G Model

FAS 1336 No. of Pages 8

Foot and Ankle Surgery xxx (2019) xxx–xxx

Contents lists available at ScienceDirect

Foot and Ankle Surgery

journal homepage: [Link]/locate/fas

Outcome after total ankle arthroplasty with a minimum of five years

follow-up: A systematic review and meta-analysis
James Randolph Onggoa,* , Mithun Nambiara,b , Kevin Phanc , Benjamin Hickeya,d ,
Michael Galvina , Harvinder Bedia
a
Department of Orthopaedic Surgery, Box Hill Hospital, 8 Arnold Street, Box Hill, Melbourne, Victoria 3128, Australia
b
Department of Surgery, The University of Melbourne, Parkville, Victoria 3010, Australia
c
NeuroSpine Surgery Research Group, The Prince of Wales Private Hospital, 320-346 Barker St, Randwick, New South Wales 2031, Australia
d
Wrexham Maelor Hospital, Croesnewydd Rd, Wrexham, Wales LL13 7TD, UK

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

Article history: Background: Total ankle arthroplasty (TAA) is increasingly gaining recognition as an alternative to ankle
Received 17 March 2019 arthrodesis in the treatment of end-stage ankle arthritis. Despite high rates of adverse events during early
Received in revised form 13 July 2019 inception, newer generations of uncemented prosthesis and design modifications have improved
Accepted 17 July 2019
outcomes. Questions remain regarding the long-term outcomes and implant survivorship of TAA.
Available online xxx
Aim: This analysis aims to establish an updated review of intermediate and long-term clinical outcome
and complication profile of TAA.
Keywords:
Patients and methods: A multi database search was performed on 14th October 2018 according to PRISMA
Total ankle arthroplasty
Total ankle replacement
guidelines. All articles that involved patients undergoing uncemented TAA with 5 years minimum follow-
Clinical outcomes up, reported clinical outcome or complication profile of TAA were included. Seventeen observational
Complications studies were included in the review, with 1127 and 262 ankles in the 5 and 10 years minimum follow-up
Survivorship groups respectively.
Long-term Results: Mean difference between pre- and post-operative AOFAS score was 43.60 (95%CI: 37.51–49.69,
Intermediate p < 0.001) at 5 years minimum follow-up. At 5 years minimum follow-up, pooled proportion (PP) of
Follow-up prostheses revision for any reason other than polyethylene exchange was 0.122 (95%CI: 0.084–0.173), all
cause revision was 0.185 (95%CI: 0.131–0.256), unplanned reoperation was 0.288 (95%CI: 0.204–0.390)
and all infection was 0.033 (95%CI: 0.021–0.051). At 10 years minimum follow-up, PP of prostheses
revision for any reason other than polyethylene exchange was 0.202 (95%CI: 0.118–0.325), all cause
revision was 0.305 (95%CI: 0.191–0.448), unplanned reoperation was 0.422 (95%CI: 0.260–0.603) and all
infection was 0.029 (95%CI: 0.013–0.066).
Conclusion: Despite good intermediate and long-term functional outcome measures, TAA has relatively
higher revision surgery prevalence with longer follow-up periods. Further research should be directed
towards identifying patient populations that would best benefit from TAA and those at greatest risk of
requiring revision surgery.
© 2019 European Foot and Ankle Society. Published by Elsevier Ltd. All rights reserved.

1. Introduction TAA had high rates of subsidence, loosening and revision. However,
subsequent advances in implant design such as uncemented
Total ankle arthroplasty (TAA) was first performed in 1970 by implants as well as fixed and mobile bearing surfaces have resulted
Lord and Marrotte [1]. It was developed as an alternative treatment in improved outcomes [2–6].
option to the “gold-standard” at the time of ankle arthrodesis for A systematic review in 2007 demonstrated similar functional
end-stage ankle degenerative joint disease. The early attempts at outcomes and complication profiles between TAA and ankle
arthrodesis [4]. An advantage of TAA includes maintaining mobility
at the ankle joint, which can potentially allow for better functional
* Corresponding author. recovery [7–9], Furthermore, TAA has been found to decrease the
E-mail addresses: jamesonggo1993@[Link] (J.R. Onggo), radiographic incidence of adjacent joint degeneration that is
mithunnambiar1@[Link] (M. Nambiar), [Link]@[Link] (K. Phan), otherwise seen in ankle arthrodesis, by preserving motion at the
drhickey@[Link] (B. Hickey), [Link]@[Link] (M. Galvin),
hsbedi1@[Link] (H. Bedi).
ankle joint and minimising load through adjacent joints [10].

