J Shoulder Elbow Surg (2021) -, 1–9

www.elsevier.com/locate/ymse

Changes in superior displacement, angulation,

and shortening in the early phase of healing for
completely displaced midshaft clavicle fractures
in adolescents: results from a prospective,
multicenter study
Andrew T. Pennock, MDa,*, Benton E. Heyworth, MDb, Tracey Bastrom, MAa,
Donald S. Bae, MDb, Kelly E. Boutelle, BSa, Michael T. Busch, MDc,
Eric W. Edmonds, MDa, Henry B. Ellis, MDd, Katelyn Hergott, MPHe,
Mininder S. Kocher, MD, MPHb, Ying Li, MDf, Elizabeth S. Liotta, MBBSe,
Nirav K. Pandya, MDg, Crystal Perkins, MDc, Coleen S. Sabatini, MD, MPHg,
David D. Spence, MDh, Samuel C. Willimon, MDc, Philip L. Wilson, MDd, FACTS Study
Group1, Jeffrey J. Nepple, MDi

a
Rady Children’s Hospital, San Diego, CA, USA
b
Department of Orthopaedic Surgery, Boston Children’s Hospital, Boston, MA, USA
c
Children’s Healthcare of Atlanta, Atlanta, GA, USA
d
Texas Scottish Rite Hospital for Children, Dallas, TX, USA
e
Division of Sports Medicine, Boston Children’s Hospital, Boston, MA, USA
f
C.S. Mott Children’s Hospital, Ann Arbor, MI, USA
g
UCSF Benioff Children’s Hospital–Oakland, Oakland, CA, USA
h
Campbell Clinic Orthopaedics, Memphis, TN, USA
i
Washington University Orthopedics, St. Louis, MO, USA

Background: Progressive displacement of diaphyseal clavicle fractures has been observed in adult patients, at times necessitating a
change from nonoperative to operative treatment. Whether this occurs in adolescent patients has not been well investigated. The purpose
of this study was to assess the rate and extent of progressive clavicle fracture displacement in adolescent patients following injury and
during the early stages of healing.
Methods: This was a multicenter study evaluating prospective data that had previously been collected as part of a larger study evalu-
ating the functional outcomes of adolescent clavicle fractures. A consecutive series of completely displaced diaphyseal clavicle fractures

Institutional review board approval was received through the University of
California, San Diego Human Research Protections Program (project no.
121458), as well as the Boston Children’s Hospital Institutional Review
Board (protocol no. P00004816).
1
This study was conducted at Rady Children’s Hospital, San Diego, CA,
and collaboratively among all sites affiliated with the FACTS (Function
After Adolescent Clavicle Trauma and Surgery) Study Group with Boston
Children’s Hospital, Boston, MA.
*Reprint requests: Andrew T. Pennock, MD, Rady Children’s Hospital,
3020 Children’s Way, MC 5062, San Diego, CA 92123, USA.
E-mail address: apennock@rchsd.org (A.T. Pennock).

1058-2746/$ - see front matter Ó 2021 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved.
https://doi.org/10.1016/j.jse.2021.05.006
2 A.T. Pennock et al.

in patients aged 10-18 years treated at 1 of 3 tertiary-care pediatric trauma centers was included; all fractures underwent standardized
imaging within 2 weeks of the date of injury and during the course of healing (5-20 weeks after injury). Measurements of clavicle short-
ening, superior displacement, and angulation were performed using validated techniques. Progressive displacement and/or interval
improvement in fracture alignment, as well as the subsequent need for surgical intervention, was noted. Patient demographic and radio-
graphic parameters were assessed as possible risk factors for interval displacement.
Results: One hundred patients met the inclusion criteria. Mean end-to-end shortening, cortex-to-cortex shortening, superior displace-
ment, and angulation at the time of injury were 24 mm, 15 mm, 15 mm, and 7
, respectively. At a mean of 10 weeks after injury, the
fracture alignment improved across all 4 measurements for the overall cohort, with mean improvements of 3.5 mm in end-to-end short-
ening, 3.3 mm in cortex-to-cortex shortening, 2.1 mm in superior displacement, and 2
in angulation. By use of a clinical threshold of a
change in shortening or displacement of 10 mm or change in angulation of 10
, 26% of fractures improved, 4% worsened, and 70%
remain unchanged. Patients with more severe fractures were more likely to have improved alignment than were patients with less dis-
placed fractures (P < .001). No patient underwent surgical intervention for progressive displacement.
Conclusion: Significant early improvements in fracture alignment were observed in a substantial percentage of adolescent patients with
completely displaced clavicle fractures. Among the most severely displaced fractures, shortening improved approximately 6 mm and
angulation improved approximately 9
. In 4% of cases, increased displacement was observed, but this tended to be mild, and in no
cases did it prompt surgical intervention. This finding indicates that the true final deformity after an adolescent clavicle fracture is
commonly less than that present at the time of injury.
Level of evidence: Level IV; Case Series; Treatment Study
Ó 2021 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved.
Keywords: Adolescent; clavicle fracture; progressive displacement; shortening; remodeling; settling

