Injury, Int. J.

Care Injured 46 (2015) 19211929

Contents lists available at ScienceDirect

Injury
journal homepage: www.elsevier.com/locate/injury

Displaced proximal humeral fractures: When is surgery necessary?

Iskandar Tamimi a,*, Guillermo Montesa b, Francisco Collado b, David Gonzalez b,
Pablo Carnero b, Facundo Rojas b, Mohamed Nagib b, Veronica Perez b,
Miguel Alvarez b, Faleh Tamimi a
a
Faculty of Dentistry, McGill University, Montreal, Canada
b
Traumatology and Orthopedic Surgery Department HRU Carlos Haya, Malaga, Spain

A R T I C L E I N F O A B S T R A C T

Article history: Background: Several therapeutic methods have been traditionally used in the treatment of displaced
Received 2 February 2015 proximal humeral fractures; however, the indication of these treatments is still controversial. The
Received in revised form 12 May 2015 purpose of this study was to compare the medium-term functional results of four methods commonly
Accepted 31 May 2015
used in the treatment of proximal humeral fractures [conservative treatment, proximal humeral nails
(PHN), percutaneous K-wiring (PKW), and locking-plates (LP)] taking into consideration the type of
Keywords: fracture and the age of the patients.
Proximal humerus fracture
Methods: We conducted a retrospective cohort study on patients with proximal humeral fractures
Locking plate
Proximal humeral nail
treated with one of the following methods: conservative treatment, PHN, PKW, or LP. Functional results
Kirschner wires were assessed using the absolute Constant score and the disabilities of the arm shoulder and hand score
Conservative treatment (DASH). The functional outcome was analysed according to age (65 years and <65 years) and fracture
Functional outcome type (displaced 2-fragment and 34-fragment fractures).
Age Results: A total of 113 patients were included in the study, with a mean age of 65.3 SD 15.2 years and
Fracture type average follow-up time of 26.2 SD 12.6 months. Patients under 65 years had higher Constant scores when
treated with PHN and PKW than those treated conservatively (77.2 vs. 54.7, p = 0.01 and 74.0 vs. 54.7,
p = 0.03, respectively). Patients above 65 years had higher Constant scores when treated with PKW
compared to PHN and conservative treatment (68.7 vs. 51.9, p = 0.02 and 68.7 vs. 55.9, p = 0.029,
respectively). In 2-fragment fractures, PKW resulted in higher Constant scores than conservative
treatment (70.4 vs. 53.9, p = 0.048). No differences were found in the nal outcome between patients
treated with LP and those treated conservatively regardless of age, and fracture type. There were also no
differences between any of the evaluated methods in the treatment of 34-fragment fractures.
Conclusion: The use of PKW was associated with better functional results than conservative treatment in
individuals of all ages, especially in patients with 2-fragment fractures; PKW also achieved better
functional results than PHN in elderly patients. PHN was superior to conservative treatment in young
individuals. No signicant differences were found between LP and conservative treatment in any of the
analysed categories.
2015 Elsevier Ltd. All rights reserved.

Introduction treatment, direct suturing of the fracture fragments, proximal

Displaced proximal humeral fractures are relatively common
in elderly and osteoporotic patients and can lead to signicant
functional incapacity [1]. Multiple treatment methods have been
used in the management of these fractures including conservative
* Corresponding author at: McGill University, 2001 McGill College Avenue, Suite
500, Montreal, Quebec, H3A 1G1 Canada. Tel.: +1 514 398 7203.
E-mail addresses: isktamimi80@yahoo.com, iskandartamimimarino@mcgill.ca
(I. Tamimi).
humeral nails (PHN), percutaneous K-wiring (PKW), percutane-
ous xed angle locking plates (LP), open reduction and internal
xation with LP, and partial or total shoulder arthroplasty [2].
However, the management of these fractures is controversial and
surgeons base their indications subjectively on a series of factors
such as degree of displacement, number of fragments, surgical
experience, baseline functional status of the patient, hand
dominance, and age.
The literature comparing these therapeutic options is limited,
and to our knowledge there are no published cohort studies that
compare four of the main treatment modalities (PHN, conservative

http://dx.doi.org/10.1016/j.injury.2015.05.049
00201383/ 2015 Elsevier Ltd. All rights reserved.

Descargado de ClinicalKey.es desde Universidad Nacional Autonoma de Mexico enero 24, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
1922 I. Tamimi et al. / Injury, Int. J. Care Injured 46 (2015) 19211929

