomy regarding posterior shoulder arthroplasty. The median (range) age of the specimens was 79 years (55 to 92). Six of the 12 specimens were right-hand dominant, and 10 specimens were male. Cadaver height was a median 171 cm (155 to 191) and weight was a median of 68 kg (59 to 125). A posterior deltoid split and internervous approach between the infraspinatus and teres minor were used. A posterior T capsulotomy was performed. The distances to important neurologic structures were measured with an electronic caliper and provided in median (range) distances in millimeters. Although not as meaningful as distance ratios accounting for a specimen’s body size, neurologic distances in millimeters are surgically practical and provide intraoperative usefulness. Surgical landmarks that can help identify the infraspinatus and teres minor plane were noted. Practical visual and tactile cues between the infraspinatus and teres minor were identified. Posterior rotator cuff tendon morphologies and widths were recorded.

Results

The closest important neurologic structure was the axillary nerve, measuring a median (range) 17 mm (9 to 19) from the inferior glenoid rim while the infraspinatus branch of the suprascapular nerve measured 21 mm (15 to 36) from the posterior glenoid rim. The axillary nerve measured 84 mm (70 to 97) from the posterior tip of the acromion in the deltoid split. Three surgical landmarks were helpful for identifying the plane between the infraspinatus and teres minor in all 12 specimens: (1) identifying the triangular teres minor tendon insertion, (2) medial palpation identifying the low point between the prominent muscle bellies of the infraspinatus and teres minor, and (3) identifying the distinct and prominent teres minor tubercle, which is well localized and palpable.

Conclusion

A major benefit of the posterior approach for shoulder arthroplasty is subscapularis preservation. Multiple practical surgical cues are consistently present and can help identify the infraspinatus and teres minor interval. We did not find the presence of fat stripes to be helpful. The suprascapular nerve is in proximity to posterior surgical dissection and differs from the deltopectoral approach. This is an important distinction from an anterior approach and requires care with dissection. Future studies are necessary to assess iatrogenic risk to the posterior rotator cuff and external rotation strength. This may entail intraoperative nerve conduction studies of the posterior rotator cuff and clinical studies assessing external rotation strength.

Clinical Relevance

Studying posterior shoulder anatomy is an initial first step to assessing the feasibility of the posterior approach for anatomic shoulder arthroplasty. Additional studies assessing the degree of glenohumeral exposure and possible iatrogenic posterior rotator cuff injury are necessary. Because of the proximity of neurologic structures, it is recommended that surgeons not perform this technique until sufficient evidence indicates that it is equivalent or superior to standard anterior approach total shoulder arthroplasty. After such evidence is available, proper training will be necessary to ensure safe use of the posterior shoulder approach.

Introduction

Anatomic total shoulder arthroplasty is an effective means of improving pain relief and function for patients with glenohumeral osteoarthritis. Traditional total shoulder arthroplasty is performed with the deltopectoral approach and involves subscapularis release and repair. If the subscapularis does not heal, reduced patient outcomes, increased pain, and persistent instability requiring revision surgery have been reported [8, 19, 20]. Clinical subscapularis dysfunction has been reported in up to 67% of patients after surgery [19]. The subscapularis has been reported to not heal in up to 47% of patients [9, 14, 24]. The inability to accommodate sling restrictions postoperatively for subscapularis healing is also a relative contraindication to anatomic shoulder arthroplasty. Therefore, surgeons have sought subscapularis-sparing approaches. These alternative approaches include a rotator interval approach [7, 15, 16, 21, 23] and a newer posterior shoulder approach [10, 11, 12]. The rotator interval approach can be challenging because of limited visualization. The rotator interval approach also may be associated with increased difficulty visualizing retroverted glenoids, retention of residual inferior humeral osteophytes, humeral head under-sizing, and differences in component position [7, 15, 21, 23]. It is also reported to have a 15.2% proportion of intraoperative approach abandonment and may result in deltoid release, anterior supraspinatus release, or muscular subscapularis release [7, 15, 16, 21, 23]. Alternatively, the posterior approach may offer a viable subscapularis-sparing approach to shoulder arthroplasty. The posterior approach has been reported for shoulder arthroplasty and uses a posterior deltoid split and internervous access between the infraspinatus and teres minor [10, 11, 12] (Fig. 1A-F). The deltoid, muscular subscapularis, or posterior rotator cuff is not released. It also offers more glenohumeral exposure than a rotator interval approach and requires less retraction force for exposure [2].

