KLEMT-2019 - Ombro Forces In Vivo

Prévia do material em texto

68 THE BONE & JOINT JOURNAL
Patients with recurrent anterior dislocation of 
the shoulder commonly have an anterior osseous 
defect in the glenoid. The prevalence of fracture 
or erosion of the rim of the glenoid ranges from 
8% (18/226) to 73% (116/158) in these patients.1,2 
As osseous defects may result in loss of concav-
ity of the glenoid, the shoulder becomes unstable 
in the mid-range of movement when stability is 
achieved by compression of the humeral head 
into the glenoid labrum through contraction of the 
rotator cuff.3 Biomechanical studies have shown 
an inverse relationship between the size of the 
defect and stability: the larger the defect, the less 
stable the shoulder.3-5
The rate of recurrent instability after arthro-
scopic Bankart repair in patients with a large 
osseous defect of the glenoid is higher than after 
bone grafting.6,7 The rationale for not perform-
ing bone grafting in all patients with a glenoid 
defect include concerns about the rate of com-
plications, the long term sequelae, and surgical 
learning curve.8 Thus, defining the critical size of 
an osseous defect that leads to persistent anterior 
instability under loading conditions expected by 
the patient during functional activities will assist 
clinical decision making.9
Currently, only three biomechanical studies 
from one group have quantified the critical size of 
an osseous defect that necessitates bone grafting.3-5 
These studies found that, for defects that corre-
spond to 19% to 21% of the length of the glenoid, 
stability cannot be maintained by a Bankart repair 
 SHOULDER & ELBOW
The critical size of a defect in the glenoid 
causing anterior instability of the shoulder 
after a Bankart repair, under physiological 
joint loading
C. Klemt, 
D. Toderita, 
D. Nolte, 
E. Di Federico, 
P. Reilly, 
A. M. J. Bull
From Imperial College 
London, London, 
United Kingdom
 C. Klemt, PhD, Post-Doctoral 
Research Fellow
 D. Toderita, Undergraduate 
Student
 D. Nolte, PhD, Post-Doctoral 
Research Fellow
 E. Di Federico, PhD, 
Post-Doctoral Research Fellow
 A. M. J. Bull, PhD, Professor 
of Musculoskeletal Mechanics; 
Head of Bioengineering 
Department
Department of Bioengineering, 
Imperial College London, 
London, UK.
 P. Reilly, MS, FRCS(Orth), 
Consultant Orthopaedic 
Surgeon, Department of 
Trauma and Orthopaedics, 
Imperial College Healthcare 
NHS Trust, London, UK.
Correspondence should 
be sent to C. Klemt; email: 
c.klemt15@imperial.ac.uk.
© 2019 The British Editorial 
Society of Bone & Joint Surgery 
doi:10.1302/0301-620X.101B1.
BJJ-2018-0974.R1 $2.00
Bone Joint J 
2019;101-B:68–74.
Aims
Patients with recurrent anterior dislocation of the shoulder commonly have an anterior 
osseous defect of the glenoid. Once the defect reaches a critical size, stability may be 
restored by bone grafting. The critical size of this defect under non-physiological loading 
conditions has previously been identified as 20% of the length of the glenoid. As the 
stability of the shoulder is load-dependent, with higher joint forces leading to a loss of 
stability, the aim of this study was to determine the critical size of an osseous defect that 
leads to further anterior instability of the shoulder under physiological loading despite a 
Bankart repair.
Patients and Methods
Two finite element (FE) models were used to determine the risk of dislocation of the 
shoulder during 30 activities of daily living (ADLs) for the intact glenoid and after creating 
anterior osseous defects of increasing magnitudes. A Bankart repair was simulated for 
each size of defect, and the shoulder was tested under loading conditions that replicate 
in vivo forces during these ADLs. The critical size of a defect was defined as the smallest 
osseous defect that leads to dislocation.
Results
The FE models showed a high risk of dislocation during ADLs after a Bankart repair for 
anterior defects corresponding to 16% of the length of the glenoid.
Conclusion
This computational study suggests that bone grafting should be undertaken for an anterior 
osseous defect in the glenoid of more than 16% of its length rather than a solely soft-tissue 
procedure, in order to optimize stability by restoring the concavity of the glenoid.
Cite this article: Bone Joint J 2019;101-B:68–74.