treatment, LP and PKW) [3,4]. On the other hand, the success rates
of the different treatment options used in the management of
displaced proximal humeral fractures may also vary according to
patients age and fracture type; as the bone quality and general
health status changes with age [5]. Accordingly, the purpose of this
study was to compare the medium-term functional results of four
methods commonly used in the treatment of proximal humeral
fractures [conservative treatment, PHN, PKW, and LP] taking into
consideration the type of fracture and the age of the patients.
Materials and methods
Patient selection
We conducted a retrospective cohort study on four treatment
methods for proximal humeral fractures [conservative treatment,
PHN, PKW, and LP]. All patients with the diagnosis of displaced
proximal humeral fracture treated in our center, between January
1st, 2008 and December 31st, 2013 were reviewed. Patients
records were withdrawn from the computerised database of the
Orthopedic Surgery Department without any exposure informa-
tion. All displaced proximal humeral fractures that occurred within
the study period were reviewed. Patients with pathological
fractures (i.e., secondary to osteomalacia, Pagets disease, primary
bone tumors or bone metastasis), treated with shoulder hemi-
arthroplasty, with a follow-up of less than 8 months, and those
who were lost or died during follow-up were excluded from the
analysis.
Clinical features
The following parameters were retrieved from patients les and
computerised records: age, gender, fracture side, pre-surgical
comorbidities (e.g., liver disease, ischemic heart disease, chronic
renal failure) type of fracture according to Neers classication (2, 3 or
4-displaced fragments), type of treatment [conservative treatment,
PHN (Expert, Synthes, Stratec Medical Ltd, Oberdorf, Switzerland), LP
(PHILOS, Synthes, Stratec Medical Ltd, Oberdorf, Switzerland), and
PKW], date of surgery, postoperative complications (i.e., hardware
migration, malunion, avascular necrosis, pseudarthrosis, infections,
and screw protrusion), and follow-up period.
Treatment options
The selection of the treatment method depended mainly on the
surgeons experience and did not follow a strict intradepartmental
protocol due to the lack of solid guidelines. When conservative
treatment was applied, patients were immobilised in a sling
bandage stabilising the arm against the chest for 2 weeks. A closed
reduction was performed if the displacement between the head
and the distal fragment was above 50% of the diaphyseal diameter;
in these cases, a control X-ray was performed after the reduction.
Follow-up X-rays were performed at 2, 4 and 6 weeks; pendular
exercises were started after 2 weeks if there was no apparent
crepitation at the fracture site (Fig. 1) [2].
PKW was performed following closed reduction under image-
intensier. Reduction was achieved by gentle arm manipulation

Fig. 1. (A) AP X-ray of the shoulder showing a displaced 3-fragment proximal humerus fracture in valgus position. (B) Transthoracic X-ray of the shoulder showing posterior
angulation (>458) of the fracture and displacement of the greater tuberosity. (C and D) AP and transthoracic X-rays showing evidence of consolidation after 5 weeks of
conservative treatment.

Descargado de ClinicalKey.es desde Universidad Nacional Autonoma de Mexico enero 24, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
 I. Tamimi et al. / Injury, Int. J. Care Injured 46 (2015) 19211929 1923

and manual traction. Two to four 2.5 mm K-wires were inserted
through the lateral and or medial epicondyles, depending on the
type of fracture [6]. K-wires were bent distally to control migration
and situated in the subchondral bone, approximately 58 mm
from the articular surface. Finally, the K-wires were left under the
skin to avoid infections and the arm was then immobilised in a
sling as previously described (Fig. 2) [2]. Metal removal was
performed once fracture consolidation was conrmed, approxi-
mately 4 months after surgery.
Patients treated with LPs underwent a deltopectoral approach;
fracture reduction was performed by a careful manipulation of the
fragments in order to avoid excessive periosteal damage. Fragments
were reduced into position by direct methods or with a K-wire used
as a joystick; reduction was conrmed by image-intensier. After
reduction and provisional stabilisation using threaded K-wires,
stable xation was performed with a LP and a minimum of 6
proximal and 3 distal screws. The arm was then immobilised in a
sling and pendular exercises were initiated (Fig. 3) [7].
Individuals that were treated with PHN xation underwent a
closed fracture reduction under image-intensier control. Then a
small deltoid-splitting and rotator cuff incision was made. The
medullary canal was opened with an awl and a guide wire was
introduced in the humeral canal. The nail connected to the
aiming arm was then inserted into the medullary canal; proximal
locking was performed using a spiral blade or a locking screw
depending on the bone quality. Finally, the rotator cuff incision
was carefully sutured and mobilisation was initiated on the
second postoperative day [8,9] (Fig. 4). In order to conrm the
correct reduction of the fracture a post-operative control X-ray
was performed in all cases regardless of the surgical treatment
applied.
Clinical evaluation
The functional assessment was performed at least 8 months
after the fracture diagnosis using the absolute Constant score, and
the disabilities of the arm shoulder and hand score (DASH) [10,11].
The absolute Constant score has a maximum punctuation of 100
points; higher scores represent a better functional outcome. The
DASH ranges from 0 to 100 points; higher scores represent a worse
functional outcome. The morbidity of the patients was evaluated
using the Charlson Comorbidity Index (CCI) [12]. Individuals were
divided into 2 different subgroups according to their preoperative
general status (CCI <3.5 and 3.5).
Functional results were analysed according to the treatment
option applied (conservative treatment, PHN, LP and PKW), age
(65 years and <65 years) and fracture type (displaced 2-fragment
and 34-fragment fractures).
Data management and statistical analysis
Data were analysed with SPSS 17.0 software (SPSS Inc, Chicago,
IL, USA), Origin Pro 8.0 (OriginLab Corporation, Northampton,
USA), and G*power 3.0.10 (Universitat Kiel, Germany). Mean
values were expressed with their corresponding standard devia-
tion (SD). Normal distribution was conrmed in each group using

Fig. 2. (A) AP X-ray of the shoulder showing a displaced surgical neck fracture of the proximal humerus, with no apparent contact between the two fragments. (B)
Transthoracic X-ray of the shoulder showing a good lateral alignment of the fracture. (C) Post-surgical X-ray conrming an adequate reduction and stabilisation of the fracture
with percutaneous Kirschner wires. (D) AP X-ray performed 3 months after the initial surgery, showing some evidence of consolidation and penetration of the articular
surface by both Kirschner wires; the patient was booked for prompt hardware removal.

