709538

research-article2017
FAIXXX10.1177/1071100717709538Foot & Ankle InternationalCody et al

Article
Foot & Ankle International

Measuring Joint Flexibility in Hallux

18
The Author(s) 2017
Reprints and permissions:
Rigidus Using a Novel Flexibility Jig sagepub.com/journalsPermissions.nav
DOI: 10.1177/1071100717709538
https://doi.org/10.1177/1071100717709538
journals.sagepub.com/home/fai

Elizabeth A. Cody, MD1, Andrew P. Kraszewski, PhD1, Anca Marinescu, BA1,

Grace C. Kunas, BA1, Sriniwasan B. Mani, BS2, Smita Rao, PT, PhD3,
Howard H. Hillstrom, PhD1, and Scott J. Ellis, MD1

Abstract
Background: The flexibility of the first metatarsophalangeal (MTP) joint in patients with hallux rigidus (HR) has not been
studied. Compared to measuring range of motion alone, measures of joint flexibility provide additional information that
may prove useful in the assessment of HR. The purpose of this study was to assess the flexibility of the hallux MTP joint in
patients with HR compared to controls using a novel flexibility device.
Methods: Fifteen patients with Coughlin stage II or III HR and 20 healthy controls were recruited prospectively. Using a
custom flexibility jig, each of 2 raters performed a series of seated and standing tests on each subject. Dorsiflexion angle
and applied torque were plotted against each other to generate 5 different parameters of flexibility. Differences between
(1) HR patients and controls and (2) the sitting and standing testing positions were assessed with t tests. Intrarater test-
retest reliability, remove-replace reliability, and interrater reliability were assessed with intraclass correlation coefficients
(ICCs).
Results: Patients in the HR group were older than patients in the control group (P < .001) and had lower maximum
dorsiflexion (P < .001). HR patients were less flexible as measured by 3 of the 5 flexibility parameters: early flexibility (first
25% of motion; P = .027), laxity angle (P < .001), and torque angle (P = .002). After controlling for age, only laxity angle
differed significantly between HR patients and controls (P < .001). Generally, patients were more flexible when seated
compared to standing, with this effect being more marked in HR patients. All parameters had good or excellent intra- and
interrater reliability (ICC 0.60).
Conclusions: Hallux MTP joint flexibility was reliably assessed in HR patients using a flexibility device. Patients with HR
had decreased flexibility of the hallux MTP joint compared to control patients.
Level of Evidence: Level II, prospective comparative study.

Keywords: hallux, hallux rigidus, metatarsophalangeal joint, flexibility, biomechanics