 THE CRITICAL SIZE OF A DEFECT IN THE GLENOID CAUSING ANTERIOR INSTABILITY OF THE SHOULDER 69
VOL. 101-B, No. 1, JANUARY 2019
Table I. Baseline properties for the finite element model of the glenohumeral joint
Anatomy Material type Parameter Value Reference
Humerus Rigid N/A N/A Terrier et al19 (2007)
Humeral cartilage Isotropic elastic E 10 MPa Buchler et al18 (2002)
υ 0.4
Glenoid Rigid N/A N/A Terrier et al19 (2007)
Glenoid cartilage Isotropic elastic E 10 MPa Buchler et al18 (2002)
υ 0.4
Labrum Transversely isotropic hyperelastic C1 3.4 Smith et al16 (2008)
C2 5.4
C3 7.0
C4 5.9
C5 5.2
C6 4.8
C7 5.7
C8 4.3
N/A, not available; E, Young’s modulus; υ, Poisson’s ratio; C1 to C8, hyperelastic labral coefficients
Fig. 1
Simulated anterior osteotomy lines, at 4%, 8%, 12%, 16%, 20%, and 24% 
of the length of the glenoid, respectively, parallel to the longitudinal axis 
of the glenoid.
4%
8%
12%
16%
20%
24%
and that stability is only restored after reconstruction of the con-
cavity of the glenoid by bone grafting. Clinical guidelines have 
taken these data as recommendations for bone grafting.10
These biochemical studies involve cadaveric experiments 
and have limitations, including non-physiological loading with 
a maximum of 50 N. In vivo measurements from instrumented 
shoulder implants have reported loads of 151% of body weight 
during activities of daily living (ADLs).11 As the stability of the 
shoulder is load-dependent, with higher forces leading to a loss 
of stability,12,13 the use of physiological loading is essential for 
the assessment of stability.
Finite element (FE) modelling may be used to overcome this 
limitation of cadaveric studies and to simulate testing under in 
vivo conditions. The aim of this study was to determine the crit-
ical size of an anterior osseous defect in the glenoid that leads to 
persistent anterior instability of the shoulder under physiologi-
cal loading despite a Bankart repair. This may assist in choosing 
the appropriate surgical procedure for patients with recurrent 
anterior dislocation.
Patients and Methods
Detailed information about the development and validation 
of two subject-specific FE models of the shoulder have been 
previously described.13 Briefly, two geometrical representations 
of the right shoulder were obtained from high-resolution slices 
of the male and female Visible Human data sets.14 The glenoid 
fossa at the lateral aspect of the scapula, the proximal humerus, 
the surfaces of the articular cartilage, and the glenoid labrum 
were manually segmented using Mimics (Mimics Research 
17.0; Materialise, Leuven, Belgium). The segmented structures 
were converted to triangular surface meshes to form a 3D model 
of the shoulder. Local coordinate systems were assigned to the 
articulating structures to position and orientate the humerus with 
respect to the glenoid fossa.15 The meshes were imported into 
FE analysis software (Marc Mentat 2015.0.0, MSC Software, 
Palo Alto, California). The articulating bones, the surfaces of 
Fig. 2
Overview of the design of the study. FE, finite element; ADL, activities of 
daily living.
- Joint position most critical to stability
- In vivo joint loading (shear and compression forces)
FE models of the shoulder
- Simulation of 2 mm to 8 mm osseous defects
- Modelling of Bankart repair
- Comparison of shoulder dislocation force with in vivo shear force
- Classification of each ADL for each defect size as stable or unstable
Risk of shoulder dislocation
Analysis of 26 functional daily activities17
70 C. KLEmT, D. TODERITA, D. NOLTE, E. DI FEDERICO, P. REILLY, A. m. J. BULL 
Follow us @BoneJointJ THE BONE & JOINT JOURNALthe articular cartilage, and the glenoid labrum were modelled as 
solid tetrahedral elements. The labrum was divided into eight 
sections,16 each of which was assigned a coordinate frame to 
define the orientation of the fibres.17
Baseline properties for each structure of the shoulder were 
assigned based on previously reported values (Table I).16-19 
Due to relatively small deformations compared with other soft 
tissues, the articulating bones were modelled as rigid materi-
als. The surfaces of the articular cartilage were assigned linear 
elastic isotropic properties, while the labrum was modelled as a 
transversely isotropic hyperelastic material due to the difference 
in modulus between the transverse plane and the circumferen-
tial direction.16 The coefficients for the hyperelastic model were 
obtained by applying the neo-Hookean constitutive equation to 
derive material properties for each labral section.20
The interfaces between the articulating cartilage surfaces 
and between the humeral cartilage and the labrum were mod-
elled using frictionless, surface-to-surface contact due to the 
low coefficient of friction in synovial joints.17 The interfaces 
between the articulating bones and the corresponding carti-
lage surfaces were modelled using tied contact. In a previous 
study, the FE models of the shoulder were validated against 
in vitro measurements of its stability for many loading condi-
tions and positions of the joint, as reported in two cadaveric 
studies.12,13
Osseous defects involving 4%, 8%, 12%, 16%, 20%, and 
24% of the length of the glenoid were created separately at its 
anterior rim. These models of a chronic defect were created 
in agreement with cadaveric testing as previously reported 
(Fig. 1).3-5 Anterior osteotomy lines were drawn parallel to a 
Fig. 3
Dislocation forces of the shoulder for the intact glenoid and after the creation of anterior osseous defects correspond-
ing to 8%, 16%, and 24% of the length of the glenoid. The forces represent those from cadaveric experiments4 and the 
results of simulation from this study. The bars represent one standard deviation from the in vitro cadaver values.