Descargado de ClinicalKey.es desde Universidad Nacional Autonoma de Mexico enero 24, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
1924 I. Tamimi et al. / Injury, Int. J. Care Injured 46 (2015) 19211929

Fig. 3. (A) AP X-ray of the shoulder showing a displaced 3-fragment fracture of the proximal humerus in valgus position. (B) Transthoracic X-ray of the shoulder showing a
good lateral alignment of the fracture. (C) Post-surgical X-ray conrming an adequate reduction and xation of the fracture with a locking plate (PHILOS). (D) AP X-ray of the
shoulder showing some evidence of consolidation 3 months after surgery.

the ShapiroWilk test. Odd ratios were presented with 95%
coefcient intervals. Differences between continuous variables
were analysed using the one-way ANOVA and LSD test, whereas
differences between binary variables were analysed using the Chi
square test. Correlation between continuous variables was
measured using the Pearson correlation coefcient. Results were
considered signicant when two-tailed P values were < 0.05.
Power analysis were performed using two-tailed post-hoc t-test
for two independent means with an a-error probability of 0.05.
Results
Patient demographics and clinical features
The total number of patients with displaced proximal humeral
fractures who were treated in our center during the study period
was 182, among whom 113 individuals [76 (67.4%) females and 37
(32.6%) males)] were included and 69 patients were excluded [3
pathological fractures, 5 hemiarthroplasties, 16 deaths, 45 lost to
follow-up].
We registered 45 (39.2%) 2-fragment fractures and 68 (60.2%)
34-fragment fractures. Conservative treatment was applied in 25
(22.1%) cases, whereas surgery was considered in 88 (77.9%)
patients [PHN xation in 19 (16.8%), LP 44 (38.9%), and PKW 25
(22.2%)]. The mean age was of 65.3 SD 15.2 years and the mean
follow-up time at functional assessment was of 26.2 SD 12.6
months (ranging from 8 to 55 months). Fractures were more
common on the right side 67 (59.3%) than on the left 46 (40.7%).
There were no signicant differences between the different
therapeutic options with regards to age, gender or injured side. In
patients above 65 years with 34-fragment fractures, PKW and
IMN were less frequently used than LP and conservative treatment
(Table 1). Regarding the preoperative general status of the patients,
individuals who underwent osteosynthesis with LP and PHN had a
better basal condition according to the CCI when compared with
patients that were treated conservatively or with PKW (Table 1).
Post-surgical results
The mean follow-up time from the date of the fracture was of
28.0 SD 8.5 months for conservative treatment, 25.9 SD 15.0 for LP,
27.6 SD 14.0 for PKW, and 22.5 SD 9.0 for PHN. No signicant
differences were found between the different groups in terms of
follow-up time (Table 2).
The total number of registered complications was 22 (19.5%)
(7 cases of hardware migration, 6 vicious consolidation, 3 avascular
necrosis, 3 pseudarthrosis, 1 supercial infection, 1 postoperative
bleeding, and 1 screw protrusion). A total of 8 (18.2%) complica-
tions were registered in patients treated with LP, 6 (24.0%) in those
treated with PKW, 7 (36.8%) in patients treated with PHN, and 1
(5%) in patients treated conservatively. There were no signicant
statistical differences in the complication rates between the
different surgical treatments. The overall functional outcome
according to the Constant score was more favourable in patients
that underwent PKW than in those treated conservatively [70.0 vs.
57.2, p = 0.012 (power 87%)]. There were no overall signicant

Descargado de ClinicalKey.es desde Universidad Nacional Autonoma de Mexico enero 24, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
 I. Tamimi et al. / Injury, Int. J. Care Injured 46 (2015) 19211929 1925

Fig. 4. (A) AP X-ray of the shoulder showing a displaced 3-fragment fracture of the proximal humerus extending to the proximal diaphysis. (B) Post-surgical X-ray conrming
an adequate reduction and xation of the fracture with a proximal humeral nail (Expert). (C and D) AP X-rays of the shoulder performed 1 months and 5 months after the
initial surgery, respectively, showing a progressive consolidation of the fracture.