Clavicle fractures account for 10%-15% of all skeletal
injuries in children, and the incidence in adolescents is
twice that in any other age group, likely owing to more
involvement in sports and other high-risk activities.10,18
Roughly 90% of clavicle fractures in adolescents are
diaphyseal, and a recent descriptive epidemiology study
revealed that over half of these (54%) will be completely
displaced with mean fracture shortening of 12-22 mm.5,17
Prior to the turn of the century, operative treatment of
this injury was uncommon in both children and adults;
Kubiak and Slongo10 reported an operative rate of 0.8% for
pediatric diaphyseal clavicle fractures over a span of 21
years. More recently, a number of randomized studies in the
adult population have suggested that operative treatment of
displaced, shortened fractures can decrease the rate of
nonunion and symptomatic malunion.3,19,21 Subsequently,
the rate of operative treatment of adolescent clavicle frac-
tures has increased substantially, largely owing to extrap-
olation from the adult data as well as parent and/or child
demand, despite a paucity of evidence-based support spe-
cific to the adolescent population.4,22
Despite this trend, current literature suggests that the
majority of adolescent clavicle fractures can be treated
nonoperatively with good functional outcomes, a high rate
of return to sport, and a low incidence of complications
such as nonunion and symptomatic malunion. Unlike adult
fractures, which have a symptomatic malunion rate be-
tween 10% and 15% with nonoperative treatment, signifi-
cant functional deficits have not been noted in adolescent
patient populations.2,6,20 There are several potential reasons
these younger patients may recover and perform better than
their adult counterparts after displaced fractures. First, the
adolescent population might be more resilient to this injury
and can better tolerate and adapt to the altered mechanics
associated with a malunited clavicle. Second, residual
growth in the clavicle, whose growth plates have been
shown to remain open up to age 25 years, may allow bony
remodeling.7,9,14 Third, ‘‘settling’’ of the fracture may
occur during the early phases of healing, which may make
an initial injury pattern that looks quite severe less dra-
matic as the fracture heals. Recent literature in adult pa-
tients, however, may suggest that this third possibility is
unlikely, given a relatively high rate of progressive
displacement and surgical conversion in this older
population.13,16
The purpose of this study was to use prospectively
collected radiographic data from a multicenter study group
to evaluate the rate and extent of change of adolescent
clavicle fracture alignment during the early phases of
healing. Furthermore, we sought to quantify the specific
changes in radiographic alignment and to identify potential
risk factors for fracture displacement.
Materials and methods
This was a multicenter study using prospectively collected data,
which were acquired as part of a larger study group evaluating the
functional outcomes of adolescent clavicle fractures. The current
study focused on a consecutive series of patients from 3 different
geographically distinct pediatric trauma centers. The inclusion
criteria included adolescent patients (aged 10-18 years) with
completely displaced diaphyseal clavicle fractures who had an
anteroposterior (AP) radiograph of the affected clavicle taken
within 14 days of their injury, as well as a second AP radiograph
taken during the intermediate phase of healing, when sufficient
advanced callus was seen traversing the fracture site, so as to
Displaced adolescent clavicle fracture alignment changes 3