Fig. 1.

Fig. 1

Open in a new tab

A-F Preoperatively, (A) AP, (B) axillary, and (C) scapular Y radiographs were taken of a shoulder with glenohumeral osteoarthritis. After anatomic shoulder arthroplasty was performed through the posterior approach, (D) AP, (E) axillary, and (F) scapular Y radiographs were taken. The deltoid was split posteriorly and the glenohumeral joint was accessed between the infraspinatus and teres minor.

However, little is known about this approach. The posterior approach for the shoulder is not routinely performed even by experienced shoulder surgeons and especially in the context of shoulder arthroplasty. Identification of the internervous plane between the infraspinatus and teres minor can be challenging due to the confluent nature of the rotator cuff and lack of familiarity. Practical guidelines for safe surgical dissection and retractor placement are unknown with the posterior approach. Although cadaveric dissection and safety evaluation were performed in more than 30 cadavers to validate the posterior shoulder approach before its use (RMG, personal communication), scientific quantification and clarification of these landmarks is necessary to educate surgeons on this technique.

Therefore, we asked: (1) What are the distances from important neurologic structures of the shoulder for arthroplasty through a posterior approach? (2) What surgical landmarks can help identify the internervous interval between the infraspinatus and teres minor?

Materials and Methods

Twelve hemitorso cadaver specimens with intact rotator cuffs were dissected by two surgeons (MSB, RMG) [22]. Age, sex, right- or left-handedness, height, and weight were recorded. Each surgeon dissected six specimens.

Description of Cadaver Specimens

The median (range) age of the specimens was 79 years (55 to 92). Six of 12 specimens were right-hand dominant, and 10 were male. The cadaver height was a median 171 cm (155 to 191), and the median weight was 68 kg (59 to 125).

Surgical Approach

A sequential dissection was performed as if a posterior shoulder arthroplasty were being performed. Specimens were placed in the lateral decubitus position and held in place with peg boards. A mayo stand and assistant were used to help position the dissected arm. A ruler was used to create a 10-cm skin incision from the posterolateral corner of the acromion down the axis of the arm. Skin flaps were raised until the posterior border of the deltoid was identified. A finger was placed deep in the deltoid to identify a raphe junction of the posterior deltoid and middle deltoid by finger palpation. The deltoid was then split, ideally between the posterior and middle deltoid heads. The deltoid was split 7 to 10 cm, and the presence of neurovascular structures (axillary nerve with its associated vasculature) in this muscle split was noted. The location of the neurovascular structures was measured from the posterolateral corner of the acromion with a digital caliper (Mitutoyo).

The posterior fascia overlying the rotator cuff’s musculature was incised, and the junction of the infraspinatus and teres minor was identified. We recorded the ability to identify the infraspinatus and teres minor interval by visualization and palpation. The presence of fat stripes as an aid to identification was also noted. We evaluated the teres minor’s insertion on the teres tubercule. The teres minor tendon triangular insertion shape was noted. The ability to palpate the prominent teres tubercle was also recorded. Surgically, we felt that palpation was an important aid in identifying the infraspinatus and teres minor interval. With internal rotation, the low point between the muscle bellies of the infraspinatus and teres minor was also palpated and recorded to aid in identifying the infraspinatus and teres minor interval. The width of the infraspinatus and teres minor at the level of the glenohumeral joint was measured. The plane between the posterior rotator cuff and capsule was then dissected. A lateral T-shaped capsulotomy was performed and carried as far inferior and superior as possible.

After visualizing the glenohumeral joint, we made a free-hand, in situ humeral head cut (Fig. 2A-B). Curved retractors were used to protect the superior rotator cuff and inferior glenohumeral joint. Each surgeon (MSB, RMG) performed an inferior capsulotomy or capsulectomy as necessary to expose the axillary nerve. For this study only, dissection of the axillary nerve and its branches was performed for measurement. The presence or absence of a space between the teres minor and subscapularis was recorded. This space was called the naked window because it exposed the axillary nerve without protecting the rotator cuff. The location of the nerves in the naked window and the distance from the glenoid rim were measured with a digital caliper and provided in median (range) distances in millimeters.

Fig. 2.