differences between LP and conservative treatment [Constant
score: p = 0.086, (power 45%), DASH: p = 0.47 (power 12%)], nor
between the other treatment methods (Table 2).
Functional results according to age
In patients under 65 years of age, the use of PHN or PKW lead to
higher Constant scores than those achieved with conservative
treatment [77.2 vs. 54.7, p = 0.01 (power 96%) and 74.0 vs. 54.7,
p = 0.031 (power 87%), respectively)] (Table 3). No signicant
differences were found in the Constant and DASH scores between
patients under 65 years of age who were treated conservatively
and those treated with LP (p = 0.09, and p = 0.47, respectively).
In patients over 65 years of age, the use of PKW had better
Constant scores than PHN and conservative treatment [68.7 vs.
51.9, p = 0.02 (power 70%) and 68.7 vs. 55.3, p = 0.029 (power 74%),
respectively] (Table 3). No signicant differences in the Constant
and DASH functional scores were found between patients treated
conservatively and those treated with LP in this group (p = 0.39,
and p = 0.80, respectively) (Table 3).
The results of PHNs were age dependent, the mean Constant
score in patients under 65 years was 77.2 vs. 51.9 in patients above
65 years [p = 0.01 (power 84%)] (Table 3) (Fig. 5). Moreover, there
were no signicant differences in the fracture type (i.e., 2-fragment
and 34 fragment fractures) distribution according to age in
patients treated with PHN OR 0.31 (CI 95% 0.042.6) (Table 1).
Functional results according to fracture type
In 2-fragment fractures the use of PKW achieved Constant
scores that were signicantly higher than conservative treatment
[70.4 vs. 53.9, p = 0.048 (power 77%)]. We did not nd signicant
differences in the Constant and DASH functional scores between
patients with 2-fragment fractures that were treated conserva-
tively and those treated with LP (p = 0.33 and 0.45, respectively). In
34-fragment fractures, no statistical signicant results were
found between any of the different treatment modalities (Table 3).
Discussion
The objectives of this study were to compare four therapeutic
methods commonly used to treat patients with displaced proximal
humeral fractures [conservative treatment, PHN, PKW, and LP],
and to analyse the effect of age and fracture type on the functional
outcome of these patients. Accordingly, we observed that PKW was
better than conservative treatment in individuals of all ages,
especially in patients with 2-fragemt fractures. PHN was superior
to conservative treatment in young individuals, and the use of PKW
achieved better functional results than PHN in elderly patients. No
signicant differences were found between LP and conservative
treatment in any of the analysed categories.
PKW vs. conservative treatment
In young patients, PKW had superior functional results
compared with conservative treatment, even though there were
no differences in the fracture type distribution between these
treatment categories (Tables 1 and 3). The use of PKW in proximal
humerus fractures has multiple advantages such as minimal blood
loss, shorter surgical time, less soft tissues stripping, and less cost
[13,14]. Nevertheless, PKW techniques are also associated with
complications like pintract infection, poor reduction, malunion,
and pin migration [14]. The insertion of metalwork in the humeral
epicondylar area can also injure neurovascular structures such as

Descargado de ClinicalKey.es desde Universidad Nacional Autonoma de Mexico enero 24, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
1926 I. Tamimi et al. / Injury, Int. J. Care Injured 46 (2015) 19211929

Table 1
Patient demographics and clinical features.

Parameter Conservative (n = 25) Locking plate (n = 44) Kirschner wires (n = 25) Proximal humeral nail (n = 19)

Patients age, y
Age <65 N (%) 7 (28) 22 (50) 8 (32) 9 (47.4)
Age 65 N (%) 18 (72) 22 (50) 17 (68) 10 (52.6)
OR (95% CI) 1 2.6 (CI 0.907.4) 1.2 (CI 0.44.1) 2.3 (CI 0.668.1)
OR (95% CI) 1 0.47 (CI 0.171.3) 0.9 (CI 0.312.6)
OR (95% CI) 1 1.9 (CI 0.566.6)

Fracture type, age <65 y N (%)

2-Fragments N (%) 2 (10.5) 6 (31.6) 6 (31.6) 5 (26.3)
34 Fragments N (%) 5 (18.5) 16 (59.3) 2 (7.4) 4 (14.8)
OR (95% CI) 1 1.1 (CI 0.167.06) 0.13 (CI 0.0131.31) 0.32 (CI 0.042.61)
OR (95% CI) 1 0.12 (CI 0.020.8)* 0.3 (CI 0.061.5)
OR (95% CI) 1 2.4 (CI 0.319.0)

Fracture type, age 65 y N (%)

2-Fragments N (%) 5 (19.2 2 (7.7) 11 (42.3) 8 (30.8)
34 Fragments N (%) 13 (31.7) 20 (48.8) 6 (14.6) 2 (4.9)
OR (95% CI) 1 3.8 (CI 0.6422.9) 0.2 (CI 0.050.87)* 0.1 (CI 0.010.61)*
OR (95% CI) 1 0.05 (CI 0.010.29)* 0.02 (CI 0.0030.2)*
OR (95% CI) 1 0.46 (CI 0.072.89)

Gender
Female N (%) 19 (76) 29 (65.9) 18 (72) 10 (52.6)
Male N (%) 6 (30) 15 (34.1) 7 (28.0) 9 (47.4)
OR (95% CI) 1 0.61 (CI 0.21.9) 0. 8 (CI 0.232.9) 0.35 (CI 0.11.3)
OR (95% CI) 1 1.3 (CI 0.453.9) 0.57 (CI 0.191.7)
OR (95% CI) 1 0.43 (CI 0.121.5)