0
2
4
6
8
10
12
14
16
18
20
A
n
te
ri
o
r 
d
is
lo
ca
ti
o
n
 f
o
rc
e 
(N
)
Intact glenoid 8% of glenoid width 16% of glenoid width 24% of glenoid width
Yamamoto et al4 (2009) Male Visible Human Female Visible Human
Table II. Classification of each size of defect for each activity of daily liv-
ing (ADL) as unstable (U) or stable (S). The sizes correspond to 4%, 8%, 
12%, 16%, 20%, and 24% of the length of the glenoid, respectively
Anterior glenoid osseous defect
Intact 4% 8% 12% 16% 20% 24%
Quarterback throw S S S S U U U
Lineout throw S S S S U U U
Extreme (reach across body) S S S S U U U
Clean upper back S S S S U U U
Volleyball serve S S S S S U U
Slow abduction S S S S S U U
Fast abduction S S S S S U U
Reach back of head S S S S S U U
Basketball free throw S S S S S S U
Sit to stand S S S S S S U
Pull S S S S S S U
Push S S S S S S U
Lift block to head height S S S S S S U
Lift block to shoulder height S S S S S S U
Reach opposite axilla S S S S S S U
Lift shopping bag from floor S S S S S S S
Lift shopping bag on lap S S S S S S S
Perineal care S S S S S S S
Reach far ahead S S S S S S S
Brush left side of head S S S S S S S
Drive slow left S S S S S S S
Drive slow right S S S S S S S
Drive fast left S S S S S S S
Drive fast right S S S S S S S
Slow flexion S S S S S S S
Fast flexion S S S S S S S
Eat with hand S S S S S S S
Drink from mug S S S S S S S
 THE CRITICAL SIZE OF A DEFECT IN THE GLENOID CAUSING ANTERIOR INSTABILITY OF THE SHOULDER 71
VOL. 101-B, No. 1, JANUARY 2019
line passing through superior and inferior contact points of the 
circle that fits the superoinferior diameter of the glenoid, with 
the 0% line being tangential to the anterior rim of the glenoid.4,5 
The defects were introduced into the models by removing FEs 
of the glenoid fossa and articular cartilage that correspond to 
each osteotomy line.
A Bankart repair was simulated for each size of defect by 
reattaching the labrum to the osteotomy line in order to re- 
establish the concavity of the glenoid. This was modelled by 
displacing the medial surface nodes of the labrum towards the 
osteotomy line until it was fully aligned and lifted to the level 
of the osteotomy line. The medial nodes of the labrum were 
fixed in that position during further simulations assessing the 
stability of the joint in order to simulate labral healing to the 
osteotomy line, as can be observed in vivo following successful 
rehabilitation of the shoulder. The other parts of the intact FE 
models were not changed. Hill–Sachs lesions were not mod-
elled in this study.
The FE models with the osseous defects in the glenoid were 
validated against in vitro measurements of anterior stability of 
the shoulder.4,5 The cadaveric measurements represent the only 
data available in the literature that assess the loss of anterior 
stability with anterior osseous defects in the glenoid of different 
magnitudes.4,5 For the validation study, the humerus was placed 
in the position of dislocation with 45° of abduction and 35° 
of external rotation in order to replicate the cadaveric experi-
ments.4,5 In the starting position, the humeral head was in contact 
with and centred on the glenoid in 0° of version and inclination. 