stabilize the fracture. These radiographs were typically obtained

6-12 weeks after injury, but the timing ranged from 5 to 20 weeks
after injury. This period for radiographic assessment was selected
to minimize the chance of any further fracture displacement or
migration but was prior to the onset of fracture remodeling known
to occur in the pediatric population. Patients were excluded if the
injury or ‘‘healing’’ radiographs did not include the entire length
of the clavicle from the sternoclavicular joint to the acromiocla-
vicular joint, if the radiographs were malrotated, or if different
levels of radiographic magnification were observed between the 2
radiographs. Patients with fracture malrotation or magnification
differences were defined as those in whom the width of the Figure 1 Radiographic measurements of end-to-end shortening in
clavicle at the level of the coracoid differed by >3 mm when the millimeters (top dashed line), cortex-to-cortex shortening in milli-
initial injury radiograph was compared with the subsequent meters (bottom dashed line), superior displacement in millimeters
healing radiograph. Patients who underwent an initial attempt at (thin solid line), and angulation in degrees (thick solid lines).
nonoperative care but whose fracture subsequently displaced on
follow-up imaging and required surgery were included in the
study so as not to underestimate the potential phenomenon of
Table I Demographic and radiographic data for 100
displacement worsening. Patient demographic data, including age,
adolescent clavicle fractures
sex, arm dominance, fracture side, and primary sport, when
applicable, were documented. The specific aspects of the nonop- Characteristic Data
erative treatment approach were left to the discretion of the Patient cohort, N 100
treating physician at each of the 3 institutions but typically con- Age, yr 14.3  1.9
sisted of relative immobilization with a simple sling for 2-3 Male sex 77 (77)
weeks, followed by intermittent sling use between 2-6 weeks after Right-sided fracture 43 (43)
injury. Typically, range-of-motion exercises were instituted by 2-4 Dominant arm 42 (42)
weeks, followed by progressive strengthening exercises. Open proximal humeral physis 78 (78)
Comminution 28 (28)
Transverse fracture configuration 19 (19)
Measurement technique Initial end-to-end shortening, mm 24.2  8.2
Initial cortex-to-cortex shortening, mm 14.6  7.3
As part of the overall study protocol, 2 upright views of the clavicle at Initial superior displacement, mm 14.6  5.2
each post-injury time point were reviewed for clinical decision Initial angulation,
7  10
making and classification, which included an AP view and a 15
-45

Data are presented as mean  standard deviation or number
AP cephalad-tilt view. Because of variability in beam tilt across pa- (percentage).
tients, time points, and institutions, it was believed that reproducible
and reliable measurements for the purposes of this study could not be
performed on these cephalad-angled projections. Therefore, all
measurements used for this study were made exclusively on the AP
radiograph. Fractures were classified in 2 ways: (1) according to the fractures. The cortex-to-cortex measurement was made from the
AO classification for diaphyseal fractures and (2) according to a fracture tip of the major medial fragment to the corresponding
descriptive classification emphasizing displacement, consisting of defect, or lucency, on the major lateral fragment or vice versa,
either (a) nondisplaced or minimally displaced, (b) partially dis- depending on what provided the closest approximation of ‘‘true’’
placed or angulated (in which at least 1 cortex was disrupted but at shortening. To allow for substratified analyses, thresholds for the
least 1 set of corresponding cortices remained in contact), or (c) ‘‘most severe’’ fracture patterns were established a priori, based
completely displaced (in which there was no cortical contact between on the injury radiographs. The ‘‘most severely’’ shortened, su-
fragments).8 For the purposes of the current study, only cases from the periorly displaced, and angulated fractures were defined by
completely displaced cohort were included. Additionally, the prox- measurements > 25 mm, > 15 mm, and > 10
, respectively. All
imal humeral physis was assessed as either open or closed. measurements were made using standardized digital measurement
A single surgeon from each of the 3 institutions from which tools. These measurements have been previously validated by Li
radiographs were used for the current study reviewed each image et al.11
from the surgeon’s respective institution to minimize variability.
The following 4 measurements were obtained at the time of
injury, as well as at the time of healing: end-to-end fracture Statistical analysis
shortening (in millimeters), cortex-to-cortex shortening (in milli-
meters), superior fracture displacement (in millimeters), and Means, standard deviations, and ranges were calculated for all
fracture angulation (in degrees) (Fig. 1). The end-to-end mea- continuous variables. These variables were also assessed for
surement was performed by measuring the distance between the normal distributions and homogeneity of variance prior to appli-
lateral-most tip of the major medial fragment and the medial-most cation of parametric statistical analyses. Pre-to-post comparisons
tip of the major lateral fragment, with appropriate adjustments in for the entire cohort were performed with the paired-sample t test.
length measurements for segmental fragments of comminuted Comparisons between groups of initial fracture severity or
4 A.T. Pennock et al.

