1874-3250/19 Send Orders for Reprints to reprints@benthamscience.

net

136

The Open Orthopaedics Journal

Content list available at: https://openorthopaedicsjournal.com

RESEARCH ARTICLE

Evaluating the Outcome of a New Functional Brace for the Management of

Humeral Shaft Fractures
Prasad Jankiram Athreya*, Vanessa Abbott, Luke Elias and Bijoy Thomas

Department of Orthopaedics, Blacktown Hospital, 18 Blacktown Rd, Blacktown, NSW 2148, Australia

Abstract:
Background:
Our practice identified several issues with the commercially available plastic off-the-shelf Sarmiento brace that is used in managing humeral shaft
fractures, with regards to comfort, moulding, and ability to hold reduction. A custom-moulded fibreglass brace was developed which was soft
padded for comfort, lightweight, and could be adjusted with changes in swelling.

Objective:
The aim of this study was to evaluate the use of this new brace.

Methods:
16 patients were identified having treatment with the brace from March 2011 to July 2013. Retrospective analysis of medical records and imaging
occurred to assess union and angulation. Prospective analysis of patient function was assessed with the CONSTANT, DASH and SPADI Shoulder
score.

Results:
Patients were initially managed with a U-Slab for an average of 26 days. Patients were followed up for an average of 70 days post brace
application. 5 patients were lost to follow up. The remaining 11 patients had the brace on for an average of 73 days. In the last follow up, 15
patients had an acceptable anterior/posterior angulation of less than 20 (1 patient - 22), varus/valgus angulation less than 20, and less than 3cm of
shortening. 12 patients had radiological evidence of union, with the other 4 demonstrating significant callus. 4 patients were recruited for
prospective analysis with DASH, SPADI and CONSTANT shoulder scores, and demonstrated minimal loss of function. There were no
complications of bracing treatment.

Conclusion:
Our new functional brace led to bony union in most patients, and from a clinical perspective, most patients were pain free and had minimal loss of
function.

Keywords: Bracing, Custom bracing, Swelling, Functional bracing, Fibreglass, Humeral shaft fractures, Sarmiento.

Article History Received: September 29, 2018 Revised: January 15, 2019 Accepted: March 07, 2019

1. INTRODUCTION Sarmiento humeral brace is a commercially available brace

used to assist the management of these injuries. We propose a
Humeral shaft fractures can occur both in the athlete and
new custom-moulded functional humeral brace which is supe-
non-athletic individual, and in both young and elderly patients.
rior to the Sarmiento /thermoplastic brace in design, construct
These fractures can be managed both operatively and non-
operatively with comparable results. Non-operative manage- and patient comfort, whilst providing the stability required for
ment typically involves the use of a functional brace, which fracture healing. This study investigates the radiological and
allows individuals to use the affected limb while it heals. The clinical outcomes of this new brace.
Historically, early treatment of humeral shaft fractures
* Address correspondence to this author at the Department of Orthopaedics,
Blacktown Hospital, 18 Blacktown Rd, Blacktown, NSW 2148, Australia; involved immobilisation, which often extended from beyond
Tel: +61400319559; E-mail: [email protected] the shoulder past the elbow. This typically resulted in stiffness

DOI: 10.2174/1874210601913010136, 2019, 13, 136-143

New Brace for Humeral Shaft Fractures The Open Orthopaedics Journal, 2019, Volume 13 137