Fig. 2

Open in a new tab

A-B (A) After humeral head exposure was obtained, (B) a free-hand humeral head cut was performed.

In addition to studying the axillary nerve, we also identified and measured the location of the suprascapular nerve. The distance between the suprascapular nerve and glenoid rim was measured at the level of the infraspinatus and teres minor split. We also measured the distance between the glenoid rim and suprascapular nerve at the spinoglenoid notch. Further, we recorded the location of the spinoglenoid notch in relation to the glenoid clock face.

Measurements

Our primary study goal was to measure the distances of important neurologic shoulder structures for arthroplasty through a posterior approach. Although not as meaningful as distance ratios, which account for a person’s size, neurologic distances are the most common and practical way to provide information about neurologic structure location surgically and they are the most common form of reporting neurologic structure location information [1, 4, 17, 18].

Our secondary goal was to identify practical surgical aids for recognizing the infraspinatus and teres minor interval. These included visual and tactile cues. The presence of these cues and how often they helped identify the infraspinatus and teres minor interval were recorded. Cues that have historically been reported to help identify the infraspinatus and teres minor interval were also evaluated.

Ethical Approval

Ethical approval was not sought given the cadaveric study material of this study.

Results

Distances to Important Neurologic Structures

The closest important neurologic structure was the axillary nerve, which measured a median (range) 17 mm (9 to 19) from the glenoid rim and occurred laterally in an exposed area between the infraspinatus and teres minor. During capsulotomy or capsulectomy, the teres minor and subscapularis provided a muscular floor for the shoulder inferior to the capsule at the inferior glenoid. In 11 of 12 specimens, there was a space without muscular protection, that is, a naked window, between the teres minor and subscapularis exposing the nerve (Fig. 3A-E). The naked window was triangle-shaped, with its base at the lateral humerus and apex at the glenoid. One specimen had a confluent teres minor and subscapularis junction without a naked window. The triceps originated proximally between the teres minor and subscapularis and marked the medial aspect of the naked window. The axillary nerve was dissected along with its branches in this naked window space. If no window was present, the tendons were spread for axillary nerve dissection. The axillary nerve was in the lateral half of the window in 12 of 12 specimens and in the lateral third of the window in 7 of 12 specimens. The axillary nerve or its closest branch to the glenoid rim was measured. Notably, if a teres minor branch was detected, it branched medially while anterior and posterior branches of the axillary nerve traveled laterally.

Fig. 3.

Fig. 3

Open in a new tab

A-E (A) A triangular window existed in the inferior glenohumeral joint where the axillary nerve was not protected by a muscle or tendon exposing the joint to the axillary nerve. (B, C, D) This naked window is marked (yellow triangle or circle) by the triceps medially, the teres minor posteriorly, and the subscapularis anteriorly. (E) Axillary nerve trifurcation (arrow) appeared to start in the naked window and continued posteriorly.

The next closest was the suprascapular nerve, measuring a median (range) 17 mm (13 to 21) from the glenoid rim in the superior-posterior quadrant. The distance from the glenoid rim to the spinoglenoid notch was also identified. The location of the spinoglenoid notch and therefore the nerve varied from the 8 o’clock to 11:30 o’clock position for a right shoulder (Fig. 4).

Fig. 4.

Fig. 4

Open in a new tab

The suprascapular nerve was a median (range) 17 mm (13 to 21) from the glenoid rim in the posterior superior quadrant. This danger zone (red) was from 8:00 o'clock to 11:30 o'clock for a right shoulder and was where the suprascapular nerve was closest to the glenoid. It was also the most susceptible to retractor damage.

The third-closest nerve was the infraspinatus branch of the suprascapular nerve at the level of the infraspinatus and teres minor interval. A branch of the infraspinatus branch of the suprascapular nerve was a median (range) 21 mm (15 to 36) from the glenoid rim at the infraspinatus and teres minor split. The closest nerve (15 mm) was in a small female specimen (153 cm high), and this specimen often provided the smallest neurovascular distance data. Neurologic distances were not normalized against specimen size.

In the posterior deltoid split, the axillary nerve measured a median (range) 84 mm (70 to 97) from the posterolateral corner of the acromion. A traversing vein was identified in 12 of 12 of the specimens in the muscle of the deltoid split. This vein was proximal and superficial to the axillary nerve and heralded the axillary nerve’s presence. As such, it was dubbed the heralding vein (Fig. 5A-B). The heralding vein was a median of 72 mm (63 to 85) distal from the posterolateral corner of the acromion.