Introduction pain occurs and whether maximum dorsiflexion appears

Hallux rigidus, or degenerative arthritis of the first metatar-
sophalangeal (MTP) joint, is a common condition charac-
terized by pain and stiffness of the first MTP joint. Patients
present with varying degrees of stiffness, osteophyte forma-
tion, and cartilage loss. Although radiographs are crucial to
diagnosis and evaluation, clinical parameters must also be
taken into account, as symptoms do not always correlate
with radiographic severity. The most widely used classifica-
tion system incorporates both clinical and radiographic fac-
tors to recommend treatment.2
Assessment of first MTP joint range of motion is an
important part of the initial assessment of hallux rigidus
(HR). As described by Coughlin and Shurnas, first MTP
joint range of motion may be measured with a goniometer,
with the plantar aspect of the foot considered neutral.
Specifically, the examiner notes where in the arc of motion
decreased.2 However, traditional range of motion measure-
ments are limited by variability in instruments and
technique.6,14 They capture only the plantarflexion and dor-
siflexion endpoints of motion. And while measuring range
of motion can identify loss in the overall arc of motion, it
does not provide any more objective information about the
function of the joint.
1
Hospital for Special Surgery, New York, NY, USA
2
Georgetown University School of Medicine, Washington, DC, USA
3
New York University Steinhardt School of Culture Education and
Human Development, New York, NY, USA
Corresponding Author:
Elizabeth A. Cody, MD, Hospital for Special Surgery, 535 East 70 Street,
New York, NY 10021, USA.
Email: codye@hss.edu
2 Foot & Ankle International 0(0)
In contrast, flexibility offers information on the mobil-
ity of the joint throughout its arc of motion, and is likely
affected by the scarring and capsular hypertrophy that
occurs in HR. A novel device has been developed to mea-
sure flexibility, which may be used in HR as an alternative
to standard range-of-motion measurements.10,12 This device
measures dorsiflexion angle and applied torque throughout
the range of motion to generate a curve. Early and late flex-
ibility, which are the slope of the curve, can then be calcu-
lated from the first 25% of motion and the last 25%,
respectively. To our knowledge, there is no other device that
measures first MTP flexibility. This flexibility jig has been
used to quantify first MTP flexibility in asymptomatic indi-
viduals, with excellent intrarater reliability.10
This device has not yet been used to evaluate patients
with HR. We believe that studying joint flexibility in
patients with HR may allow clinicians to better understand
the condition. Validating these flexibility measurements in
patients with HR is a necessary first step. In this study, we
aimed to (1) prospectively assess first MTP joint flexibility
in patients with Coughlin grade II and III HR compared to
asymptomatic controls, and (2) establish intra- and interra-
ter reliability for these measurements. We hypothesized that
patients with HR would have less flexibility compared to
controls and that intra- and interrater reliability would be
good to excellent.
Methods
All procedures were approved by the institutional review
board at the authors institution and written informed con-
sent was obtained from all enrolled patients.
Subject Enrollment
Two cohorts were prospectively recruited: a study group
and a control group. For the study group, patients were eli-
gible if they were between the ages of 18 and 70, had radio-
graphic Coughlin stage II or III HR, and were indicated for
cheilectomy following clinical evaluation by one of 2 fel-
lowship-trained orthopedic foot and ankle surgeons.
Patients with unilateral and bilateral symptomatic HR were
included. In the case of bilateral HR, the more symptomatic
foot was selected for testing. Grading of HR was based on
clinical evaluation and radiographic findings as outlined by
Coughlin and Shurnas.3 Exclusion criteria included prior
foot surgery, hallux valgus (defined by an intermetatarsal
angle greater than 9 degrees), any connective tissue disor-
der (such as Ehlers-Danlos or Marfan syndrome), or other
significant foot pathology.
An asymptomatic control group that lacked HR or other
foot pathology was recruited via the approved Clinical
Trials Registry at the authors institution. Standard bilateral
anteroposterior (AP) and lateral weightbearing radiographs
Figure 1. The flexibility jig.
were performed and the images were reviewed by a muscu-
loskeletal radiologist and the senior author to confirm the
lack of any significant pathology. In addition, the senior
author performed a clinical assessment prior to enrollment
to further ensure the absence of any foot pathology. The lat-
erality for testing was randomly selected.
Fifteen patients with HR (8 right feet and 7 left feet) met
the inclusion requirements and consented to the study pro-
cedure. Ten were female and 5 were male; the mean age was
53.4 (range, 44-66) years. Out of this group, 3 patients
elected to continue conservative treatment, with the remain-
der undergoing cheilectomy with or without Moberg oste-
otomy following the study testing. Twenty controls (6 left
feet and 14 right feet) were also enrolled. Nine were female
and 11 male. The control group was significantly younger
than the study group with an average age of 32.1 (range,
19-68) years (P < .001). One control patient did not com-
plete the third seated flexibility trial; otherwise, all patients
completed the full testing protocol. Extreme outlier data
points from individual trials were filtered out to avoid inac-
curate measurements (eg, negative values for early or late
flexibility). Out of 7560 measurements, 312 (4.1%) were
excluded prior to data analysis.
Motion and Flexibility Testing
A specially constructed flexibility jig (Figure 1) was used to
measure load-displacement properties of the first MTP joint
using a previously described technique.10,12 Prior to testing,
the flexibility jig was connected to TracerDAQ Pro software
(Measurement Computing Corporation, Norton, MA) to col-
lect rater-applied torque and range-of-motion measurements
at 100 samples per second. The patient was instructed to sit
on a chair, which was adjusted in height to ensure that the
patients knees were flexed to 90 degrees, with thighs paral-
lel to the floor. The foot selected for testing was secured in
the jig apparatus, with the contralateral foot resting adjacent
to it at the same level. Prior to data recording, each subjects
MTP joint was cyclically loaded 10 times in the flexibility
Cody et al 3