at both the shoulder and elbow joints, an outcome which was The commercially available sarmiento brace is a plastic
not beneficial. In 1977, Sarmiento et al. first described func- off-the-shelf brace which is commonly used for the manage-
tional bracing, a form of immobilisation which promoted frac- ment of humeral shaft fractures. Although effective, our prac-
ture healing and encouraged motion at the elbow and shoulder tice has recognised several issues:
joint [1].
Patient Comfort
Thus, non-operative management through the use of a Plastic edges can cause skin problems and
functional brace is the predominant treatment choice for these pain
injuries, with acceptable healing and functional outcomes in Brace is bulky and conspicuous
the majority of patients. Consequently, functional humeral Expensive
bracing has become the frontline clinical choice for definitive Application
management of the humeral shaft fractures. Sizing - Does not fit all arms
Less ability to mould to patient’s arm, especia-
Functional humeral braces should, thus, be designed with lly over time as swelling resolves
several important considerations in mind: Fracture reduction and the outcome
may be affected
For fracture healing
Having a well designed anterior component After recognising these problems and their impact on
with a bicipital contour patient compliance with bracing, and consequent suboptimal
Having a well designed posterior component results, a custom brace was developed (Fig. 1). This allowed:
with a flat mould for the triceps muscle
Be able to maintain adequate compression Custom moulding to an individual patient’s arm, with
over time (by use of adjustable straps) an ability to remake the brace as arm size changed with
Provides support and fracture union swelling
Allowing unimpeded elbow flexion and Cheaper and cost-effective materials - fibreglass
extension A brace that was easier to mould than thermoplastic
Allowing unimpeded shoulder movement materials
Application Soft padded edges to reduce the risk of skin injuries
Be simple and easy to apply, and adjust from brace
A much more circumferential brace to promote comp-
Be cost effective
ression and fracture reduction
For patient comfort Reduced weight and smaller profile allowing patients
Have skin care in mind to have clothing over brace
Be lightweight Easier to remove and apply by patients and physicians
Be easy to manage for showering, hygiene and / physiotherapists
clothing Quick, easy and clean application technique

Fig. (1). Our Custom Moulded Brace.

138 The Open Orthopaedics Journal, 2019, Volume 13 Athreya et al.

2. MATERIAL AND METHODS 2. A prospective assessment of patient functional status

and range of motion:
The aim of this study was to evaluate the use of a specific
functional brace in the management of patients with closed
[a] Current patient clinical and functional status
humeral shaft fractures. The evaluation included:
[i] Disabilities of the Arm, Shoulder and Hand
An assessment of radiological fracture healing and al- Score (DASH)
ignment [ii] Shoulder Pain and Disability Index (SPADI)
An assessment of patient clinical and functional [b] Physiotherapy assessment of elbow and shoulder range
outcome of motion, compared with the unaffected side
[i] CONSTANT Shoulder Score
17 patients, aged between 11 and 91 years, were identified [ii] Elbow range of motion
from medical records as having treatment with the brace,
March 2011 to July 2013. Administration of the SPADI/DASH/CONSTANT ques-
tionnaires, and measurements of the elbow and shoulder range
Inclusion Criteria of motion was performed by one senior physiotherapist. Range
Patients who sustained a closed humeral shaft of motion was measured using a goniometer, and for the
fracture who were treated with the above hum- strength assessment component of the CONSTANT score, a
eral brace hand-held force dynamometer was used.
Exclusion Criteria:
Patients who initially had a different brace
(e.g. sarmiento) prior to the above brace being
fitted
Patients with segmental fractures
Patients with fractures extending into the hum-
eral surgical neck
Patients with open fractures
Patients with radial nerve injury initial pre-
sentation

Patients were initially evaluated and managed in the

emergency department, and placed in a Plaster of Paris U slab,
before being referred to the outpatient fracture clinic for further
review and assessment.
When swelling was deemed acceptable, the functional
brace was applied by qualified experienced senior physio-
therapists (Fig. 2). Active movement of the elbow was
encouraged, with pendular exercises at the shoulder commen-
ced at 3 weeks post application of brace.
Fig. (2). Application of our custom brace.
The functional brace was removed upon confirmation of
clinical OR radiographic union of the fracture. Radiographic
union was defined as being present when osseous bridging Outcomes measured included:
between the main fragments was observed on at least one
radiograph, and clinical union was defined as there is no Fracture Union
pain/tenderness at the fracture site. Radiographic
The study involved two components: Clinical
Functional outcomes (with SPADI and DASH scores)
1. A retrospective analysis of patient records and imaging,
Range of motion of elbow
via patient paper and electronic record, recording data
Range of motion of shoulder (with CONSTANT
including:
scores)
Patient age Ethics approval from local health district was obtained.
Patient gender
Mechanism of fracture 3. RESULTS
Type of fracture Of the 17 patients identified as meeting inclusion criteria, 9
Initial management patients were unable to be contacted, 3 patients did not wish to
Duration of initial management participate in the prospective component of the study, and 1
Duration of brace treatment patient died during treatment (due to lung co-morbidities) (Fig.
Complications (e.g. pressure areas, rash, skin injuries) 3). Consequently, a total of 16 patients were evaluated for the
Progressive radiograph appearances of fractures and retrospective component of the study and 4 patients for the
healing prospective component of the study.
New Brace for Humeral Shaft Fractures The Open Orthopaedics Journal, 2019, Volume 13 139

Fig. (3). Recruitment of Patients for Prospective Component.