Table II Fracture shortening, angulation, and superior displacement data for 100 adolescent clavicle fractures
Injury radiograph Healing radiograph Interval D P value
End-to-end shortening, mm 24.2  8.2 20.7  6.7 3.5  4.4 <.001
Cortex-to-cortex shortening, mm 14.6  7.3 11.4  5.5 3.3  4.7 <.001
Superior displacement, mm 14.6  5.2 12.4  4.5 2.1  3.8 <.001
Angulation,
7  10 6  7 2  8 .055
Data are presented as mean  standard deviation.

Figure 2 Distribution of the change in fracture end-to-end shortening (A) and cortex-to-cortex shortening (B) from the time of injury to
the time of fracture healing for the entire cohort of 100 patients. Positive values represent an improvement in fracture shortening, whereas
negative values represent a worsening.
Displaced adolescent clavicle fracture alignment changes
Figure 3 Representative example of a 13-year-old female pa-
tient with a shortened clavicle fracture (26.0 mm) (dashed line, A)
that improved by 9.8 mm (dashed line, B) during the early phases
of healing. Given the comminution, the patient’s cortex-to-cortex
and end-to-end shortening were the same.
magnitude of change in radiographic appearance were performed
with 1-way analysis of variance. Categorical variables were
compared by the c2 or Fisher exact test. Alpha was set at P  .05
to declare significance on the univariate analyses. The identifica-
tion of predictors of significant radiographic change (>10
or >10
mm) was analyzed using binary forward stepwise logistic
regression. The level of probability for model entry was set at .05,
and the level of probability for removal was set at .10. Analyses
were conducted using IBM SPSS Statistics for Windows (version
26.0 [2019 release]; IBM, Armonk, NY, USA).
Results
In the prospective database, 102 consecutive fractures were
initially identified, of which 100 fractures in 100 patients
met our inclusion and exclusion criteria. The patient
demographic data and initial fracture pattern data are
presented in Table I. The initial injury radiograph was taken
at a mean of 1.2  2.8 days after injury, and the healing
radiograph was taken at a mean of 10.9  4.8 weeks after
injury. In the study cohort, no patients underwent conver-
sion from nonoperative management to surgical interven-
tion owing to fracture displacement.
Shortening
At the time of the healing radiograph, end-to-end fracture
shortening improved by a mean of 3.5  4.4 mm (range, 5.3
mm of worsening to 15.6 mm of improvement), representing
an overall reduction in fracture shortening by 15% (P < .001).
5
Similarly, cortex-to-cortex shortening improved by a mean of
3.3  4.7 mm (range, 5.8 mm of worsening to 18.1 mm of
improvement), representing an overall reduction in fracture
shortening by 22% (P < .001) (Table II). The distribution of
change in shortening is depicted in Figure 2. Twelve percent of
the cohort was noted to have a change in shortening > 10 mm,
with the changes in all of these cases leading to an improve-
ment in the overall alignment of the fracture. Another 22% of
the cohort was noted to have a change in shortening > 5 mm
but < 10 mm, and nearly all of these changes (21 of 22)
favored an improvement in alignment (Fig. 3). Among the
most severely displaced fractures (end-to-end shortening > 25
mm, n ¼ 44), a greater mean improvement in fracture short-
ening was observed (5.9  4.6 mm) relative to the less
shortened fractures (1.6  3.1 mm, P < .001).
Superior displacement
At the time of the healing radiograph, superior displacement
had improved by a mean of 2.1  3.8 mm (range, 6.4 mm of
worsening to 18.5 mm of improvement), representing an
overall reduction in displacement by 15% (P < .001) (Table
II). The distribution of change in superior displacement is
depicted in Figure 4. Two percent of the cohort was noted to
have a change in superior displacement > 10 mm, leading to
improvement in the overall alignment of the fracture (Fig. 5).
Another 21% of the cohort was noted to have a change in
superior displacement > 5 mm but < 10 mm, and nearly all of
these changes (18 of 21) favored an improvement in align-