[Link]
1268-7731/© 2019 European Foot and Ankle Society. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: J.R. Onggo, et al., Outcome after total ankle arthroplasty with a minimum of five years follow-up: A systematic
review and meta-analysis, Foot Ankle Surg (2019), [Link]
G Model
FAS 1336 No. of Pages 8

2 J.R. Onggo et al. / Foot and Ankle Surgery xxx (2019) xxx–xxx

TAA can however be complicated by prosthesis failure, due to data were not available. All p-values were two-sided. R studio 3.4
infection, loosening or subsidence requiring revision of prosthesis and Review Manager (version 5.3, Copenhagen, The Nordic
or conversion to arthrodesis. Follow-up studies after TAA report Cochrane Centre, The Cochrane Collaboration, 2014) were used
rates of revision up to 49.4% [11–13]. However, a previous for statistical analysis.
systematic review [14] that included 1105 TAA with a minimum
follow-up of 2 years reported an overall failure rate of 10% at 5 3. Results
years with a wide range (0–32%). Unfortunately, no longer-term
data was presented. 3.1. Literature search
While TAA is fast becoming a suitable alternative to ankle
arthrodesis [15], questions remain regarding long-term functional A total of 878 studies were identified, of which 56 duplicates
outcomes, implant survivorship and complications. This review and 63 non-English language articles were excluded. Titles and
aims to provide an up-to-date analysis of the intermediate and abstracts of the remaining studies were screened in accordance to
long-term outcomes of TAA in terms of functional outcome and the pre-defined inclusion criteria. (Fig. 1) Seventeen studies
complication profile. [2,12,13,18–31] were included for meta-analysis (Table 2).

2. Methods 3.2. Demographics

2.1. Literature search strategy Our analysis included a total of 1127 cases with a minimum of 5
years follow-up and 262 cases with a minimum of 10 years follow-
Our study was conducted according to the PRISMA guidelines up. Demographic data is summarised in Table 1. The most
[16]. Electronic databases were searched from the date of inception commonly used implant was the Scandinavian Total Ankle
to 14th October 2018. The databases included Medline, PubMed, Arthroplasty (STAR, Stryker GmbH or Waldemar Link) prosthesis
EMBASE and Cochrane Database of Systematic Reviews. In order to used in 698 ankles. All implants in the included studies were
increase the sensitivity of our search, we combined terms in the uncemented. Follow-up periods in individual studies ranged
following manner, [(‘Ankle arthroplasty’ OR ‘Ankle replacement’) between a minimum of 5 and maximum of 22.8 years [26].
AND ‘Joint prosthesis’ AND (‘Treatment outcome’ OR ‘Long term’)],
as MeSH terms. The identified articles, as well as their references, 3.3. Clinical outcome scores
were reviewed according to the selection criteria for consideration
of inclusion in the study. The AOFAS and Kofoed were two of the most commonly used
scoring system to determine the magnitude of effect an interven-
2.2. Selection criteria tion has on the ankle [32]. The mean difference between the pre-
and post-operative AOFAS score was 43.60 (95%CI: 37.51–49.69,
From the literature search, we included all studies that: 1) p < 0.001) in the minimum 5 years group analysis. (Fig. 2) There
involved patients that underwent uncemented TAA, 2) all patients was insufficient raw data from individual studies to conduct a
who had a follow-up period of at least five years, 3) reported the reasonable analysis of the AOFAS score at 10 years minimum
implant revision rates or the pre- and post-operative functional follow-up and the Kofoed score at both 5 and 10 years minimum
outcome. Our study focused on uncemented prostheses in order to follow-up, with only two of the included studies reporting the
exclude results from older outdated cemented prostheses types, latter [21,27].
which may impact on results. Conference papers, case reports,
letters to the editor and abstracts were excluded. Only English 3.4. Survivorship
language articles were included in the study to avoid discrepancies
in translation. The literature defines implant survivorship in various ways. In
this review, survivorship was defined as either: (1) conversion to
2.3. Data extraction and critical appraisal arthrodesis or revision of any prosthetic components excluding liner
exchange in a mobile bearing TAA, and (2) conversion to arthrodesis
Articles from the literature search were independently or revision of any prosthetic components including liner exchange in
reviewed by two authors (JO, MN) with regard to suitability for a mobile bearing TAA. When using definition (1), the PP of revision at
inclusion in the study. The critical review checklist of the Dutch 5 and 10 years minimum follow-up was 0.122 (95%CI: 0.084–0.173)
Cochrane Centre proposed by MOOSE was used to evaluate studies and 0.202 (95%CI: 0.118–0.325) respectively (Fig. 3.1A and 1B). With
[17] (Table 1). Data was extracted from the article text, tables and definition (2), the PP of revision at 5 and 10 years minimum follow-up
graphs. Parameters included in the analysis were clinical outcome was 0.185 (95%CI: 0.131–0.256) and 0.305 (95%CI: 0.191–0.448)
measures, implant survivorship and complication rates. respectively. (Fig. 3.2A and 2B). The reported 5, 10 and 15 years
survivorship of implants are shown in Table 3.
2.4. Statistical analysis
3.5. Range of motion
A statistician (KP) performed a meta-analysis based on
weighted pooled proportions (PP) and weighted mean differences. The most commonly used parameter was the range of motion
Random-effects models were used, where it was assumed that (ROM) in the sagittal plane of the ankle. Only three individual
there were variations between studies. Chi squared tests were used studies reported the pre-operative and final follow-up ROM, but
to study heterogeneity between trials. I2 statistic was used to this was inadequate to conduct a reasonable meta-analysis for
estimate the percentage of total variation across studies, owing to ankle ROM [21–23].
heterogeneity rather than chance, with values greater than 50%
considered as substantial heterogeneity. I2 can be calculated as: 3.6. Complications
I2 = 100%  (Q df)/Q, with Q defined as Cochrane's heterogeneity
statistics and df defined as degree of freedom. Specific analyses Reported complications associated with TAA included hetero-
considering confounding factors were not possible because raw topic ossification, arthrofibrosis, adjacent joint osteoarthritis,