jig to overcome any initial stiffness. Two raters, a physical

therapist and a biomedical engineer, were designated as rater
1 and rater 2. The 2 raters alternated flexibility testing
between each trial and between each patient.
Each trial consisted of cyclically loading the first MTP
joint to maximum dorsiflexion and then restoring it to neutral
position 3 times. For a given rater, the first trial was repeated
after a short rest to satisfy a test-retest recording, so that
raters 1 and 2 completed 2 consecutive trials each. A remove-
replace recording was then performed in which the patient
fully removed his or her foot from the jig and then refixed it
in the jig, after which each rater performed a third flexibility
trial. The test-retest was intended to investigate the imme-
diate measurement consistency; the remove-replace was
intended to simulate a pre-/postoperative experimental para-
digm. After the seated trials, the patient was instructed to
stand on the jig in his or her normal stance with equal weight
on both feet. In standing position, 3 flexibility trials were per-
formed by each rater as previously described.
Data Analysis
The recorded voltage signals were processed with custom
MATLAB (MathWorks; Natick, MD) code. The raw volt-
age signals corresponding to range and load were converted
to angle (degrees) and torque (Ncm) signals then baseline
offset, rectified, and smoothed with a low-pass Butterworth
filter set at a 6.0-Hz cutoff frequency. Each cycle within a
trial was identified with a pattern recognition algorithm
based on the normalized angle signal. In each cycle, the first
5.0% and last 7.5% of the ascending portion of the signal
were truncated, leaving the remaining central 87.5% for
flexibility parameter extraction.
Dorsiflexion angle and applied torque were plotted
together on the vertical and horizontal axes, respectively, to
create a flexibility curve (Figure 2). All flexibility parame-
ters were calculated from the ascending portion of a cycle,
which corresponded to increasing dorsiflexion. Early and
late flexibility (degrees/Ncm) of the first MTP joint were
defined as the slope in the first 25% and last 25% of the
joints flexibility curve, respectively. The intersection of the
early and late flexibility lines yielded a point from which 2
additional parameters, the laxity angle (degrees) and
laxity torque (Ncm), were generated (Figure 2).12 Finally,
the mean laxity torque value among controls during sitting
(intended as a standard baseline value of laxity torque) was
used to generate a corresponding torque angle from each
patients flexibility curve. Maximum dorsiflexion (degrees)
was also calculated from the full range of motion signal.
Statistical Analysis
Patient age, early and late flexibility, laxity angle, laxity
torque, torque angle, and maximum dorsiflexion were
Figure 2. A diagram of the angle-torque flexibility curve is
shown as a thick black line. The intersection of the early and
late flexibility slope lines yields a point at which the x- and
y-coordinates are defined as the laxity torque and laxity angle,
respectively. The torque angle was defined for each patient as
the angle (y-coordinate) at which the average laxity torque value
of the control patients intersected his or her flexibility curve.
compared between the study group and the control group
using independent samples 2-tailed t tests with significance
set at alpha = .05. In a secondary analysis, age was used as
a co-varying variable to control for differences between the
groups. The internal consistency reliability of each flexibil-
ity parameter was calculated with intraclass correlation
coefficient (ICC) statistics. Intrarater reliabilities were
established per rater using an ICC(2,1) statistic11equiva-
lent to the Cronbach alpha statisticfrom trial 1 and trial 2
data (test-retest) and trial 2 and trial 3 data (remove-replace).
Interrater reliability for each parameter was calculated with
an ICC(2,2) statistic11 based on rater 1 and rater 2 measure-
ment data from the first flexibility trial. Reliability was
graded on the following scale: less than 0.40, poor; 0.40 to
0.59, fair; 0.60 to 0.74, good; 0.75 or greater, excellent.5
Each reliability coefficient was tested against the null
hypothesis that it was greater than 0 (single-tailed) at the
alpha = .05 significance level. All statistical analyses were
performed in SPSS version 22.0 (IBM Corp, Armonk, NY).
Results
Overall, HR patients were less flexible than control patients
(Figure 3). For both cases and controls, early flexibility
was greater than late flexibility (P < .001 for both). Early
flexibility, laxity angle, torque angle, and maximum dorsi-
flexion were significantly lower in study patients compared
to controls, both in sitting and standing positions (Table 1;
4 Foot & Ankle International 0(0)

