Accepted Manuscript Title: Predictors of chronic ankle instability after an index lateral ankle sprain: A systematic review Author: Fereshteh Pourkazemi Claire E. Hiller Jacqueline Raymond Elizabeth J. Nightingale Kathryn M. Refshauge PII: DOI: Reference:
S1440-2440(14)00024-3 http://dx.doi.org/doi:10.1016/j.jsams.2014.01.005 JSAMS 989
To appear in:
Journal of Science and Medicine in Sport
Received date: Accepted date:
8-12-2013 25-1-2014
Please cite this article as: Pourkazemi F, Hiller CE, Raymond J, Nightingale EJ, Refshauge KM, Predictors of chronic ankle instability after an index lateral ankle sprain: A systematic review, Journal of Science and Medicine in Sport (2014), http://dx.doi.org/10.1016/j.jsams.2014.01.005 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Predictors of chronic ankle instability after an index lateral ankle sprain: A systematic
2
review
3 Fereshteh Pourkazemi,a Claire E. Hiller,a Jacqueline Raymond,b Elizabeth J Nightingale,a Kathryn M.
5
Refshauge a
ip t
4
6 a
Discipline of Physiotherapy, Faculty of Health Sciences, University of Sydney, Australia
8
b
Discipline of Exercise and Sport Science, Faculty of Health Sciences, University of Sydney, Australia
cr
7
Corresponding author: Fereshteh Pourkazemi
11
e-mail: [email protected]
an
10
us
9
Word count: 3037
14
The Abstract word count: 234
15
Number of tables: 2
16
Number of figures: 1
17
Supplemental files: 2
te
Ac ce p
18
d
13
M
12
1
Page 1 of 21
18
Predictors of chronic ankle instability after an index lateral ankle sprain: A systematic
19
review
20
Abstract
22
Objective: To identify the predictors of CAI after an index lateral ankle sprain.
23
Design: Systematic review
24
Methods: The data bases of MEDLINE, CINAHL, AMED, Scopus, SPORTDiscus, Embase, Web of
25
Science, PubMed, PEDro, and Cochrane Register of Clinical Trials were searched from the earliest
26
record until May 2013. Prospective studies investigating any potential intrinsic predictors of chronic
27
ankle instability (CAI) after an index ankle sprain were included. Eligible studies had a prospective
28
design (follow-up of at least three months), participants of any age with an index ankle sprain, and
29
had assessed ongoing impairments associated with CAI. Eligible studies were screened and data
30
extracted by two independent reviewers.
31
Results: Four studies were included. Three potential predictors of CAI, i.e., postural control,
32
perceived instability, and severity of the index sprain, were investigated. Decreased postural control
33
measured by number of foot lifts during single-leg stance with eyes closed and perceived instability
34
measured by Cumberland Ankle Instability Tool (CAIT) were not predictors of CAI. While the results
35
of one study showed that the severity of the initial sprain was a predictor of re-sprain, another study
36
did not.
37
Conclusion: Of the three investigated potential predictors of CAI after an index ankle sprain, only
38
severity of initial sprain (grade II) predicted re-sprain. However, concerns about validity of the
39
grading system suggest that these findings should be interpreted with caution.
40
Key words: ankle sprains, joint instability, sprain severity, postural balance
Ac ce p
te
d
M
an
us
cr
ip t
21
2
Page 2 of 21
41 42
1. Introduction
43 Ankle sprain injuries, particularly sprain of the lateral ligaments, are among the most common lower
45
limb injuries in the general1,2 and athletic populations3 as well as among military personnel.4 High
46
recurrence rate, along with residual impairments, are the consequence of ankle sprain for up to 54% of
47
individuals.5 The most common residual impairments include re-sprain, perceived instability and
48
episodes of giving way (also referred to as functional instability), joint laxity (also referred to as
49
mechanical instability), pain, swelling, a feeling of weakness and subsequently reduced level of
50
physical activity.5 These residual impairments, alone or in combination, are frequently termed chronic
51
ankle instability (CAI).6 Prevention of CAI, particularly re-sprain, is the main treatment goal for many
52
studies of ankle sprain,7 however, prevention is only possible if people at risk of developing CAI can
53
be identified. That is, to determine the most effective prevention strategy it is essential to understand
54
the underlying causes leading to CAI, and the factors increasing the risk of CAI.
55
History of a previous sprain is the most frequently reported risk factor for lateral sprain.8-10 Basketball
56
players with history of an ankle sprain were found to be five times more likely to re-sprain, although
57
the reasons for the increased risk are unknown.9 Hertel11 suggested that ankle sprains cause various
58
sensorimotor deficits that can lead to instability, and that the presence of instability increases the risk
59
of further sprain. Many studies have investigated this theory, but the findings are inconsistent.12-14
60
There are few studies of predictors of CAI, however two systematic reviews have evaluated closely
61
related questions and some information can be derived from their findings. One systematic review
62
found decreased dorsiflexion range of motion to be a strong predictor of lateral ankle sprain.15 The
63
second systematic review investigated the clinical course of acute ankle sprain,5 finding only one
64
study that reported on prognostic factors for development of CAI16. Based on this study,16 athletes
65
competing at a higher level of competition were at greater risk of persisting impairments after an
66
acute sprain than athletes competing at lower levels. Severity of pain, number of re-sprains, as well as
67
level of perceived instability and self-reported recovery after an acute lateral ankle sprain were found
68
to be independent of the severity of the initial sprain. Furthermore, men were reported to have a
Ac ce p
te
d
M
an
us
cr
ip t
44
3
Page 3 of 21
greater risk of developing residual impairments than women. However, not all studies included in
70
these two systematic reviews investigated the risk factors contributing to CAI after an index ankle
71
sprain.
72
Research to date has tended to focus more on changes associated with CAI rather than predictors of
73
CAI. Sensorimotor, functional, anatomical or biomechanical changes associated with CAI have been
74
analysed in a number of systematic reviews.14,17,18 Postural instability,17,18 prolonged time to balance
75
after a jump,14 lower limb muscle weakness or muscle imbalance19 were found to be associated with
76
CAI. Whether these impairments cause CAI or develop as a result of CAI is not clear. Therefore, the
77
objective of the present systematic review was to identify predictors of CAI (e.g., age, sex, body mass
78
index (BMI), level of physical activity, balance, postural control, proprioception, motor control,
79
severity of ankle sprain, perceived ankle instability, feeling of giving way, ligament laxity, pain, or
80
swelling) after an index ankle sprain. Prospective studies investigating any of these variables as
81
potential predictors of CAI after an index ankle sprain might enable health providers to design more
82
effective treatments to prevent ongoing problems.
84
2. Methods
Ac ce p
85
te
83
d
M
an
us
cr
ip t
69
86
The study protocol was developed based on the framework outlined in the guidelines provided by the
87
PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement.20 The
88
protocol of this systematic review is registered on PROSPERO (registration number
89
CRD42012002990).
90
2.1. Eligibility criteria
91
Studies were included if they met the following inclusion criteria: i) longitudinal design, ii) follow-up
92
length of at least three months since the sprain, iii) participants of any age who had sustained an index
93
ankle sprain only, iv) measuring at least one of the potential predictors of CAI, v) reporting on any re-
94
sprain or residual symptoms after the initial ankle sprain during the follow-up period, vi) published
95
full paper; abstracts were included if the authors provided the raw data for further analysis. Papers
4
Page 4 of 21
96
where it was possible that at least some of the participants had suffered an index sprain were
97
considered for inclusion, however if the data from this sub-group could not be isolated, then the paper
98
was excluded.
99
For randomised controlled trials, we included the minimal intervention group (e.g. conservative treatment of lateral ankle sprains, such as modified footwear and associated supports, taping, adapted
101
training programmes, and education). Randomised clinical trials were excluded if there was no
102
minimal intervention group e.g., if surgery or immobilisation for more than three days were
103
investigated.