Please cite this article in press as: J.R. Onggo, et al., Outcome after total ankle arthroplasty with a minimum of five years follow-up: A systematic
review and meta-analysis, Foot Ankle Surg (2019), [Link]
 FAS 1336 No. of Pages 8
G Model
review and meta-analysis, Foot Ankle Surg (2019), [Link]
Please cite this article in press as: J.R. Onggo, et al., Outcome after total ankle arthroplasty with a minimum of five years follow-up: A systematic

Table 1
Patient demographics.

Author Publication Year of Country Manufacturer of Prosthesis Indications No. of No. of Mean age at index Mean BMI Mean follow-
year study prosthesis patients ankles TAA (years) (kg/m2) up (years)
Primary Secondary RA Others
OA OA
Bonnin 2011 1997– France Tornier Salto Total Ankle Arthroplasty 14a 55a 27a 0a 96 98 56  13 24 8.9 (6.8–11.1)
2000
Brunner 2013 1996– Switzerland Waldemar Link Scandinavian Total Ankle Arthroplasty 30 70 0 0 72 77 56.9  13.9 28.3  3.7 12.4 (10.8–
2000 14.9)
Caravaggi 2014 2004– Italy Finsbury BOX Ankle 10 0 0 0 10 10 55 25.7 5.0c

J.R. Onggo et al. / Foot and Ankle Surgery xxx (2019) xxx–xxx
2007 Orthopaedics
Daniels 2015 2001– Canada Waldemar Link Scandinavian Total Ankle Arthroplasty 16 68 22 5 98 111 61.9  11.7 27.9  5.5 9.0  1.0
2005
Di Iorio 2017 2003– France Biomet Ankle Evolutive System TAR 3 43 2 2 47 50 56  13 28  4 10.0 (9.0–13.0)
2006
b
Frigg 2017 1996– Switzerland Stryker Scandinavian Total Ankle Arthroplasty 6 39 5 0 46 50 58 26.1 14.6 (12.9–
2006 16.4)b
Giannini 2017 2003– Italy Finsbury BOX Ankle 7 60 4 4 73 75 62.5  13.2 – 6.5 (5.0–9.0)
2008 Orthopaedics
Jastifer 2015 1998– USA Stryker Scandinavian Total Ankle Arthroplasty 3 14 1 0 18 18 60.9 – 12.6 (10.2–
2003 14.6)
Karantana 2010 1999– UK Waldemar Link Scandinavian Total Ankle Arthroplasty – - - - 45 52 62 - 6.7 (5.0–9.2)
2002
Kerkhoff 2016 1999– Netherlands Waldemar Link Scandinavian Total Ankle Arthroplasty 34 43 56 1 124 134 59  12.5 - 7.5c
2008
Kofoed 2004 1990– Denmark Waldemar Link Scandinavian Total Ankle Arthroplasty 22 3 0 25 25 58 – 9.5  1.7
1995
Kraal 2013 1988– Netherlands DePuy, Endotec LCS mobile-bearing TAR, Buechel- 0 0 93 76 93 57.6 – 14.8 (10.7–
1999 Pappas mobile-bearing TAR 22.8)
Palanca 2018 1998– USA Waldemar Link Scandinavian Total Ankle Arthroplasty – – – – 24 24 73.7 – 15.7 (15.0–
2000 17.7)
San 2006 1990– USA Endotec Buechel-Pappas TAR 0 0 31 0 23 31 61 24.2 8.3 (5.0–12.0)
Giovanni 1997
Stewart 2017 2007– USA Integra Salto Talaris implant 7 62 2 1 72 72 61.9 29.6 6.8 (5.0–9,6)
2012
Wood 2000 – UK Waldemar Link Scandinavian Total Ankle Arthroplasty 0 0 7 0 6 7 62.9 – 5.4 (5.0–6.0)
Wood 2008 – UK Waldemar Link Scandinavian Total Ankle Arthroplasty 56 25 119 184 200 N.R. – 7.33 (5.0–13.0)