Injured arm
Right N (%) 15 (60) 28 (63.6) 14 (44) 11 (57.9)
Left N (%) 10 (40) 16 (36.4) 11 (56) 8 (42.1)
OR (95% CI) 1 1.2 (CI 0.423.2) 0.85 (CI 0.282.6) 0.92 (CI 0.273.1)
OR (95% CI) 1 0.73 (CI 0.271.98) 0.79 (CI 0.262.4)
OR (95% CI) 1 1.1 (CI 0.323.6)

CCI
CCI <3.5 8 (32) 27 (61) 8 (32) 12 (63)
CCI 3.5 17 (68) 17 (38) 17 (68) 7 (36.8)
OR (95% CI) 1 3.4 (CI 1.29.5)* 1.0 (CI 0.313.28) 3.6 (CI 1.0412.8)*
OR (95% CI) 1 0.3 (CI 0.10.84)* 1.1 (CI 0.363.3)
OR (95% CI) 1 3.6 (CI 1.0412.8)*

Abbreviations: CI, condence interval; y, years; OR, odds ratio; CCI, Charlson Comorbidity Index; N (%), values are expressed as percentages. *Signicant p-values (<0.05) for OR.

the ulnar nerve [15]. Fortunately, most of these complications are
relatively minor and PKW can still be considered a relatively safe
technique that can be particularly useful in patients with
compromised health status and in situations were open surgery
is contraindicated [16].
In elderly individuals, PKW was also more benecial than
conservative treatment. These ndings should be interpreted
with caution as these two groups had different fracture type
distributions, and the use of PKW in osteoporotic bone is known
to cause poorer fracture reductions, and higher pin migration
rates [14,16]. Nevertheless, our observations are in agreement
with a previous retrospective study performed on 51 elderly
patients who were treated either with PKW or conservatively
[16].
In 2-fragment fractures, PKW achieved better functional results
than conservative treatment in patients of similar age (Tables 1
and 3). These results are also consistent with previous studies that
reported satisfactory functional results associated to the use of
PKW in the treatment of 2-fragment fractures [14,17].
PHN vs. Conservative treatment
The functional outcome of PHN in individuals with similar
fracture types appeared to be age dependent. In young patients,

Table 2
Post-surgical results.

Group characteristics Conservative (n = 25) Locking plate (n = 44) Kirschner wires (n = 25) Proximal humeral nail (n = 19)

Follow-up, months (SD) 28.0 (SD 8.5) 25.9 (SD 15) 27.6 (SD 14) 22.5 (SD 9.0)
<25 N (%) 9 (36) 23 (52.3) 10 (40) 10 (52.6)
25 N (%) 16 (64) 21 (45.5) 15 (60) 9 (47.4)
OR (95% CI) 1 1.9 (CI 0.75.3) 1.19 (CI 0.455.4) 2.0 (CI 0.596.7)
OR (95% CI) 1 0.61 (CI 0.21.6) 1.0 (CI 0.34 3.0)
OR (95% CI) 1 1.7 (CI 0.55.6)

Complications N (%) 1 (4) 8 (18.2) 6 (24.0) 7 (36.8)

DASH N (SD) 38.4 (19.2) 34.5 (20.3) 28.9 (20.9) 34.9 (26.5)
CS N (SD) 57.2 (12.7)a 62.9 (16.8) 70.0 (18.8) a
63.9 (23.6)

Abbreviations: CI, condence interval; OR, odds ratio; CS, absolute Constant-Murley score; DASH, the disabilities of the arm shoulder and hand score; N (%), values are
expressed as percentages.
a
Indicates signicant differences between two values with the same letter, p = 0.012 (power 87%). The rest of p values are not signicant.

Descargado de ClinicalKey.es desde Universidad Nacional Autonoma de Mexico enero 24, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
 I. Tamimi et al. / Injury, Int. J. Care Injured 46 (2015) 19211929 1927

Table 3
Functional results according to age and fracture type.

Clinical feature Conservative (n = 25) Locking plate (n = 44) Kirschner wires (n = 25) Proximal humeral nail (n = 19)

DASH CS DASH CS DASH CS DASH CS

Fracture type
2-Fragments 41.7 (SD 30.6) 53.9 (SD 12.6)a 32.1 (SD 15.1) 63.0 (SD 4) 27.2 (SD 20.3) 70.4 (SD 17.4)a 36.0 (SD 29.4) 63.3 (SD 24.9)
34 Fragments 37.1 (SD 13.5) 55.7 (SD 14.3) 35.0 (SD 21.6) 62.9 (SD 18.4) 32.4 (SD 23) 70.5 (SD 23.0) 32.4 (SD 20.9) 65.2 (SD 22.6)

Age group
Patients <65 41.7 (SD 22.0) 54.7 (SD 4.6)c,b 33.6 (SD 22) 65.6 (SD 15.6) 23.9 (SD 20.7) 74.0 (SD 20.8)c 22.9 (SD 20.6) 77.2 (SD 20.6)b, e

Patients 65 37.1 (SD 18.5) 55.3 (SD 15.9)f 35.4 (SD 19.2) 60.2 (SD 17.7) 31.2 (SD 21.1) 68.7 (SD 18.3)d, f
45.7 (SD 27.4) 51.9 (SD 19.9)d, e

Abbreviations: CS, absolute Constant-Murley score; DASH, the disabilities of the arm shoulder and hand score.
af
Indicate signicant differences between two values with the same letter. ap = 0.048 (power 77%), bp = 0.01 (power 96%), cp = 0.031 (power 87%), dp = 0.020 (power 70%),
e
p = 0.014 (power 84%), fp = 0.029 (power 74%). The rest of p values are not signicant.