With the surface of the glenoid being fixed in all degrees of free-
dom during the simulation, the boundary conditions involved 
the application of a 50 N compressive force through the centre 
of the humeral head, perpendicular to the plane of the articu-
lating surface of the glenoid in order to simulate the loading 
used in the cadaveric studies. Under permanent compression, 
the humeral head was subsequently translated towards the ante-
rior rim of the glenoid until dislocation occurred. The maximum 
force required to dislocate the shoulder was defined as the dislo-
cation force. These forces were determined for the intact glenoid 
and after introducing defects of 8%, 16%, and 24% of the length 
of the glenoid to validate the results of modelling against in vitro 
measurements, as reported in the literature.5
The critical size of a defect was investigated during 30 
functional activities by determining the dislocation forces 
of the intact glenoid and after creating defects involving 4%, 
8%, 12%, 16%, 20%, and 24% of the length of the glenoid, 
respectively. For each activity, the forces were determined in 
Table III. Anterior dislocation forces of the intact glenoid and after creating defects of 4%, 8%, 12%, 16%, 
20%, and 24% of the length of the glenoid, and the in vivo anterior shear force for each functional task. The 
forces were tested in the position of the joint which was the most susceptible to instability
Anterior dislocation forces, N In vivo shear 
force, N
Intact 4% 8% 12% 16% 20% 24%
Quarterback throw 632 595 557 522 489 397 294 516
Lineout throw 453 415 393 366 342 274 220 359
Extreme (reach across body) 103 95 89 84 78 64 49 81
Clean upper back 98 90 85 81 74 60 46 79
Volleyball serve 324 294 266 241 210 153 112 201
Slow abduction 180 162 148 132 119 87 61 112
Fast abduction 174 161 144 128 114 85 54 108
Reach back of head 110 106 102 96 90 74 57 84
Basketball free throw 278 254 236 215 196 142 116 132
Sit to stand 437 401 369 313 281 193 120 171
Pull 86 82 78 73 68 53 41 48
Push 85 80 76 72 67 54 41 46
Lift block to head height 204 190 176 159 147 106 89 98
Lift block to shoulder height 197 184 171 155 141 98 83 93
Reach opposite axilla 82 77 72 67 63 51 44 47
Lift shopping bag from floor 180 171 160 149 137 99 83 80
Lift shopping bag on lap 212 197 183 171 158 127 86 83
Perineal care 108 103 98 91 85 71 58 51
Reach far ahead 190 175 163 151 139 112 74 66
Brush left side of head 109 105 101 96 90 74 57 52
Drive slow left 109 105 100 94 88 71 56 51
Drive slow right 111 108 103 96 91 75 60 49
Drive fast left 102 97 9185 79 65 49 43
Drive fast right 99 95 89 84 79 65 49 43
Slow flexion 202 190 178 163 149 118 77 68
Fast flexion 189 180 167 151 138 107 72 66
Eat with hand 85 80 76 71 65 52 44 39
Drink from mug 83 79 75 70 63 51 42 37
72 C. KLEmT, D. TODERITA, D. NOLTE, E. DI FEDERICO, P. REILLY, A. m. J. BULL 
Follow us @BoneJointJ THE BONE & JOINT JOURNAL
the position of the joint that was most susceptible to instability 
as quantified in a previous study,21 where the ratio of shear-to-
joint compression force was maximal. The dislocation forces 
for each activity were also determined under in vivo loading 
conditions that were quantified previously, when in vivo con-
tact forces were analyzed during 26 tasks.21 For the purpose of 
this study, four throwing activities were added using the same 
experimental approach as described by Klemt et al.21
In order to quantify the critical size of a defect that leads 
to persistent instability despite a Bankart repair, the disloca-
tion forces for each ADL as obtained by the FE models were 
compared with in vivo shear forces during these activities, as 
previously quantified by the UK National Shoulder Model (UK 
NSM) (Fig. 2).21 Based on the comparison of in vivo shear force 
with dislocation forces for each task, each ADL was classified 
for each size of defect as stable, when the dislocation force of 
the glenoid is larger than the in vivo shear force of the ADL, or 
unstable, when the dislocation force is smaller than the shear 
force of the ADL. Finally, the critical size of a defect was 
quantified as the size of the smallest defect that leads to anterior 
dislocation in all ADLs.
