Arch Orthop Trauma Surg (2013) 133:15–22
DOI 10.1007/s00402-012-1631-3
ORTHOPAEDIC SURGERY
Risk factors for nonunion in 337 displaced midshaft clavicular
fractures treated with Knowles pin fixation
Chih-Lung Wu • Hui-Chin Chang • Ko-Hsiu Lu
Received: 9 April 2012 / Published online: 19 October 2012
Ó Springer-Verlag Berlin Heidelberg 2012
Abstract
Background Clavicular fractures account for nearly 10 %
of all fractures, and the majority of those fractures involve
the midshaft. Historically, these fractures were treated
nonoperatively; however, recent data suggest an increased
risk of nonunion and symptomatic malunion for displaced,
comminuted midshaft clavicular fractures treated conser-
vatively. Surgical intervention via plate osteosynthesis or
intramedullary fixation with pins, nails, or screws has been
shown to reduce, but not eliminate, this risk. Identification
of risk factors predictive of nonunion would improve the
overall management of displaced, comminuted midshaft
clavicular fractures.
Methods The medical records of 337 consecutive patients
who underwent Knowles pin fixation and supplemental
cerclage for the treatment of displaced, comminuted
C.-L. Wu
K.-H. Lu
Institute of Medicine, Chung Shan Medical University,
Taichung, Taiwan
C.-L. Wu
K.-H. Lu
School of Medicine, Chung Shan Medical University,
Taichung, Taiwan
C.-L. Wu
K.-H. Lu (&)
Department of Orthopedics, Chung Shan Medical University
Hospital, 110, Section 1, Chien-Kuo N. Road,
Taichung 402, Taiwan
e-mail: cshy034@csh.org.tw
H.-C. Chang
Department of Public Health, Chung Shan Medical University,
Taichung, Taiwan
H.-C. Chang
Library, Chung Shan Medical University Hospital,
Taichung, Taiwan
midshaft clavicular fractures between April 2007 and
March 2009 were retrospectively reviewed. The records of
the mechanism of injury, side of injury, Robinson fracture
classification, presence of associated injuries, cerclage
material, and patient-related variables, including diabetes
mellitus, hypertension and smoking, were analyzed. Vari-
ables were assessed by univariate and multivariate analysis
to identify those factors significantly associated with the
development of fracture nonunion.
Results A total of 19 nonunions occurred. Increasing age
and use of wire for supplemental cerclage fixation were
significantly associated with an increased risk for fracture
nonunion (p \ 0.001). Although suggested as predictors of
nonunion in other studies, female gender and fracture
severity were not significantly associated with nonunion.
Conclusions Nonunion remains a significant complica-
tion in the treatment of displaced, comminuted midshaft
clavicular fractures even with intramedullary fixation. Use
of absorbable suture in place of wire for cerclage fixation
and careful selection of treatment strategy in the elderly
may reduce the risk for nonunion.
Keywords Displaced midshaft clavicular fracture

Nonunion
Malunion
Cerclage wire
Knowles pin
Introduction
Clavicular fractures account for about 2.5–10 % of all
fractures, and most commonly result from a traffic injury or
a fall onto an outstretched arm [1, 2]. Because the middle-
third of the clavicle lies directly under the skin without the
protection of soft tissue or muscle attachments, this thin-
nest area of the clavicle is the most vulnerable to trauma.
Approximately 75 % of clavicle fractures involve the
123
16
midshaft region [3, 4], with the majority of those fractures
involving displacement of the fracture fragments. Dis-
placement occurs mainly from the combined effect of the
sternocleidomastoid muscle pulling the medial fragment
superiorly and posteriorly, and the pectoralis major and
deltoid muscles pulling the lateral fragment inferiorly and
anteriorly. Despite the high prevalence of displaced frac-
tures, historically most patients were treated nonopera-
tively [5–7]. Open reduction and internal fixation (ORIF)
for midshaft clavicular fractures had been reserved for
cases with significant displacement, shortening, or com-
minution; cases with open or impending open fractures;
cases involving neurovascular compromise; and cases
involving polytrauma [3, 8, 9].