Fig. 5.

Fig. 5

Open in a new tab

A-B (A) The presence of a traversing vein (arrow) was a median (range) 72 mm (63 to 85) from the posterolateral acromion. This vein was consistently present (in 12 of 12 specimens) and heralded the axillary nerve, which lay distal and deep to the presence of this “heralding vein.” (B) The axillary nerve (marked at the tip of dissection scissors) was at a median 84 mm (70 to 97) from the posterolateral acromion and was distal and deep to the heralding vein (arrow).

Landmark Identifying the Plane Between the Infraspinatus and Teres Minor

A surgical landmark to identify the plane between the infraspinatus and teres minor was visualization of the triangular teres minor tendinous insertion, seen in 12 of 12 of the specimens. The superior margin of the teres minor tendon could have a white triangular shape with a distinct corner that could help identify the infraspinatus and teres minor interval (Fig. 6A-B). In other specimens, the triangular tendon was more central, with superior muscle fibers marking the superior border of the teres minor (Fig. 6C-D).

Fig. 6.

Fig. 6

Open in a new tab

A-D (A) The upper border of the teres minor often appeared as a white triangular tendon. (B) The triangle could be superior-based and could help define the superior margin of the teres minor. (C-D) At other times, it was more central, with superior muscle fibers.

A second surgical landmark to help identify the infraspinatus and teres minor interval was medial palpation, found in 12 of 12 specimens. With internal rotation, the posterior rotator cuff was placed in tension and the infraspinatus and teres minor interval could be felt as a valley between the peaks of the prominent muscle bellies of the infraspinatus and teres minor (Fig. 7). We feel that palpation is an important aid in confirming the correct location of the infraspinatus and teres minor interval intraoperatively, and palpation is almost always performed by the surgeon to confirm the location of the infraspinatus and teres minor interval.

Fig. 7.

Fig. 7

Open in a new tab

The interval between the infraspinatus and teres minor could be palpated medially as a valley (arrow) between the two prominent muscle bellies.

Third, identifying the distinct and prominent teres minor tubercle helped recognize the infraspinatus and teres minor interval in 12 of 12 specimens. The teres minor tubercle was a discrete prominence that was readily palpable and visible (Fig. 8A-B). This tubercle may be more aptly named a mesa or volcano because it often appears to have a flat top and often a central dimple or hollowing. The infraspinatus tendon inserts onto a more gradual prominence of the greater tuberosity, an infraspinatus ridge. Identifying the teres minor tubercle in relation to the infraspinatus ridges helped localize the infraspinatus and teres minor interval in between, or the valley between two peaks (Fig. 9A-B). We did not feel that the presence of a fat stripe was a reliable aid to identifying the infraspinatus and teres minor interval because of a lack of any consistent pattern or presence.

Fig. 8.

Fig. 8

Open in a new tab

A-B The teres minor and infraspinatus insertions were palpable prominences. (A) The teres minor tubercle (arrow) was typically shaped like a mesa (a discrete prominence with a flat top), (B) while the infraspinatus inserted onto a posterolateral facet with its lateral border marked by a palpable ridge.

Fig. 9.

Fig. 9

Open in a new tab

A-B (A) A visual valley marked the interval between the teres minor and infraspinatus and (B) could be seen as the midpoint between the two peaks of the teres minor tubercle and infraspinatus insertion.

At the level of the glenohumeral joint, the median (range) infraspinatus width was 27 mm (20 to 36) in 10 of 12 specimens, while that of the teres minor was 38 mm (23 to 50) in all 12 specimens. The mean teres minor was wider than the infraspinatus at the level of the glenohumeral joint. Appreciating the relative sizes of the posterior rotator cuff tendons and muscles also aided in identifying the infraspinatus and teres minor interval.