dorsiflexion differed significantly between HR patients

and controls (Table 1). Of the standing measurements, only
laxity angle and torque angle remained significantly differ-
ent between groups after adjusting for age.
The position in which testing was conducted had a sig-
nificant effect on most measurements. Early flexibility and
torque angle were lower in the standing position compared
to the seated position (P < .01 for all comparisons; Figures
4 and 5). Late flexibility did not differ between sitting and
standing (P = .41 for cases and P = .12 for controls; Figure
4). Laxity angle was significantly lower when standing
compared to sitting for HR patients (P = .03) but not for
controls (P = .43; Figure 5). Laxity torque, correspondingly,
was significantly higher when standing compared to sitting
for both cases and controls (P < .001 for both; Figure 5).
Maximum dorsiflexion was significantly lower in the stand-
ing position for cases (P = .024) but not for controls (P =
.11; Figure 6).
Most measurements had excellent reliability (Table 2).
Intra- and interrater reliability was excellent for early flex-
ibility, laxity angle, torque angle, and maximum dorsiflex-
Figure 3. First MTP joint flexibility in the sitting position is
ion in both the seated and standing positions. Late flexibility,
shown for HR patients compared to controls. The solid lines
represent the mean (thick line) and 1.0 standard deviation laxity angle, and laxity torque measurements were gener-
(thin lines) of the control group flexibility. The dashed lines ally less reliable than the other measurements. Both seated
represent the HR group flexibility. The horizontal axis and standing measurements had excellent reliability.
represents torque normalized to each cycle maximum as a
percentage. HR, hallux rigidus; MTP, metatarsophalangeal.
Discussion
Table 1. P Values for Comparisons Between Hallux Rigidus In summary, we demonstrated that first MTP joint flexibil-
Patients and Controls. ity could be reliably assessed in patients with HR using a
custom flexibility jig. Testing was simple to perform and
P Value, Age-
was tolerated well by all patients. In contrast to simply mea-
P Value Adjusted Model
suring range of motion, measuring flexibility adds addi-
Seated tional depth to the assessment of HR. Decreased early
Early flexibility .027* .476 flexibilityor less movement per applied torque in the first
Late flexibility .960 .144 25% of motionis an indication that the pathology of HR
Laxity angle <.001* <.001* affects the whole joint, and not just maximal dorsiflexion.
Laxity torque .143 .115 After controlling for age, early flexibility was no longer sig-
Torque angle .002* .127 nificantly different between groups, but laxity anglemore
Maximum dorsiflexion <.001* <.001* reflective of overall joint flexibilityremained signifi-
Standing cantly lower in HR patients.
Early flexibility .017* .800
Only a few prior studies have quantified flexibility of the
Late flexibility .067 .650
first MTP joint. The first investigated MTP joint flexibility
Laxity angle <.001* <.001*
with and without custom orthoses in patients with early HR,
Laxity torque .372 .078
but no validation of the flexibility measurements was per-
Torque angle <.001* .001*
Maximum dorsiflexion .017* .747
formed.12 The second, more recent study investigated the
relationship between first MTP joint flexibility and plantar
*P < .05. loads in asymptomatic patients with different arch struc-
tures. In this study, intrarater reliability was excellent; how-
Figures 4-6). Late flexibility did not differ between cases ever, interrater reliability was not assessed.10
and controls (P = .07 standing, P = .96 sitting; Figure 4), Advantages of our study compared to these prior stud-
nor did laxity torque (P = .37 standing, P = .14 sitting; ies include our assessment of test-retest as well as remove-
Figure 5). After adjusting for age on the seated measure- replace reliability. Our findings of excellent test-retest
ment comparisons, only laxity angle and maximum reliability suggest there was little variability occurring
Cody et al 5

Figure 4. Early and late flexibility measurements are shown for the control group compared to the study group, in sitting and
standing positions. R1, rater 1; R2, rater 2.