TAA = total ankle arthroplasty; Primary OA = no idenfiable aetiology; Secondary OA = any identifiable aetiology, including post trauma, infection etc.; ‘–’ = not specified.
a
Based on number of patients.
b
Median values.
c
Minimum values.

3
G Model
FAS 1336 No. of Pages 8

4 J.R. Onggo et al. / Foot and Ankle Surgery xxx (2019) xxx–xxx

Fig. 1. PRISMA flowchart.

Table 2
MOOSE evaluation table.

Articles Clear definition of Clear definition of outcomes Independent assessment Sufficient No selective loss Important confounders and
study population and outcome assessment of outcome parameters duration of during follow-up prognostic factors identified
follow-up
Bonnin Yes Yes Yes Yes Yes Yes
2011
Brunner Yes Yes Yes Yes Yes Yes
2013
Caravaggi Yes Yes Yes Yes Yes No
2014
Daniels Yes Inadequate No Yes Yes Inadequate
2015
Di Iorio Yes Inadequate No Yes Yes No
2017
Frigg 2017 Yes Yes No Yes Yes Inadequate
Giannini Yes Yes No Yes Yes No
2017
Jastifer Yes Yes Yes Yes Yes Yes
2015
Karantana No Yes Yes Yes Yes No
2010
Kerkhoff Yes Yes No Yes Yes Inadequate
2016
Kofoed Yes Yes Yes Yes Yes Inadequate
2004
Kraal 2013 Yes Inadequate No Yes Yes Yes
Palanca No Yes Yes Yes Yes Yes
2018
San Yes Yes No Yes Yes No
Giovanni 2006
Stewart Yes Yes No Yes Yes No
2017
Wood Yes Inadequate No Yes Yes No
2000
Wood Yes Yes No Yes Yes No
2008

prosthesis loosening, polyethylene liner fracture and wear, to theatre. The PP of unplanned reoperation for any reason was
prosthesis subsidence, periprosthetic fracture and valgus or varus 0.288 (95%CI: 0.204–0.390) and 0.422 (95%CI: 0.260–0.603) for
deformity. Other general operative complications include deep and minimum 5 and 10 years follow-up respectively. Specific to TAA,
superficial infection, deep vein thrombosis and unplanned return the most prevalent complication was heterotopic ossification with

Please cite this article in press as: J.R. Onggo, et al., Outcome after total ankle arthroplasty with a minimum of five years follow-up: A systematic
review and meta-analysis, Foot Ankle Surg (2019), [Link]
G Model
FAS 1336 No. of Pages 8

J.R. Onggo et al. / Foot and Ankle Surgery xxx (2019) xxx–xxx 5

Fig. 2. AOFAS score.