Results
In agreement with cadaveric experiments, the dislocation forces 
determined by the male and female FE models showed reduced 
stability of the shoulder with increasing size of defect in the 
glenoid.4 The dislocation forces quantified by the models com-
pare well with those experimentally measured with predictions 
being within one standard deviation of the experimental values 
(Fig. 3).
The classification of ADLs with respect to stability for each 
size of defect is shown in Table II, with the corresponding dis-
location forces and joint positions that yielded this classification 
being shown in Tables III and IV. As can be seen in Table II, the 
critical size of a defect that leads to persistent instability under 
physiological loading despite a Bankart repair was predicted by 
the models to be 16% of the length of the glenoid.
There were four loading conditions with high anterior shear 
forces: a quarterback throw, a lineout throw, reaching across 
the body, and cleaning the upper back. These show a high like-
lihood of instability, with a defect of 16% of the length of the 
glenoid (Table II). This is due to in vivo shear forces for these 
tasks that exceed the anterior dislocation forces of the glenoid 
with a defect corresponding to 16% of its length (Table III). 
There are four functional daily activities with large ranges of 
abduction and external rotation: slow abduction, fast abduction, 
reaching the back of the head, and a volleyball serve. These 
show a great risk of instability with a defect of 20% of the 
length of the glenoid (Tables II and III). Most ADLs such as 
eating, drinking, reaching opposite axilla, flexion, driving, and 
a basketball throw only affect the stability for defects of more 
than 24% of the length of the glenoid (Table II). These activities 
involve less abduction and external rotation and loading with 
anterior in vivo shear forces for these tasks not exceeding the 
dislocation forces for defects of more than 24% of the length of 
the glenoid (Tables II, III, and IV).
Discussion
In this study, two FE models of the shoulder with anterior gle-
noid osseous defects of different magnitudes were validated 
against in vitro measurements of anterior stability of the shoul-
der.4 The dislocation forces predicted by the models were within 
one standard deviation of experimentally measured values. The 
validated FE models quantified the critical size of a defect 
that leads to persistent instability under physiological loading 
despite a Bankart repair, to be 16% of the length of the glenoid. 
These findings suggest that if there is a defect of this size, bone 
grafting should be considered rather than solely a soft-tissue 
procedure, to optimize stability.
Previous authors have shown that slightly larger defects 
of between 19% and 21% of the length of the glenoid were 
required to produce significant loss of stability.3-5 These results 
have been used as an indication for bone grafting to restore sta-
bility,10 despite the in vitro experiments being conducted under 
low non-physiological loading. As the stability of the shoul-
der is load dependent, with higher forces leading to a loss of 
stability,12,13 the results of the FE models strongly suggest that 
Table IV. The positions of the shoulder that served to determine the 
dislocation forces. These positions were shown to be most susceptible 
to instability due to the largest ratio of shear-to-joint compression force. 
The angles represent forward flexion (positive sign indicates forward 
flexion), abduction (positive sign indicates abduction), and rotation 
(positive sign indicates internal rotation)
Functional activity Shoulder position, °
Flexion Abduction Rotation
Quarterback throw 94 71 -39
Lineout throw 75 -61 -53
Extreme (reach across 
body)
111 60 -48
Clean upper back -45 49 -41
Volleyball serve 107 54 -37
Slow abduction 120 83 -74
Fast abduction 102 70 -71
Reach back of head 67 -44 -56
Basketball free throw 86 59 -47
Sit to stand -12 -14 -4
Pull 12 8 -9
Push 21 3 -12
Lift block to head height 70 27 -32
Lift block to shoulder 
height
52 22 -24
Reach opposite axilla 60 23 23
Lift shopping bag from 
floor
-7 -12 -14
Lift shopping bag on lap 29 -9 -22
Perineal care -19 -5 8
Reach far ahead 76 24 -27
Brush left side of head 60 16 19
Drive slow left 39 -2 -19
Drive slow right 27 2 -14
Drive fast left 33 -3 -17
Drive fast right 20 -4 -10
Slow flexion 114 42 -37
Fast flexion 98 39 -35
Eat with hand 34 -7 -18
Drink from mug 24 -18 -22
 THE CRITICAL SIZE OF A DEFECT IN THE GLENOID CAUSING ANTERIOR INSTABILITY OF THE SHOULDER 73
VOL. 101-B, No. 1, JANUARY 2019
slightly smaller defects of 16% of the length of the glenoid lead 
to persistent instability under physiological loading despite a 
Bankart repair. Similar findings have recently been reported.22,23 
These in vitro studies showed significant changes in function 
of the shoulder following the creation of defects correspond-
ing to between 13.5% and 15% of the length of the glenoid. 