Figure 5. The laxity angle (degrees) and laxity torque (Ncm) are shown for both groups, in sitting and standing positions. R1, rater 1;
R2, rater 2.

Figure 6. Torque angle and maximum dorsiflexion are shown for both groups, in sitting and standing positions. R1, rater 1; R2, rater 2.
6 Foot & Ankle International 0(0)

Table 2. Intra- and Interrater Reliability.a

Test-Retest Intrarater Remove-Replace

ICC Intrarater ICC Interrater ICC

Controls Cases Controls Cases Controls Cases

Early flexibility (/N cm) Sitting 0.973 0.934 0.893 0.960 0.979 0.965
Standing 0.981 0.968 0.958 0.805 0.968 0.958
Late flexibility (/N cm) Sitting 0.928 0.915 0.710 0.893 0.810 0.772
Standing 0.956 0.961 0.806 0.828 0.879 0.891
Laxity angle () Sitting 0.872 0.959 0.894 0.885 0.839 0.820
Standing 0.837 0.793 0.856 0.752 0.780 0.771
Laxity torque (N cm) Sitting 0.938 0.888 0.910 0.807 0.954 0.761
Standing 0.781 0.778 0.816 0.772 0.613 0.929
Torque angle () Sitting 0.965 0.965 0.908 0.945 0.970 0.972
Standing 0.979 0.947 0.960 0.754 0.975 0.916
Maximum dorsiflexion () Sitting 0.981 0.982 0.932 0.973 0.967 0.852
Standing 0.981 0.957 0.971 0.944 0.955 0.887

Abbreviation: ICC, intraclass correlation coefficient.

a
Intrarater ICCs were averaged between raters, and all ICCs were significant (P < .05).