104
2.2. Information Sources
105
Studies were identified through a search of the MEDLINE, CINAHL, AMED, Scopus,
106
SPORTDiscus, Embase, Web of Science, PubMed, PEDro, and Cochrane Register of Clinical Trials
107
to May 2013. Abstracts from the International Ankle Symposium (IAS) and International Foot and
108
Ankle Biomechanics (i-FAB) conference proceedings, and bibliographies of eligible studies were
109
hand searched. In addition, relevant experts were contacted to identify any unpublished studies that
110
may exist and to review the list of identified studies for completeness. No language restriction was
111
imposed, however, if a translation could not be arranged, the paper was excluded.
112
2.3. Search Strategy
113
Key terms used in our search strategy are presented in supplemental file 1. Search protocols were
114
specifically designed to target prospective studies.
115
2.4. Study Selection
116
All studies identified by the search strategy were screened against the eligibility criteria independently
117
by the first author (FP) and by one other author (CH, JR or KR). Titles and abstracts were inspected
118
and clearly ineligible studies were removed. Full copies of potentially eligible papers were retrieved.
119
Any inconsistencies regarding inclusion of trials were resolved by consensus.
120
2.5. Potential Predictors of CAI
121
The following predictors, identified from previous studies of risk factors for lateral ankle sprain, were
122
considered as potential risk factors of developing CAI: age, sex, height, weight, BMI, leg dominance,
123
foot type, foot and lower limb alignment, ankle joint laxity, general ligamentous laxity, postural
Ac ce p
te
d
M
an
us
cr
ip t
100
5
Page 5 of 21
control, muscle weakness or muscle strength/power imbalance in lower limbs, lower limb muscle
125
reaction time, range of motion at lower limb joints, physical fitness, gait biomechanics, and severity
126
of symptoms after the initial sprain (e.g., pain, swelling, laxity, feeling of instability).
127
2.6. Outcome measures
128
Any measures that assessed ongoing impairments associated with CAI after the initial sprain during
129
the follow up period (e.g., re-sprain, swelling, pain, mechanical instability, perceived instability,
130
feeling of giving way, or feeling of weakness) were considered as outcome measures.
131
2.7. Data Extraction Process
132
Data were extracted independently by two authors (FP and CH) and confirmed by one other author
133
(either JR or KR). Any discrepancies were settled by further discussion and consensus. Study
134
characteristics extracted were study type, target population (setting, sex, age), sample size, inclusion
135
criteria, follow-up duration, prognostic factors measured, interventions (if any) and all reported
136
outcome measures.
137
2.8. Quality Assessment
138
Risk of bias and methodological quality of included studies were assessed using the quality
139
assessment tool developed by Pengel21 et al. This tool consists of 7 items rated as either ‘yes’, ‘no’ or
140
‘N/A’ (not applicable). Four items relate to control of bias (items 1-4), two to appropriate
141
measurement of variables (items 5 and 6) and one to control of confounding variables (item 7). Two
142
raters (FP and JR) independently assessed the quality and a third author (CH) resolved disagreements.
143
2.9. Synthesis of Results
144
Study outcomes were statistically pooled if the studies were considered to be sufficiently
145
homogeneous. For homogeneous studies, raw data were used in a direct logistic regression to assess
146
whether the predictor variables increased the likelihood of re-sprain. If the studies were considered
147
too heterogeneous, data were not pooled and the outcomes were described.
Ac ce p
te
d
M
an
us
cr
ip t
124
148 149
6
Page 6 of 21
149 150
3. Results
151 The search strategy identified 8085 titles. Following title and abstract screening, 210 potentially
153
relevant articles were identified, of which 16 met all the inclusion criteria. The data required for
154
analysis were reported in two of the included studies.22,23 The authors of the remaining 14 articles,
155
which included participants with one or more ankle sprains, were contacted for data related to the
156
participants with only an index ankle sprain. Seven authors replied, of whom only two were able to
157
provide raw data for analysis.24,25 Therefore, four studies22-25 were included in this review (Figure 1).
158
3.1. Study Characteristics
159
All included studies were prospective cohort trials, with follow-up periods that varied between 8 and
160
24 months. Participants were adults in two studies,23,24 adolescents in one study25 and children and
161
adolescents in the fourth study.22 In two studies the inclusion criterion was an acute index ankle sprain
162
of less than two days duration.22,23 The other two studies24,25 included a heterogeneous group of
163
participants, with a history that varied between multiple ankle sprains to no history of sprain, and the
164
time since sprain was not standardised. Separate data of the participants with an index ankle sprain
165
were provided by the authors of these two studies.24,25 Participants were recruited from various
166
settings, including a university,24 dance school,25 and primary care practice(s).22,23 The predictors of
167
CAI measured were perceived instability and balance24,25 in two of the studies and severity of index
168
sprain22,23 in the other two. The severity of injury was measured by either clinical symptoms and
169
anterior drawer stress x-ray23 or clinical symptoms and magnetic resonance imaging (MRI).22 The
170
outcomes reported were re-sprain in three studies23-25 and pain, mobility and perceived instability in
171
the remaining study.22 Details of the included studies are presented in Table 1.
172
3.2. Quality Assessment
173
No study was rated as having excellent methodological quality. Methodological quality was moderate
174
in three studies, with scores of 5 25 and 4.23,24 The fourth study22 had low methodological quality, with
175
a score of 3 (supplemental file 2).None of these studies used statistical adjustment for potentially
176
confounding factors. Only one study used blinded outcome assessor(s).22
Ac ce p
te
d
M
an
us
cr
ip t
152
7
Page 7 of 21
Ac ce p
te
d
M
an
us
cr
ip t
177
8
Page 8 of 21
177 3.4. Predictive Factors of CAI
179
Data for two of the studies24,25 were pooled, because they investigated perceived instability and
180
balance as predictors for re-sprain. We were not able to pool data for the two other studies22,23
181
because, although they both investigated severity of initial sprain, they used different grading systems
182
and reported on different outcome measures.
183
Perceived instability and balance: Two studies24,25 investigated perceived instability and static balance
184
on both legs as a predictor of re-sprain. Perceived instability was measured using the Cumberland
185
Ankle Instability Tool (CAIT) and balance was measured recording the number of foot movements
186
during single-leg stance with eyes closed for 30 seconds. Hiller25 et al and de Noronha24 et al provided
187
raw data for 44 participants with history of an index ankle sprain. After the follow-up periods of 12-
188
13 months, 11 participants had re-sprained their ankles. Raw data from these two studies were pooled
189
and logistic regression was performed, finding that the predictive capacity of the model was not
190
statistically significant (Table 2). That is, CAIT score and number of foot lifts during single-leg stance
191
with eyes close did not predict re-sprain.
192
Severity of the index ankle sprain: Two studies investigated severity of the index sprain as a predictor
193
of CAI. Malliaropoulos23 et al used an expanded grading system26 (I, II, IIIA and IIIB) to determine
194
severity of sprain. They found significant differences in rate of re-sprain between athletes with
195
different grades of injury during a 24 month follow-up period. Elite track and field athletes with less
196
severe ankle sprain (grade I or II) had a higher rate of re-sprain than athletes with more severe ankle
197
sprain (grade IIIA or IIIB). To further analyse the results of this study, we performed direct logistic
198
regression to assess the impact of the severity of index sprain on the likelihood of re-sprain. The
199
predictive capacity of the model was statistically significant, X²(3, n=202) = 14.71, P=0.002. The
200
severity of sprain (grades I to IIIB) explained between 7% (Cox and Snell R2) and 11.4% (Nagelkerke
201
R2) of the variance in re-sprain status, and correctly classified 81.7% of cases. However, only grade I
202
and II made a unique statistically significant contribution to the model (Table 2). The strongest
203
predictor of re-sprain was grade II injury, recording an odds ratio (OR) of 2.6. This indicates that
204
athletes with grade II injury were 2.6 times more likely to re-sprain their ankles.