PP of 0.215 (95%CI: 0.072–0.490) and 0.204 (95%CI: 0.062–0.497) at long-term clinical efficacy of TAA. It is interesting to note that
5 and 10 years minimum follow-up, while the least common AOFAS score post TAA tends to gradually reduce over a longer
complication was polyethylene insert fracture (PP = 0.039, 95%CI: follow-up period as reported by Stewart et al. [30]. There are two
0.016–0.090) and polyethylene insert wear (PP = 0.054, 95%CI: likely causes for this gradual decrease in AOFAS scores. Firstly, long
0.016–0.166) at 5 and 10 years minimum follow-up. The prevalence term wear of the polyethylene liner with subsidence and loosening
of infection was low with a PP of 0.033 (95%CI: 0.021–0.051) and of prosthesis may lead to declining functional performance of the
0.029 (95%CI: 0.013–0.066) at 5 and 10 years minimum follow-up ankle joint. Furthermore, with the passage of time, issues related to
respectively. The prevalence of deep periprosthetic joint infections stability, pain or development of heterotopic ossification might
was equal at 5 (PP = 0.026, 95%CI: 0.016–0.045) and 10 (PP = 0.026, arise, leading to a decrease in clinical outcome score. However, a
95%CI: 0.010–0.063) years minimum follow-up. Meta-analysis of decrease in AOFAS score over longer period of follow-up was not
all complications for minimum 5 and 10 years follow-up is detailed universal. Kraal et al. [26] noted an increase in AOFAS scores over
in Appendices 1 and 2 respectively. the years, but this may be attributed to the high attrition rate for
various reasons, from an initial pool of 87 ankles at 1 year post-
4. Discussion operation, to only 17 ankles at 15 years follow-up.

To our knowledge, this is the most recent and up to date review 4.2. Range of motion
of the intermediate to long-term outcomes of TAA, aiming to
provide information to assist in clinical decision-making. A Post-operative ankle ROM in the saggital plane was similar to
previous analysis performed by Zaidi et al. [33] in 2013 included pre-operative values [21–23]. However, one should note that this
studies only up to December 2012. Eleven of 17 studies in our preservation of ROM following TAA is notably better than the loss
review have been published since then [12,18–23,25,26,28,30]. of ROM associated with ankle arthrodesis. Although there was no
Furthermore, the classification of follow-up periods used in Zaidi significant improvement in ankle ROM with TAA, the prosthesis
et al. [33] was based on the mean follow-up periods reported in still preserved the physiological movements and functional
individual studies, without taking into account the minimum conditions of the ankle and showed statistically significant
follow-up period. Hence, some of these studies also included improvement in spatio-temporal parameters such as gait speed,
short-term results in their reports despite the higher mean follow- cadence and cycle time (p < 0.05) [18]. This suggests improved
up study period. Our analysis has excluded studies, which do not postoperative mobility and potentially earlier functional rehabili-
have a minimum follow-up period of 5 years. More importantly, tation.
Zaidi et al. [33] also reported a substantially higher overall
survivorship of up to 89% at ten years as compared to the 69.5% 4.3. Survivorship
(including liner exchange) and 79.8% found in this present study.
This difference may be explained by the strict inclusion criterion of The issue of implant survivorship is of great importance in ankle
having a minimum of 5 and 10 years follow-up periods to calculate arthroplasty. Revision surgery entails additional risk of infection,
associated survival rates. Unlike the rates reported by Zaidi et al. morbidity and mortality, as well as further hospitalisation and
[33], which are estimates, the results in this study are probably rehabilitation. These contribute to becoming an economic burden
more reflective of the true implant survivorship. for the health system [36,37]. Although the 10 years survivorship
rates of hip and knee arthroplasties have been reported to be as
4.1. Clinical outcome high as 95% [38,39], the existing data on TAA survivorship varies
widely across different study groups, with registries reporting 10
Although the AOFAS score is not a validated scoring system, it is years survivorship as low as 62% (95%CI: 52–72%) [40], while
the most commonly used clinical outcome score in foot and ankle another study reported as high as 95% [27]. Registry data may not
surgery [32] since its first description in 1994 [34]. The AOFAS be an accurate comparison since registries are known to have
score has also been shown to have enough discriminatory capacity issues with follow-up and other potential confounders, such as
to assess the post-operative improvement in patients with TAA variable endpoints, which can impact on the reporting of
[35]. The mean difference in AOFAS score between pre-operative survivorship data [41]. Furthermore, it is noted that there is a
and at final follow-up was statistically significant and suggests the wide variation of revision rates with different country registry data

Please cite this article in press as: J.R. Onggo, et al., Outcome after total ankle arthroplasty with a minimum of five years follow-up: A systematic
review and meta-analysis, Foot Ankle Surg (2019), [Link]
G Model
FAS 1336 No. of Pages 8

6 J.R. Onggo et al. / Foot and Ankle Surgery xxx (2019) xxx–xxx

Fig. 3. Conversion to arthrodesis or revision of any component excluding liner exchange at (1A) 5 years and (1B) 10 years minimum follow-up; Conversion to arthrodesis or
revision of any component including polyethylene change at (2A) 5 years and (2B) 10 years follow-up.