between immediately successive measurements. Our
findings of excellent remove-replace reliability suggest
that patients can be tested on separate occasions (for
instance, preoperatively and postoperatively) with high
confidence. Moreover, we used 2 raters from different
professional backgroundsa physical therapist (S.R.)
and a biomedical engineer (H.H.)to perform the test-
ing. The overall excellent interrater reliability between
these 2 raters suggests that individuals with disparate and
even nonclinical professional backgrounds can perform
testing on the flexibility jig and still record comparable
results.
One related study described a method of measuring stiff-
ness of the first MTP joint. A tactile pressure sensing system
was used to measure applied force, with angular displace-
ment collected by video analysis.7 No reliability testing was
performed. By measuring stiffness, these authors used a
curve that is the inverse of the flexibility curve used in our
study, making angular displacement the independent vari-
able and torque the dependent variable. However, we
believe that flexibility is a more useful measure than stiff-
ness. First, the ubiquitous tool at a clinicians disposal is a
goniometer rather than a torque sensor or dynamometer.
Therefore, it is more clinically relevant to think of assessing
a joints angular laxity (output) as the dependent outcome of
the clinicians self-judged applied load (input), which is
modeled by the flexibility curve. Second, as applied torque
(x-axis of the flexibility curve) increases, angular displace-
ment (y-axis) plateaus, and as such the curves slope
approaches zero. Conversely, a stiffness curve has a slope
that becomes very large quickly as angular displacement
(x-axis) reaches its maximum. This renders measurement of
a slope (stiffness) at the end range of motion less tangible
than an equivalent flexibility measurement.
The reason we chose to measure flexibility in the first
25% and the last 25% of the range-of-motion arc was to
reflect the different biomechanical phenomena occurring in
each of these ranges. We chose the first and last 25%, rather
than 10% for instance, because the flexibility data tended to
be more erratic at the extremes of motion. The biomechani-
cal load-response behavior of most joints is commonly
characterized by an early linear-elastic region in the early to
middle range of motion and later by a nonlinear-viscoelastic
region in the end range of motion.1 From a clinical perspec-
tive, we thought it would be useful and interesting to under-
stand what the slope of the flexibility curve is at both
regions. As expected, early flexibility significantly exceeded
late flexibility in both cases and controls. The fact that early
flexibility was significantly lower in patients with HR com-
pared to controls suggests that in HR the biomechanics of
the joint are altered even in the midrange of motion.
Flexibility measurements are interesting in part because
they are affected by dynamic soft tissue tension, not just
bony anatomy and ligamentous structure. Flavin etal have
proposed that HR is caused by increased tension in the plan-
tar fascia and other soft tissue structures, which results in
abnormal stress on the articular cartilage.4 Our finding of
decreased flexibility early in the arc of motion of patients
with HR supports that theory.
Other authors have pointed to an entity labeled func-
tional hallux rigidus in which increased soft tissue tension
specifically in the weightbearing position is the major
source of pain.8 Such patients might have abnormal early
flexibility measurements in the standing position, but
Cody et al 7
normal measurements in the nonweightbearing, seated
position. In our study, we saw a general decrease in flexibil-
ity in the standing position. This may reflect increased soft
tissue tension in both HR patients and controls as a result of
muscle activation and/or a tightened plantar fascia.
However, HR patients seemed to be more affected by posi-
tioning than control patients: maximum dorsiflexion and
laxity angle were both significantly lower in the standing
position for HR patients, but not for controls. This finding
again suggests that abnormal soft tissue tension may be
associated with HR. We advocate the use of early flexibility
measurements in the nonweightbearing, seated position to
identify pathology without the confounding factor of mus-
cle activation; however, comparing them to standing mea-
surements may help identify the relative effect of the
weightbearing stance and potential functional impact.
It is reasonable to assume that the hallux MTP joint, like
most other joints in the body, loses motion with age, irre-
spective of the presence of hallux rigidus. This loss of
motion is likely accompanied by a loss of flexibility, which
may explain why some of the differences between HR
patients and controls were no longer significant after con-
trolling for age. However, the laxity angle parameteran
indication of overall joint flexibilityremained strongly
significant after controlling for age. Therefore, despite the
large difference in age between the 2 groups in this study,
the pathology of hallux rigidus still was a major contributor
to decreased joint flexibility.
The flexibility jig we used also enabled reliable mea-
surement of maximum dorsiflexion. Other authors have
attempted to find reliable methods of measuring first MTP
joint range of motion.9,13,14 Vulcano etal14 compared clini-
cal and radiographic measurements of maximum dorsiflex-
ion of the hallux MTP in patients with HR. They found
excellent intra- and interrater reliability with both methods
