The Journal of Foot & Ankle Surgery 57 (2018) 982–986

Contents lists available at ScienceDirect

The Journal of Foot & Ankle Surgery

j o u r n a l h o m e p a g e : w w w. j f a s . o r g

Comparative Outcomes of Cast and Removable Support in Fracture

Fifth Metatarsal Bone: Systematic Review and Meta-Analysis
Kwanchai Pituckanotai, MD 1, Alisara Arirachakaran, MD 1, Peerapong Piyapittayanun, MD 1,
Harit Tuchinda, MD 2, Ekachot Peradhammanon, MD 3, Jatupon Kongtharvonskul, MD, PhD 4
1Orthopedist, Orthopedics Department, Police General Hospital, Bangkok, Thailand
2Orthopedist, Orthopedic Department, Bangkok Metropolitan Administration General Hospital, Bangkok, Thailand
3Orthopedist, Orthopedics Department, Phrachomklao Hospital, Pedchburi, Thailand
4
Orthopedist, Sport and Orthopaedic Department, Samitivej Hospital, Bangkok, Thailand

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

Level of Clinical Evidence: 2 Fractures of the metatarsals account for 35% of all foot fractures. Conservative management of fractures
proximal to the metaphyseal–diaphyseal junction of the fifth metatarsal bone (pseudo-Jones) is pro-
Keywords:
tected weightbearing. The methods of protected weightbearing include a short-leg cast and splint (boot
bandage
foot score cast, Jones bandage, and elastic bandage). However, no consensus has yet been reached regarding which
fracture fifth metatarsal bone method is most suitable. We conducted a systematic review and meta-analysis to compare the out-
meta-analysis comes of a short-leg cast and splint for pseudo-Jones metatarsal fractures. We searched the PubMed and
nonunion Scopus databases up to October 29, 2016. Five of 104 studies (3 comparative studies and 2 randomized
pseudo-Jones controlled trials; n = 246 patients) were eligible. Of the studies, 3, 5, and 4 were included in pooling of
short leg cast early (within 1 month) and last follow-up foot function scale scores and fracture nonunion, respective-
splint
ly. The unstandardized mean difference of early (within 1 month) and last follow-up foot scores for the
systematic review
short leg cast were −14.58 (95% confidence interval [CI] −24.12 to −5.04) and −3.89 (95% CI −6.30 to −1.49),
significantly lower than the scores for the splint (bandage or boot support) for pseudo-Jones fracture of
the fifth metatarsal bone. The risk of nonunion of the fifth metatarsal bone fracture of the patients who
were treated with short leg cast method was insignificantly greater at 1.57 times (95% CI 0.29 to 8.49)
that compared with the splint. The treatment of fracture of the pseudo-Jones fifth metatarsal bone with
a splint (boot or bandage) resulted in foot function scale scores better than those with short leg cast treat-
ment and a lower nonunion rate.
© 2018 by the American College of Foot and Ankle Surgeons. All rights reserved.

Fractures of the base of the fifth metatarsal are a common injury
of the foot originally described by Sir Robert Jones in 1902 (1,2).
Since then, virtually all fractures involving the proximal aspect of
the fifth metatarsal have been classified as “Jones” fractures. Several
investigators, however, have recognized the existence of ≥2 major
patterns of fracture at the base of the fifth metatarsal: (1) an avul-
sion fracture of a variably sized portion of the tuberosity or the
most proximal part of the metatarsal; and (2) a transverse fracture
through the proximal diaphysis of the metatarsal within 1.5 cm of
the tuberosity, which has been called a “pseudo-Jones avulsion frac-
ture” (1–11) Several nonoperative treatments have been studied,
including elastic bandaging and wearing a hard-soled shoe to immo-
Financial Disclosure: None reported.
Conflict of Interest: None reported.
Address correspondence to: Jatupon Kongtharvonskul, MD, PhD, Sport and
Orthopaedic Department, Samitivej Hospital, Bangkok 10500, Thailand.
E-mail address: Jatupon_kong@hotmail.com (J. Kongtharvonskul).
bilization in a cast, focused rigidity casting, or a walking boot
(2,4,12–19). Several comparative studies have compared short leg
casting and splinting, including elastic or compression bandaging,
and a walking boot (2,14,17–19). These included 2 randomized con-
trolled trials (RCTs) (18,19) and 3 cohort studies (2,14,17). However,
no consistent results have been reported from these trials. Some
studies (2,17,19) have reported advantages to using removable sup-
ports over short leg casts, and other studies (14,18) have not. To the
best of our knowledge, no meta-analysis or systematic review has
compared short leg casting and splinting for treatment of pseudo-
Jones avulsion fractures proximal to the metaphyseal–diaphyseal
junction of the fifth metatarsal bone. The effect of splinting on
nonoperative outcomes for treatment of pseudo-Jones metatarsal
fractures has been debated. Therefore, we performed a meta-
analysis and systematic review of comparative and randomized clinical
studies to establish the best evidence to address this controversy.
The present systematic review and meta-analysis compared the out-
comes of short leg casting and splinting for treatment of pseudo-Jones