ment. Among the most severely superiorly displaced fractures
(>15 mm, n ¼ 39), a greater mean improvement in superior
displacement was observed (4.0  3.4 mm) compared with the
less severely displaced fractures (1.0  3.4 mm, P < .001).
Angulation
At the time of healing, the overall fracture angulation of the
cohort improved by a mean of 2
 8
, corresponding to an
overall reduction in angulation by 21% (P ¼ .055) (Table
II). The distribution of change in fracture angulation is
depicted in Figure 6. Seventeen percent of the cohort was
noted to have a change in fracture angulation > 10
, and in
the majority of these patients (76%), this led to an
improvement in the overall alignment of the fracture
(Fig. 6). Among the most severely angulated fractures (n
¼ 31), a significantly greater mean improvement in angu-
lation of 9
 8
was observed when compared with the
less angulated fractures (–1
 5
, P < .001) (Fig. 7).
Risk factors for fracture alignment change
Overall, 75% of the cohort did not experience a significant
change (>10
or >10 mm) in the radiographic variables
(shortening, vertical displacement, and/or angulation).
Among the 25% of the cohort that did have a significant
6 A.T. Pennock et al.
Figure 4 Distribution of the change in superior fracture displacement from the time of injury to the time of fracture healing for the entire
cohort of 100 patients. Positive values represent an improvement in superior displacement, whereas negative values represent a worsening.
Figure 5 Representative example of a 14-year-old female pa-
tient with 20.7 mm of superior displacement of the clavicle
fracture (solid line, A) that improved by 5.2 mm (solid line, B)
during the first 6 weeks of healing.
change, 21 of 25 patients (84%) showed an overall
improvement in the fracture alignment whereas 4 of 25
(16%) represented a worsening in the overall alignment.
Changes in fracture alignment were not associated with any
specific demographic variables (sex, age, fracture side, or
arm dominance) or fracture comminution but were associ-
ated with the initial fracture shortening and angulation
(P < .05), with the more displaced fractures undergoing
greater improvements in alignment during the healing
process (Table III).
Discussion
In this series of adolescent patients with completely dis-
placed diaphyseal clavicle fractures, 25% of patients had
significant changes in fracture alignment during the early
phases of healing. An important finding was that the large
majority of these fractures with changing radiographic
parameters had interval improvement in their fracture
alignment. Among the most severely displaced fractures,
many of which would meet traditional adult parameters for
surgical treatment, fracture shortening, displacement, and
angulation all improved, by 6 mm, 4 mm, and 10
,
respectively. This finding is helpful in predicting the final
position of the clavicle at healing in the adolescent and has
not been previously reported. This observation may help
explain why adolescent patients seem to fare better after
nonoperative management than their adult counterparts,
even with severely displaced clavicle fractures.
To date, there are limited data in the literature quanti-
fying interval fracture displacement beyond the first week
or two from the patient’s injury. In 2011, Plocher et al16
were the first authors to draw attention to progressive
displacement of clavicle fractures in the early post-injury
period. Their study retrospectively identified 15 clavicle
fractures in patients with an average age of 35.5 years that
were displaced <2 cm and were all initially treated non-
operatively. During follow-up, serial radiographs, obtained
approximately 14 days after injury, showed progressive
horizontal shortening in 67% of these patients, averaging
14.3 mm, and progressive vertical displacement in 87%,
averaging 13.1 mm. All 15 patients went on to surgical
intervention, representing 27% of operative cases at the
authors’ institution over a 5-year period. More recently,
Onizuka et al15 prospectively evaluated the interval
Displaced adolescent clavicle fracture alignment changes 7

