ORIGINAL ARTICLE

Anatomical considerations for safe scapular resection in snappingscapula syndrome

Anjali Aggarwal • Pratima Wahee •

Aditya K. Aggarwal • Harjeet Kaur •

Daisy Sahni

Received: 7 February 2011 / Accepted: 25 May 2011 / Published online: 9 June 2011

� Springer-Verlag 2011

Abstract

Background The resection of the superomedial angle of

the scapula in snapping scapula syndrome is associated

with potential risk of injury to the suprascapular nerve. The

aim of the present study was to determine the distance of

site of resection on the upper border of the scapula from the

suprascapular notch during arthroscopic resection of the

superomedial angle using standard superior Bell’s and

medial portals at the middle of medial border.

Materials and methods The study was conducted on 92

intact dry adult scapulae of unknown sex. The junction of

the medial two-thirds and the lateral one-third of an

imaginary line joining the superomedial angle and the

lateral margin of the acromion was marked as superior

Bell’s portal. The middle of the medial border was con-

sidered as the arthroscopic resection target. The site of

resection on the upper border was found out with the help

of these two points and other relevant measurements were

also taken.

Results The distance between the suprascapular notch

and the lateral edge of resection on the upper border of the

scapula was more than 10 mm in 85.9% cases and less than

10 mm in 14.1%. A statistically significant positive cor-

relation was found between this distance and the distance

between the superior angle and acromion angle of the

scapula.

Conclusion Our study suggests that during arthroscopic

resection using superior Bell’s portal and medial portal,

suprascapular notch hence the suprascapular nerve would

have safe margin of more than 10 mm from the resection

site on upper border in 85.9% cases and would be vul-

nerable to injury in 14% cases. The procedure would be

safer in patients with a wider scapula.

Keywords Suprascapular nerve � Suprascapular notch �Superomedial angle � Arthroscopy

Introduction

Snapping scapula syndrome is a disorder of scapulocostal

mechanism characterized by audible painful grating during

scapulothoracic motion. Cases of snapping secondary to

bony incongruity at the superomedial angle, not responding

to conservative management are usually treated by surgical

resection of superomedial corner [1, 7, 9]. The resection of

superomedial corner has traditionally been done by an open

procedure [8]. However, the large incision necessary to

visualize the scapula and its underlying bursa often has

poor cosmetic results. Ciullo in 1992 [4] described a

technique of arthroscopic treatment of snapping scapula,

which was later evolved further by Harper et al. [5] and

Chan et al. [3]. This procedure not only allows resection of

bursa and bony prominence but also helps in early reha-

bilitation and yields better cosmetic results. During

arthroscopic or open surgical resection of superomedial

corner, it is imperative to protect the suprascapular nerve

from iatrogenic injury. Arthroscopic resection is techni-

cally more demanding as compared to open resection as no

standard bony landmark is available to predict the position

of the suprascapular nerve at the upper border of scapula

A. Aggarwal (&) � P. Wahee � H. Kaur � D. Sahni

Department of Anatomy, Post Graduate Institute of Medical

Education and Research, # 123-C Type V Sector 24-A,

Chandigarh 160023, India

e-mail: anjli_doc@yahoo.com; agg_adityadoc@yahoo.co.in

A. K. Aggarwal

Department of Orthopaedic Surgery, Post Graduate Institute

of Medical Education and Research, Chandigarh 160012, India

123

Surg Radiol Anat (2012) 34:43–47

DOI 10.1007/s00276-011-0835-5

and difficulty is experienced in judging the amount of bone

to be resected.

For the arthroscopic resection of the superomedial cor-

ner in the treatment of snapping scapula syndrome, Chan

et al. in 2002 [3] suggested a portal called Bell’s portal

located at the junction of the medial two-thirds and lateral

one-third of line joining superomedial angle and lateral

margin of acromion. Bell and Van Riet in 2008 [2] sup-

plemented the superior Bell’s portal with arthroscopic

resection target (ART) fixed at the middle of medial border

of scapula and concluded that if these were used together

there would be a safe distance between suprascapular nerve

and site of resection on the upper border. We designed our

study on the guidelines proposed by Bell and Van Riet [2].

The purpose of this anatomic study was to find out the

distance of lateral rim of resection on upper border from

suprascapular notch in Indian population, as knowledge of

this measurement may help in preventing injury to the

nerve.

Materials and methods

The present study was conducted on 92 dry, intact adult

scapulae of unknown sex procured from departmental

collection. On each bone, the fixed points as mentioned

below were marked (Figs. 1, 2).

A, Junction of lateral and medial margin of acromion

process.

B, Superior angle at the junction of superior and medial

border.

C, Medial edge of suprascapular notch.