Table 1. Patient demographics.

Patient Current Age Gender Date of Injury Age at Mechanism Side AO Type Duration of U- Duration of Brace
Injury Slab (Days) (Days)
1 22 Male 28/11/2012 22 Fall, standing height Right 12-A1 21 84
2 91 Female 16/11/2012 91 Fall, standing height Right 12-A1 12 68
3 59 Female 26/12/2012 58 Fall, standing height Left 12-B1 20 67
4 61 Female 13/06/2011 59 Fall, standing height Left 12-B1 60 31
5 20 Female 9/09/2011 19 Assault Left 12-B1 26 Lost to Follow Up
6 21 Female 23/12/2011 20 Motor Vehicle Accident Left 12-B1 32 63
7 29 Male 19/06/2011 27 Wrestling Right 12-A1 31 34
8 11 Male 31/03/2013 11 Fall, 1m Right 12-A3 30 Lost to Follow Up
9 30 Male 14/06/2012 32 Assault Right 12-A1 6 65
10 42 Female 30/12/2011 41 Fall, standing height Right 12-A1 38 63
11 15 Female 6/01/2012 14 Motor Vehicle Accident Left 12-A3 19 Lost to Follow Up
12 81 Male 29/01/2013 80 Fall, standing height Left 12-A2 14 99
13 66 Female 27/05/2012 65 Fall, standing height Left 12-C1 23 165
14 53 Male 5/01/2013 53 Fall, ladder Left 12-A3 25 Lost to Follow Up
15 44 Male 5/03/2011 42 Fall, horse Right 12-A1 25 Lost to Follow Up
16 70 Female 23/06/2012 69 Fall, standing height Left 12-A1 32 61

Demographic information for the 16 patients in the study is 3.1. Fracture Union
detailed in Table 1. All patients were initially managed in a U- All patients had their brace removed once evidence of
Slab for which they were in for an average of 26 days before union was present on x-ray, or when no pain or tenderness was
the functional brace was applied. 5 patients were lost to follow elicited at fracture site. At their final visit, all but two patients,
up, and it is unclear when these patients had their brace who were lost to follow up, proceeded to either radiographic or
removed. The functional brace was in place for an average of clinical union or both (Table 2). At last follow up, 15 patients
73 days or 10.5 weeks before being removed (n = 11). had acceptable anterior/posterior angulation of less than 20 (1
patient - 22), varus/valgus angulation less than 20, and less
There were no complications (e.g. pressure areas, wounds) than 3cm of shortening. 12 patients had radiological evidence
of the bracing treatment. of union, with the other 4 demonstrating significant callus.
140 The Open Orthopaedics Journal, 2019, Volume 13 Athreya et al.

Fig. (4). Patient Radiographs.

Table 2. Radiographic outcomes.

Patient Anterior / Posterior Apex Varus/Valgus Angulation Position Radiological Evidence of Clinical Evidence of Union
Angulation Union
1 1 Posterior 3 Valgus Yes Yes
2 20 Posterior 15 Valgus Yes Yes
3 6 Posterior 7 Varus Yes Yes
4 5 Posterior 3 Valgus Yes Yes
5 22 Posterior 15 Varus Yes Lost to Follow Up
6 15 Posterior 15 Varus Yes Yes
7 8 Posterior 13 Varus Yes Yes
8 14 Posterior 19 Varus Yes Lost to Follow Up
9 11 Posterior 9 Varus Yes Yes
10 13 Posterior 5 Varus Yes Yes
11 19 Posterior 6 Varus No Lost to Follow Up
12 7 Anterior 16 Varus No Yes
13 7 Anterior 7 Varus Yes Yes
4 4 Posterior 15 Varus No Lost to Follow Up
15 12 Posterior 15 Varus Yes Lost to Follow Up
16 16 Posterior 16 Varus No Yes
New Brace for Humeral Shaft Fractures The Open Orthopaedics Journal, 2019, Volume 13 141