Figure 6 Distribution of the change in fracture angulation from the time of injury to the time of fracture healing for the entire cohort of
100 patients. Positive values represent an improvement in fracture angulation, whereas negative values represent a worsening.

Table III Demographic and radiographic variables associ-

ated with change in fracture alignment > 10 mm or > 10
from
time of initial injury radiograph to time of healing
No change in Change in P value
alignment alignment
(n ¼ 75) (n ¼ 25)
Age, yr 14.3  1.8 14.2  2.3 .714
Male sex, % 80 68 .270
Right-sided fracture, % 41 48 .560
Dominant arm, % 43 40 .812
Physis open, % 80 72 .403
Comminution, % 24 40 .123
Time to healing 10.9  5.2 11.0  3.2 .724
radiograph, weeks
Initial shortening, mm 23.1  7.2 27.5  10.3 .021*
Initial superior 14.2  4.7 15.6  6.4 .256
displacement, mm
Initial angulation,
58 14  12 <.001*
Data are presented as mean  standard deviation unless otherwise
indicated.
* Statistically significant (P  .05).
Figure 7 Representative example of a 15-year-old male patient
with 23
of clavicle fracture angulation (arrow, A) that improved
by 10
(arrow, B) 12 weeks after the injury.
study, in which significant worsening of overall alignment
displacement of 50 clavicle fractures in patients with a occurred in only 4% of the entire cohort and only 1% of the
mean age of 43 years with serial radiographs ranging from most severe fractures. We hypothesize that this difference is
3-9 weeks after injury. Their results showed progressive largely attributable to our mean patient age being much
displacement in 32%, with 18% of the cohort treated with younger, at 14.3 years.
surgery. Progressive displacement was associated with There are several possible explanations that may help
older patient age and ipsilateral girdle or chest wall injury. explain the improvements in fracture alignment that were
Their study did not report any cases in which the fracture observed during the early weeks of healing in our cohort of
alignment improved during the healing process. The results adolescent patients. First, at the time of the initial injury
of these 2 studies differ substantially from those of our radiograph, many patients are experiencing pain and will
8 A.T. Pennock et al.
undergo splinting of their extremity, which may increase the
fracture displacement and shortening. Specifically, regarding
superior migration, we hypothesize that the sternocleidomas-
toid muscle, as well as other muscles of the shoulder girdle, may
spasm immediately following injury and contribute to superior
migration of the medial clavicle fragment. As the period after
the injury increases, these muscles relax, allowing the medial
fragment to settle and lessen the overall displacement and
angulation of the fracture. Additionally, our healing radiographs
were taken at a mean of 10.9 weeks after injury, which was
substantially longer than in the previously mentioned adult
studies. This may provide additional time for true fracture
settling to occur. Previous studies may not have investigated the
post-injury progression for an adequate duration to observe the
true extent of the phenomenon. A final likely contributing factor
is that the periosteum in adolescent patients is significantly
thicker than that in adults.8 After a fracture, a larger portion of
the stronger periosteal sleeve in these younger patients is likely
to remain intact, helping the fracture settle into a more anatomic
position during the early phases of healing.
An important logistical and clinical consideration when
assessing clavicle fractures is the nature of the radiographic
assessment, particularly at the time of injury. Studies have
shown that significant differences in the positioning and beam
angulation of clavicle radiographs commonly occur, despite
standardized radiographic protocols.1,12 Our study used only
the AP clavicle radiograph and excluded the angled (cephalic-
tilt) view, given that greater variability in patient positioning
and beam angles across institutions made measurements
from these images unreliable. This study also used the most
common radiographic measurements in clinical practice,
including measurements of shortening. Bilateral clavicle ra-
diographs are advocated by some clinicians to better perform
assessments of clavicle shortening but were not available in
our study. Additionally, computed tomography imaging
would give a more comprehensive assessment of any residual
deformity, but our study protocol did not utilize these alter-
native imaging approaches in an attempt to minimize radiation
exposure for these youthful patients. Another limitation of the
current study is that we focused on radiographic changes and
outcomes and did not evaluate clinical data such as whether
the fracture had pierced the trapezial fascia or the ultimate
patient-reported outcomes.
Conclusion
Significant early improvements in fracture alignment