1067-2516/$ - see front matter © 2018 by the American College of Foot and Ankle Surgeons. All rights reserved.
https://doi.org/10.1053/j.jfas.2018.03.018
 K. Pituckanotai et al. / The Journal of Foot & Ankle Surgery 57 (2018) 982–986
avulsion fractures proximal to the metaphyseal–diaphyseal junction
of the fifth metatarsal bone.
Materials and Methods
The Medline and Scopus databases were used to identify relevant studies re-
ported in English since the date of inception to October 29, 2016. The PubMed and Scopus
search engines were used to locate studies using the following search terms: “frac-
ture metatarsal bone” AND “clinical trial.” References from the reference lists of the
included trials were also explored.
Study Selection
The identified studies were first selected according to the titles and abstracts by 2
independent authors (K.P., A.A.). The full report was retrieved if a decision could not
be made from the abstract. Disagreements were resolved by consensus and discus-
sion with a third author (J.K.). The reasons for ineligibility or exclusion of the studies
were recorded and described.
Inclusion Criteria
RCTs and comparative studies that had compared the clinical outcomes between
short leg casting and splinting for the treatment pseudo-Jones avulsion fractures prox-
imal to the metaphyseal–diaphyseal junction of the fifth metatarsal bone were eligible
if they had met the following criteria:
• Comparison of clinical outcomes between short leg casting and splinting for treat-
ment of pseudo-Jones avulsion fractures proximal to the metaphyseal–diaphyseal
junction of the fifth metatarsal bone
• Comparison of ≥1 of the following outcomes: foot function scales (American Or-
thopaedic Foot and Ankle Society [AOFAS] ankle-hindfoot scale and visual analog
scale [VAS] for the foot and ankle), VAS for pain, and fracture nonunion
• Sufficient data available to extract and pool (i.e., reported mean, standard devia-
tion), the number of subjects stratified by treatment for the continuous outcomes
and the number of patients stratified by treatment for the dichotomous outcomes
Data Extraction
Two of us (K.P., A.A.) independently performed the data extraction using standard-
ized data extraction forms. The general characteristics of the study (i.e., mean age, gender,
level of evidence, mean follow-up time, and foot function scale scores at baseline) were
extracted. The number of subjects and means ± standard deviation for continuous out-
comes (i.e., foot function scale scores [AOFAS ankle-hindfoot scale or VAS for the foot
and ankle]) between groups were extracted. Cross-tabulated frequencies between treat-
ment and all dichotomous outcomes (nonunion of fracture) were also extracted. Any
disagreements were resolved by discussion and consensus with a third author (J.K.).
Quality Assessment
Two of us (J.K., A.A.) independently assessed the risk of bias for RCTs using the Pre-
ferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (20). Six
domains were assessed, including sequence generation, allocation concealment, blind-
ing (participant, personnel, and outcome assessors), incomplete outcomes data, selective
outcomes reporting, and other sources of bias. Assessment of the risk of bias for cohort
studies was performed using the modified Newcastle-Ottawa scale (21). Five domains
were assessed, including representativeness of cohorts, ascertainment of exposure and
outcome, adjustment for confounders, and follow-up duration. The risk of bias in each
domain was classified as low, high, or unclear (Appendix S1). Disagreements between
2 authors were resolved by consensus and discussion with a third author (J.K.). The level
of agreement for each domain and the overall domains was assessed using kappa statistics.
Outcomes of Interest
The outcomes of interests included foot function scores (AOFAS ankle-hindfoot scale
or VAS for the foot and ankle) and fracture nonunion. These outcomes were mea-
sured as reported in the original studies, including a foot function scale score from 0
to 100, with higher values indicating better outcomes. The rate of fracture nonunion
was also considered.
Statistical Analysis
For continuous outcomes (i.e., foot function scales, including the AOFAS ankle-
hindfoot scale and VAS for the foot and ankle), the mean difference between short leg