3.2. Patient Outcomes (Table 3) Differences in length of initial management and follow
up
Fig. (4) shows some examples of some of the study
patient’s radiographs.
Questionnaires being utilised at a single point in time,
thus preventing any comparative analysis over time.
Prospectively, patients (n = 4) were assessed at an average
of 510 days (18 months) post brace removal. However, despite these inherent limitations, most fractures
went onto bony union, and most patients were pain-free, and
The average SPADI score at the time of assessment was
3.3, and the average DASH score was 32.1. The analysis of the
has reasonable clinical function (as demonstrated by DASH,
pain component of the SPADI questionnaire revealed that three SPADI and CONSTANT scores).
patients were pain-free at the time of assessment, with one Fractures of the humeral shaft can occur from high energy
patient having some residual pain, specifically when lying on injuries, such as car or cycling accidents, or from low energy
the affected limb, and reaching for an object on a high shelf. events such as falls from standing height. Injuries in the young
Similarly, more detailed analysis of the DASH scores population are typically of higher energy, whilst elderly
revealed that after brace treatment, three patients (the same patients sustain fractures from low energy accidents such as
three without residual pain above), showed no or mild falls.
difficulty in performing all questioned tasks. The fourth patient Humeral shaft fractures can be managed either non-
had only mild or moderate difficulty in overhead activities, operatively, with use of a brace, or surgically, with intrame-
such as changing a lightbulb or placing objects on high shelves. dullary nails or plate fixation. Functional bracing works by
All patients had less than 11 points difference in Constant providing a hydraulic compressive centripetal force applied
Scores between the affected and unaffected limbs, indicating more or less uniformly throughout the diaphyseal segment of
an ‘Excellent’ score. Elbow range of motion in the affected the limb [2]. The cylindrical brace effectively compresses the
limb in all patients was comparable to the unaffected limb, biceps and triceps muscles, with an anterior and posterior shell
with no limitations or residual stiffness. that is contoured to accommodate the arm musculature, whilst
allowing early shoulder and elbow motion [2, 3].
Overall these scores indicate a positive functional and pain
related outcome for these patients, with three patients having As described by Sarmiento et al., the fracture callus
had an almost complete return to previous function. created through functional activity during the healing process is
more robust and is mechanically stronger than that gained
4. DISCUSSION through rigid immobilisation [2]. The advantage of this type of
Our practice identified several issues with the commer- bracing is that it avoids unnecessary immobilization of other
cially available functional Sarmiento brace, prompting the joints and allows for earlier restoration of motion and function
development of a custom moulded synthetic cast brace used in to the injured extremity [2].
this study.
In 2000, Sarmiento et al. reported on 620 patients treated
Being a small predominantly retrospective study, we were with a functional brace for both closed and open humeral shaft
limited by: fractures, between 1978 and 1990 [3]. They demonstrated that
98% of all closed injuries and 94% of all open fractures
A small sample size proceeded to union [3]. Similarly, in their analysis of 195
A lack of control (i.e. comparing with the off-the-shelf fractures in 1995, Ostermann et al., reported a nonunion rate of
Sarmiento brace, or even comparing with surgical 2% with treatment using functional humeral bracing [4]. High
management) rates of union have been demonstrated extensively over the

Table 3. DASH, SPADI, CONSTANT scores, and elbow range of motion.

Patient SPADI DASH Elbow Range of Motion Elbow Range of Motion

Affected Limb Unaffected Limb
13 0.0 24.2 0-140 -5-140
12 0.0 38.3 0-130 0-130
10 11.9 40.0 -5-150 -5-150
16 1.3 25.8 0-130 -5-130
Average 3.3 32.1 - -
CONSTANT Score Affected Limb Unaffected Limb Difference Interpretation
13 85 81 -4 Excellent
12 74 74 0 Excellent
10 82 90 8 Excellent
16 98 92 -6 Excellent
Average 84.75 84.25 -0.5 Excellent
142 The Open Orthopaedics Journal, 2019, Volume 13 Athreya et al.