Several recent studies have suggested unfavorable long-
term results following the nonoperative treatment of dis-
placed midshaft clavicular fractures. Studies have cited
nonunion rates as high as 33 % among patients treated
nonoperatively [10, 11]. Surgical treatment has been rec-
ommended in these cases to minimize the risk for nonunion
or symptomatic malunion [3, 12–16]. Operative treatment
for displaced midshaft clavicular fractures includes plate
osteosynthesis and intramedullary fixation with nails, pins,
or wire. Knowles pin fixation has been advocated over
extramedullary plating by some authors [8, 17–19], who
cited shorter operative times, smaller surgical scars, less
pain medication usage, lower complications rates, and
easier removal; however, nonunion has remained a concern
even with this technique.
The aim of this retrospective study was to identify risk
factors for the development of nonunion in displaced,
comminuted midshaft clavicular fractures treated with
Knowles pin fixation and supplemental cerclage. Identifi-
cation of those patients at increased risk for the develop-
ment of nonunion would facilitate the selection of the
optimal treatment approach for each patient.
Patients and methods
Patients
A retrospective review was performed on 337 consecutive
patients with closed, displaced, comminuted, midshaft
clavicular fractures treated with intramedullary Knowles
pins and supplemental cerclage fixation between April
2007 and March 2009. Approximately two-thirds (212/337)
of the patients were male with an average age of
38.71 ± 14.19 years (range 15–79 years). The females
(125/337) had an average age of 44.11 ± 15.63 years
(range 15–70 years). Type 2 diabetes was noted among
8.9 % of the patients. Alcohol use, defined as at least one
drink per day, was identified in 16 % of patients. Tobacco
123
Arch Orthop Trauma Surg (2013) 133:15–22
use, defined as smoking one pack of cigarettes per day, was
identified in 18 % of patients (Table 1).
The medical records were also reviewed to identify the
mechanism of injury, side of injury, Robinson fracture
classification, presence of associated injuries, cerclage
material, and patient-related variables, including diabetes
mellitus, hypertension, and smoking (Table 2). None of the
patients had a previous history of clavicle fracture, and all
were diagnosed with closed, displaced, comminuted mid-
shaft clavicular fractures based on clinical examination and
confirmed by plain radiographs. Patients with simple, dis-
placed; simple, nondisplaced; and transverse clavicular
fractures were excluded from the study because these
fracture patterns are not traditionally treated with supple-
mental cerclage fixation.
Method
The technique for ORIF of the midshaft clavicular fractures
performed on all patients in this study by one of six surgeons
was previously described by Chu et al [8]. Briefly, the
intramedullary canal of the proximal and distal fractured
fragments was predrilled with a 3-mm Kirschner wire. Then,
the Kirschner wire was penetrated through the posterolateral
cortex of distal fragment to make an entrance. A Knowles pin
(Zimmer 3.8 mm) was inserted from the posterolateral
entrance of the clavicle using a 2 mm suction tip as a guide.
After reduction, the Knowles pin was advanced into the
proximal fragment. In our study, a cerclage with wire or
absorbable suture was added to reinforce the fixation for all
displaced, comminuted fractures. In cases where the frag-
ments were too small to accept fixation, absorbable suture
was used to loosely tie the fragments into place. Bone
grafting was not performed in any patient in this study.
Patients were discharged 1–3 days following surgery,
and followed-up monthly in clinic for the first 3 months.
Radiographs were obtained each month and assessed for
evidence of healing. Failure to achieve union at 3 months
warranted continued monthly surveillance with radiographs
Table 1 Baseline patient characteristics
Variables Mean ± SD/number (%)
Age (years) 40.71 ± 14.95
Gender
Male 212 (62.91)
Female 125 (37.09)
Diabetes mellitus 30 (8.90)
Smoking 62 (18.40)
Drinking 54 (16.02)
Data were presented as mean ± SD for continuous variables or count
(percentage) for categorical variables
Arch Orthop Trauma Surg (2013) 133:15–22 17

Table 2 Characteristics of the clavicle fractures nonunion. Severely comminuted fractures (2B2) resulted in
Variables Number (%)
nonunion twice as often as less extensive comminution
(2B1) (9.1 vs. 4.5 %, respectively). Approximately 95 % of
Mechanism of injury the patients with nonunion received a supplemental cer-
Traffic accident 323 (95.85) clage fixation with wire.