Discussion

Subscapularis-sparing approaches for anatomic shoulder arthroplasty are appealing and include the rotator interval approach and posterior approach. The posterior approach offers better visualization of the glenohumeral joint with less rotator cuff retraction forces than the rotator interval approach but is the less well known of the two approaches [2, 7, 12, 15, 16, 21, 23]. Identification of the posterior internervous plane between the infraspinatus and teres minor can be challenging due to the confluent nature of the rotator cuff and lack of familiarity. Practical guidelines for safe surgical dissection and retractor placement are unknown with the posterior approach. The information from this study provides initial surgical guidelines for dissection and retractor placement. The deltoid split is limited due to the axillary nerve but sufficient for glenohumeral exposure. An inferior glenohumeral capsulotomy or capsulectomy can be posteriorly preformed similar to traditional deltopectoral approaches by staying close to the inferior glenoid rim. Surgical dissection or retractor placement should not pass the glenoid rim posteriorly more than 1 cm due to the suprascapular nerve. The internervous plane between the infraspinatus and teres minor can be reliably identified by three key landmarks: palpating the teres tubercle, identifying the triangular teres minor tendon, and medially palpating the low spot between the infraspinatus and teres minor.

Limitations

One limitation of the study is that it is a cadaveric study of only 12 fresh hemitorso specimens with intact rotator cuffs. A larger number of specimens would yield more measurements. However, a similar study on the posterior approach also used 12 specimens [2]. As a cadaver study documenting anatomic features, we felt a precedent of 12 specimens was an acceptable first step to studying posterior anatomy. Our next step is a larger study with more specimens to confirm and understand the range of values better. Our specimens were also predominantly male (10 of 12 specimens). Specimen sex was not controlled. We do not believe sex-based differences of the specimens would specifically affect structural shoulder anatomy with the infraspinatus and teres minor interval question in mind. However, more cadaveric dissections in females are required to ensure distance measurement apply equally to females and males. Neurologic distances are not normalized against specimen size, and smaller patients will have smaller neurologic distances. Therefore, the range of distance measurements are presented, and we recommend surgeons consider using the lower range of numbers depending on the patient’s size with an additional margin of caution. For example, the infraspinatus branch of the suprascapular nerve measured a median (range) of 21 mm (15 to 36) from the glenoid rim. Therefore, we do not recommend placing a retractor past 1 cm from the glenoid margin. Two surgeons performed the dissection, which may have affected the reproducibility of the dissection procedure or measurements made. However, given the novelty of the approach, we believed that having multiple surgeons may also have increased our ability to identify important novel surgical cues or landmarks for the dissection.

Distances to Important Neurologic Structures

The location of neurologic structures in the posterior approach allows the axillary and suprascapular nerves to be managed similarly to the deltopectoral approach but requires additional care for the infraspinatus branch of the suprascapular nerve and the posterior axillary nerve. Inferior capsular release can be performed in a similar fashion with regard to the axillary nerve. The closest nerve to the field of dissection was the axillary nerve, which was a median (range) of 17 mm (9 to 19) from the inferior glenoid rim. This value is a similar distance to the mean distance of 11 to 14 mm when dissection is performed using the deltopectoral approach [1, 17, 18]. Therefore, inferior capsular dissection may be performed in a similar fashion with an attempt to stay as close to the glenoid as possible inferiorly [17, 18]. Potentially, there may be less risk to the axillary nerve with the posterior approach because the surgeon releases the inferior capsule from a posterior to anterior direction as the axillary nerve travels medial to lateral and is further away from the glenoid rim posterior than anteriorly. Similar to the deltopectoral approach, we do not recommend long, spiked Bankart or Hohmann retractors in the posterior superior glenoid quadrant out of respect for the suprascapular nerve, which was 17 mm (13 to 21) from the glenoid rim. This value is similar to a previous reported value of a mean distance of 18 mm (14 to 25) from the glenoid rim to the suprascapular nerve at the scapular spine [4]. Unlike the deltopectoral approach, the infraspinatus branch of the suprascapular nerve and the posterior axillary nerve are more present in the posterior surgical field and potentially at higher risk of injury. The infraspinatus branch of the suprascapular nerve is 21 mm (15 to 36) from the infraspinatus and teres minor interval at the posterior glenoid rim. We do not recommend posterior dissection greater than 1 cm from the glenoid rim. Glenoid exposure past 1 cm does not appear to be necessary for this approach for arthroplasty, but the imminent presence of the suprascapular nerve is an important and new distinction from the deltopectoral approach. Although the infraspinatus and teres minor are not divided, intraoperative retraction of the posterior external rotators may lead to short-term and possible long-term external rotation weakness. We believe that the possibility of long-term external rotation weakness is low, but additional studies are required to evaluate this possibility. As performed similar to the deltopectoral approach, intraoperative electromyographic and nerve conduction studies are important to evaluating possible injury to the posterior rotator cuff. Finally, in the posterior deltoid split, the axillary nerve measured 84 mm (70 to 97) from the posterior tip of the acromion. Its presence was heralded by a traversing vein proximal and superficial. This heralding vein and was the effective stopping landmark for dissection of the deltoid split. Dissection past the heralding vein is not necessary but limits the posterior approach; the posterior approach is not extensile. These distance results were obtained with specimens in the lateral decubitus position and may differ if measurements are made in the beach chair position. Because of prior studies demonstrating landmarks and nerve distances in cadavers [1, 4, 17, 18] and due to the relatively small distances in this study, the authors felt that ratios relative to cadaver size would not be as useful in providing practical surgical data as distance measurements with ranges.