years with acceptable alignment and healing occurring in more From economic perspective, the cost of our custom brace
than 90% of cases [4 - 9]. was AU$10.00. It took approximately thirty minutes to apply,
and some patients had braces redone over the course of their
Proponents of surgical management report rates of union
treatment due to changes in swelling, and general wear and
and more anatomical reduction than functional bracing [10].
tear. The cheap cost of materials for our brace enable a
Given the mobility of the shoulder and elbow, which is
reapplication of the brace as often as needed (maximum was
encouraged in functional bracing, angular deformities occur.
three times for one patient in our study). The off-the-shelf
However, these angular deformities are generally well tolerated
Sarmiento brace retails for AU$47.70, which is approximately
with minimal functional impairment [1, 2, 11 - 13]. Matsunaga
4.5 times the cost of our brace.
et al., compared surgical fixation and functional bracing and
found patients who were fixed had higher DASH scores, higher There have been numerous studies about the success of
rates of union, and less anteroposterior residual angulation functional bracing in humeral shaft fractures. However, to our
[14]. However, there were no differences with regards to the knowledge, most of the braces used in these studies are
SF-36 score, pain level, and Constant score. Surgical complica- commercial plastic or thermoplastic designs, which we feel are
tions included superficial infections, scarring and transient limited with regards to their ‘moulding’ and customised
radial nerve neuropraxias. support capacity. Limitations to functional bracing do exist and
Operative treatment may provide more predictable need to be taken into consideration when determining the
alignment (which may not be functionally important) as well as appropriate treatment strategy for each patient. Patient compli-
immediate stability (which may lead to more rapid restoration ance is the main limiting factor, as bracing can be a long
of function), but it is associated with additional operative risks. treatment option, thus patients who are non-compliant with
bracing are at a higher risk of failure. Non-compliance, in our
The brace used by Matsunaga et al., was a thermoplastic experience, has been due to issues with comfort, high profile
type, and 2 patients in the bracing group did not tolerate the and heavy braces, and poor pain control due to poor moulding.
brace for reasons undisclosed, and 5 patients developed contact By using synthetic semi-rigid casting material as our moulding
dermatitis. Our custom-moulded fibreglass brace had no such material, and providing a lighter weight and soft padded brace,
complications. we feel that our brace achieves adequate fracture union with
Radiological parameters which have been deemed accep- good clinical outcomes.
table for fracture reduction are [1, 2, 11 - 13]:
CONCLUSION
30 of varus/valgus angulation With these promising results, we hope to recruit more
20 of anterior/posterior bowing patients in a future prospective randomised trial, comparing our
Up to 15 of internal rotation brace with other products or treatment interventions, with
Less than 3cm of shortening intent to demonstrate that our custom functional brace is
economic, better tolerated by patients, and effective in treating
Beyond these limits, deformity and functional impairment humeral shaft fractures.
may be shown clinically.
ETHICAL APPROVAL AND CONSENT TO
The design of our brace allowed unimpeded elbow and
PARTICIPATE
shoulder joint motion, with comparable elbow range of motion
demonstrated in all our patients, and ‘excellent’ shoulder range NSW health western sydney, Local health district approved
of motion (CONSTANT scores) in all patients. this study.
We recognise that drawing conclusions from our DASH
HUMAN AND ANIMAL RIGHTS
and SPADI scores is difficult, as questionnaires were adminis-
tered at one point in time, with no ability, thus to detect any No Animals were used in this research. All human research
changes over time. Despite this, our study did demonstrate procedures followed were in accordance with the ethical
benefits in both pain and functional outcomes, which reflects standards of the committee responsible for human experi-
directly on the efficacy of the custom functional brace. mentation (institutional and national), and with the Helsinki
Declaration of 1975, as revised in 2013.
Excluding two patients who were lost to follow up, all
patients proceeded to either clinical or radiological union, and CONSENT FOR PUBLICATION
analysis of radiographs revealed all but one fracture in
acceptable alignment. Written and informed consent was obtained from patients
to use radiograph images in Fig. (4).
Our results are similar to Swellengrebel et al., who
compared a thermoplastic humeral brace with an above elbow CONFLICT OF INTEREST
fibreglass cast in 75 patients [15]. They found no differences in
union rate, and detected no major complications associated The authors declare no conflict of interest, financial or
with the cast treatment. We believe our brace design is slightly otherwise.
superior in that it is not circumferential, can be adjusted using
ACKNOWLEDGEMENTS
the velcro straps, allows easier inspection of the skin and soft
tissue, and does not immobilise the elbow. Declared none.
New Brace for Humeral Shaft Fractures The Open Orthopaedics Journal, 2019, Volume 13 143