Ac ce p
te
d
M
an
us
cr
ip t
178
9
Page 9 of 21
In the second study, Endele22 et al divided 30 children with an acute ankle sprain into three groups
206
based on the severity of their clinical symptoms (I, II or III). Only children with grade II and III were
207
followed up for eight months. Endele22 et al reported no significant differences between children with
208
grades II or III of ankle sprain, with regard to limitation of mobility, pain or instability, three months
209
after the index injury. These outcome measures were not reported at eight months follow up. Due to
210
the small number of participants in this study (n=15) no further analysis was performed.
ip t
205
4. Discussion
us
212
cr
211
213
The present systematic review found that only three potential predictors of CAI (severity of sprain,
215
balance and perceived instability) had been investigated in longitudinal studies after an index ankle
216
sprain. Although recent evidence from cross-sectional studies shows that postural sway, time to
217
balance after a jump, foot position during gait, strength, and proprioception are altered in participants
218
with CAI,14,15,19 none of these factors has been investigated in a longitudinal study after the index
219
sprain.
220
Two studies investigated the severity of the index ankle sprain as a potential predictor of CAI, with
221
conflicting results. Although there are factors limiting comparisons between these studies, including
222
the use of different grading systems, different age groups (adults23 vs. children22), and different
223
duration of follow-up (24 months23 vs. 8 months22), a critique of the two studies highlights some
224
critical issues.
225
Malliaropoulos23 et al, classified the severity of index sprains into four grades based on the patients’
226
clinical symptoms (dorsiflexion range of motion and swelling). Participants were referred for stress
227
radiography only if they were diagnosed with a grade III sprain. In the second study, Endele22 et al
228
also classified the sprains as grades I, II or III based on the clinical symptoms (weight bearing status
229
and swelling), however, they performed additional MRI investigations. Interestingly, they found that
230
clinical symptoms were not well correlated with the grade of injury. The amount of swelling and
231
ability to weight bear were highly correlated with bone bruising, based on their MRI findings.
Ac ce p
te
d
M
an
214
10
Page 10 of 21
Furthermore, six (50%) children with sprains classified as grade II based on the clinical symptoms,
233
had complete ligament rupture and three had a Salter I injury. These findings raise concerns about the
234
validity of the grading systems based on symptoms, alone.
235
The second study22 provides valuable information regarding the severity of ankle sprain in 30
236
children. In this study however, only children with sprains classified as grade II or III were followed
237
up, and the follow-up was at eight months, potentially an inadequate period when investigating re-
238
sprain and CAI. In this study, anatomical changes were reported in the MRIs of 10 children at eight
239
months, but Endele 22 et al did not re-assess pain, mobility or instability at eight months follow up.
240
Since the main aim of this study was to compare symptoms with MRI findings, the outcomes of pain,
241
instability or mobility were not reported.
242
Based on the results of these two studies, it is not possible to conclude definitively that ankle sprains
243
classified as grades I or II predict development of CAI. To be able to determine whether the severity
244
of an initial sprain predicts CAI, a valid grading system, a large sample size and adequate follow-up
245
are required.
246
Pooled data from the two other studies24,25 showed that CAIT scores do not predict re-sprain. CAIT is
247
a measure of perceived instability27 and previous studies demonstrated that perceived ankle instability
248
is related to ankle sprain.28 However, based on the new model of sub-groups of CAI developed by
249
Hiller29 et al, there is a sub-group of participants with CAI who have perceived instability without
250
recurrence of sprain. The findings of Hiller25 et al and de Noronha24 et al support the proposed model,
251
and explain the lack of predictive value of the CAIT score. Decreased CAIT score, or presence of
252
perceived instability, may increase risk of re-sprain in some subgroups, but not others.
253
The results of this systematic review also show that postural control, measured as the number of foot
254
lifts during single-leg stance with eyes closed was not a predictor of CAI. Many studies have
255
investigated postural control as a predictor of acute ankle sprain.25,30,31 These studies however, used
256
different measurement methods and found different results. While most of these studies found
257
decreased postural control as a predictor of acute ankle sprains,25,32 several found no relationship
258
between postural control scores and the risk of ankle sprain.30,31,33 In a recent systematic review,
259
Witchalls34 et al reported that the studies in which postural control was tested by scoring the number
Ac ce p
te
d
M
an
us
cr
ip t
232
11
Page 11 of 21
of errors during a test, did not show that poorer postural control increased the risk of ankle sprain.
261
They suggested that the increased subjectivity of these methods may increase the variability in scores
262
and render them less meaningful for measuring postural control.
263
While it seems the number of foot lifts during single-leg stance might not predict re-sprain, these
264
results should be interpreted with caution. Firstly, the different populations in these two studies
265
(adolescent dancers25 vs. active university students24) may have been a confounding factor. In
266
addition, the time from the index ankle sprain was not controlled for, potentially constituting another
267
confounding factor. Finally, the aim of these studies was not to predict CAI after an index sprain, and
268
therefore data were a subset. This resulted in a small number of participants in these two studies with
269
consequent lack of power for our re-analysis.
270
The strength of this systematic review was that no language limitations were imposed. Where
271
required, a translation in different languages was made (e.g. German, Italian, Japanese, and French).
272
We also included the grey literature. However, several studies could not be included because
273
reporting was insufficient and the data were not available for further analysis.
an
M
Ac ce p
276
d
5. Conclusion
te
274 275
us
cr
ip t
260
277
We found that of the three potential predictors of CAI that have been investigated prospectively after
278
an index ankle sprain, only severity of initial sprain (grade II) predicted re-sprain. However, concerns
279
about validity of the grading system suggest that these findings should be interpreted with caution.
280
Consequently, there is a large gap in the literature. Anatomical, central, physiological and
281
psychosocial changes have yet to be investigated after an initial sprain and the relationship of such
282
variables to re-injury and other residual symptoms remains unknown.
283
12
Page 12 of 21
283 284
Practical Implications
285
Severity of the initial ankle sprain does not necessarily predict the likelihood of
ip t
•
286
developing chronic ankle instability. •
288
Perceived ankle instability and static balance do not appear to predict future ankle
cr
287
sprains. •
290
Currently there is limited research investigating predictors of chronic ankle instability after
us
289
an index ankle sprain.
291
293
an
292 Acknowledgment
M
294
The authors would like to thank Dr Markus Hübscher, Ms Naomi Kusano, and Ms Alexandra Di Lallo
296
for their assistance in translation of non-English papers.
Ac ce p
te
d
295
13
Page 13 of 21
297 298
References
299
States. J Bone Joint Surg Am. 2010;92(13):2279-2284.
301
302
Waterman BR, Owens BD, Davey S, et al. The epidemiology of ankle sprains in the United
ip t
1.
2.
cr
300
Bridgman SA, Clement D, Downing A, et al. Population based epidemiology of ankle sprains attending accident and emergency units in the West Midlands of England, and a survey of UK
304
practice for severe ankle sprains. Emerg Med J. 2003;20(6):508-510.
5.
6.
7.
Eiff MP, Smith AT, Smith GE. Early Mobilization Versus Immobilization in the Treatment of Lateral Ankle Sprains. Am J Sports Med. 1994;22(1):83-88.
314
315
Delahunt E, Coughlan GF, Caulfield B, et al. Inclusion criteria when investigating
insufficiencies in chronic ankle instability. Med Sci Sports Exerc. 2010;42(11):2106-2121.
312
313
van Rijn RM, van Os AG, Bernsen RM, et al. What is the clinical course of acute ankle
sprains? A systematic literature review. Am J Med. 2008;121(4):324-331 e326.
310
311
M
United States Military Academy. Am J Sports Med. 2010;38(4):797-803.
308
309
Waterman BR, Belmont PJ, Jr., Cameron KL, et al. Epidemiology of ankle sprain at the
d
4.
te
307
Fong DT, Hong Y, Chan LK, et al. A systematic review on ankle injury and ankle sprain in sports. Sports Med. 2007;37(1):73-94.
306
an
3.
Ac ce p
305
us
303
8.
Milgrom C, Shlamkovitch N, Finestone A, et al. Risk factors for lateral ankle sprain: a prospective study among military recruits. Foot Ankle. 1991;12(1):26-30.
316
14
Page 14 of 21
9.
factors. Br J Sports Med. 2001;35(2):103-108.
318
10.
J Rehabil Res Dev. 2006;43(7):819-824.
13.
14.
15.
de Noronha M, Refshauge KM, Herbert RD, et al. Do voluntary strength, proprioception,
range of motion, or postural sway predict occurrence of lateral ankle sprain? Br J Sports Med.
330
2006;40(10):824-828; discussion 828.