Fall from height 9 (2.67) The relationships between such factors as patient age,
Other 5 (1.48) gender, Robinson fracture classification, fracture side,
Injury side presence of associated injuries, cerclage material, presence
Right 168 (49.85) of diabetes, and alcohol and tobacco use, and the devel-
Left 169 (50.15) opment of nonunion are displayed in Table 3. Among these
Robinson fracture classification variables, only patient age and cerclage material were
Type 2B1 230 (68.25) significantly associated with nonunion (p \ 0.001) in the
Type 2B2 107 (31.75) univariate analysis. Multivariate analysis confirmed the
Associated injuries 121 (35.91) significant association of both the cerclage material and
Cerclage material patient age with the development of nonunion (p \ 0.001)
Wire 131 (38.87) (Figs. 1, 2).
Absorbable suture 206 (61.13) Every additional 1 year increases the risk of nonunion
by 7 % [OR 0.93, 95 % confidence interval (CI)
Data were presented as count (percentage)
0.91–0.96]. Cerclage with absorbable suture resulted in a
lower risk of nonunion (1/205) compared to cerclage with
until union was reached or until the 6-month follow-up. wire (18/113). The OR for nonunion using absorbable
Patients with evidence of union (i.e., callus formation suture was 0.03 (OR 0.03, 95 % CI from 0.01 to 0.13).
without fracture gap and no pin migration visualized on
X-ray) were followed-up in another 6 months. The pin was
removed 6–12 months after surgery. Patients with evidence Discussion
of nonunion (i.e., fracture gap or pin migration visualized
on X-ray) at the 6-month follow-up underwent subsequent The primary goal in the treatment of clavicular fracture is
pin removal and plate fixation. to restore shoulder function, which historically was
Approval from the Institutional Review Board was accomplished via nonoperative treatment [5, 20]. However,
obtained before the start of this study. increasing evidence suggests that the incidence of non-
union and symptomatic malunion after conservative treat-
Statistical analysis ment of displaced midshaft clavicular fractures is more
prevalent than previously reported [7, 12, 21, 22]. Unsat-
Data were presented as mean ± standard deviation (SD) isfactory results, including nonunion, following the non-
for continuous variables or as count (percentage) for cat- operative treatment of displaced midshaft clavicular
egorical variables. Associations between factors and the fractures have been reported in up to 31 % of cases [3, 23].
nonunion of midshaft clavicular fractures were assessed Symptomatic malunion with shortening may present with
using univariate logistic regression. Factors with a p value pain, weakness, and a reduced range of motion affecting
B0.2 in the univariate analysis were then assessed by the involved shoulder, easy fatigability, paresthesias con-
multivariate logistic regression. The strength of each sistent with thoracic outlet syndrome and brachial plexus
association was represented by the odds ratio (OR). Sta- impingement, inability to sleep in the supine position, and
tistical significance was set at p \ 0.05. Statistical analyses cosmetic deformity [3, 20, 24]. A recent meta-analysis by
were performed using SAS 9.2 statistics software (SAS McKee et al. [16] found that primary operative fixation of
Institute Inc., Cary, NC, USA). clavicular fractures provided a more rapid return of func-
tion and reduced early disability, and the rates of nonunion
and malunion were significantly higher with nonoperative
Results treatment (14.5/1.42 %) as compared to operative treat-
ment (8.5/0 %). However, the authors also reported that the
Nonunion occurred in 19 patients (5.6 %) at the 6-month overall functional improvement at 1-year follow-up was
follow-up. The mean age of the patients experiencing minimally different between the groups (5-point improve-