casting and removable support groups for the treatment of acute fifth metatarsal frac-
983
ture was estimated for each study. The unstandardized mean difference was applied
to pool the outcomes across studies. Before pooling, intervention effects were as-
sessed by whether they varied or were heterogeneous across the included studies.
Heterogeneity of the mean differences was checked using the Q statistic, and the degree
of heterogeneity was also quantified using the I2 statistic. If heterogeneity was signif-
icant or I2 >25%, the unstandardized mean difference was estimated using a random
effects model, otherwise a fixed effects model was applied.
For dichotomous outcomes, the odds ratio (OR) for fracture nonunion was esti-
mated for each study. The heterogeneity of ORs across studies was assessed using the
same method stated previously. If heterogeneity was present, the random effects using
the Dersimonian and Laird method was applied to pool the ORs, otherwise the fixed
effects by inverse variance method was applied.
Metaregression was applied to explore the cause of heterogeneity by fitting a
covariable (e.g., mean age, gender, level of evidence, mean follow-up time, and foot func-
tion scores at baseline) in the meta-regression model. A subgroup or sensitivity analysis
was then performed according to the results of the meta-regression. Publication bias
was assessed using contour funnel plots (22,23) and Egger tests (24). Asymmetry of
the funnel plot might have resulted from some missing studies in which the negative
results might not have been reported and thus could not be identified. The metatrim
and fill method was used to estimate the number of studies that might have been missing
and to adjust the pooled estimate (25). All analyses were performed using STATA, version
14.0 (26). A p value < .05 was considered statistically significant, except for the test of
heterogeneity for which p < .10 was used.
Results
We identified 46 and 81 studies from Medline and Scopus, respec-
tively (Fig. 1). Of the identified studies, 17 were duplicates, leaving
104 studies for review of the titles and abstracts. Of these, 5 studies
underwent a full review and data were extracted. The characteris-
tics of the 5 studies (2,14,17–19) are described in Table 1. These 5
studies, including 3 (2,14,17) cohort studies and 2 (18,19) RCTs, re-
ported the foot function scale scores (AOFAS ankle-hindfoot scoring
system or VAS for the foot and ankle) and fracture nonunion rates.
Foot function was reported using the AOFAS midfoot scoring system
in 3 studies (2,14,18) and a VAS for the foot and ankle in 2 studies
(17,19). Nonunion was reported in 4 studies (2,14,18,19). The mean
age and mean follow-up duration for the participants was 37.5 to 49.7
years and 1 to 2 years, respectively. The proportion of males ranged
from 28% to 54%. All 5 studies used short leg casting; the compara-
tor was elastic bandaging in 3 studies (17–19), Jones bandage in 1 study
(2), and boot splinting in 1 study (14).
Risk of Bias in Included Studies
The risk of bias in the 5 studies is presented in Table 2.
Outcomes
Early Outcomes (Follow-Up Point Within 1 Month)
Three studies reported the mean function scores within 1 month
of follow-up comparing short leg casting and splinting, with 67 and
69 patients, respectively (Table 3). The pooled unstandardized mean
difference varied moderately across the studies (χ2 = 5.14, df = 2, p =
.077, I2 = 61.1%) and was −14.58 (95% confidence interval [CI] −24.12
to −5.04), indicating that the short leg casting group had foot func-
tion scores that were significantly worse statistically compared with
those from the splinting group, including Jones bandage, elastic
bandage, and boot splint; Fig. 2). None of the covariables could explain
the heterogeneity. No evidence of publication bias was found using
Egger’s test or contour funnel plot (coefficient = −0.27, standard error
= 2.36, p = .928).
Late Outcomes (Last Follow-Up Point)
Five studies reported the mean foot function scores comparing short
leg casting and splinting, with 102 and 107 patients, respectively
(Table 3). The pooled unstandardized mean difference was homoge-
nous across the studies (χ2 = 3.61, df = 0.4, p = .426, I2 = 0%) and was
984 K. Pituckanotai et al. / The Journal of Foot & Ankle Surgery 57 (2018) 982–986