REFERENCES Functional treatment of humeral shaft fractures: Indications and

results. J Orthop Trauma 1997; 11(4): 283-7.
[1] Sarmiento A, Kinman PB, Galvin EG, Schmitt RH, Phillips JG. [http://dx.doi.org/10.1097/00005131-199705000-00011] [PMID: 9258
827]
Functional bracing of fractures of the shaft of the humerus. J Bone
[9] Zagorski JB, Latta LL, Zych GA, Finnieston AR. Diaphyseal fractures
Joint Surg Am 1977; 59(5): 596-601.
of the humerus. Treatment with prefabricated braces. J Bone Joint
[http://dx.doi.org/10.2106/00004623-197759050-00004] [PMID: 873
Surg Am 1988; 70(4): 607-10.
955]
[http://dx.doi.org/10.2106/00004623-198870040-00018] [PMID: 3356
[2] Walker M, Palumbo B, Badman B, Brooks J, Van Gelderen J, Mighell
728]
M. Humeral shaft fractures: A review. J Shoulder Elbow Surg 2011;
[10] Denard A Jr, Richards JE, Obremskey WT, Tucker MC, Floyd M,
20(5): 833-44.
Herzog GA. Outcome of nonoperative vs operative treatment of
[http://dx.doi.org/10.1016/j.jse.2010.11.030] [PMID: 21393016]
humeral shaft fractures: A retrospective study of 213 patients.
[3] Sarmiento A, Zagorski JB, Zych GA, Latta LL, Capps CA. Functional
Orthopedics 2010; 33(8)
bracing for the treatment of fractures of the humeral diaphysis. J Bone
[PMID: 20704103]
Joint Surg Am 2000; 82(4): 478-86. [11] Foulk DA, Szabo RM. Diaphyseal humerus fractures: Natural history
[http://dx.doi.org/10.2106/00004623-200004000-00003] [PMID: and occurrence of nonunion. Orthopedics 1995; 18(4): 333-5.
10761 938] [PMID: 7603916]
[4] Ostermann PA, Ekkernkamp A, Muhr G. Functional bracing of shaft [12] McMaster WC, Tivnon MC, Waugh TR. Cast brace for the upper
fractures of the humerus - an analysis of 195 cases Orthop Trans 1993; extremity. Clin Orthop Relat Res 1975; (109): 126-9.
17(937) [http://dx.doi.org/10.1097/00003086-197506000-00018] [PMID: 1132
[5] Koch PP, Gross DF, Gerber C. The results of functional (Sarmiento) 195]
bracing of humeral shaft fractures. J Shoulder Elbow Surg 2002; [13] Tytherleigh-Strong G, Walls N, McQueen MM. The epidemiology of
11(2): 143-50. humeral shaft fractures. J Bone Joint Surg Br 1998; 80(2): 249-53.
[http://dx.doi.org/10.1067/mse.2002.121634] [PMID: 11988725] [http://dx.doi.org/10.1302/0301-620X.80B2.8113] [PMID: 9546454]
[6] Rutgers M, Ring D. Treatment of diaphyseal fractures of the humerus [14] Matsunaga FT, Tamaoki MJ, Matsumoto MH, Netto NA, Faloppa F,
using a functional brace. J Orthop Trauma 2006; 20(9): 597-601. Belloti JC. Minimally invasive osteosynthesis with a bridge plate
[http://dx.doi.org/10.1097/01.bot.0000249423.48074.82] [PMID: versus a functional brace for humeral shaft fractures: A randomized
17088 660] controlled trial. J Bone Joint Surg Am 2017; 99(7): 583-92.
[7] Sharma VK, Jain AK, Gupta RK, Tyagi AK, Sethi PK. Non-operative [http://dx.doi.org/10.2106/JBJS.16.00628] [PMID: 28375891]
treatment of fractures of the humeral shaft: A comparative study. J [15] Swellengrebel HJC, Saper D, Yi P, Weening AA, Ring D, Jawa A.
Indian Med Assoc 1991; 89(6): 157-60. Nonoperative treatment of closed extra-articular distal humeral shaft
[PMID: 1748787] fractures in adults: A comparison of functional bracing and above-
[8] Wallny T, Westermann K, Sagebiel C, Reimer M, Wagner UA. elbow casting Am J Orthop (Belle Mead NJ) 2018; 47(5)

© 2019 Athreya et al.

This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0), a copy of which is
available at: https://creativecommons.org/licenses/by/4.0/legalcode. This license permits unrestricted use, distribution, and reproduction in any medium, provided the
original author and source are credited.