Landmark Identifying the Plane Between the Infraspinatus and Teres Minor

The interval between the infraspinatus and teres minor could be identified by three consistent landmarks. The first was visualization of the triangular teres minor tendon insertion. Traditionally, the teres minor is pictured as rectangular or square [3, 6, 13]. However, the morphology of the teres minor tendon is not a major focus in many studies [3, 6]. Understanding that the teres minor tendon is actually triangular will help identify the infraspinatus and teres minor interval. The triangular teres minor tendon has two variations. In one variation, the triangular tendon is more superior and abuts the infraspinatus and teres minor interval. Alternatively, the triangular tendon is more central in the teres minor insertion, and red muscular slips form the superior margin of the infraspinatus and teres minor interval. In the second variation, medial palpation helps identify the infraspinatus and teres minor interval. With internal rotation, the infraspinatus and teres minor muscles are made more prominent medially. Running a finger in a superior to inferior direction allows identification of a low spot between the muscle bellies that corresponds to the infraspinatus and teres minor interval. Surgically, medial confirmation with palpation is performed as routine secondary confirmation of the proper infraspinatus and teres minor interval location. Third, the teres minor inserts onto a discrete and prominent tubercle. Palpating or visualizing this landmark in reference to the infraspinatus insertion allows localization of the infraspinatus and teres minor interval. The teres minor tubercle is discrete, well located, and prominent to palpation. It has also not been well described [3-6]. The infraspinatus inserts onto a ridge of the posterolateral corner of the greater tuberosity and is not as discrete. Visually, the infraspinatus and teres minor interval appears as a valley between two peaks. The infraspinatus and teres minor interval may also feel more superior than expected, given the larger width of the teres minor compared with that of the infraspinatus. The teres minor is unipennate and the infraspinatus is multipennate, and the direction of the muscle fibers or muscle bellies may be an additional clue to identifying the infraspinatus and teres minor interval [3]. We did not find fat stripes to be helpful for identifying the infraspinatus and teres minor interval. As a final examination, the posterior joint line can be palpated to help confirm proper positioning.

Conclusion

The posterior approach is a subscapularis-sparing approach that allows internervous access to the glenohumeral joint between the infraspinatus and teres minor (Fig. 10A-B). This plane can be reliably identified with several novel key practical surgical cues. However, unlike the deltopectoral approach, the infraspinatus branch of the suprascapular nerve is present within the surgical field and requires additional care. Additional studies are necessary to assess the utility of this approach and its potential pitfalls, which include possible short-term or long-term external rotator weakness, suprascapular nerve injury, posterior instability, lack of humeral intramedullary shaft access, lack of traditional orthogonal exposure to the humeral head, and performing arthroplasty in the lateral decubitus position. Because the posterior approach does not dislocate the glenohumeral joint, it provides less direct humeral head and humeral shaft access. The lack of less humeral shaft access can be managed with stemless humeral head implants [10, 11, 12]. However, stemless fixation requires good quality bone in the proximal humerus. Both authors have used stemmed humeral components, but this is more difficult in the posterior approach. Avoiding stemmed implants obviates the possibility of humeral shaft fractures and medullar fixation complications. However, there is also less visualization or orthogonal humeral head exposure with the posterior approach. A study with a similar posterior approach reported that orthogonal access to the center of the humeral head was allowed in all specimens, but access to the entire humeral head cut surface was present in 80% of the cadavers [12]. Clinically, orthogonal humeral head access may be additionally reduced in patients with an elevated BMI or patients with substantial stiffness or deformity. This may require nonorthogonal humeral head preparation or changes in implant fixation design. Robot-assisted or computer navigation may aid nonorthogonal humeral head bone preparation. However, posterior glenoid visualization is improved compared with the deltopectoral approach. Correction of a posterior glenoid deformity may be easier with the posterior approach. Performing arthroplasty in the lateral decubitus position also provides new challenges by reorienting the glenoid axis from the traditional beach chair or supine position. An assessment of humeral version also requires new landmarks. Clinically, humeral version is assessed with the forearm externally rotated 20° to 40° to the ground level in the lateral decubitus position to perform an anatomic humeral cut. These orientation challenges may be improved with navigation or robot-assisted technology. Given the many unknowns of this approach, we do not recommend it for shoulder arthroplasty at this time. It should only be performed in the setting of prospective research trials and after sufficient training. The posterior approach offers subscapularis preservation, but additional studies are necessary regarding efficacy and safety.