When considering the clinical relevance of this study, it should 
be noted that recurrent instability commonly presents in rela-
tion to sport rather than during ADLs.24,25 While we included 
sports involving throwing to identify the critical size of a defect 
that leads to persistent instability under physiological loading 
despite a Bankart repair, we acknowledge that the further inclu-
sion of contact and impact sports might suggest a smaller crit-
ical size of defect due to the increased loads on the shoulder 
during these activities. However, such data are not available 
due to the difficulty in measuring the load on the shoulder dur-
ing these activities. Previous studies have not provided details 
of anterior shear forces in the shoulder during tackles in con-
tact sports.26 When considering the clinical significance of this 
study, it should be noted that the rate of recurrent dislocation 
is lower after bone grafting compared with a Bankart repair6,7, 
and performing bone grafting on patients with smaller osse-
ous defects will reduce the rate of further recurrent instability. 
However, bone grafting involves increased invasiveness, com-
plication rates, and longer rehabilitation.8,27 As we found a high 
likelihood of instability for the glenoid with a defect of 16% 
of its length during activitieswith large anterior shear forces 
at increased abduction and external rotation, slightly smaller 
defects may be considered suitable for grafting in patients with 
high functional demands. The amount of glenoid bone that 
is lost can be determined from CT scans with an accuracy of 
less than 3% of its length,28 thus our findings may help deci-
sion making by suggesting that slightly smaller defects of 16% 
of the length of the glenoid should be reconstructed to ensure 
restoring stability. As we did not investigate Hill–Sachs lesions, 
the findings only represent anterior defects in the absence of 
these lesions.
A novel aspect of this study is the use of loading the shoul-
der with in vivo compressive and shear contact forces that have 
been quantified by musculoskeletal modelling for a wide range 
of ADLs.21,29 The comparison of these forces with the forces 
that are predicted to induce dislocation allows a mechanical 
assessment of stability through the classification of each ADL 
as either stable with a force of the glenoid being larger than the 
in vivo shear force of the ADL, or unstable with a force of the 
glenoid being smaller than the force of the ADL, for each size of 
defect. This mechanical assessment of stability to quantify the 
critical size of the defect that leads to further instability despite 
a Bankart repair represents the first approach of its kind, with 
previous in vitro studies3-5 using statistically significant changes 
in stability as predictors of the critical size of the defect.
The study has limitations. The anatomy of the defects in 
the glenoid was idealized as being parallel to the longitudinal 
axis, in order to replicate in vitro experiments. This does not 
represent all in vivo defects. It has, however, been shown that 
the margin of a defect is usually linear and most defects are 
located at the anterior rim of the glenoid.30 Also, the defects 
were created in increments of 4% of the length of the glenoid, 
thus representing a different proportion of loss for scapulae of 
different sizes.31 This was partially taken into account by using 
one large and one small scapula. The dislocation forces for each 
ADL were also determined in the position most susceptible to 
instability, without considering movement of the humeral head 
in different positions of the arm. As the humeral head remains 
mainly centred on the glenoid fossa with translations of only a 
few millimetres during ADLs, the effect of this simplification is 
assumed to be small.
The stability provided by the bony articulating structures is 
only one stabilizing mechanism in the shoulder. Other compo-
nents, such as capsuloligamentous structures, need to be consid-
ered in future studies. These results therefore only represent the 
mid-range of movement of the shoulder, where the capsuloliga-
mentous structures are lax. Despite these limitations, this is the 
first study to determine the critical size of an anterior defect of 
the glenoid that leads to further instability of the shoulder under 
physiological loading conditions despite a Bankart repair. These 
findings suggest reasons why instability persists in some 
patients who have had soft-tissue stabilization only. We propose 
that bone grafting should be used for defects of the glenoid of 
16% of its length.
Take home message
- This is the first study to assess the critical size of an anterior 
osseous defect in the glenoid that leads to further anterior 
dislocation of the shoulder under physiological loading dur-
ing daily activities.
- We identified the critical size of the lesion that necessitates bone graft-
ing to restore stability.