331
332
Hiller CE, Nightingale EJ, Lin C-WC, et al. Characteristics of people with recurrent ankle sprains: a systematic review with meta-analysis. Br J Sports Med. 2011;45(8):660-672.
328
329
d
postural control in athletes with a history of ankle sprain. J Athl Train. 2013;48(2):203-208.
326
327
Steib S, Zech A, Hentschke C, et al. Fatigue-induced alterations of static and dynamic
M
324
325
Akbari M, Karimi H, Farahini H, et al. Balance problems after unilateral lateral ankle sprains.
an
12.
us
Instability. J Athl Train. 2002;37(4):364-375.
322
323
Hertel J. Functional Anatomy, Pathomechanics, and Pathophysiology of Lateral Ankle
cr
11.
ip t
Sports Exerc. 1983;15(3):267-270.
320
321
Ekstrand J, Gillquist J. Soccer injuries and their mechanisms: a prospective study. Med Sci
te
319
McKay GD, Goldie PA, Payne WR, et al. Ankle injuries in basketball: injury rate and risk
Ac ce p
317
16.
Linde F, Hvass I, Jurgensen U, et al. Early mobilizing treatment in lateral ankle sprains.
333
Course and risk factors for chronic painful or function-limiting ankle. Scand J Rehabil Med.
334
1986;18(1):17-21.
335
17.
Arnold BL, De La Motte S, Linens S, et al. Ankle instability is associated with balance impairments: A meta-analysis. Med Sci Sports Exerc. 2009;41(5):1048-1062.
336
15
Page 15 of 21
18.
instability: A systematic review with meta-analysis. J Sci Med Sport. 2010;13(1):2-12.
338
19.
prognosis. BMJ. 2003;327(7410):323.
22.
23.
24.
25.
Hiller CE, Refshauge KM, Herbert RD, et al. Intrinsic predictors of lateral ankle sprain in adolescent dancers: a prospective cohort study. Clin J Sport Med. 2008;18(1):44-48.
352
353
de Noronha M, Franca LC, Haupenthal A, et al. Intrinsic predictive factors for ankle sprain in
active university students: A prospective study. Scand J Med Sci Sports. 2012.
350
351
Malliaropoulos N, Ntessalen M, Papacostas E, et al. Reinjury after acute lateral ankle sprains in elite track and field athletes. Am J Sports Med. 2009;37(9):1755-1761.
348
349
d
children. Foot Ankle Int. 2012;33(12):1063-1068.
346
347
Endele D, Jung C, Bauer G, et al. Value of MRI in diagnosing injuries after ankle sprains in
M
344
345
Pengel LH, Herbert RD, Maher CG, et al. Acute low back pain: systematic review of its
an
21.
us
meta-analyses: the PRISMA statement. J Clin Epidemiol. Oct 2009;62(10):1006-1012.
342
343
Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and
cr
20.
ip t
Functional Ankle Instability: A Meta-Analysis. J Athl Train. 2009;44(6):653-662.
340
341
Arnold BL, Linens SW, de Ia Motte SJ, et al. Concentric Evertor Strength Differences and
te
339
Munn J, Sullivan SJ, Schneiders AG. Evidence of sensorimotor deficits in functional ankle
Ac ce p
337
26.
Malliaropoulos N, Papacostas E, Papalada A, et al. Acute Lateral Ankle Sprains in Track and Field Athletes: An Expanded Classification. Foot Ankle Clin. 2006;11(3):497-507.
354
16
Page 16 of 21
27.
of validity and reliability testing. Arch Phys Med Rehabil. 2006;87(9):1235-1241.
28.
instability measures. Foot Ankle Int. 2011;32(12):1140-1146.
29.
preliminary analysis. Isokinet Exerc Sci. 2009;17(3):155-160.
31.
Prospective Study. Am J Sports Med. 2005;33(3):415-423.
366
367
33.
Engebretsen AH, Myklebust G, Holme I, et al. Intrinsic risk factors for acute ankle injuries
among male soccer players: a prospective cohort study. Scand J Med Sci Sports.
368
2010;20(3):403-410.
369
370
Willems TM. Intrinsic Risk Factors for Inversion Ankle Sprains in Male Subjects: A
te
32.
d
Med. 2007;37(6):547-556.
364
365
Hrysomallis C. Relationship between balance ability, training and sports injury risk. Sports
M
362
363
Hadzic V, Sattler T, Topole E, et al. Risk factors for ankle sprain in volleyball players: A
an
30.
us
Athl Train. 2011;46(2):133-141.
360
361
Hiller CE, Kilbreath SL, Refshauge KM. Chronic ankle instability: evolution of the model. J
cr
358
359
Donahue M, Simon J, Docherty CL. Critical review of self-reported functional ankle
ip t
356
357
Hiller CE, Refshauge KM, Bundy AC, et al. The Cumberland ankle instability tool: a report
Ac ce p
355
34.
Witchalls J, Blanch P, Waddington G, et al. Intrinsic functional deficits associated with
371
increased risk of ankle injuries: a systematic review with meta-analysis. Br J Sports Med.
372
2012;46(7):515-523.
373 374 17
Page 17 of 21
i cr 375
Table 1. Characteristics of included studies
us
Tables
Type of Study
Followup length
Participants
Inclusion/Exclusion criteria
Predictor
Interventions
Outcome measures
Hiller et al 2008
Prospective Cohort
13 mths
33 adolescents dancers with an index sprain Age: 14 ± 1.8 yrs (total study participants: 115)
Inclusion: Dance and ballet students Exclusion criteria: None
Cumberland Ankle Instability tool, Foot movement during singleleg stance with eyes closed (30 s)
None
Re-sprain
Malliaropoulos et al 2009
Prospective Cohort
24 mths
202 elite Greek track and field athletes Age: 19 ± 4.1 yrs
Inclusion criteria: Acute index lateral ankle sprain, no history of ankle and foot injury Exclusion criteria: syndesmotic injury or a lower limb fracture
Severity of the initial sprain (I, II, IIIA or IIIB)
Common rehabilitation program
Re-sprain
de Noronha et al 2011
Prospective Cohort
12 mths
11 active university students with an index sprain Age: 20.9 ± 2.7 yrs (total study participants: 125)
Inclusion criteria: regular exercise (≥2/week) Exclusion criteria: sprain ≤ 1 month
CAIT, Foot movement during single-leg stance with eyes closed (30 s)
None
Re-sprain
8 mths
30 children presenting to ER(15 followed up for 8 mths ) Age: 11.2 yrs (range: 7-15)
Inclusion criteria: Acute index inversion sprain (≤ 2 days) with open epiphyseal plate Exclusion criteria: fracture or recurrent sprains
Severity of the initial sprain (I,II, III)
Initially:elastic compression wraps, AirSport Ankle brace, forearm crutches (if required). After 2 weeks: common rehabilitation program
Pain, mobility and instability* MRI
Ac
ce
pt
ed
M
Author, Year
an
374
Endele et al 2012
Prospective Cohort
376
CAIT: Cumberland ankle instability tool, ER: Emergency room, MRI: magnetic resonance imaging, Mths: months, S: seconds, Yr: year
377
*The method of measurement for pain, mobility and instability was not explained in the study 18
Page 18 of 21
378
Table 2. Odds ratio of re-sprains associated with CAIT scores, balance, and severity of the initial sprain Variables
OR
95% CI for odds ratio
P-value
0.839
0.664-1.061
0.142
CAIT score (non-injured side)
1.037
0.800-1.346
0.782
Foot-lifts (injured side)
0.958
0.897-1.057
0.391
Foot-lifts (non-injured side) Constant
1.017
0.914-1.132
0.756 0.403
15.887
Severity of the initial sprain
0.004
Grade II
2.603
1.175-5.768
0.018
Grade III
0.364
0.065-1.466
0.139
Constant 0.162 OR: odds ratio, CI: confidence interval
< 0.01
Ac ce p
te
d
379
an
Grade I
M
Step 1
cr
CAIT score (injured side)
us
Step 1
ip t
CAIT scores and balance
19
Page 19 of 21
380 381
Figure Legends Figure 1: Flow chart of the review process
Ac ce p
te
d
M
an
us
cr
ip t
382
20
Page 20 of 21
Ac ce p
te
d
M
an
us
cr
ip t
Figure(s)
Page 21 of 21
S1440-2440(14)00024-3 http://dx.doi.org/doi:10.1016/j.jsams.2014.01.005 JSAMS 989
To appear in:
Journal of Science and Medicine in Sport
Received date: Accepted date:
8-12-2013 25-1-2014
Please cite this article as: Pourkazemi F, Hiller CE, Raymond J, Nightingale EJ, Refshauge KM, Predictors of chronic ankle instability after an index lateral ankle sprain: A systematic review, Journal of Science and Medicine in Sport (2014), http://dx.doi.org/10.1016/j.jsams.2014.01.005 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Predictors of chronic ankle instability after an index lateral ankle sprain: A systematic