D, Center of root of spine on the medial border.

E, Inferior angle at the junction of lateral and medial

border.

F, Point on the medial border midway between D and E.

Points A and B were chosen because they were easily

palpable fixed bony landmarks. The distance AB was

divided into three equal parts. The junction of medial one-

third and lateral two-thirds of the imaginary line AB was

marked as point O (Fig. 1). Point ‘O’ represented superior

Bell’s portal [3]. Point ‘F’ was equivalent to ART as used

by Bell and Van Riet [2]. A metallic wire ZZ0 was placed

on the costal surface of the scapula passing through the

points O and F. The point at which this wire cut across the

superior border was marked as the site of resection (Fig. 1).

The bone was then placed on a paper with its costal surface

facing down. Two tangential lines YY0 and ZZ0 (Fig. 2)

were drawn on the paper along the medial border of

scapula intersecting at the center of root of spine (D). The

angle of medial border YDZ0 (angle X) as shown in Fig. 2

was measured with a protractor. Parameters as described in

Table 1 were measured using digital caliper of precision

0.02 mm (Mitutoyo, Japan). The first two authors per-

formed all measurements twice, independently.

Statistical analysis

Statistical analysis was performed using SPSS (SPSS 15

for Windows, Chicago, Il, USA). The Kolmogorov–Smir-

nov test was applied to see normality of variables. Pear-

son’s correlation coefficient was obtained to see any

correlation between various parameters. Correlation was

considered significant at the p value of 0.05 and highly

significant at 0.01.

Results

All the variables tested by Kolmogorov–Smirnov test were

found to be normally distributed. Results of all parameters

measured in this study are shown in the Table 2. The dis-

tance between the suprascapular notch and the point of

resection on the upper border (CG) was further categorized

Fig. 1 Dorsal aspect of scapula demonstrating various anatomical

landmarks and measurements A lateral edge of acromion process;

B Superior angle; C Medial edge of suprascapular notch; D Center of

root of spine; E Inferior angle; F midway between the center of root

of spine and the inferior angle on the medial border; O Superior portal

at the junction of medial one-third and lateral two-thirds of imaginary

line AB; PP0 Imaginary line passing from superior portal ‘O’ to

medial portal at F; G point on the upper border intersected by PP0

44 Surg Radiol Anat (2012) 34:43–47

123

into three groups. It was less than 10 mm in 14.1% cases,

between 10.1 and 20 mm in 59.8% cases and more than

20 mm in 26.1% cases (Fig. 3). This distance showed

positive correlation with the distance between the superior

angle and acromion angle (AB) with a correlation coeffi-

cient of 0.374 (p value \0.01) as well as with angle of

medial border (X) with correlation coefficient of 0.245

(p value \0.05).

Similarly, the length of superior border (BC) as well as

the angle of medial border (X) showed positive correlation

with distance between superior angle and acromion angle

with a correlation coefficient of 0.805 (p value\0.01) and

0.411 (p value \0.01), respectively. Significant negative

correlation of angle of medial border (X) with the length of

medial border (BD) was observed (correlation coefficient

-0.324; p \ 0.01).

Discussion

During partial scapulectomy, open or arthroscopic, diffi-

culty is usually encountered in deciding the extent of

resection on the superior border so as to protect the

suprascapular nerve. This study is focused on a particular

surgical risk. However, this is not the only one with this

procedure, even the accessory nerve and the dorsal scapular

nerve are also at risk during arthroscopic resection. How-

ever, the authors conducted the present study using the

suprascapular notch as a reference point.

The risk for nerve depends on the distance from portals

and on surgical procedure. Chan et al. [3] reported that if

arthroscopic trocar is passed from a portal situated 2–3 cm

medial to middle border at the level of spine of scapula

towards Bell’s portal, a minimum distance of 12 mm was

always there between the nerve and the Bell’s portal.

Subsequently, Bell and Van Riet [2] suggested that this

distance might not be a sufficient distance in the region of

suprascapular notch, especially, if suction is used which

could potentially move the nerve towards the shaver.