- Based on a large number of daily activities and sporting activities, these 
findings help to explain why instability of the shoulder is seen in the pres-
ence of lesions involving more than 16% of the length of the glenoid.
References
	 1.	 Hovelius	L,	Eriksson	K,	Fredin	H,	et	al. Recurrences after initial dislocation of 
the shoulder. Results of a prospective study of treatment. J Bone Joint Surg [Am] 
1983;65-A:343–349. 
	 2.	 Norlin	R. Use of Mitek anchoring for Bankart repair: a comparative, randomized, pro-
spective study with traditional bone sutures. J Shoulder Elbow Surg 1994;3:381–385. 
	 3.	 Itoi	E,	Lee	SB,	Berglund	LJ,	Berge	LL,	An	KN. The effect of a glenoid defect on 
anteroinferior stability of the shoulder after Bankart repair: a cadaveric study. J Bone 
Joint Surg [Am] 2000;82-A:35–46. 
	 4.	 Yamamoto	N,	Itoi	E,	Abe	H,	et	al. Effect of an anterior glenoid defect on anterior 
shoulder stability: a cadaveric study. Am J Sports Med 2009;37:949–954. 
	 5.	 Yamamoto	 N,	 Muraki	 T,	 Sperling	 JW,	 et	 al. Stabilizing mechanism in bone- 
grafting of a large glenoid defect. J Bone Joint Surg [Am] 2010;92-A:2059–2066. 
	 6.	 Boileau	P,	Villalba	M,	Héry	JY,	et	al. Risk factors for recurrence of shoulder insta-
bility after arthroscopic Bankart repair. J Bone Joint Surg [Am] 2006;88-A:1755–1763. 
	 7.	 Chen	 AL,	 Hunt	 SA,	 Hawkins	 RJ,	 Zuckerman	 JD. Management of bone loss 
associated with recurrent anterior glenohumeral instability. Am J Sports Med 
2005;33:912–925. 
	 8.	 Kim	SH,	Ha	KI,	Kim	SH. Bankart repair in traumatic anterior shoulder instability: 
open versus arthroscopic technique. Arthroscopy 2002;18:755–763. 
	 9.	 Amirthanayagam	TD,	Emery	RJ. Primum non nocere: risk assessment for new sur-
gical procedures. J Shoulder Elbow Surg 2014;23:1417–1418. 
	10.	 Bigliani	LU,	Kelkar	R,	Flatow	EL,	Pollock	RG,	Mow	VC. Glenohumeral stabil-
ity. Biomechanical properties of passive and active stabilizers. Clin Orthop Relat Res 
1996;330:13–30. 
	11.	 Bergmann	 G,	 Graichen	 F,	 Bender	 A,	 et	 al. In vivo glenohumeral contact 
 forces--measurements in the first patient 7 months postoperatively. J Biomech 
2007;40:2139–2149. 
	12.	 Halder	AM,	Kuhl	SG,	Zobitz	ME,	Larson	D,	An	KN. Effects of the glenoid labrum 
and glenohumeral abduction on stability of the shoulder joint through concavity-com-
pression: an in vitro study. J Bone Joint Surg [Am] 2001;83-A:1062–1069. 
74 C. KLEmT, D. TODERITA, D. NOLTE, E. DI FEDERICO, P. REILLY, A. m. J. BULL 
Follow us @BoneJointJ THE BONE & JOINT JOURNAL
	13.	 Klemt	C,	Nolte	D,	Grigoriadis	G,	et	al. The contribution of the glenoid labrum to 
glenohumeral stability under physiological joint loading using finite element analy-
sis. Comput Methods Biomech Biomed Engin 2017;20:1613–1622. 
	14.	 Ackerman	 MJ. The Visible Human Project: from body to bits. IEEE Pulse 
2017;8:39–41. 
	15.	 Wu	G,	van	der	Helm	FCT,	Veeger	HEJ,	et	al. ISB recommendation on definitions 
of joint coordinate systems of various joints for the reporting of human joint motion–
Part II: shoulder, elbow, wrist and hand. J Biomech 2005;38:981–992. 
	16.	 Smith	CD,	Masouros	SD,	Hill	AM,	et	al. Tensile properties of the human glenoid 
labrum. J Anat 2008;212:49–54. 
	17.	 Gatti	CJ,	Maratt	JD,	Palmer	ML,	Hughes	RE,	Carpenter	JE. Development and 
validation of a finite element model of the superior glenoid labrum. Ann Biomed Eng 
2010;38:3766–3776. 