nonunion was 54 years. Among those 19 patients, 8 were ment in DASH and CS scores in the operative group as
female (42 %), and the majority did not use alcohol or compared to the nonoperative group). The authors did
tobacco. Diabetes was present in only one patient with acknowledge a number of limitations with the study in

123
18 Arch Orthop Trauma Surg (2013) 133:15–22

Table 3 Factors associated with nonunion among patients with midshaft clavicular fractures
Variables Nonunion (n = 19) Union (n = 318) Univariate analysis Multivariate analysis
Odds ratio (95 % CI) P value Odds ratio (95 % CI) P value

Age 54.00 ± 10.93 39.92 ± 14.80 1.07 (1.03, 1.11) \0.001* 1.07 (1.04, 1.11) \0.001*
Gender
Male 11 (6.40) 201 (93.60) 1.25 (0.49, 3.20) 0.642
Female 8 (5.19) 117 (94.81) 1 –
Classification
2B2 9 (8.41) 98 (91.59) 2.02 (0.80, 5.13) 0.139 1.92 (0.95, 3.88) 0.070
2B1 10 (4.35) 220 (95.65) 1 – 1 –
Fracture side
Left 11 (6.51) 158 (93.49) 1.40 (0.55, 3.55) 0.489
Right 8 (4.76) 160 (95.24) 1 –
Associated chest injury
Yes 4 (4.55) 84 (95.45) 0.74 (0.24, 2.30) 0.607
No 15 (6.02) 234 (93.98) 1 –
Cerclage material
Wire 18 (13.74) 113 (86.26) 32.65 (4.30, 247.72) \0.001* 31.52 (7.90, 125.78) \0.001*
Absorbable suture 1 (0.49) 205 (99.51) 1 – 1 –
Diabetes mellitus
Yes 1 (3.33) 29 (96.67) 0.55 (0.07, 4.30) 0.572
No 18 (94.14) 289 (5.86) 1 –
Hypertension
Yes 4 (8.16) 45 (5.21) 1.62 (0.51, 5.10) 0.411
No 15 (91.84) 273 (94.79) 1 –
Smoking
Yes 4 (6.45) 58 (93.55) 1.20 (0.38, 3.73) 0.759
No 15 (5.45) 260 (94.55) 1 –
Drinking
Yes 5 (9.26) 49 (90.74) 1.96 (0.68, 5.69) 0.216
No 14 (4.95) 269 (95.05) 1 –
Univariate and multivariate analysis were performed by logistic regression
CI confidence interval, OR odds ratio
* P value \0.05. The factors with P value \0.2 at univariate analysis were selected into multivariate analysis

terms of number of studies included, heterogeneity, lack of
a standard operative protocol, and incomplete data, and
ultimately concluded that displaced midshaft clavicular
fractures treated nonoperatively are at higher risk of non-
union and malunion, but there is no evidence that surgery
improves long-term functional outcomes. Other studies
have indicated that benefits of operative intervention
include more rapid return to function, slightly better long-
term functional outcomes, and increased cost-effectiveness
for select patients [25, 26].
Knowles pin fixation is an older surgical procedure, and
many recent papers report on use of a modified locking
plate instead. Plating increases both the operative time and
the hospital cost, making this treatment option less cost
effective than Knowles pin fixation [19]. Another new
method of intramedullary nail fixation uses titanium elastic
nails; however, the procedure is technically demanding
with longer operative times and is associated with higher
rates of nonunion and malunion for comminuted fractures
compared to Knowles pin fixation [26]. Furthermore, the
titanium elastic nail is ten times more expensive compared
to the Knowles pin (NT$9000 vs. NT$930, respectively)
[26].
Operative treatment clearly reduces the rates of malun-
ion and nonunion. Specifically, intramedullary stabilization
has been shown to provide a risk reduction of 87 % for
nonunion in displaced clavicular fractures [15]. A recent
meta-analysis reported a 2.2 % nonunion rate with plate
fixation compared to 15 % nonunion rate for nonoperative
treatment [1]. Basamania, a leading proponent of intra-
medullary fixation of clavicular fractures indicates that
plate or intramedullary fixation can be considered for

123
Arch Orthop Trauma Surg (2013) 133:15–22 19

Fig. 1 Case 1: a A 50-year-old

female patient with a displaced
right midshaft clavicular
fracture. b Open reduction with
Knowles pin and cerclage wire
fixation of displaced right
midshaft clavicular fracture.