The use of superior Bell’s portal [3, 10, 11] and medial

portal [2] has been reported to make resection of the su-

peromedial corner much easier and safer. Bell and Van

Riet [2] attempted to combine superior Bell’s portal with

inferior portal at inferior angle of scapula in nine cases. In

other nine cases, he combined superior Bell’s portal with

ART fixed at the middle of medial border. The proximity

of the suprascapular nerve to the resection site on the upper

border was studied in both the groups and the results were

compared. The distance between the resected margin on

the upper border and the suprascapular notch ranged from

25 to 50 mm (31 ± 9 mm) in second group and 10 to

30 mm (21 ± 7 mm) in the first group. As evident, a

minimum distance of 25 mm was always present between

the resected margin and the suprascapular notch if ART

was chosen at the middle of medial border. Therefore, this

was the recommended method over other methods as it had

a wide margin of safety for the suprascapular nerve. We

conducted our study on 92 scapulae using the points on the

scapula corresponding to those used by Bell and Van Riet

[2] in the second group of their study. The distance of the

resection site on the upper border from the suprascapular

notch ranged from 4.6 to 28.8 mm with an average value of

Fig. 2 Dorsal aspect of scapula illustrating linear and angular

measurements D Center of root of spine; YY0 Tangential line passing

along upper part of medial border; ZZ0 Tangential line passing along

lower part of medial border; Angle X (YDZ0) Angle of medial border

subtended at D; BD length of supraspinous portion of medial border;

BC length of upper border

Table 1 Parameters studied

AB Distance between superior angle and lateral margin of acromion process

BC Length of upper border (distance between superior angle and medial edge of supra scapular notch)

BD Length of supraspinous portion of medial border (distance between superior angle and center of root of spine)

GC Length of safe zone (distance between suprascapular notch and point G)

BG Distance between superior angle and point G (length of upper border to be resected in arthroscopic resection)

Angle X Angulation of medial border (angle between supraspinous portion and infraspinous portion of medial border)

Surg Radiol Anat (2012) 34:43–47 45

123

16.48 ± 5.32 mm. In 14.1% cases, the distance was less

than 10 mm; in 59.8% cases, the distance was between

10.1 and 20 mm; and in 26.1% cases, it was more than

20 mm (Fig. 4). Our data suggests that in 14% cases the

margin of safety distance was very small. Hence, these

cases may be labeled as a high-risk group and may require

extra precautions during arthroscopic resection for avoid-

ing injury to the suprascapular nerve. In 26% cases, the

notch, and thus the nerve would be at quite safe distance

from the site of resection with distance between 20.1 and

28.8 mm. In Bell and Van Riet’s study [2], the minimum

distance of 25 mm was always available when the shaver

was directed from the superior portal toward ART. This

distance observed in our study was much lower than that

found in Bell and Van Riet’s [2] study. The authors

used landmarks on bones, not on cadavers, without the

imprecision due to soft tissue. Other studies are anatomic

studies with dissection, nerve exposure, and measure on a

3D analysis, not on a plane analysis [2]. The smaller dis-

tance in our study is attributed probably to the

measurements done on the bare bones without the soft

tissues and secondly to the narrow body build of Asians.

Thus, the narrow distance of safety for the suprascapular

nerve observed in the Indian population for scapular

resection indicates that the Bell’s portal may not be safe

during operative intervention in all the cases.

As the sex of the scapulae used in the present study was

unknown, therefore, the sex ratio for the scapulae and the

influence of sex on the dimensions of the scapula could not

be commented on.

In our study, the length of segment of bone to be

resected, as measured from superior angle to the resection

site on the upper border ranged from 16 to 38.9 mm

(average 28 ± 4.79 mm; Fig. 4). In comparison, this

length of segment of bone ranged from 30 to 45 mm

(38 ± 5 mm) in the study conducted by Bell and Van

Riet’s [2]. A narrow dimension observed in our study

seems to be due to the difference in technique used for

measuring length of the resected segment. We measured

this on the upper border of scapula whereas Bell and Van

Table 2 Measurements of various parameters studied

Parameters Mean ± SD (mm) Range (mm)

Length of superior border (BC) 43.66 ± 6.99 23.7–59.7

Length of upper part of medial border (BD) 45.73 ± 6.78 32–62

Distance between lateral margin of acromion process and superior angle (AB) 101.84 ± 10.59 69.1–129.9

Distance between superior angle and lateral edge of resection (BG) 28 ± 4.79 16–38.9

Length of safe zone (distance between suprascapular notch and point G- GC) 16.48 ± 5.32 4.6–28.8

Angulation of medial border (angle X) 132.64�± 6.82� 111–148�

5.0 10.0 15.0 20.0 25.0

Length of safe zone on the upper border

0.0

2.5

5.0

7.5

10.0

Co

un

t

Fig. 3 Histogram showing distribution of safe distance of supra-

scapular notch in mm from the site of resection on the upper border

20.0 25.0 30.0 35.0

Length of resected segment of upper border

0

5

10

15

20

Co

un

t

Fig. 4 Histogram showing distribution of length of resected segment

of upper border in mm

46 Surg Radiol Anat (2012) 34:43–47

123

Riet’s [2] measured the oblique distance from upper border

to the spine of scapula. Another observation in our study

was a statistically significant positive correlation of dis-

tance between acromion and superior angle with the dis-

tance of suprascapular notch from the resection site (GC in

Fig. 1). This positive correlation suggests that there would

be greater margin of safety in wider scapula. Thus, the

technique would be safer in scapulae in which the distance

between acromion and superior angle is wide.