Fig. 1. Flow of study selection.

Table 1
Characteristics of included studies

Investigator Year Study Type Level of Evidence Intervention Follow-Up (mo) Male (%) Fracture Type Age (y) Outcome Measure

Wiener et al (2) 1997 Cohort II SLC and JB 12 52 I 37.5 Foot scale score and nonunion
Zenios et al (19) 2005 RCT I SLC and B 12 54 NR 43.5 Foot scale score and nonunion
Gray et al (14) 2008 Cohort II SLC and BS 12 41 I 43 Foot scale score and nonunion
Shahid et al (17) 2013 Cohort II SLC and B 12 33 I 49.7 Foot scale score
Akimau et al (18) 2016 RCT I SLC and B 24 28 I 42 Foot scale score and nonunion

Abbreviations: B, bandage; BS, boot splint; JB, Jone’s bandage; NR, not reported; RCT, randomized controlled trial; SLC, short leg cast.

−3.89 (95% CI −6.30 to 01.49), indicating that the short leg casting group
had foot function scores that were significantly lower statistically than
those of the splinting group (Fig. 2). We found no evidence of publi-
cation bias using Egger’s test or contour funnel plot (coefficient = −0.72,
standard error = 1.23, p = .6).
Fracture Nonunion
Four studies reported nonunion of the metatarsal bone fracture in
the short leg casting and splinting groups (Table 4). The risk of frac-
ture nonunion in the short leg casting group (χ2 = 1.75, df = 3, p = .610,
I2 = 0%, Fig. 3) revealed a pooled relative risk of 1.57 (95% CI 0.29 to
8.49) compared with the splinting group, indicating that the risk of
fracture nonunion was greater by ~57% in the short leg casting group
compared with the splinting group. Neither the contour funnel nor
Egger’s test suggested evidence of a publication bias.
Table 3
Comparisons of mean differences between short leg cast and removable splint

Investigator Short Leg Cast Splint

Table 2
Patients (n) Mean ± SD Patients (n) Mean ± SD
Risk of bias assessment
Foot function scale score at
Variable Zenios Akimau Wiener Gray Shahid
last follow-up point*
et al (19) et al (18) et al (2) et al (14) et al (17)
Wiener et al (2) 30 86 ± 16 30 92 ± 16
Sequence generation Yes Yes NR NR NR Zenios et al (19) 25 80 ± 11.55 25 89.5 ± 13.56
Allocation concealment Yes No NR NR NR Gray et al (14) 20 87.5 ± 5.82 17 90.5 ± 5.2
Blinding No No NR NR NR Shahid et al (17) 23 93 ± 7 16 96 ± 7
Incomplete outcomes data No No NR NR NR Akimau et al (18) 24 93 ± 19.83 36 93 ± 23.69
Selective outcomes report Yes Yes NR NR NR Foot scale score within 1 mo†
Representativeness of cohort NR NR Yes Yes Yes Gray et al (14) 20 54.5 ± 2.98 17 71 ± 8.7
Ascertainment of exposure NR NR Yes Yes Yes Shahid et al (17) 23 29 ± 47 16 65 ± 47
Ascertainment of outcome NR NR Yes Yes Yes Akimau et al (18) 24 65 ± 18.21 36 72 ± 18.01
Adjustment for confounder NR NR No No No
Abbreviation: SD, standard deviation.
Follow-up duration NR NR Yes Yes Yes
* Unstandardized mean difference −3.89 (95% confidence interval 6.30 to −1.49).
Abbreviation: NR, not reported. † Unstandardized mean difference −14.58 (95% confidence interval −4.12 to −5.04).
 K. Pituckanotai et al. / The Journal of Foot & Ankle Surgery 57 (2018) 982–986 985

Fig. 2. Comparison of mean foot scale scores between short leg cast and removable splint groups.