PHN was associated with better results than conservative
treatment, yet the outcome gradually deteriorated with age
(Fig. 5) (Tables 1 and 3). A previous study on a series of 40
patients, reported that conservative treatment had better func-
tional results than PHN; however, it did not analyse the effect age
may have on the functional outcome considering that the age of the
studied individuals varied signicantly (i.e., 2787 years) [18]. The
use of PHN has the advantage of causing less soft tissue damage
and fewer infections at the fracture site than other surgical
procedures [19]. However, the conventional antegrade approach
used in PHN can damage the subacromial space and rotator cuff
tendons, and cause postoperative shoulder pain [20]. The
prevalence of partial or complete tears of the rotator cuff tendons
in patients over the age of 60 years is approximately of 54% [21].
Elderly patients with debilitated rotator cuff tendons could be less
tolerant to the iatrogenic injuries caused by the anterograde
surgical approach used in PHN. In addition, poor bone quality in
elderly individuals could also condition the mechanical stability of
proximal humeral fractures leading to decient primary xations
and implant failure [22,23]. These factors could explain why PHN
seems to achieve better results in younger individuals.
PKW vs. PHN
The use of PKW in the elderly achieved better functional results
than PHN in patients with a comparable fracture distribution
(Tables 1 and 3). As mentioned above, these results could be
explained by the additional damage to rotator cuff tendons caused
by the insertion of PHN in older individuals [22,23]. Accordingly,
the use of PKW could be a better option than PHN in older patients.
LP vs. Conservative
The use of LP did not prove to be more benecial than
conservative treatment in neither young nor elderly individuals
(Table 3) (Fig. 5). These results are consistent with a previous work
that reported no differences in the functional outcome between
conservative and LP in a randomised controlled trial conducted on
patients above 60 years of age with displaced 3 and 4-fragment
[24]. Some studies have also warned of potential complications
associated with the use of LP such as humeral head necrosis, plate
impingement and screw cut out [25,26]. These complications can
be related to the surgical technique, the position of the implant and
locking screws, and the characteristics and severity of the fracture
[25,26]. LP is currently used in the treatment of displaced 2, 3 and
4-fragment fractures, proximal humerus pseudarthrosis, and when
the bone quality is poor [2,27,28]. However, these indications may
change as new evidence does not nd signicant advantages of LP
over conservative treatment [22,29,30].
On the other hand, conservative treatment is a safe therapeutic
option for the management of proximal humeral fractures of
different complexity, with reasonable functional results, and few
complications [5]. Conservative treatment is currently indicated in
non-displaced fractures and fractures of the greater tuberosity
displaced less than 5 mm superiorly or 10 mm posteriorly [2].
However, this concept may change in the near future as authors
continue to report evidence supporting the use of this therapeutic
option in displaced 34-fragment fractures [24,3133]. Further-
more, studies conducted on elderly patients treated either
conservatively or with hemiarthroplasty have not found differ-
ences in the nal functional outcome between these two treatment
options [32,33].

Strengths and weaknesses

Fig. 5. The functional outcome (Constant-Murley score) of patients treated with
proximal humeral nails vs. age (R Square = 0.41, p = 0.003).
This study points out how age may inuence the outcome of
methods used in the treatment of patients with proximal humerus
fractures. Moreover, this is the rst cohort study that compares four
techniques commonly used in the treatment of proximal humerus
fractures, and provides novel ndings that challenge some of the
current thoughts on aggressive surgical management of these
fractures. Another strength of this study is the homogeneity of the
study groups regarding the age and gender of the patients, which
increases the reliability of our results. In addition, despite the small
cohort size we managed to detect signicant differences between
different treatment options with a statistical power that ranged
between 70 and 96% and an a-error of 0.05.
On the other hand, there are a series of possible limitations
associated with this work. Firstly, the small sample size in each
study group limited the statistical power to detect b errors (e.g.,
stating that there were no signicant differences between LP vs.
conservative treatment). We must also add that there could be a