The postoperative X-ray
showed good reduction and
rigid fixation. c Knowles pin
was removed 10 months later
after assuming fracture union
had occurred. d Bony separation
(i.e., fracture nonunion) at the
original fracture site was noted.
If the clavicular fracture was
nonunion, bony separation will
develop later after pin removal

clavicular fractures; however, he believes that plate fixa-
tion is better suited for transverse simple fractures and
nonunions with bone loss than intramedullary fixation,
whereas for all other clavicular fractures intramedullary
fixation should be considered [27, 28].
In our study with Knowles pin fixation, the rates of
malunion and nonunion were 0 and 5.6 %, respectively.
Our rate of nonunion was higher than the rates reported by
Ferran et al. [29] (0 %) and Zlowodzki et al. [15] (2 %),
but lower than the rate reported by Chu et al. [8] (6.4 %).
For this reason, logistic regression was utilized to identify
risk factors for nonunion in patients with displaced, com-
minuted midshaft clavicular fractures undergoing Knowles
pin fixation and supplemental cerclage. Analyzed variables
included patient age, gender, fracture severity, cerclage
material, fracture side, presence of an associated chest
injury, presence of diabetes mellitus, and tobacco and
alcohol use. Increasing patient age and cerclage with wire
were significantly associated with an increased risk of
nonunion.
Robinson et al. [10] reported that the risk of nonunion
was significantly increased by advancing age; however, the
study involved a cohort of patients with clavicular fractures
treated nonoperatively. Our results also revealed that
increasing patient age is a significant predictor of non-
union, despite surgical intervention via Knowles pin fixa-
tion and supplemental cerclage. Although Ferran et al. [29]
reported a 0 % nonunion rate for Rockwood pin fixation,
the mean patient age (23.8 years) of that study cohort was
markedly lower than our mean patient age (40.71 years)
who underwent Knowles pin fixation and supplemental
cerclage. The risk for nonunion may increase with
advancing age due to diminished blood supply to the
fracture site [30] and decreased bone marrow cellularity
[31]. In addition, although fixation with a Knowles pin can
produce inter-fragmentary compression, this biomechani-
cal advantage was inhibited by the increased width of bone
marrow and osteoporosis in patients with advancing age.
Therefore, intramedullary fixation with a Knowles pin is
not an appropriate treatment for midshaft clavicular frac-
tures in elderly patients, and plate fixation may be a better
treatment option.
Because the clavicle is not a weight-bearing bone,
extremely rigid fixation is not required. Additional fixation

123
20
Fig. 2 Case 2: a A 61-year-old
female patient with a displaced
left midshaft clavicular fracture.
b Open reduction with Knowles
pin and cerclage wire fixation of
the displaced left midshaft
clavicular fracture. The
postoperative X-ray
demonstrated good fixation.
c Two months later, loss of
reduction occurred when the
Knowles pin backed out
via cerclage with wire or absorbable suture has been
reported in the operative treatment of comminuted clavic-
ular fractures to supplement Knowles pinning [8, 19], and
atrophic nonunion can develop in these cases from the
inappropriate usage of cerclage wire. The clavicle has no
single-specific intramedullary nutrient artery, so interrup-
tion of the surrounding blood supply may lead to devas-
cularization of the bony fragments [32]. In addition, the
wire may interfere with bone healing if it migrates into the
fracture gap. However, use of the absorbable suture for
cerclage would likely not affect long-term healing even if it
slid into the fracture gap. Accordingly, our study indicated
that the use of wire for supplemental cerclage fixation for
displaced, comminuted midshaft clavicular fractures was
significantly associated with an increased risk of nonunion.
In cases where additional fixation is necessary, absorbable
suture should be used in place of wire.