Table 4 the chance of these results not actually having a difference is only 1%.
Comparisons of dichotomous outcomes between short leg cast and removable splint In contrast, the fracture nonunion rate between the 2 groups was not
Investigator Nonunion RR*† 95% CI significantly different statistically.
Short Leg Cast Splint The present study had several strengths. First, the present study
included 2 RCTs and 3 cohort studies in the pooling of relevant clin-
Yes No Yes No
ical outcomes (i.e., foot functional scores, union rates) for short leg
Wiener et al (2) 0 30 0 30 1.00 0.02 to 48.82 casting and splinting. Second, the possible causes of heterogeneity were
Zenios et al (19) 3 22 0 25 7.00 0.38 to 128.87
explored if covariate data at baseline (e.g., mean age, gender, level of
Gray et al (14) 0 17 1 19 0.39 0.02 to 8.97
Akimau et al (18) 0 24 0 36 1.48 0.29 to 72.17 evidence, mean follow-up time, and foot function scale scores at base-
line) were available. The publication bias for each outcome was also
Abbreviations: CI, confidence interval; RR, relative risk.
* Pooled RR 1.57 (95% CI 0.29 to 8.49).
assessed. The present study had some limitations. We did not pool
† p < .05 considered to indicate a statistically significant difference. important outcomes such as pain scores (VAS score for pain), com-
plications (compartment syndrome or pressure sores), and cost-
effective analysis because of insufficient data. Finally, too many
Discussion treatment variations were found within the first and second week after
the initial fracture, with some studies using casting or splinting or not
The results of the present review suggest that foot and ankle splint- reporting the modality in either group. Also, none of the studies re-
ing (boot splint or compression bandage) results in foot function scores ported the pain scores, functional scores, or complications within 2
of −3.89 (95% CI −6.30 to −1.49) and −14.58 (95% CI −24.12 to 5.04), a weeks after treatment.
statistically significant difference at the 1-month and last follow-up In conclusion, the foot function scores at the ≥1-month follow-up
points compared with the short leg casting group. Patients treated with point in the splinting group were greater than those in the short leg
short leg casting were approximately 1.6 times (95% CI 0.29 to 8.49) casting group. The former group also had lower nonunion rates. Further
more likely to experience fracture nonunion at the fifth metatarsal bone RCT research that assesses pain, function scores, and complications
compared with patients treated with splinting. Our meta-analysis of within 1 to 2 weeks after treatment is necessary to evaluate short leg
comparative studies, including 122 and 124 patients, had a type I error casting and splinting in the treatment of pseudo-Jones avulsion frac-
of 1% for detecting the 3.89-score and 14.58-score difference at the tures of the fifth metatarsal bone. Moreover, a cost-effective analysis
1-month and last follow-up points for short leg casting versus splint- should be performed.
ing for acute fracture of the fifth metatarsal bone. This suggests that
Supplementary Material

Supplementary material associated with this article can be found

in the online version at www.jfas.org (https://doi.org/10.1053/
j.jfas.2018.03.018).