were observed in a substantial percentage of adolescent
patients with completely displaced clavicle fractures.
Among the most severely displaced fractures, a 6-mm
improvement in shortening and a 9
improvement in
angulation were identified. Contrary to findings in adult
studies, worsening displacement was observed in a small
minority of cases; however, it tended to be mild, and in
no case did it prompt surgical intervention. This finding
indicates that the true final deformity present after an
adolescent clavicle fracture is commonly less than that
present at the time of injury. Such findings may help
explain the exceedingly low rate of symptomatic mal-
union in adolescent clavicle fractures and the excellent
functional outcomes reported in other studies investi-
gating patient-reported outcomes following nonopera-
tive treatment.
Disclaimer
No external funding was received for this study, and no
financial biases exist. This study was supported by the
Rady Children’s Hospital, San Diego Division of
Orthopedics.
Andrew T. Pennock is a board or committee member of
the American Orthopaedic Society for Sports Medicine
and Pediatric Orthopaedic Society of North America
(POSNA); owns stock or stock options in Imagen Tech-
nologies; is a paid consultant for OrthoPediatrics; re-
ceives publishing royalties and financial or material
support from Wolters Kluwer Health–Lippincott Wil-
liams & Wilkins.
Benton E. Heyworth receives other financial or ma-
terial support from AlloSource and Vericel; owns stock
or stock options in Imagen Technologies; is a board or
committee member of POSNA and Pediatric Research in
Sports Medicine (PRISM); and receives publishing
royalties and financial or material support from Springer.
Donald S. Bae is a board or committee member of the
American Academy of Orthopaedic Surgeons (AAOS),
American Society for Surgery of the Hand, and POSNA; is
on the editorial or governing board of the Journal of Pe-
diatric Orthopedics; receives publishing royalties and
financial or material support from Lippincott Williams &
Wilkins and Springer; and is a paid consultant for
Orthopediatrics.
Michael T. Busch receives other financial or material
support from Arthrex, Orthopediatrics, and Smith &
Nephew and is a paid presenter or speaker for Arthrex.
Eric W. Edmonds receives speaker fees from Arthrex;
receives consulting fees from OrthoPediatrics; and is a
board or committee member of AAOS and POSNA.
Henry B. Ellis is a board or committee member of
AAOS, POSNA, and PRISM.
Mininder S. Kocher is a paid consultant for Best
Doctors, OrthoPediatrics, Ossur, and Smith & Nephew; is
a board or committee member of Harvard Medical
School, Harvard School of Public Health, POSNA,
PRISM, and Steadman Philippon Research Institute; re-
ceives intellectual property royalties from OrthoPediatrics
Displaced adolescent clavicle fracture alignment changes
and Ossur; and receives publishing royalties and financial
or material support from Saunders/Mosby-Elsevier and
Wolters Kluwer Health–Lippincott Williams & Wilkins.
Ying Li is a paid consultant for Medtronic and is a
board or committee member of POSNA and Scoliosis
Research Society (SRS).
Nirav K. Pandya is a paid consultant for Orthopedi-
atrics and is a board or committee member of POSNA
and PRISM.
Coleen S. Sabatini is a board or committee member
of AAOS, J. Robert Gladden Society, MiracleFeet
Medical Advisory Board, and POSNA.
David D. Spence receives publishing royalties and
financial or material support from Elsevier; receives
research support from Orthopediatrics; and is a board or
committee member of POSNA.
Samuel C. Willimon is a paid consultant for Smith &
Nephew Endoscopy.
Philip L. Wilson receives research support from Allo-
Source and Ossur; receives publishing royalties and financial
or material support from Elsevier; is on the editorial or
governing board of the Journal of Pediatric Orthopedics;
and is a board or committee member of POSNA.
The FACTS group has received funding from: Boston
Children’s Hospital Center for Program for Patient Safety
and Quality Research Grant (2013), Boston Children’s
Hospital Family Trust Private Donation (2015), and
POSNA Directed Research Grant (2015-2018).
Jeffrey J. Nepple is on the editorial or governing
board of Arthroscopy; is a board or committee member
of PRISM Society; is a paid consultant for Responsive
Arthroscopy and Smith & Nephew; receives research
support from Smith & Nephew and Zimmer; and is a
paid presenter or speaker for Smith & Nephew.
The other authors, their immediate families, and any
research foundations with which they are affiliated have
not received any financial payments or other benefits from
any commercial entity related to the subject of this article.
References
1. Backus JD, Merriman DJ, McAndrew CM, Gardner MJ, Ricci WM.
Upright versus supine radiographs of clavicle fractures: does posi-