2
review
3 Fereshteh Pourkazemi,a Claire E. Hiller,a Jacqueline Raymond,b Elizabeth J Nightingale,a Kathryn M.
5
Refshauge a
ip t
4
6 a
Discipline of Physiotherapy, Faculty of Health Sciences, University of Sydney, Australia
8
b
Discipline of Exercise and Sport Science, Faculty of Health Sciences, University of Sydney, Australia
cr
7
Corresponding author: Fereshteh Pourkazemi
11
e-mail: [email protected]
an
10
us
9
Word count: 3037
14
The Abstract word count: 234
15
Number of tables: 2
16
Number of figures: 1
17
Supplemental files: 2
te
Ac ce p
18
d
13
M
12
1
Page 1 of 21
18
Predictors of chronic ankle instability after an index lateral ankle sprain: A systematic
19
review
20
Abstract
22
Objective: To identify the predictors of CAI after an index lateral ankle sprain.
23
Design: Systematic review
24
Methods: The data bases of MEDLINE, CINAHL, AMED, Scopus, SPORTDiscus, Embase, Web of
25
Science, PubMed, PEDro, and Cochrane Register of Clinical Trials were searched from the earliest
26
record until May 2013. Prospective studies investigating any potential intrinsic predictors of chronic
27
ankle instability (CAI) after an index ankle sprain were included. Eligible studies had a prospective
28
design (follow-up of at least three months), participants of any age with an index ankle sprain, and
29
had assessed ongoing impairments associated with CAI. Eligible studies were screened and data
30
extracted by two independent reviewers.
31
Results: Four studies were included. Three potential predictors of CAI, i.e., postural control,
32
perceived instability, and severity of the index sprain, were investigated. Decreased postural control
33
measured by number of foot lifts during single-leg stance with eyes closed and perceived instability
34
measured by Cumberland Ankle Instability Tool (CAIT) were not predictors of CAI. While the results
35
of one study showed that the severity of the initial sprain was a predictor of re-sprain, another study
36
did not.
37
Conclusion: Of the three investigated potential predictors of CAI after an index ankle sprain, only
38
severity of initial sprain (grade II) predicted re-sprain. However, concerns about validity of the
39
grading system suggest that these findings should be interpreted with caution.
40
Key words: ankle sprains, joint instability, sprain severity, postural balance
Ac ce p
te
d
M
an
us
cr
ip t
21
2
Page 2 of 21
41 42
1. Introduction
43 Ankle sprain injuries, particularly sprain of the lateral ligaments, are among the most common lower
45
limb injuries in the general1,2 and athletic populations3 as well as among military personnel.4 High
46
recurrence rate, along with residual impairments, are the consequence of ankle sprain for up to 54% of
47
individuals.5 The most common residual impairments include re-sprain, perceived instability and
48
episodes of giving way (also referred to as functional instability), joint laxity (also referred to as
49
mechanical instability), pain, swelling, a feeling of weakness and subsequently reduced level of
50
physical activity.5 These residual impairments, alone or in combination, are frequently termed chronic
51
ankle instability (CAI).6 Prevention of CAI, particularly re-sprain, is the main treatment goal for many
52
studies of ankle sprain,7 however, prevention is only possible if people at risk of developing CAI can
53
be identified. That is, to determine the most effective prevention strategy it is essential to understand
54
the underlying causes leading to CAI, and the factors increasing the risk of CAI.
55
History of a previous sprain is the most frequently reported risk factor for lateral sprain.8-10 Basketball
56
players with history of an ankle sprain were found to be five times more likely to re-sprain, although
57
the reasons for the increased risk are unknown.9 Hertel11 suggested that ankle sprains cause various
58
sensorimotor deficits that can lead to instability, and that the presence of instability increases the risk
59
of further sprain. Many studies have investigated this theory, but the findings are inconsistent.12-14
60
There are few studies of predictors of CAI, however two systematic reviews have evaluated closely
61
related questions and some information can be derived from their findings. One systematic review
62
found decreased dorsiflexion range of motion to be a strong predictor of lateral ankle sprain.15 The
63
second systematic review investigated the clinical course of acute ankle sprain,5 finding only one
64
study that reported on prognostic factors for development of CAI16. Based on this study,16 athletes
65
competing at a higher level of competition were at greater risk of persisting impairments after an
66
acute sprain than athletes competing at lower levels. Severity of pain, number of re-sprains, as well as
67
level of perceived instability and self-reported recovery after an acute lateral ankle sprain were found
68
to be independent of the severity of the initial sprain. Furthermore, men were reported to have a
Ac ce p
te
d
M
an
us
cr
ip t
44
3
Page 3 of 21
greater risk of developing residual impairments than women. However, not all studies included in
70
these two systematic reviews investigated the risk factors contributing to CAI after an index ankle
71
sprain.
72
Research to date has tended to focus more on changes associated with CAI rather than predictors of
73
CAI. Sensorimotor, functional, anatomical or biomechanical changes associated with CAI have been
74
analysed in a number of systematic reviews.14,17,18 Postural instability,17,18 prolonged time to balance
75
after a jump,14 lower limb muscle weakness or muscle imbalance19 were found to be associated with
76
CAI. Whether these impairments cause CAI or develop as a result of CAI is not clear. Therefore, the
77
objective of the present systematic review was to identify predictors of CAI (e.g., age, sex, body mass
78
index (BMI), level of physical activity, balance, postural control, proprioception, motor control,
79
severity of ankle sprain, perceived ankle instability, feeling of giving way, ligament laxity, pain, or
80
swelling) after an index ankle sprain. Prospective studies investigating any of these variables as
81
potential predictors of CAI after an index ankle sprain might enable health providers to design more
82
effective treatments to prevent ongoing problems.
84
2. Methods
Ac ce p
85
te
83
d
M
an
us
cr
ip t
69
86
The study protocol was developed based on the framework outlined in the guidelines provided by the
87
PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement.20 The
88
protocol of this systematic review is registered on PROSPERO (registration number
89
CRD42012002990).
90
2.1. Eligibility criteria
91
Studies were included if they met the following inclusion criteria: i) longitudinal design, ii) follow-up
92
length of at least three months since the sprain, iii) participants of any age who had sustained an index
93
ankle sprain only, iv) measuring at least one of the potential predictors of CAI, v) reporting on any re-
94
sprain or residual symptoms after the initial ankle sprain during the follow-up period, vi) published
95
full paper; abstracts were included if the authors provided the raw data for further analysis. Papers
4
Page 4 of 21
96
where it was possible that at least some of the participants had suffered an index sprain were
97
considered for inclusion, however if the data from this sub-group could not be isolated, then the paper
98
was excluded.
99
For randomised controlled trials, we included the minimal intervention group (e.g. conservative treatment of lateral ankle sprains, such as modified footwear and associated supports, taping, adapted
101
training programmes, and education). Randomised clinical trials were excluded if there was no
102
minimal intervention group e.g., if surgery or immobilisation for more than three days were
103
investigated.
104
2.2. Information Sources
105
Studies were identified through a search of the MEDLINE, CINAHL, AMED, Scopus,
106
SPORTDiscus, Embase, Web of Science, PubMed, PEDro, and Cochrane Register of Clinical Trials
107
to May 2013. Abstracts from the International Ankle Symposium (IAS) and International Foot and
108
Ankle Biomechanics (i-FAB) conference proceedings, and bibliographies of eligible studies were
109
hand searched. In addition, relevant experts were contacted to identify any unpublished studies that
110
may exist and to review the list of identified studies for completeness. No language restriction was
111
imposed, however, if a translation could not be arranged, the paper was excluded.