Descargado de ClinicalKey.es desde Universidad Nacional Autonoma de Mexico enero 24, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
1928 I. Tamimi et al. / Injury, Int. J. Care Injured 46 (2015) 19211929
selection bias, as surgeons at this institution selected the treatment
method based on their own experience and clinical expertise.
Therefore, treatment allocation in this study was not randomised
and the indication for treatment could be related to uncontrolled
risk factors such as demographic differences between groups (e.g.,
differences in the fracture distribution between the different
groups). In addition, we decided to include displaced 3 and 4-
fragment fractures in the same category to gain additional
statistical power; consequently, we could not identify differences
between these two fracture types.
Another limitation was the fact that we did not include other
relevant surgical methods that are routinely used in many centers
such as hemiarthroplasty and percutaneous xed angle LP. This
was due to the limited number of cases that were registered in our
center. Finally, the mean follow-up period was of 26.2 SD 12.6
months, which can be considered medium-term. Therefore, the
analysis of long term complications such or avascular necrosis
post-traumatic arthritis was relatively limited.
Future research
This study is not a denitive opinion since it lacks the rigor of
prospective randomised trials but it indicates that further work
needs to be done in the eld. Accordingly, future clinical trials
should be performed in order to conrm our ndings and reduce
the limitations that we confronted in this work. Moreover, future
studies could be designed to compare the use of PKW, percutane-
ous xed angle LP and hemiarthroplasy vs. conservative treatment
taking into consideration the fracture type and age of the patients.
Clinical message
In this study PKW was superior to conservative treatment in
individuals of all ages, especially in patients with 2-fragment
fractures. PKW was also a better option than PHN in the elderly.
PKW is a non-invasive technique, relatively safe, and cost effective,
and therefore should be considered as an appropriate option in the
previously mentioned situations.
We believe that PHN could be of particular use in young
patients, but should be avoided in patients above the age of 65
years who frequently have a poor bone quality and deteriorated
rotator cuffs.
Finally, no differences were observed between LP and
conservative treatment in any of the analysed categories.
Therefore, we cannot encourage the use of LP in the treatment
of proximal humerus fractures.
Conclusions
In our hands, patients under the age of 65 years with displaced
2-fragment fractures had better functional results when treated
with PKW or PHN. The use of PKW in the treatment of patients
above the age of 65 years with displaced 2-fragment fractures was
associated with superior functional results than conservative
treatment. The use of PHN in elderly patients was associated with
poor functional results. Finally, there were no differences in the
functional outcome between conservative treatment and surgery
in the treatment of patients with displaced 34-fragment
fractures.
Conict of interest
There are neither nancial nor non-nancial competing
interests.
Descargado de ClinicalKey.es desde Universidad Nacional Autonoma de Mexico enero 24, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
Acknowledgements
XiXi Wu DDSa.
References
[1] Calvo E, Morcillo D, Foruria AM, Redondo-Santamaria E, Osorio-Picorne F,
Caeiro JR. Nondisplaced proximal humeral fractures: high incidence among
outpatient-treated osteoporotic fractures and severe impact on upper extrem-
ity function and patient subjective health perception. J Shoulder Elbow Surg/
Am Shoulder Elbow Surg 2011;20:795801.
[2] Rockwood CA, Bucholz RW, Green DP, Court-Brown CM, Heckman JD, Tornetta
P. Rockwood and Greens fractures in adults: Wolters Kluwer health/Lippincott
Williams & Wilkins; 2010.
[3] Handoll HH, Ollivere BJ, Rollins KE. Interventions for treating proximal hu-
meral fractures in adults. Cochrane Database Syst Rev 2012;12:Cd000434.
[4] Lekic N, Montero NM, Takemoto RC, Davidovitch RI, Egol KA. Treatment of two-
part proximal humerus fractures: intramedullary nail compared to locked
plating. HSS J: Musculoskeletal J Hospital Special Surg 2012;8:8691.
[5] Hauschild O, Konrad G, Audige L, de Boer P, Lambert SM, Hertel R, et al.
Operative versus non-operative treatment for two-part surgical neck fractures
of the proximal humerus. Arch Orthop Trauma Surg 2013;133:138593.
[6] Sankar B, Nicholson S, Henman PD. Centromedullary manipulation and stabi-
lization of completely displaced proximal humerus fractures in adolescents.
Orthopedics 2012;35:85660.
[7] Erasmo R, Guerra G, Guerra L. Fractures and fracture-dislocations of the
proximal humerus: a retrospective analysis of 82 cases treated with the
Philos((R)) locking plate. Injury 2014;45(Suppl. 6):S438.
[8] Atesok K, Liebergall M, Sucher E, Temper M, Mosheiff R, Peyser A. Treatment of
pathological humeral shaft fractures with unreamed humeral nail. Ann Surg
Oncol 2007;14:14938.
[9] Lopiz Y, Garcia-Coiradas J, Garcia-Fernandez C, Marco F. Proximal humerus
nailing: a randomized clinical trial between curvilinear and straight nails. J
Shoulder Elbow Surg/Am Shoulder Elbow Surg 2014;23:36976.
[10] Constant CR, Gerber C, Emery RJ, Sojbjerg JO, Gohlke F, Boileau P. A review of
the Constant score: modications and guidelines for its use. J Shoulder Elbow
Surg/Am Shoulder Elbow Surg 2008;17:35561.
[11] Beaton DE, Wright JG, Katz JN. Development of the QuickDASH: comparison
of three item-reduction approaches. J Bone Joint Surg Am Vol 2005;87:
103846.
[12] Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying
prognostic comorbidity in longitudinal studies: development and validation.
J Chron Dis 1987;40:37383.
[13] Elidrissi M, Bensaad S, Shimi M, Elibrahimi A, Elmrini A. [Surgical treatment of
proximal humeral fractures anatomical plate versus palm tree pinning (26
cases)]. Chirurgie de la main 2013;32:259.
[14] Magovern B, Ramsey ML. Percutaneous xation of proximal humerus frac-
tures. Orthop Clin North Am 2008;39:40516. v.
[15] Rasool MN. Ulnar nerve injury after K-wire xation of supracondylar humerus
fractures in children. J Pediat Orthoped 1998;18:68690.
[16] Innocenti M, Carulli C, Civinini R, Matassi F, Tani M, Muncibi F. Displaced
fragility fractures of proximal humerus in elderly patients affected by severe
comorbidities: percutaneous xation and conservative treatment. Aging Clin
Exp Res 2013;25:44752.
[17] Matziolis D, Kaeaeb M, Zandi SS, Perka C, Greiner S. Surgical treatment of two-
part fractures of the proximal humerus: comparison of xed-angle plate
osteosynthesis and Zifko nails. Injury 2010;41:10416.
[18] van den Broek CM, van den Besselaar M, Coenen JM, Vegt PA. Displaced
proximal humeral fractures: intramedullary nailing versus conservative treat-
ment. Arch Orthop Trauma Surg 2007;127:45963.
[19] Lin J. Treatment of humeral shaft fractures with humeral locked nail and
comparison with plate xation. J Trauma 1998;44:85964.
[20] Cox MA, Dolan M, Synnott K, McElwain JP. Closed interlocking nailing of
humeral shaft fractures with the Russell-Taylor nail. J Orthopaed Trauma
2000;14:34953.
[21] Sher JS, Uribe JW, Posada A, Murphy BJ, Zlatkin MB. Abnormal ndings on
magnetic resonance images of asymptomatic shoulders. J Bone Joint Surg Am
Vol 1995;77:105.
[22] Faraj D, Kooistra BW, Vd Stappen WA, Werre AJ. Results of 131 consecutive
operated patients with a displaced proximal humerus fracture: an analysis
with more than two years follow-up. Eur J Orthopaed Surg Traumatol: Orthop
Traumatol 2011;21:712.
[23] Euler SA, Hengg C, Kolp D, Wambacher M, Kralinger F. Lack of fth anchoring
point and violation of the insertion of the rotator cuff during antegrade
humeral nailing: pitfalls in straight antegrade humeral nailing. Bone Joint J
2014;96-b:24953.
[24] Fjalestad T, Hole MO, Hovden IA, Blucher J, Stromsoe K. Surgical treatment
with an angular stable plate for complex displaced proximal humeral fractures
in elderly patients: a randomized controlled trial. J Orthopaed Trauma
2012;26:98106.
[25] Yang H, Li Z, Zhou F, Wang D, Zhong B. A prospective clinical study of proximal
humerus fractures treated with a locking proximal humerus plate. J Orthop
Trauma 2011;25:117.
 I. Tamimi et al. / Injury, Int. J. Care Injured 46 (2015) 19211929 1929