due to heterogeneity of data collection methods and definitions of Register (n = 780) reporting at 69% at 10 years [45]. Hence, we have
revision cases. These include the Finnish Arthroplasty Register excluded any joint registries from our analysis to avoid these
(n = 515) reporting 5-years survivorship at 83% [42], the National issues.
Joint registry for England, Wales, Northern Ireland (n = 4687) The classification of revision by articles in our study was
reporting 91.3% at 7-years [43], the New Zealand Joint registry heterogeneous. This analysis has clearly defined the parameters as
(n = 1380) reporting 79.5% at 11-years [44] and the Swedish Ankle above to avoid any confusion. Some studies [21,23,28] also

Please cite this article in press as: J.R. Onggo, et al., Outcome after total ankle arthroplasty with a minimum of five years follow-up: A systematic
review and meta-analysis, Foot Ankle Surg (2019), [Link]
G Model
FAS 1336 No. of Pages 8

J.R. Onggo et al. / Foot and Ankle Surgery xxx (2019) xxx–xxx 7

Table 3
Implant survivorship.

Author Implant survivorship, excluding polyethylene liner Implant survivorship, including polyethylene liner

5 y survival (%) 10 y survival (%) 15 y survival (%) 5 y survival (%) 10 y survival (%) 15 y survival (%)
Bonnin – – – – 84.8 –
Brunner – 70.7 – – – –
Daniels – 88 – – – –
Di Iorio – 68 – – 68 –
Frigg 92 90 87 90 78 70
Giannini 97.3 – – 97.3 – –
Jastifer – 94.4 – – – –
Karantana – – – 90 – –
Kerkhoff – 78 – – – –
Kofoed 2004 100 95 – 100 95 –
Kraal – – – – 81 73
Palanca 96 90 73 – – –
San Giovanni 93.4 93.4 – 93.4 93.4 –
Stewart 95.8 – – – – –
Wood 2008 93.3 80.3 – 93.3 80.3 –

‘–’ = not specified.

reported the routine exchange of polyethylene liners when an Unfortunately, the contact stresses and loading on an area of the
arthrotomy was performed for any reason. These routine liner may cause wearing and fractures that could result in
exchanges will reduce the impact of polyethylene wear by persistent pain or failure of the TAA, thus requiring revision
upgrading the in situ liner to the latest generation of polyethylene surgery.
component and thus could lead to under reporting of polyethylene
wear as a complication [23]. It is important to note that there was 4.5. Limitations
an increase in revisions when polyethylene liner exchange was
included in analysis rather than when it was excluded (5 years: This study was a systematic review of prospective and
PP = 0.185 vs 0.122, 10 years: PP = 0.305 vs 0.202). This large retrospective studies, with heterogeneity in their indications for
increase in revision may be caused by polyethylene fracture, wear surgery, implants designs and surgical technique. We have
or infection, and raises questions around the quality and design of attempted to minimise these differences and maintain the
the polyethylene liner used and concerns about them being the relevance of this review by selecting only uncemented prosthesis
weakest link among the TAA components. It is thus prudent for designs.
future studies to investigate a more optimal design for the metallic The quality of this study was limited by the quality of some
and polyethylene components, as well as determining the included studies, which lacked prospective follow-up and random-
appropriate indications and patient selections for TAA surgery. isation. Hence, inherent biases such as recall, publication and
selection bias cannot be excluded.
4.4. Complication profile Factors such as technical skills, degree of hindfoot deformity
and the complexity of treated cases could not be addressed as these
While this review did not investigate the indications for were not reported by most studies. It was also not possible to
revision surgeries specifically, aseptic loosening has been de- accurately assess and conduct an analysis on the causes for revision
scribed to be the highest cause of revision in 6 of the 12 included because not all individual studies were specific in reporting the
studies that reported on the prevalence of aseptic loosening indications for each case. With the growing favour of TAA as
[12,21,23,25,26,31]. Time to revision indicated by aseptic loosening treatment of end-stage ankle arthritis, it is important to have
had a mean of 109 months and ranged between 35 and 200 months further studies, which directly compare the procedure with ankle
across three studies that reported the time to each revision arthrodesis, with regards to not only clinical outcomes but also a
procedure [12,23,26]. The mean implant survival time to revision cost-benefit analysis.
for any cause was between 48 and 86 months [12,19,20,27,31].
Some modifications that have helped to improve implant 5. Conclusion
survivorship have included use of uncemented prosthesis types,
fixed-bearing and semi-constrained systems and minimising bony TAA provides good intermediate and long-term functional
resection [3]. Nevertheless, despite the use of newer generations of outcome measures for end-stage ankle degenerative joint disease.
implants, studies included in this review still reports a loosening TAA has relatively higher revision surgery prevalence than hip and
PP of 0.081 (95%CI: 0.056–0.116) and 0.117 (95%CI: 0.081–0.165) in knee arthroplasties at longer follow-up periods. Given the
studies with a minimum of 5 and 10 years of follow-up increasing recognition of ankle arthroplasty as a suitable alterna-
respectively. With the evolution of newer uncemented implants, tive to ankle arthrodesis, studies should directly compare the
further long term surveillance and results review will be required procedure with ankle arthrodesis, further identifying patient
to assess the durability of the prosthesis. populations that would best benefit from TAA, as well as long term
While polyethylene liner fractures and wear was found to be the surveillance and results review to assess the long term sustain-
least common TAA related complication in our analysis, it still has a ability of uncemented prostheses.
role in accounting for revisions, which is notably higher when the
definition of revision includes exchange of liner. This is due to the Conflict of interest
functional significance of the polyethylene liner in the success of
TAA. The polyethylene liner between the tibial and talar metallic Each author certifies that he or she has no commercial
component provides a smooth articular contact surface. associations (e.g., consultancies, stock ownership, equity interest,