112
2.3. Search Strategy
113
Key terms used in our search strategy are presented in supplemental file 1. Search protocols were
114
specifically designed to target prospective studies.
115
2.4. Study Selection
116
All studies identified by the search strategy were screened against the eligibility criteria independently
117
by the first author (FP) and by one other author (CH, JR or KR). Titles and abstracts were inspected
118
and clearly ineligible studies were removed. Full copies of potentially eligible papers were retrieved.
119
Any inconsistencies regarding inclusion of trials were resolved by consensus.
120
2.5. Potential Predictors of CAI
121
The following predictors, identified from previous studies of risk factors for lateral ankle sprain, were
122
considered as potential risk factors of developing CAI: age, sex, height, weight, BMI, leg dominance,
123
foot type, foot and lower limb alignment, ankle joint laxity, general ligamentous laxity, postural
Ac ce p
te
d
M
an
us
cr
ip t
100
5
Page 5 of 21
control, muscle weakness or muscle strength/power imbalance in lower limbs, lower limb muscle
125
reaction time, range of motion at lower limb joints, physical fitness, gait biomechanics, and severity
126
of symptoms after the initial sprain (e.g., pain, swelling, laxity, feeling of instability).
127
2.6. Outcome measures
128
Any measures that assessed ongoing impairments associated with CAI after the initial sprain during
129
the follow up period (e.g., re-sprain, swelling, pain, mechanical instability, perceived instability,
130
feeling of giving way, or feeling of weakness) were considered as outcome measures.
131
2.7. Data Extraction Process
132
Data were extracted independently by two authors (FP and CH) and confirmed by one other author
133
(either JR or KR). Any discrepancies were settled by further discussion and consensus. Study
134
characteristics extracted were study type, target population (setting, sex, age), sample size, inclusion
135
criteria, follow-up duration, prognostic factors measured, interventions (if any) and all reported
136
outcome measures.
137
2.8. Quality Assessment
138
Risk of bias and methodological quality of included studies were assessed using the quality
139
assessment tool developed by Pengel21 et al. This tool consists of 7 items rated as either ‘yes’, ‘no’ or
140
‘N/A’ (not applicable). Four items relate to control of bias (items 1-4), two to appropriate
141
measurement of variables (items 5 and 6) and one to control of confounding variables (item 7). Two
142
raters (FP and JR) independently assessed the quality and a third author (CH) resolved disagreements.
143
2.9. Synthesis of Results
144
Study outcomes were statistically pooled if the studies were considered to be sufficiently
145
homogeneous. For homogeneous studies, raw data were used in a direct logistic regression to assess
146
whether the predictor variables increased the likelihood of re-sprain. If the studies were considered
147
too heterogeneous, data were not pooled and the outcomes were described.
Ac ce p
te
d
M
an
us
cr
ip t
124
148 149
6
Page 6 of 21
149 150
3. Results
151 The search strategy identified 8085 titles. Following title and abstract screening, 210 potentially
153
relevant articles were identified, of which 16 met all the inclusion criteria. The data required for
154
analysis were reported in two of the included studies.22,23 The authors of the remaining 14 articles,
155
which included participants with one or more ankle sprains, were contacted for data related to the
156
participants with only an index ankle sprain. Seven authors replied, of whom only two were able to
157
provide raw data for analysis.24,25 Therefore, four studies22-25 were included in this review (Figure 1).
158
3.1. Study Characteristics
159
All included studies were prospective cohort trials, with follow-up periods that varied between 8 and
160
24 months. Participants were adults in two studies,23,24 adolescents in one study25 and children and
161
adolescents in the fourth study.22 In two studies the inclusion criterion was an acute index ankle sprain
162
of less than two days duration.22,23 The other two studies24,25 included a heterogeneous group of
163
participants, with a history that varied between multiple ankle sprains to no history of sprain, and the
164
time since sprain was not standardised. Separate data of the participants with an index ankle sprain
165
were provided by the authors of these two studies.24,25 Participants were recruited from various
166
settings, including a university,24 dance school,25 and primary care practice(s).22,23 The predictors of
167
CAI measured were perceived instability and balance24,25 in two of the studies and severity of index
168
sprain22,23 in the other two. The severity of injury was measured by either clinical symptoms and
169
anterior drawer stress x-ray23 or clinical symptoms and magnetic resonance imaging (MRI).22 The
170
outcomes reported were re-sprain in three studies23-25 and pain, mobility and perceived instability in
171
the remaining study.22 Details of the included studies are presented in Table 1.
172
3.2. Quality Assessment
173
No study was rated as having excellent methodological quality. Methodological quality was moderate
174
in three studies, with scores of 5 25 and 4.23,24 The fourth study22 had low methodological quality, with
175
a score of 3 (supplemental file 2).None of these studies used statistical adjustment for potentially
176
confounding factors. Only one study used blinded outcome assessor(s).22
Ac ce p
te
d
M
an
us
cr
ip t
152
7
Page 7 of 21
Ac ce p
te
d
M
an
us
cr
ip t
177
8
Page 8 of 21
177 3.4. Predictive Factors of CAI
179
Data for two of the studies24,25 were pooled, because they investigated perceived instability and
180
balance as predictors for re-sprain. We were not able to pool data for the two other studies22,23
181
because, although they both investigated severity of initial sprain, they used different grading systems
182
and reported on different outcome measures.
183
Perceived instability and balance: Two studies24,25 investigated perceived instability and static balance
184
on both legs as a predictor of re-sprain. Perceived instability was measured using the Cumberland
185
Ankle Instability Tool (CAIT) and balance was measured recording the number of foot movements
186
during single-leg stance with eyes closed for 30 seconds. Hiller25 et al and de Noronha24 et al provided
187
raw data for 44 participants with history of an index ankle sprain. After the follow-up periods of 12-
188
13 months, 11 participants had re-sprained their ankles. Raw data from these two studies were pooled
189
and logistic regression was performed, finding that the predictive capacity of the model was not
190
statistically significant (Table 2). That is, CAIT score and number of foot lifts during single-leg stance
191
with eyes close did not predict re-sprain.
192
Severity of the index ankle sprain: Two studies investigated severity of the index sprain as a predictor
193
of CAI. Malliaropoulos23 et al used an expanded grading system26 (I, II, IIIA and IIIB) to determine
194
severity of sprain. They found significant differences in rate of re-sprain between athletes with
195
different grades of injury during a 24 month follow-up period. Elite track and field athletes with less
196
severe ankle sprain (grade I or II) had a higher rate of re-sprain than athletes with more severe ankle
197
sprain (grade IIIA or IIIB). To further analyse the results of this study, we performed direct logistic
198
regression to assess the impact of the severity of index sprain on the likelihood of re-sprain. The
199
predictive capacity of the model was statistically significant, X²(3, n=202) = 14.71, P=0.002. The
200
severity of sprain (grades I to IIIB) explained between 7% (Cox and Snell R2) and 11.4% (Nagelkerke
201
R2) of the variance in re-sprain status, and correctly classified 81.7% of cases. However, only grade I
202
and II made a unique statistically significant contribution to the model (Table 2). The strongest
203
predictor of re-sprain was grade II injury, recording an odds ratio (OR) of 2.6. This indicates that
204
athletes with grade II injury were 2.6 times more likely to re-sprain their ankles.
Ac ce p
te
d
M
an
us
cr
ip t
178
9
Page 9 of 21
In the second study, Endele22 et al divided 30 children with an acute ankle sprain into three groups
206
based on the severity of their clinical symptoms (I, II or III). Only children with grade II and III were
207
followed up for eight months. Endele22 et al reported no significant differences between children with
208
grades II or III of ankle sprain, with regard to limitation of mobility, pain or instability, three months
209
after the index injury. These outcome measures were not reported at eight months follow up. Due to
210
the small number of participants in this study (n=15) no further analysis was performed.
ip t
205
4. Discussion
us
212
cr
211
213
The present systematic review found that only three potential predictors of CAI (severity of sprain,
215
balance and perceived instability) had been investigated in longitudinal studies after an index ankle
216
sprain. Although recent evidence from cross-sectional studies shows that postural sway, time to
217
balance after a jump, foot position during gait, strength, and proprioception are altered in participants
218
with CAI,14,15,19 none of these factors has been investigated in a longitudinal study after the index
219
sprain.