Unlike the study by Ferran et al. [29] (which reported a
0 % nonunion rate following Rockwood pin fixation where
the majority of patients did not experience high-energy
trauma), 95 % of the displaced midshaft clavicular
123
Arch Orthop Trauma Surg (2013) 133:15–22
fractures in our study occurred as a result of high-energy
trauma (i.e., traffic accidents and falls from significant
heights), and nearly one-third were associated with other
injuries. Greater soft tissue destruction at the fracture site
with subsequent reduction in blood supply to the bone may
account for the poorer fracture healing in our study. In
addition, it is not uncommon to find severe head trauma
(e.g., intracerebral hemorrhage) and/or severe chest injuries
(e.g., hemothorax, rib fracture) associated with severely
displaced clavicular fractures in the trauma patient. No
significant difference in the nonunion rate was identified in
this group of patients in our study, even when operative
treatment was delayed to address other more severe inju-
ries. Operative intervention for displaced midshaft clavic-
ular fractures results in improved chest wall motion with
less overall pain, and a reduced risk of complications (i.e.,
pneumonia) [33].
Several classification systems have been developed in an
attempt to guide intervention for the treatment of clavicular
fractures. Developed in 1998, the Robinson classification
system provided specific consideration for the degree of
Arch Orthop Trauma Surg (2013) 133:15–22
displacement between the fracture fragments and demon-
strated satisfactory levels of both inter-observer and intra-
observer reliability and reproducibility [4]. Robinson
divided the midshaft clavicular fractures into type 2A
(cortical alignment fractures) and type 2B (displaced
fractures). In an effort to provide direction for both treat-
ment and prognosis, Robinson further divided 2A and 2B
into subgroups: 2A1 (nondisplaced), 2A2 (angulated), 2B1
(simple or wedge comminuted), and 2B2 (isolated or
comminuted segmental). Fracture severity has been sig-
nificantly associated with an increased risk of nonunion in
femur and pediatric humeral lateral condyle fractures [34–
36]. In general, highly comminuted midshaft clavicular
fractures (2B2) involve extensive surgical intervention and
are technically demanding to fix. Although the severity of
the clavicular fracture was not a significant predictor for
nonunion in our study, a trend toward a higher rate of
nonunion was observed for 2B2 fractures (p \ 0.07).
Female gender has been mentioned in several studies as
a risk factor for nonunion, not only in clavicular fractures,
but in other fractures as well [2, 10, 30]. However, multi-
variate logistic regression performed in the current study
failed to identify gender as a significant predictor of non-
union with displaced midshaft clavicular fractures.
Few papers mention the possible association between
either diabetes or alcohol and the development of non-
union. Our study failed to identify any significant associ-
ation between either of those factors and nonunion of
displaced midshaft clavicular fractures. Although tobacco
use has been cited as a risk factor for fracture nonunion
[31], our study failed to identify it as a significant predictor
for nonunion in the fixation of displaced, comminuted
midshaft clavicular fractures with Knowles pinning and
supplemental cerclage.
Conclusion
Growing evidence suggests that operative treatment should
be strongly considered for displaced midshaft clavicular
fractures due to a higher rate of symptomatic nonunion [2,
3, 5, 16, 37]. In addition to reducing complications, oper-
ative treatment has been shown to reduce the length of
hospital stay, improve functional outcome, and reduce
overall healthcare costs [1, 12, 14, 25, 38]. Previous studies
reported that the outcome of Knowles pin fixation was
excellent in the treatment of displaced midshaft clavicle
fractures; however, our results indicated 5.6 % of patients
developed fracture nonunion after surgery. The result of
multivariate logistic regression showed that increasing
patient age and the intraoperative application of cerclage
wire were significant predictors for postoperative fracture
21
nonunion. In addition, absorbable sutures can be used to
replace wire for supplemental cerclage fixation.
Conflict of interest The authors declare that they have no conflict
of interest.