References

Fig. 3. Comparison of risk of fracture nonunion of fifth metatarsal bone between short
leg cast and removable splint groups.
1. Jones R. I. Fracture of the base of the fifth metatarsal bone by indirect violence.
Ann Surg 35:697–700, 1902.
2. Wiener BD, Linder JF, Giattini JFG. Treatment of fractures of the fifth metatarsal:
a prospective study. Foot Ankle Int 18:267–269, 1997.
3. Anderson LD. Injuries of the forefoot. Clin Orthop Relat Res 122:18–27, 1977.
4. Dameron TB Jr. Fractures and anatomical variations of the proximal portion of the
fifth metatarsal. J Bone Joint Surg Am 57:788–792, 1975.
5. DeLee JC, Evans JP, Julian J. Stress fracture of the fifth metatarsal. Am J Sports Med
11:349–353, 1983.
6. Kavanaugh JH, Brower TD, Mann RV. The Jones fracture revisited. J Bone Joint Surg
Am 60:776–782, 1978.
7. Kitaoka HB. Salvage of nonunion following ankle arthrodesis for failed total ankle
arthroplasty. Clin Orthop Relat Res 268:37–43, 1991.
986 K. Pituckanotai et al. / The Journal of Foot & Ankle Surgery 57 (2018) 982–986
8. Peltier LF. Eponymic fractures: Robert Jones and Jones fracture. Surgery 71:522–526,
1972.
9. Richli WR, Rosenthal DI. Avulsion fracture of the fifth metatarsal: experimental study
of pathomechanics. AJR Am J Roentgenol 143:889–891, 1984.
10. Stewart IM. Jones fracture: fracture of base of fifth metatarsal. Clin Orthop
16:190–198, 1960.
11. Torg JS. Fractures of the base of the fifth metatarsal distal to the tuberosity.
Orthopedics 13:731–737, 1990.
12. Clapper MF, O’Brien TJ, Lyons PM. Fractures of the fifth metatarsal: analysis of a
fracture registry. Clin Orthop Relat Res 315:238–241, 1995.
13. Egol K, Walsh M, Rosenblatt K, Capla E, Koval KJ. Avulsion fractures of the fifth
metatarsal base: a prospective outcome study. Foot Ankle Int 28:581–583, 2007.
14. Gray AC, Rooney BP, Ingram R. A prospective comparison of two treatment options
for tuberosity fractures of the proximal fifth metatarsal. Foot (Edin) 18:156–158,
2008.
15. Holubec KD, Karlin JM, Scurran BL. Retrospective study of fifth metatarsal fractures.
J Am Podiatr Med Assoc 83:215–222, 1993.
16. Lawrence SJ, Botte MJ. Jones fractures and related fractures of the proximal fifth
metatarsal. Foot Ankle 14:358–365, 1993.
17. Shahid MK, Punwar S, Boulind C, Bannister G. Aircast walking boot and below-knee
walking cast for avulsion fractures of the base of the fifth metatarsal: a comparative
cohort study. Foot Ankle Int 34:75–79, 2013.
18. Akimau PI, Cawthron KL, Dakin WM, Chadwick C, Blundell CM, Davies MB.
Symptomatic treatment or cast immobilisation for avulsion fractures of the base
of the fifth metatarsal: a prospective, randomised, single-blinded non-inferiority
controlled trial. Bone Joint J 98:806–811, 2016.
19. Zenios M, Kim WY, Sampath J, Muddu BN. Functional treatment of acute metatarsal
fractures: a prospective randomised comparison of management in a cast versus
elasticated support bandage. Injury 36:832–835, 2005.
20. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, Clarke M,
Devereaux PJ, Kleijnen J, Moher D. The PRISMA statement for reporting systematic
reviews and meta-analyses of studies that evaluate health care interventions:
explanation and elaboration. PLoS Med 6:e1000100, 2009.
21. Wells GASB, O’Connell D, Peterson J, Welch V, Losos M, Tugwell P. The Newcastle-
Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-
analyses. Available at: http://www.ohri.ca/programs/clinical/epidemiology/
oxford.asp. Accessed November 29, 2016.
22. Palmer TM, Peter JL, Sutton AJ, Moreno SG. Contour-enhanced funnel plots for
meta-analysis. Stata J 8:242–254, 2008.
23. Peters JL, Sutton AJ, Jones DR, Abrams KR, Rushton L. Contour-enhanced meta-
analysis funnel plots help distinguish publication bias from other causes of
asymmetry. J Clin Epidemiol 61:991–996, 2008.
24. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected
by a simple, graphical test. BMJ 315:629, 1997.
25. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of testing
and adjusting for publication bias in meta-analysis. Biometrics 56:455–463, 2000.
26. StataCorp. Stata Statistical Software: Release 14, StataCorp LP, College Station, TX,
2015.