of dorsiflexion measurement. However, there was a signifi-
cant difference between measurements from the 2 methods,
with clinical measurements yielding lower dorsiflexion
angles than radiographic measurements.14 In our study,
maximal dorsiflexion measurements were calculated con-
sidering the floor as neutral, and were within the range of
the radiographic measurements of Vulcano etal, which
were made considering the long axis of the first metatarsal
as neutral. We believe that dorsiflexion measurements rela-
tive to the floor may be more pertinent because they neu-
tralize the effect of an elevated or plantarflexed first ray.
Regardless, these findings show that different methods of
measuring dorsiflexion cannot be used interchangeably.
Our study is not without limitations. First, our sample
size was relatively small in both cohorts; however, our
numbers are similar to those used in prior studies.10,12
Second, there was a significant difference in age between
the study group and the control group. While this difference
limits our ability to compare the 2 groups, we did control
for age in secondary analyses. Third, we did not compare
our range of motion measurements against clinical or radio-
graphic methods of measuring range of motion. Finally, we
did not standardize the rate of torque application or the
magnitude of applied torque during the flexibility trials.
This could have inflated the variability of some parameters
derived from the curve, but despite this, all flexibility
parameters were very consistent within and between raters.
Further research will be required to determine the clini-
cal utility of these measurements. Physical therapists might
be able to use flexibility measurements to guide their
approach to each patient. If flexibility can be improved
through manual therapy, it is possible that pain might
improve as well. It is also possible that flexibility measure-
ments may be used to predict which patients will benefit
from cheilectomy, to help surgeons better indicate patients
for this surgery. The authors use current radiographic and
clinical staging systems along with a lack of pain in mid-
range of motion as justification to perform cheilectomy
instead of arthrodesis, but outcomes are not always predict-
able. It is possible that patients with greater flexibility pre-
operatively are more likely to have good outcomes from
cheilectomy. Conversely, patients with lower flexibility
may be more likely to fail cheilectomy and ultimately
require arthrodesis. Future studies will also have to com-
pare flexibility measurements before and after surgery and
correlate improvement in these parameters with standard-
ized clinical outcome scores. It is our hope that continuing
this work will help us better understand the pathology and
appropriate treatment of HR.
In conclusion, to our knowledge, this is the first study to
demonstrate a reliable method of measuring first MTP joint
flexibility in patients with HR. We found that flexibility,
even early in the arc of motion, is impaired in patients with
HR. Moreover, significant differences between sitting and
standing measurements suggest that soft tissue tension may
be a major contributor to this finding. We do not know yet
how flexibility of the joint relates to symptomatology, or if
the surgeries performed for HR affect flexibility.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research,
authorship, and/or publication of this article.
References
1. Burstein AH, Wright TM. Fundamentals of Orthopaedic
Biomechanics. Baltimore, MD: Williams & Wilkins; 1994.
2. Coughlin MJ, Shurnas PS. Hallux rigidus. Grading and long-
term results of operative treatment. J Bone Joint Surg Am.
2003;85(11):2072-2088.
8 Foot & Ankle International 0(0)
3. Coughlin MJ. Hallux valgus. J Bone Joint Surg Am.
1996;78(6):932-966.
4. Flavin R, Halpin T, OSullivan R, FitzPatrick D, Ivankovic A,
Stephens MM. A finite-element analysis study of the metatar-
sophalangeal joint of the hallux rigidus. J Bone Joint Surg Br.
2008;90(10):1334-1340.
5. Fleiss JL, Levin B, Paik MC. Statistical Methods for Rates
and Proportions. Hoboken, NJ: John Wiley & Sons; 2003.
6. Gajdosik RL, Bohannon RW. Clinical measurement of range
of motion. Review of goniometry emphasizing reliability and
validity. Phys Ther. 1987;67(12):1867-1872.
7. Heng ML, Kong PW. A novel technique of quantifying first
metatarsophalangeal (1st MPJ) joint stiffness. J Foot Ankle
Res. 2014;7(suppl 1):A32.
8. Maceira E, Monteagudo M. Functional hallux rigidus and
the Achilles-calcaneus-plantar system. Foot Ankle Clin.
2014;19(4):669-699.
9. Otter SJ, Agalliu B, Baer N, etal. The reliability of a smart-
phone goniometer application compared with a traditional
goniometer for measuring first metatarsophalangeal joint dor-
siflexion. J. Foot Ankle Res. 2015;8(1):30.
10. Rao S, Song J, Kraszewski A, etal. The effect of foot struc-
ture on 1st metatarsophalangeal joint flexibility and hallucal
loading. Gait Posture. 2011;34(1):131-137.
11. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing
rater reliability. Psychol Bull. 1979;86(2):420-428.
12. Song J, Whitney K, Heilman B, Kim E, Hillstrom HJ. First
metatarsal phalangeal joint flexibility: a quantitative tool
for evaluation of hallux limitus. Clin Biomech. 2008;23(5):
704-705.
13. Swanson JE, Stoltman MG, Oyen CR, etal. Comparison of
2D-3D measurements of hallux and first ray sagittal motion
in patients with and without hallux valgus. Foot Ankle Int.
2016;37(2):227-232.
14. Vulcano E, Tracey JA 3rd, Myerson MS. Accurate mea-
surement of first metatarsophalangeal range of motion in
patients with hallux rigidus. Foot Ankle Int. 2016;37(5):
537-541.