References
1. Kulshrestha V, Roy T, Audige L (2011) Operative versus non-
operative management of displaced midshaft clavicle fractures: a
prospective cohort study. J Orthop Trauma 25(1):31–38
2. Smekal V, Oberladstaetter J, Struve P, Krappinger D (2009) Shaft
fractures of the clavicle: current concepts. Acta Orthop Trauma
Surg 129(6):807–815
3. Hillen RJ, Burger BJ, Pöll RG, de Gast A, Robinson CM (2010)
Malunion after midshaft clavicle fractures in adults. Acta Orthop
81(3):273–279
4. Robinson CM (1998) Fractures of the clavicle in the adult.
Epidemiology and classification. J Bone Jt Surg Br 80(3):
476–484
5. Nordqvist A, Petersson CJ, Redlund-Johnell I (1998) Mid-clavi-
cle fractures in adults: end result study after conservative treat-
ment. J Orthop Trauma 12(8):572–576
6. Post M (1989) Current concepts in the treatment of fractures of
the clavicle. Clin Orthop Relat Res 245:89–101
7. Smekal V, Deml C, Kamelger F, Dallapozza C, Krappinger D
(2010) Corrective osteotomy in symptomatic midshaft clavicular
malunion using elastic stable intramedullary nails. Arch Orthop
Trauma Surg 130:681–685
8. Chu CM, Wang SJ, Lin LC (2002) Fixation of mid-third cla-
vicular fractures with Knowles pins. Acta Orthop Scand 73(2):
134–139
9. Zenni EJ Jr, Krieg JK, Rosen MJ (1981) Open reduction and
internal fixation of clavicular fractures. J Bone Jt Surg Am
63(1):147–151
10. Robinson CM, Court-Brown CM, McQueen MM, Wakefield AE
(2004) Estimating the risk of nonunion following nonoperative
treatment of a clavicular fracture. J Bone Jt Surg Am 86-A(7):
1359–1365
11. Brinker MR, Edwards TB, O’Connor DP (2005) Estimating the
risk of nonunion following nonoperative treatment of a clavicular
fracture (letter). J Bone Jt Surg Am 87:676–677
12. Altamimi SA, McKee MD (2008) Nonoperative treatment com-
pared with plate fixation of displaced midshaft clavicular frac-
tures. Surgical technique. J Bone Jt Surg Am 90(Suppl 2 Pt
1):1–8
13. Lenza M, Belloti JC, Gomes Dos Santos JB, Matsumoto MH,
Faloppa F (2009) Surgical interventions for treating acute frac-
tures or non-union of the middle third of the clavicle. Cochrane
Database Syst Rev (4):CD007428
14. McKee MD (2010) Clavicle fractures in 2010: sling/swathe or
open reduction and internal fixation? Orthop Clin North Am
41(2):225–231
15. Zlowodzki M, Zelle BA, Cole PA, Jeray K, McKee MD,
Evidence-Based Orthopaedic Trauma Working Group (2005)
Treatment of acute midshaft clavicle fractures: systematic
review of 2144 fractures: on behalf of the Evidence-Based
Orthopaedic Trauma Working Group. J Orthop Trauma 19(7):
504–507
16. McKee RC, Whelan DB, Schemitsch EH, McKee MD (2012)
Operative versus nonoperative care of displaced midshaft cla-
vicular fractures: a meta-analysis of randomized clinical trials.