Fig. 10.

Fig. 10

Open in a new tab

A-B (A) Humeral and (B) glenoid exposure were obtained during the posterior approach.

Acknowledgments

We thank Jarett Michaelson PhD for his assistance with the study grant and lab assistance and Vince Wolff BFA for the illustrations.

Footnotes

The institution of one or more of the authors (MSB, RMG) has received, during the study period, funding from Stryker Corp.

One of the authors certifies that he (MSB), or a member of his immediate family, has received or may receive payments or benefits, during the study period, in an amount of USD 10,000 to USD 100,000 from Stryker Corp; and he has a patent pending for shoulder arthroplasty.

One of the authors certifies that he (RMG), or a member of his immediate family, has received or may receive payments or benefits, during the study period, in an amount of USD 10,000 to USD 100,000 from Stryker Corp; in an amount of less than USD 10,000 from Linvatec, outside the submitted work; and he has a patent pending for a posterior approach to total shoulder arthroplasty.

All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request.

Clinical Orthopaedics and Related Research® neither advocates nor endorses the use of any treatment, drug, or device. Readers are encouraged to always seek additional information, including FDA approval status, of any drug or device before clinical use.

Ethical approval was not sought for the present study.

This work was performed at Homer Stryker Center, Mahwah, NJ, USA.