	18.	 Buchler	P,	Ramaniraka	NA,	Rakotomanana	LR,	Iannotti	JP,	Farron	A. A finite 
element model of the shoulder: application to the comparison of normal and osteoar-
thritic joints. Clin Biomech (Bristol, Avon) 2002;17:630–639. 
	19.	 Terrier	 A,	 Reist	 A,	 Vogel	 A,	 Farron	 A. Effect of supraspinatus deficiency on 
humerus translation and glenohumeral contact force during abduction. Clin Biomech 
2007 22:645–651.
	20.	 Quapp	KM,	Weiss	JA. Material characterization of human medial collateral liga-
ment. J Biomech Eng 1998;120:757–763. 
	21.	 Klemt	 C,	 Prinold	 JA,	 Morgans	 S,	 et	 al. Analysis of shoulder compressive and 
shear forces during functional activities of daily life. Clin Biomech (Bristol, Avon) 
2018;54:34–41. 
	22.	 Shin	SJ,	Ko	YW,	Scott	J,	McGarry	MH,	Lee	TQ. The effect of defect orientation 
and size on glenohumeral instability: a biomechanical analysis. Knee Surg Sports 
Traumatol Arthrosc 2016;24:533–539.23.	 Shin	SJ,	Kim	RG,	Jeon	YS,	Kwon	TH. Critical value of anterior glenoid bone loss 
that leads to recurrent glenohumeral instability after arthroscopic Bankart repair. Am 
J Sports Med 2017;45:1975–1981. 
	24.	 Mahure	SA,	Mollon	B,	Capogna	BM,	et	al. Risk factors for recurrent instability or 
revision surgery following arthroscopic Bankart repair. Bone Joint J 2018;100-B:324–330. 
	25.	 Olds	M,	Ellis	R,	Donaldson	K,	Parmar	P,	Kersten	P. Risk factors which predis-
pose first-time traumatic anterior shoulder dislocations to recurrent instability in 
adults: a systematic review and meta-analysis. Br J Sports Med 2015;49:913–922. 
	26.	 Usman	J,	McIntosh	AS,	Fréchède	B. An investigation of shoulder forces in active 
shoulder tackles in rugby union football. J Sci Med Sport 2011;14:547–552. 
	27.	 Owens	BD,	Harrast	JJ,	Hurwitz	SR,	Thompson	TL,	Wolf	JM. Surgical trends in 
Bankart repair: an analysis of data from the American Board of Orthopaedic Surgery 
certification examination. Am J Sports Med 2011;39:1865–1869. 
	28.	 Bois	AJ,	Fening	SD,	Polster	J,	Jones	MH,	Miniaci	A. Quantifying glenoid bone 
loss in anterior shoulder instability: reliability and accuracy of 2-dimensional and 
3-dimensional computed tomography measurement techniques. Am J Sports Med 
2012;40:2569–2577. 
	29.	 Charlton	IW,	Johnson	GR. A model for the prediction of the forces at the gleno-
humeral joint. Proc Inst Mech Eng H 2006;220:801–812. 
	30.	 Griffith	JF,	Antonio	GE,	Yung	PSH,	et	al. Prevalence, pattern, and spectrum of 
glenoid bone loss in anterior shoulder dislocation: CT analysis of 218 patients. AJR 
Am J Roentgenol 2008;190:1247–1254. 
	31.	 Ohl	X,	Billuart	F,	Lagacé	PY,	et	al. 3D morphometric analysis of 43 scapulae. Surg 
Radiol Anat 2012;34:447–453. 
Author contributions:
C. Klemt: Designed the study, Collected, analyzed, and interpreted the data, 
Wrote the manuscript.
D. Toderita: Collected and analyzed the data.
D. Nolte: Analyzed the data.
E. Di Federico: Critically reviewed the manuscript.
P. Reilly: Designed the study, Interpreted the results, Critically reviewed the 
manuscript.
A. M. J. Bull: Designed the study, Interpreted the results, Critically reviewed 
the manuscript.
Funding statement:
This work was supported by [Engineering and Physical Research Council, 
JRI Orthopaedics] grant number [EP/M507878/1].
No benefits in any form have been received or will be received from a com-
mercial party related directly or indirectly to the subject of this article.
Disclosure statement:
Supporting data are available on request by contacting the corresponding 
author.
This article was primary edited by J. Scott.