Lehtinen et al. [6] found it difficult to judge the amount

of bone to be removed, arthroscopically. Inadequate partial

resection of superomedial corner of scapula is known to be

associated with persistence of symptom. If superomedial

corner is excessively angulated anteriorly, more radical

resection might be necessary to adequately decompress the

scapulothoracic articulation. Lehtinen et al. [6] suggested

that 1–7 cm of bone could be safely excised. Oizumi et al.

[9] reported complete resection measuring 2 9 7 cm of

superior angle, whereas according to Harper et al. [5], the

extent of arthroscopic resection should be as much as

required. There was no consensus regarding the amount of

bone to be resected. Hence, precise knowledge of the

anatomical dimensions of the suprascapular portion of

scapula is necessary before attempting open or arthroscopic

resection. In our study, the superior border (BC in Fig. 2)

varied in length from 23.7 to 59.7 mm with average dis-

tance of 43.66 ± 6.99 mm and the length of medial border

(BD in Fig. 2) varied from 32 to 62 mm

(45.73 ± 6.78 mm). To the best of our knowledge, mea-

surement of these borders of the scapula in relation to

resection has not been described in the literature until so

far.

We attempted to see a correlation between the angula-

tion of the medial border subtended at the level of root of

spine with distance of site of resection on upper border

with the suprascapular notch. Our observations suggested

that wider the angle, greater would be the margin of safety

for suprascapular nerve from the site of resection on the

upper border. A negative correlation of this angle with the

length of supraspinous portion of medial border (BD)

suggests that in scapulae with wider angle, length of

resected segment along the medial border would be less.

Such correlations have not been reported in the literature

until date.

The present study delineates the dimensions of scapula

as applicable to its surgical resection for the treatment of

snapping scapula and describes about the safe distance to

prevent iatrogenic injury to suprascapular nerve. Bell’s

portal should be combined with arthroscopic resection

target (ART) at the middle of the medial border in those

cases with wide scapula (wide distance between acromion

and superior angle) during arthroscopic resection of su-

peromedial angle of scapula in snapping scapula syndrome.

The paucity of data on the safe zone in Indian population

should pave way for more extensive research in this field.

Conflict of interest The authors declare that they have no conflict

of interest.

References

1. Aggarwal A, Wahee P, Harjeet, Aggarwal AK, Sahni D (2011)

Variable osseous anatomy of costal surface of scapula and its

implications in relation to snapping scapula syndrome. Surg

Radiol Anat 33:135–140

2. Bell SN, Van Riet RP (2008) Safe zone for arthroscopic resection

of the superomedial scapular border in the treatment of snapping

scapular syndrome. J Shoulder Elbow Surg 17:647–649

3. Chan BK, Chakrabarti AJ, Bell SN (2002) An alternative portal

for scapulothoracic arthroscopy. J Shoulder Elbow Surg

11:235–238

4. Ciullo JV (1992) Subscapular bursitis. Treatment of ‘‘snapping

scapula’’ or ‘‘washboard syndrome’’. Arthroscopy 8:412–413

5. Harper GD, Mcllroy S, Bayley JIL, Calvert PT (1999) Arthro-

scopic partial resection of the scapula for snapping scapula: a new

technique. J Shoulder Elbow Surg 8:53–57

6. Lehtinen JT, Macy JC, Cassinelli E, Warner JJP (2004) The

painful scapulothoracic articulation: surgical management. Clin

Orthop 423:99–105

7. Lehtinen JT, Tingart MJ, Apreleva M, Warner JP (2005) Quan-

titative morphology of the scapula: normal variations of the su-

peromedial scapular angle and superior and inferior pole

thickness. Orthopedics 28(5):481–486

8. Milch H, Burman MS (1933) Snapping scapula and humerus

varus: report of six cases. Arch Surg 26:570–588

9. Oizumi N, Suenaga N, Minami A (2004) Snapping scapula

caused by abnormal angulation of the superior angle of the

scapula. J Shoulder Elbow Surg 13:115–118

10. Pavlik A, Ang K, Coghlan J, Bells S (2003) Arthroscopic treat-

ment of painful snapping of scapula by using a new superior

portal. Arthroscopy 19(6):608–612

11. Van Riet RP, Bell SN (2006) Scapulothoracic arthroscopy. Tech

Shoulder Elbow Surg 7:143–146

Surg Radiol Anat (2012) 34:43–47 47

123