Please cite this article in press as: J.R. Onggo, et al., Outcome after total ankle arthroplasty with a minimum of five years follow-up: A systematic
review and meta-analysis, Foot Ankle Surg (2019), [Link]
G Model
FAS 1336 No. of Pages 8

8 J.R. Onggo et al. / Foot and Ankle Surgery xxx (2019) xxx–xxx

patent/licensing arrangements, etc.) that might pose a conflict of [22] Giannini S, Romagnoli M, Barbadoro P, Marcheggiani Muccioli GM, Cadossi
interest in connection with the submitted article. M, Grassi A, et al. Results at a minimum follow-up of 5 years of a ligaments-
compatible total ankle replacement design. Foot Ankle Surg: Official
Journal of the European Society of Foot and Ankle Surgeons 2017;23
Appendix A. Supplementary data (2):116–21.
[23] Jastifer JR, Coughlin MJ. Long-term follow-up of mobile bearing total ankle
arthroplasty in the United States. Foot Ankle Int 2015;36(2):143–50.
Supplementary data associated with this article can be found, in [24] Karantana A, Hobson S, Dhar S. The scandinavian total ankle replacement:
the online version, at [Link] survivorship at 5 and 8 years comparable to other series. Clin Orthop Relat Res
2010;468(4):951–7.
[25] Kerkhoff YR, Kosse NM, Metsaars WP, Louwerens JW. Long-term functional
References and radiographic outcome of a mobile bearing ankle prosthesis. Foot Ankle Int
2016;37(12):1292–302.
[1] Lord G, Marotte JH. Total ankle prosthesis. Technic and 1st results Apropos of [26] Kraal T, van der Heide HJ, van Poppel BJ, Fiocco M, Nelissen RG, Doets HC. Long-
12 cases. Revue de chirurgie orthopedique et reparatrice de l’appareil moteur term follow-up of mobile-bearing total ankle replacement in patients with
1973;59(2):139–51. inflammatory joint disease. Bone Joint J 2013;95-b(12:165–261.
[2] Bonnin M, Gaudot F, Laurent JR, Ellis S, Colombier JA, Judet T. The Salto total [27] Kofoed H. Scandinavian total ankle replacement (STAR). Clin Orthopaed Relat
ankle arthroplasty: survivorship and analysis of failures at 7 to 11 years. Clin Res 2004;424:73–9.
Orthop Relat Res 2011;469(1):225–36. [28] Palanca A, Mann RA, Mann JA, Haskell A. Scandinavian total ankle
[3] Gougoulias NE, Khanna A, Maffulli N. History and evolution in total ankle replacement: 15-year follow-up. Foot Ankle Int 2018;39(2):135–42.
arthroplasty. Brit Med Bull 2009;89(1):111–51. [29] San Giovanni TP, Keblish DJ, Thomas WH, Wilson MG. Eight-year results of a
[4] Haddad SL, Coetzee JC, Estok R, Fahrbach K, Banel D, Nalysnyk L. Intermediate and minimally constrained total ankle arthroplasty. Foot Ankle Int 2006;27
long-term outcomes of total ankle arthroplasty and ankle arthrodesis. A systematic (6):418–26.
review of the literature. J Bone Joint Sur Am Vol 2007;89(9):1899–905. [30] Stewart MG, Green CL, Adams Jr. SB, DeOrio JK, Easley ME, Nunley 2nd. JA.
[5] Mann JA, Mann RA, Horton E. STAR ankle: long-term results. Foot Ankle Int Midterm results of the Salto Talaris total ankle arthroplasty. Foot Ankle Int
2011;32(5):S473–84. 2017;38(11):1215–21.
[6] Saltzman CL, Mann RA, Ahrens JE, Amendola A, Anderson RB, Berlet GC, et al. [31] Wood PL, Prem H, Sutton C. Total ankle replacement: medium-term results in