[26] Schulte LM, Matteini LE, Neviaser RJ. Proximal periarticular locking plates in
proximal humeral fractures: functional outcomes. J Shoulder Elbow Surg
2011;20:123440.
[27] Duralde XA, Leddy LR. The results of ORIF of displaced unstable proximal
humeral fractures using a locking plate. J Shoulder Elbow Surg/Am Shoulder
Elbow Surg 2010;19:4808.
[28] Jaeger M LF, Li W. Proximal humerus. Switzerland: AO Foundation; 2011.
[29] Zhu Y, Lu Y, Shen J, Zhang J, Jiang C. Locking intramedullary nails and locking
plates in the treatment of two-part proximal humeral surgical neck fractures:
a prospective randomized trial with a minimum of three years of follow-up.
J Bone Joint Surg Am Vol 2011;93:15968.
[30] Bachelier F, Pizanis A, Schwitalla J, Pohlemann T, Kohn D, Wirbel R. Treatment
for displaced proximal humerus fractures: comparison of interlocking plate
xation versus minimal invasive techniques. Eur J Orthop Surg Traumatol:
Orthoped Traumatol 2014;24:70714.
[31] Olerud P, Ahrengart L, Ponzer S, Saving J, Tidermark J. Internal xation versus
nonoperative treatment of displaced 3-part proximal humeral fractures in
elderly patients: a randomized controlled trial. J Shoulder Elbow Surg/Am
Shoulder Elbow Surg 2011;20:74755.
[32] Olerud P, Ahrengart L, Ponzer S, Saving J, Tidermark J. Hemiarthroplasty versus
nonoperative treatment of displaced 4-part proximal humeral fractures in
elderly patients: a randomized controlled trial. J Shoulder Elbow Surg/Am
Shoulder Elbow Surg 2011;20:102533.
[33] Boons HW, Goosen JH, van Grinsven S, van Susante JL, van Loon CJ. Hemi-
arthroplasty for humeral four-part fractures for patients 65 years and older: a
randomized controlled trial. Clin Orthopaed Related Res 2012;470:348391.

Descargado de ClinicalKey.es desde Universidad Nacional Autonoma de Mexico enero 24, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.