References

  • 1.Apaydin N, Uz A, Bozkurt M, et al. The anatomic relationships of the axillary nerve and surgical landmarks for its localization from the anterior aspect of the shoulder. Clin Anat. 2007;20:273-277. [DOI] [PubMed] [Google Scholar]
  • 2.Amirthanayagam TD, Amis AA, Reilly P, Emery RJ. Rotator cuff-sparing approaches for glenohumeral joint access: an anatomic feasibility study. J Shoulder Elbow Surg. 2017;26:512-520. [DOI] [PubMed] [Google Scholar]
  • 3.Bacle G, Gregoire JM, Patat F, et al. Anatomy and relations of the infraspinatus and the teres minor muscles: a fresh cadaver dissection study. Surg Radiol Anat. 2017;39:119-126. [DOI] [PubMed] [Google Scholar]
  • 4.Bigliani LU, Dalsey RM, McCann PD, et al. An anatomical study of the suprascapular nerve. Arthroscopy. 1990;6:301-305. [DOI] [PubMed] [Google Scholar]
  • 5.Chafik D, Galatz LM, Keener JD, et al. Teres minor muscle and related anatomy. J Shoulder Elbow Surg. 2013;22:108-114. [DOI] [PubMed] [Google Scholar]
  • 6.Curtis AS, Burbank KM, Tierney JJ, et al. The insertional footprint of the rotator cuff: an anatomic study. Arthroscopy. 2006;22:603-609. [DOI] [PubMed] [Google Scholar]
  • 7.Ding DY, Mahure SA, Akuoko JA, et al. Total shoulder arthroplasty using a subscapularis-sparing approach: a radiographic analysis. J Shoulder Elbow Surg .2015;24:831-837. [DOI] [PubMed] [Google Scholar]
  • 8.Edwards B, Boulahia A, Kempf JF, et al. The influence of rotator cuff disease on the results of shoulder arthroplasty for primary osteoarthritis: results of a multicenter study. J Bone Joint Surg Am .2002;84:2240-2248. [DOI] [PubMed] [Google Scholar]
  • 9.Gerber C, Pennington SD, Yian EH, et al. Lesser tuberosity osteotomy for total shoulder arthroplasty. Surgical technique. J Bone Joint Surg Am. 2006;88:170-177. [DOI] [PubMed] [Google Scholar]
  • 10.Greiwe RM. Posterior rotator cuff-sparing total shoulder arthroplasty: three cases. J Am Acad Orthop Surg Glob Res Rev. 2017;31:e1-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Greiwe RM, Hill MA, Boyle MS, et al. Posterior approach total shoulder arthroplasty: a retrospective analysis of short-term results. Orthopedics. 2020;43:e15-20. [DOI] [PubMed] [Google Scholar]
  • 12.Greiwe RM, Witzig SA, Kohrs BJ, et al. Posterior approach shoulder arthroplasty: a cadaveric study assessing access. J Shoulder Elbow Surg . Published online November 19, 2020. DOI: 10.1016/j.jse.2020.10.017. [DOI] [PubMed]
  • 13.Hoppenfeld S, deBoer P. The Shoulder. In: Hurley R, Wolfberg E, Mallard M, eds. Surgical Exposures in Orthopaedics the Anatomic Approach, 3rd ed.Lippincott Williams and Wilkins; 2003:1-66. [Google Scholar]
  • 14.Jackson JD, Cil A, Smith J, et al. Integrity and function of the subscapularis after total shoulder arthroplasty. J Shoulder Elbow Surg. 2010;19:1085-1090. [DOI] [PubMed] [Google Scholar]
  • 15.Kwon YW, Zuckerman JD. Subscapularis-sparing total shoulder arthroplasty: a prospective, double-blinded, randomized clinical trial. Orthopedics. 2019;42:e61-67. [DOI] [PubMed] [Google Scholar]
  • 16.Lafosse L, Schnaser E, Haag M, et al. Primary total shoulder arthroplasty performed entirely thru the rotator interval: technique and minimum two-year outcomes. J Shoulder Elbow Surg. 2009;18:864-873. [DOI] [PubMed] [Google Scholar]
  • 17.Leschinger T, Hackl M, Buess E, et al. The risk of suprascapular and axillary nerve injury in reverse total shoulder arthroplasty: an anatomic study. Injury. 2017;48:2042-2049. [DOI] [PubMed] [Google Scholar]
  • 18.McFarland EG, Caicedo JC, Guitterez MI, et al. The anatomic relationship of the brachial plexus and axillary artery to the glenoid. Implications for anterior shoulder surgery. Am J Sports Med. 2001;29:729-733. [DOI] [PubMed] [Google Scholar]
  • 19.Miller SL, Hazrati Y, Klepps S, et al. Loss of subscapularis function after total shoulder replacement: a seldom recognized problem. J Shoulder Elbow Surg. 2003;12:29-34. [DOI] [PubMed] [Google Scholar]
  • 20.Moeckel BH, Altchek DW, Warren RF, et al. Instability of the shoulder after arthroplasty. J Bone Joint Surg Am. 1993;75:492-497. [DOI] [PubMed] [Google Scholar]
  • 21.Ransom EF, Adkison DP, Woods DP, et al. Subscapularis sparing total shoulder arthroplasty through a superolateral approach: a radiographic study. J Shoulder Elbow Surg .2020;29:814-820. [DOI] [PubMed] [Google Scholar]
  • 22.Sher JS, Uribe JW, Posada A, et al. Abnormal findings on magnetic resonance images of asymptomatic shoulders. J Bone Joint Surg Am. 1995;77:10-15. [DOI] [PubMed] [Google Scholar]
  • 23.Simovitch R, Fullick R, Kwon Y, et al. Use of the subscapularis preserving technique in anatomic total shoulder arthroplasty. Bull Hosp Jt Dis. 2013;71:94-100. [PubMed] [Google Scholar]
  • 24.Small KM, Siegel EJ, Miller LR, et al. Imaging characteristics of lesser tuberosity osteotomy after total shoulder replacement: a study of 220 patients. J Shoulder Elbow Surg .2014;23:1318-1326. [DOI] [PubMed] [Google Scholar]

Articles from Clinical Orthopaedics and Related Research are provided here courtesy of The Association of Bone and Joint Surgeons

RESOURCES