19.1 Introduction

Scapular notching in reverse shoulder arthroplasty has been a well-documented orthopaedic phenomenon since the development and implementation of the Grammont-style reverse shoulder arthroplasty in 1985. In early reports, the incidence ranged from 44 to 96%. ¹ , ² , ³ , ⁴ Scapular notching is identified as the osseous defect of the lateral scapular pillar and neck that is created by impingement of the humeral component when the arm is adducted. The scapular defect and resultant osteolysis is potentially compounded by the biological response to polyethylene wear debris. ² , ⁵ This osseous defect typically occurs inferior to the glenosphere, but can also be found anterior or posterior depending on the implant design and positioning of the glenosphere and humeral component. ⁶

The Nerot-Sirveaux classification ¹ is the most commonly used grading system to define the extent of scapular notching in reverse shoulder arthroplasty. In this classification, grade 1 notching concerns only the scapular pillar. Grade 2 is demarcated by erosion to the inferior screw of the baseplate. Grade 3 describes erosion beyond the inferior screw, and grade 4 describes progression of the defect to the central baseplate peg (► Fig. 19.1). It should be noted that some authors ³ prefer to simplify this classification, by regrouping Sirveaux grades 1 and 2 into a single group and grades 3 and 4 into another group.

Radiographic evaluation of scapular notching includes true anteroposterior (AP) and axillary roentgenograms. The Grashey view or true AP of the shoulder in the scapular plane allows direct examination of the scapular neck, without overlap of the humeral prosthesis. The axillary view may demonstrate notching of the anterior or posterior scapular neck that may otherwise go unrecognized (► Fig. 19.2). It should also be recognized that the presence of a bony spur medial to the notch may make the notch appear larger than it truly is.

Scapular notching is usually observed within the first year postoperatively, with most authors reporting scapular neck osteolysis between 1 and 14 months. ⁴ Previously, rates of scapular notching ranged from 44 to 96%. ¹ , ² , ³ , ⁴ Based on a review of prior studies, Roche et al ⁷ calculated the weighted mean average of notching for reverse shoulder arthroplasty to be 68%, with a weighted mean average of 21% for grade 3 and grade 4 notches. However, with improvements in surgical technique and prosthetic design, the rate of scapular notching has been on the decline in recent years. In the past 5 years, reported rates have ranged from 0 to 19%. ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹²

19.2 Is Scapular Notching a Problem?

While scapular notching is generally observed within the first year following reverse shoulder arthroplasty, it remains unclear whether notching progresses with time. In their review of one prosthesis, Werner et al ² and Simovitch et al ⁴ suggested that scapular notching plateaus with time. Werner et al reported that 79% of detected scapular notches did not show evidence of progression over 1 year, while the remaining 21% increased by a maximum of 1 grade by 38 months postoperatively. ² Simovitch et al noted that scapular notching plateaued at 18 months with no evidence of progression at 24 months. ⁴ More recently, Al-Hadithy et al ¹³ found that in 28 shoulders with scapular notching, all notching was visible at 12 months, and only 3 progressed with time.

In contrast, Lévigne et al ³ found that the presence and the grade of notching directly correlated with the length of follow-up. In this series, notching was detected in 48% of shoulders at 1 year, 60% at 2 years, and 68% at 3 years. Notch progression varied from case to case. These authors described three groups of patients: those who would never develop notching, those with notches that plateaued within 2 years, and those with notches that continued to progress after 3 years. They also noted an increase in both grade 3 and grade 4 notches with longer postoperative follow-up. In a series of Favard et al, ¹⁴ the percentage of grade 3 and grade 4 notches increased from 35% at 5 years to 49% at 9 years postoperatively.

As mentioned earlier, notch progression may be secondary to polyethylene wear–related osteolysis. ¹⁵ Damage modes observed in retrieved reverse shoulder arthroplasty have been reported. ¹⁶ In the largest retrieval study to date, Wiater et al ¹⁷ found that the grade of scapular notching was positively and significantly correlated with dishing, delamination, and embedding of third-body particles in the articular surface of retrieved polyethylene components. Furthermore, the authors detected a relationship between notching grade and damage to the articular and rim surfaces of the polyethylene liners. This wear pattern was consistent with impingement of the humeral polyethylene at the lateral edge of the scapula. Despite these concerning findings, the real impact of scapular notching on implant survivorship remains unknown. Very few cases of glenoid component loosening directly attributable to scapular notching have been reported within the literature, and longer-term follow-up is required to better understand the consequences of scapular notch progression.

The clinical relevance of scapular notching is also unclear. Lévigne et al ³ found no relationship between notching and pain or Constant scores with a mean follow-up of 51 months. Similarly, Werner et al ² found no correlation between notching and clinical outcome. Favard et al ¹⁴ concluded that scapular notching, no matter the grade, did not influence clinical outcomes in their retrospective review of 527 reverse shoulder arthroplasties. Recently, Al-Hadithy et al ¹³ reported no association between scapular notch progression and Constant or Oxford scores at 24 and 60 months. In contrast, several authors have associated scapular notching with poorer functional outcomes, lower patient satisfaction, and more limited shoulder motion. Sirveaux et al ¹ reported that patients with more severe notching had lower postoperative Constant scores. Similarly, Simovitch et al ⁴ found that patients with notching had lower Constant scores, a decreased subjective shoulder score, inferior shoulder strength, and worse postoperative range of motion.

While scapular notching tends to develop early in the postoperative period, the progression and clinical relevance of notching remains controversial. Furthermore, while the theoretical development of osteolysis secondary to polyethylene wear debris may contribute to its progression, very few cases of glenoid component failure have been directly attributed to notching alone. Longer follow-up and larger series are needed to better elucidate the etiologies and clinical impact of scapular notching.