tioning matter? J Orthop Trauma 2014;28:636-41. https://doi.org/10.
1097/BOT.0000000000000129
2. Bae DS, Shah AS, Kalish LA, Kwon JY, Waters PM. Shoulder motion,
strength, and functional outcomes in children with established mal-
union of the clavicle. J Pediatr Orthop 2013;33:544-50. https://doi.org/
10.1097/BPO.0b013e3182857d9e
3. Canadian Orthopaedic Trauma S. Nonoperative treatment compared
with plate fixation of displaced midshaft clavicular fractures. A
multicenter, randomized clinical trial. J Bone Joint Surg Am 2007;89:
1-10. https://doi.org/10.2106/JBJS.F.00020
4. Carry PM, Koonce R, Pan Z, Polousky JD. A survey of physician
opinion: adolescent midshaft clavicle fracture treatment preferences
among POSNA members. J Pediatr Orthop 2011;31:44-9. https://doi.
org/10.1097/BPO.0b013e3181ff67ce
9
5. Ellis HB, Li Y, Bae DS, Kalish LA, Wilson PL, Pennock AT, et al.
Descriptive epidemiology of adolescent clavicle fractures: results from
the FACTS (Function after Adolescent Clavicle Trauma and Surgery)
prospective, multicenter cohort study. Orthop J Sports Med 2020;8:
2325967120921344. https://doi.org/10.1177/2325967120921344
6. Herzog MM, Whitesell RC, Mac LM, Jackson ML, Culotta BA,
Axelrod JR, et al. Functional outcomes following non-operative versus
operative treatment of clavicle fractures in adolescents. J Child Orthop
2017;11:310-7. https://doi.org/10.1302/1863-2548.11.160267
7. Hughes JL, Newton PO, Bastrom T, Fabricant PD, Pennock AT. The
clavicle continues to grow during adolescence and early adulthood.
HSS J 2020;16:372-7. https://doi.org/10.1007/s11420-020-09754-8
8. Jacobsen FS. Periosteum: its relation to pediatric fractures. J Pediatr
Orthop B 1997;6:84-90.
9. Jit I, Kulkarni M. Times of appearance and fusion of epiphysis at the
medial end of the clavicle. Indian J Med Res 1976;64:773.
10. Kubiak R, Slongo T. Operative treatment of clavicle fractures in
children: a review of 21 years. J Pediatr Orthop 2002;22:736-9.
11. Li Y, Donohue KS, Robbins CB, Pennock AT, Ellis HB Jr, Nepple JJ,
et al. Reliability of radiographic assessments of adolescent midshaft
clavicle fractures by the FACTS Multicenter Study Group. J Orthop
Trauma 2017;31:479-84. https://doi.org/10.1097/BOT.
0000000000000877
12. Malik A, Jazini E, Song X, Johal H, O’Hara N, Slobogean G, et al.
Positional change in displacement of midshaft clavicle fractures: an
aid to initial evaluation. J Orthop Trauma 2017;31:e9-12. https://doi.
org/10.1097/BOT.0000000000000727
13. Marsh JL, Slongo TF, Agel J, Broderick JS, Creevey W, DeCoster TA,
et al. Fracture and dislocation classification compendiumd2007:
Orthopaedic Trauma Association classification, database and out-
comes committee. J Orthop Trauma 2007;21:S1-6. https://doi.org/10.
1097/00005131-200711101-00001
14. McGraw MA, Mehlman CT, Lindsell CJ, Kirby CL. Postnatal growth
of the clavicle: birth to 18 years of age. J Pediatr Orthop 2009;29:937-
43. https://doi.org/10.1097/BPO.0b013e3181c11992
15. Onizuka N, Anderson JP, Gilbertson JA, MacCormick LM, Cole PA.
Displacement of diaphyseal clavicle fractures related to patient posi-
tion and progressive displacement in the peri-injury period. J
Shoulder Elbow Surg 2018;27:667-73. https://doi.org/10.1016/j.jse.
2018.01.004
16. Plocher EK, Anavian J, Vang S, Cole PA. Progressive displacement of
clavicular fractures in the early postinjury period. J Trauma 2011;70:
1263-7. https://doi.org/10.1097/TA.0b013e3182166a6f
17. Postacchini F, Gumina S, De Santis P, Albo F. Epidemiology of
clavicle fractures. J Shoulder Elbow Surg 2002;11:452-6. https://doi.
org/10.1067/mse.2002.126613
18. Robinson CM. Fractures of the clavicle in the adult. Epidemiology and
classification. J Bone Joint Surg Br 1998;80:476-84.
19. Robinson CM, Goudie EB, Murray IR, Jenkins PJ, Ahktar MA,
Read EO, et al. Open reduction and plate fixation versus nonoperative
treatment for displaced midshaft clavicular fractures: a multicenter,
randomized, controlled trial. J Bone Joint Surg Am 2013;95:1576-84.
https://doi.org/10.2106/JBJS.L.00307
20. Schulz J, Moor M, Roocroft J, Bastrom TP, Pennock AT. Functional
and radiographic outcomes of nonoperative treatment of displaced
adolescent clavicle fractures. J Bone Joint Surg Am 2013;95:1159-65.
https://doi.org/10.2106/JBJS.L.01390
21. Smekal V, Irenberger A, Struve P, Wambacher M, Krappinger D,
Kralinger FS. Elastic stable intramedullary nailing versus nonopera-
tive treatment of displaced midshaft clavicular fracturesda random-
ized, controlled, clinical trial. J Orthop Trauma 2009;23:106-12.
https://doi.org/10.1097/BOT.0b013e318190cf88
22. Suppan CA, Bae DS, Donohue KS, Miller PE, Kocher MS,
Heyworth BE. Trends in the volume of operative treatment of midshaft
clavicle fractures in children and adolescents: a retrospective, 12-year,
single-institution analysis. J Pediatr Orthop B 2016;25:305-9. https://
doi.org/10.1097/BPB.0000000000000301