220
Two studies investigated the severity of the index ankle sprain as a potential predictor of CAI, with
221
conflicting results. Although there are factors limiting comparisons between these studies, including
222
the use of different grading systems, different age groups (adults23 vs. children22), and different
223
duration of follow-up (24 months23 vs. 8 months22), a critique of the two studies highlights some
224
critical issues.
225
Malliaropoulos23 et al, classified the severity of index sprains into four grades based on the patients’
226
clinical symptoms (dorsiflexion range of motion and swelling). Participants were referred for stress
227
radiography only if they were diagnosed with a grade III sprain. In the second study, Endele22 et al
228
also classified the sprains as grades I, II or III based on the clinical symptoms (weight bearing status
229
and swelling), however, they performed additional MRI investigations. Interestingly, they found that
230
clinical symptoms were not well correlated with the grade of injury. The amount of swelling and
231
ability to weight bear were highly correlated with bone bruising, based on their MRI findings.
Ac ce p
te
d
M
an
214
10
Page 10 of 21
Furthermore, six (50%) children with sprains classified as grade II based on the clinical symptoms,
233
had complete ligament rupture and three had a Salter I injury. These findings raise concerns about the
234
validity of the grading systems based on symptoms, alone.
235
The second study22 provides valuable information regarding the severity of ankle sprain in 30
236
children. In this study however, only children with sprains classified as grade II or III were followed
237
up, and the follow-up was at eight months, potentially an inadequate period when investigating re-
238
sprain and CAI. In this study, anatomical changes were reported in the MRIs of 10 children at eight
239
months, but Endele 22 et al did not re-assess pain, mobility or instability at eight months follow up.
240
Since the main aim of this study was to compare symptoms with MRI findings, the outcomes of pain,
241
instability or mobility were not reported.
242
Based on the results of these two studies, it is not possible to conclude definitively that ankle sprains
243
classified as grades I or II predict development of CAI. To be able to determine whether the severity
244
of an initial sprain predicts CAI, a valid grading system, a large sample size and adequate follow-up
245
are required.
246
Pooled data from the two other studies24,25 showed that CAIT scores do not predict re-sprain. CAIT is
247
a measure of perceived instability27 and previous studies demonstrated that perceived ankle instability
248
is related to ankle sprain.28 However, based on the new model of sub-groups of CAI developed by
249
Hiller29 et al, there is a sub-group of participants with CAI who have perceived instability without
250
recurrence of sprain. The findings of Hiller25 et al and de Noronha24 et al support the proposed model,
251
and explain the lack of predictive value of the CAIT score. Decreased CAIT score, or presence of
252
perceived instability, may increase risk of re-sprain in some subgroups, but not others.
253
The results of this systematic review also show that postural control, measured as the number of foot
254
lifts during single-leg stance with eyes closed was not a predictor of CAI. Many studies have
255
investigated postural control as a predictor of acute ankle sprain.25,30,31 These studies however, used
256
different measurement methods and found different results. While most of these studies found
257
decreased postural control as a predictor of acute ankle sprains,25,32 several found no relationship
258
between postural control scores and the risk of ankle sprain.30,31,33 In a recent systematic review,
259
Witchalls34 et al reported that the studies in which postural control was tested by scoring the number
Ac ce p
te
d
M
an
us
cr
ip t
232
11
Page 11 of 21
of errors during a test, did not show that poorer postural control increased the risk of ankle sprain.
261
They suggested that the increased subjectivity of these methods may increase the variability in scores
262
and render them less meaningful for measuring postural control.
263
While it seems the number of foot lifts during single-leg stance might not predict re-sprain, these
264
results should be interpreted with caution. Firstly, the different populations in these two studies
265
(adolescent dancers25 vs. active university students24) may have been a confounding factor. In
266
addition, the time from the index ankle sprain was not controlled for, potentially constituting another
267
confounding factor. Finally, the aim of these studies was not to predict CAI after an index sprain, and
268
therefore data were a subset. This resulted in a small number of participants in these two studies with
269
consequent lack of power for our re-analysis.
270
The strength of this systematic review was that no language limitations were imposed. Where
271
required, a translation in different languages was made (e.g. German, Italian, Japanese, and French).
272
We also included the grey literature. However, several studies could not be included because
273
reporting was insufficient and the data were not available for further analysis.
an
M
Ac ce p
276
d
5. Conclusion
te
274 275
us
cr
ip t
260
277
We found that of the three potential predictors of CAI that have been investigated prospectively after
278
an index ankle sprain, only severity of initial sprain (grade II) predicted re-sprain. However, concerns
279
about validity of the grading system suggest that these findings should be interpreted with caution.
280
Consequently, there is a large gap in the literature. Anatomical, central, physiological and
281
psychosocial changes have yet to be investigated after an initial sprain and the relationship of such
282
variables to re-injury and other residual symptoms remains unknown.
283
12
Page 12 of 21
283 284
Practical Implications
285
Severity of the initial ankle sprain does not necessarily predict the likelihood of
ip t
•
286
developing chronic ankle instability. •
288
Perceived ankle instability and static balance do not appear to predict future ankle
cr
287
sprains. •
290
Currently there is limited research investigating predictors of chronic ankle instability after
us
289
an index ankle sprain.
291
293
an
292 Acknowledgment
M
294
The authors would like to thank Dr Markus Hübscher, Ms Naomi Kusano, and Ms Alexandra Di Lallo
296
for their assistance in translation of non-English papers.
Ac ce p
te
d
295
13
Page 13 of 21
297 298
References
299
States. J Bone Joint Surg Am. 2010;92(13):2279-2284.
301
302
Waterman BR, Owens BD, Davey S, et al. The epidemiology of ankle sprains in the United
ip t
1.
2.
cr
300
Bridgman SA, Clement D, Downing A, et al. Population based epidemiology of ankle sprains attending accident and emergency units in the West Midlands of England, and a survey of UK
304
practice for severe ankle sprains. Emerg Med J. 2003;20(6):508-510.
5.
6.
7.
Eiff MP, Smith AT, Smith GE. Early Mobilization Versus Immobilization in the Treatment of Lateral Ankle Sprains. Am J Sports Med. 1994;22(1):83-88.
314
315
Delahunt E, Coughlan GF, Caulfield B, et al. Inclusion criteria when investigating
insufficiencies in chronic ankle instability. Med Sci Sports Exerc. 2010;42(11):2106-2121.
312
313
van Rijn RM, van Os AG, Bernsen RM, et al. What is the clinical course of acute ankle
sprains? A systematic literature review. Am J Med. 2008;121(4):324-331 e326.
310
311
M
United States Military Academy. Am J Sports Med. 2010;38(4):797-803.
308
309
Waterman BR, Belmont PJ, Jr., Cameron KL, et al. Epidemiology of ankle sprain at the
d
4.
te
307
Fong DT, Hong Y, Chan LK, et al. A systematic review on ankle injury and ankle sprain in sports. Sports Med. 2007;37(1):73-94.
306
an
3.
Ac ce p
305
us
303
8.
Milgrom C, Shlamkovitch N, Finestone A, et al. Risk factors for lateral ankle sprain: a prospective study among military recruits. Foot Ankle. 1991;12(1):26-30.
316
14
Page 14 of 21
9.
factors. Br J Sports Med. 2001;35(2):103-108.
318
10.
J Rehabil Res Dev. 2006;43(7):819-824.
13.
14.
15.
de Noronha M, Refshauge KM, Herbert RD, et al. Do voluntary strength, proprioception,
range of motion, or postural sway predict occurrence of lateral ankle sprain? Br J Sports Med.
330
2006;40(10):824-828; discussion 828.
331
332
Hiller CE, Nightingale EJ, Lin C-WC, et al. Characteristics of people with recurrent ankle sprains: a systematic review with meta-analysis. Br J Sports Med. 2011;45(8):660-672.