J Bone Jt Surg Am 94(8):675–684
123
22
17. Neviaser RJ, Neviaser JS, Neviaser TJ, Neviaser JS (1975) A
simple technique for internal fixation of the clavicle. A long term
evaluation. Clin Orthop Relat Res 109:103–107
18. Lee YS, Lin CC, Huang CR, Chen CN, Liao WY (2007) Oper-
ative treatment of midclavicular fractures in 62 elderly patients:
Knowles pin versus plate. Orthopedics 30(11):959–964
19. Lee YS, Huang HL, Lo TY, Hsieh YF, Huang CR (2008) Sur-
gical treatment of midclavicular fractures: a prospective com-
parison of Knowles pinning and plate fixation. Int Orthop
32(4):541–545
20. Neer CS 2nd (1960) Nonunion of the clavicle. JAMA
172:1006–1011
21. McKee MD, Wild LM, Schemitsch EH (2004) Midshaft mal-
unions of the clavicle. Surgical technique. J Bone Jt Surg Am
86-A(Suppl 1):37–43
22. Kim W, McKee MD (2008) Management of acute clavicle
fractures. Orthop Clin North Am 39(4):491–505
23. Liu HH, Chang CH, Chia WT, Chen CH, Tarng YW, Wong CY
(2010) Comparison of plates versus intramedullary nails for
fixation of displaced midshaft clavicular fractures. J Trauma
69(6):E82–E87
24. Kitsis CK, Marino AJ, Krikler SJ, Birch R (2003) Late compli-
cations following clavicular fractures and their operative man-
agement. Injury 34(1):69–74
25. Pearson AM, Tosteson AN, Koval KJ, McKee MD, Cantu RV,
Bell JE, Vicente M (2010) Is surgery for displaced, midshaft
clavicle fractures in adults cost-effective? Results based on a
mulitcenter randomized, controlled trial. J Orthop Trauma
24(7):426–433
26. Frigg A, Rillmann P, Perren T, Gerber M, Ryf C (2009) Intra-
medullary nailing of clavicular midshaft fractures with the tita-
nium elastic nail: problems and complications. Am J Sports Med
37(2):352–359
27. Basamania CJ (1999) Claviculoplasty and intramedullary fixation
of malunited shortened clavicle fractures. J Should Elbow Surg
8:540 (abstract)
28. Basamania CJ (2003) Treatment of clavicular fractures and
malunions. In: Levine WN, Marra G, Bigliani LU (eds) Fractures
of the shoulder girdle. Informa Healthcare, UK, pp 222–248
123
Arch Orthop Trauma Surg (2013) 133:15–22
29. Ferran NA, Hodgson P, Vannet N, Williams R, Evans RO (2010)
Locked intramedullary fixation versus plating for displaced and
shortened mid-shaft clavicle fractures: a randomized trial.
J Should Elbow Surg 19(6):783–789
30. Parker MJ, Raghavan R, Gurusamy K (2007) Incidence of frac-
ture-healing complications after femoral neck fractures. Clin
Orthop Relat Res 458:175–179
31. Bidula JC, Boehm C, Powell K, Barsoum W, Nakamoto C,
Mascha E, Muschler GF (2006) Osteogenic progenitors in bone
marrow aspirates from smokers and nonsmokers. Clin Orthop
Relat Res 442:252–259
32. Chang IL, Lee TS, Ku MC (1994) Treatment of femoral fracture
with nonunion. Kao-Hsiung i Hsueh Ko Hsueh Tsa Chih (Ka-
ohsiung J Med Sci) 10(10):543–549
33. Solberg BD, Moon CN, Nissim AA, Wilson MT, Marqulies DR
(2009) Treatment of chest wall implosion injuries without tho-
racotomy: technique and clinical outcomes. J Trauma 67(1):8–13
34. Noumi TK, Yokoyama K, Ohtsuka H, Nakamura K, Itoman M
(2005) Intramedullary nailing for open fractures of the femoral
shaft: evaluation of contributing factors on deep infection and
nonunion using multivariate analysis. Injury 36(9):1085–1093
35. el Moumni M, Leenhouts PA, ten Duis HJ, Wendt KW (2009)
The incidence of non-union following unreamed intramedullary
nailing of femoral shaft fractures. Injury 40(2):205–208
36. Weiss JM, Graves S, Yang S, Mendelsohn E, Kay RM, Skaggs
DL (2009) A new classification system predictive of complica-
tions in surgically treated pediatric humeral lateral condyle
fractures. J Pediatr Orthop 29(6):602–605
37. O’Neill BJ, Hirpara KM, O’Briain D, McGarr C, Kaar TK (2011)
Clavicle fractures: a comparison of five classification systems and
their relationship to treatment outcomes. Int Orthop 35(6):
909–914
38. Böhme JA, Bonk A, Bacher GO, Wilharm A, Hoffmann R, Josten
C (2011) Current treatment concepts for mid-shaft fractures of the
clavicle—results of a prospective multicentre study. Zeitschrift
fur Orthopadie & Unfallchirurgie 149(1):68–76