328
329
d
postural control in athletes with a history of ankle sprain. J Athl Train. 2013;48(2):203-208.
326
327
Steib S, Zech A, Hentschke C, et al. Fatigue-induced alterations of static and dynamic
M
324
325
Akbari M, Karimi H, Farahini H, et al. Balance problems after unilateral lateral ankle sprains.
an
12.
us
Instability. J Athl Train. 2002;37(4):364-375.
322
323
Hertel J. Functional Anatomy, Pathomechanics, and Pathophysiology of Lateral Ankle
cr
11.
ip t
Sports Exerc. 1983;15(3):267-270.
320
321
Ekstrand J, Gillquist J. Soccer injuries and their mechanisms: a prospective study. Med Sci
te
319
McKay GD, Goldie PA, Payne WR, et al. Ankle injuries in basketball: injury rate and risk
Ac ce p
317
16.
Linde F, Hvass I, Jurgensen U, et al. Early mobilizing treatment in lateral ankle sprains.
333
Course and risk factors for chronic painful or function-limiting ankle. Scand J Rehabil Med.
334
1986;18(1):17-21.
335
17.
Arnold BL, De La Motte S, Linens S, et al. Ankle instability is associated with balance impairments: A meta-analysis. Med Sci Sports Exerc. 2009;41(5):1048-1062.
336
15
Page 15 of 21
18.
instability: A systematic review with meta-analysis. J Sci Med Sport. 2010;13(1):2-12.
338
19.
prognosis. BMJ. 2003;327(7410):323.
22.
23.
24.
25.
Hiller CE, Refshauge KM, Herbert RD, et al. Intrinsic predictors of lateral ankle sprain in adolescent dancers: a prospective cohort study. Clin J Sport Med. 2008;18(1):44-48.
352
353
de Noronha M, Franca LC, Haupenthal A, et al. Intrinsic predictive factors for ankle sprain in
active university students: A prospective study. Scand J Med Sci Sports. 2012.
350
351
Malliaropoulos N, Ntessalen M, Papacostas E, et al. Reinjury after acute lateral ankle sprains in elite track and field athletes. Am J Sports Med. 2009;37(9):1755-1761.
348
349
d
children. Foot Ankle Int. 2012;33(12):1063-1068.
346
347
Endele D, Jung C, Bauer G, et al. Value of MRI in diagnosing injuries after ankle sprains in
M
344
345
Pengel LH, Herbert RD, Maher CG, et al. Acute low back pain: systematic review of its
an
21.
us
meta-analyses: the PRISMA statement. J Clin Epidemiol. Oct 2009;62(10):1006-1012.
342
343
Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and
cr
20.
ip t
Functional Ankle Instability: A Meta-Analysis. J Athl Train. 2009;44(6):653-662.
340
341
Arnold BL, Linens SW, de Ia Motte SJ, et al. Concentric Evertor Strength Differences and
te
339
Munn J, Sullivan SJ, Schneiders AG. Evidence of sensorimotor deficits in functional ankle
Ac ce p
337
26.
Malliaropoulos N, Papacostas E, Papalada A, et al. Acute Lateral Ankle Sprains in Track and Field Athletes: An Expanded Classification. Foot Ankle Clin. 2006;11(3):497-507.
354
16
Page 16 of 21
27.
of validity and reliability testing. Arch Phys Med Rehabil. 2006;87(9):1235-1241.
28.
instability measures. Foot Ankle Int. 2011;32(12):1140-1146.
29.
preliminary analysis. Isokinet Exerc Sci. 2009;17(3):155-160.
31.
Prospective Study. Am J Sports Med. 2005;33(3):415-423.
366
367
33.
Engebretsen AH, Myklebust G, Holme I, et al. Intrinsic risk factors for acute ankle injuries
among male soccer players: a prospective cohort study. Scand J Med Sci Sports.
368
2010;20(3):403-410.
369
370
Willems TM. Intrinsic Risk Factors for Inversion Ankle Sprains in Male Subjects: A
te
32.
d
Med. 2007;37(6):547-556.
364
365
Hrysomallis C. Relationship between balance ability, training and sports injury risk. Sports
M
362
363
Hadzic V, Sattler T, Topole E, et al. Risk factors for ankle sprain in volleyball players: A
an
30.
us
Athl Train. 2011;46(2):133-141.
360
361
Hiller CE, Kilbreath SL, Refshauge KM. Chronic ankle instability: evolution of the model. J
cr
358
359
Donahue M, Simon J, Docherty CL. Critical review of self-reported functional ankle
ip t
356
357
Hiller CE, Refshauge KM, Bundy AC, et al. The Cumberland ankle instability tool: a report
Ac ce p
355
34.
Witchalls J, Blanch P, Waddington G, et al. Intrinsic functional deficits associated with
371
increased risk of ankle injuries: a systematic review with meta-analysis. Br J Sports Med.
372
2012;46(7):515-523.
373 374 17
Page 17 of 21
i cr 375
Table 1. Characteristics of included studies
us
Tables
Type of Study
Followup length
Participants
Inclusion/Exclusion criteria
Predictor
Interventions
Outcome measures
Hiller et al 2008
Prospective Cohort
13 mths
33 adolescents dancers with an index sprain Age: 14 ± 1.8 yrs (total study participants: 115)
Inclusion: Dance and ballet students Exclusion criteria: None
Cumberland Ankle Instability tool, Foot movement during singleleg stance with eyes closed (30 s)
None
Re-sprain
Malliaropoulos et al 2009
Prospective Cohort
24 mths
202 elite Greek track and field athletes Age: 19 ± 4.1 yrs
Inclusion criteria: Acute index lateral ankle sprain, no history of ankle and foot injury Exclusion criteria: syndesmotic injury or a lower limb fracture
Severity of the initial sprain (I, II, IIIA or IIIB)
Common rehabilitation program
Re-sprain
de Noronha et al 2011
Prospective Cohort
12 mths
11 active university students with an index sprain Age: 20.9 ± 2.7 yrs (total study participants: 125)
Inclusion criteria: regular exercise (≥2/week) Exclusion criteria: sprain ≤ 1 month
CAIT, Foot movement during single-leg stance with eyes closed (30 s)
None
Re-sprain
8 mths
30 children presenting to ER(15 followed up for 8 mths ) Age: 11.2 yrs (range: 7-15)
Inclusion criteria: Acute index inversion sprain (≤ 2 days) with open epiphyseal plate Exclusion criteria: fracture or recurrent sprains
Severity of the initial sprain (I,II, III)
Initially:elastic compression wraps, AirSport Ankle brace, forearm crutches (if required). After 2 weeks: common rehabilitation program
Pain, mobility and instability* MRI
Ac
ce
pt
ed
M
Author, Year
an
374
Endele et al 2012
Prospective Cohort
376
CAIT: Cumberland ankle instability tool, ER: Emergency room, MRI: magnetic resonance imaging, Mths: months, S: seconds, Yr: year
377
*The method of measurement for pain, mobility and instability was not explained in the study 18
Page 18 of 21
378
Table 2. Odds ratio of re-sprains associated with CAIT scores, balance, and severity of the initial sprain Variables
OR
95% CI for odds ratio
P-value
0.839
0.664-1.061
0.142
CAIT score (non-injured side)
1.037
0.800-1.346
0.782
Foot-lifts (injured side)
0.958
0.897-1.057
0.391
Foot-lifts (non-injured side) Constant
1.017
0.914-1.132
0.756 0.403
15.887
Severity of the initial sprain
0.004
Grade II
2.603
1.175-5.768
0.018
Grade III
0.364
0.065-1.466
0.139
Constant 0.162 OR: odds ratio, CI: confidence interval
< 0.01
Ac ce p
te
d
379
an
Grade I
M
Step 1
cr
CAIT score (injured side)
us
Step 1
ip t
CAIT scores and balance
19
Page 19 of 21
380 381
Figure Legends Figure 1: Flow chart of the review process
Ac ce p
te
d
M
an
us
cr
ip t
382
20
Page 20 of 21
Ac ce p
te
d
M
an
us
cr
ip t
Figure(s)
Page 21 of 21