Prospective controlled trial of STAR total ankle replacement versus ankle 200 Scandinavian total ankle replacements. J Bone Joint Surg Brit Vol 2008;90
fusion: initial results. Foot Ankle Int 2009;30(7):579–96. (5):605–9.
[7] Piriou P, Culpan P, Mullins M, Cardon JN, Pozzi D, Judet T. Ankle replacement versus [32] Naal FD, Impellizzeri FM, Rippstein PF. Which are the most frequently used
arthrodesis: a comparative gait analysis study. Foot Ankle Int 2008;29(1):3–9. outcome instruments in studies on total ankle arthroplasty? Clin Orthop Relat
[8] Bonnin MP, Laurent JR, Casillas M. Ankle function and sports activity after total Res 2010;468(3):815–26.
ankle arthroplasty. Foot Ankle Int 2009;30(10):933–44. [33] Zaidi R, Cro S, Gurusamy K, Siva N, Macgregor A, Henricson A, et al. The
[9] Naal FD, Impellizzeri FM, Loibl M, Huber M, Rippstein PF. Habitual physical outcome of total ankle replacement: a systematic review and meta-analysis.
activity and sports participation after total ankle arthroplasty. Am J Sports Med Bone Joint J 2013;95-b(11:1500–7.
2009;37(1):95–102. [34] Kitaoka HB, Alexander IJ, Adelaar RS, Nunley JA, Myerson MS, Sanders M.
[10] SooHoo NF, Zingmond DS, Ko CY. Comparison of reoperation rates following Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes.
ankle arthrodesis and total ankle arthroplasty. J Bone Joint Surg Am Vol Foot Ankle Int 1994;15(7):349–53.
2007;89(10):2143–9. [35] Pena F, Agel J, Coetzee JC. Comparison of the MFA to the AOFAS outcome tool in
[11] Hopgood P, Kumar R, Wood PL. Ankle arthrodesis for failed total ankle a population undergoing total ankle replacement. Foot Ankle Int 2007;28
replacement. J Bone Joint Surg Brit Vol 2006;88(8):1032–8. (7):788–93.
[12] Brunner S, Barg A, Knupp M, Zwicky L, Kapron AL, Valderrabano V, et al. The [36] Vanhegan IS, Malik AK, Jayakumar P, Ul-Islam S, Haddad FS. A financial analysis
Scandinavian total ankle replacement: long-term, eleven to fifteen-year, of revision hip arthroplasty: the economic burden in relation to the national
survivorship analysis of the prosthesis in seventy-two consecutive patients. J tariff. J Bone Joint Surg Brit Vol 2012;94(5):619–23.
Bone Joint Surg Am Vol 2013;95(8):711–8. [37] Lavernia CJ, Drakeford MK, Tsao AK, Gittelsohn A, Krackow KA, Hungerford DS.
[13] Wood PLR, Clough TM, Jari S. Clinical comparison of two total ankle Revision and primary hip and knee arthroplasty. A cost analysis. Clin Orthop
replacements. Foot Ankle Int 2000;21(7):546–50. Relat Res 1995;311:136–41.
[14] Gougoulias N, Khanna A, Maffulli N. How successful are current ankle [38] Hopley CD, Crossett LS, Chen AF. Long-term clinical outcomes and survivorship
replacements?.: a systematic review of the literature. Clin Orthop Relat Res after total knee arthroplasty using a rotating platform knee prosthesis: a meta-
2010;468(1):199–208. analysis. J Arthroplasty 2013;28(1)68–77 e1–3.
[15] Singh JA, Ramachandran R. Time trends in total ankle arthroplasty in the USA: [39] Kandala NB, Connock M, Pulikottil-Jacob R, Sutcliffe P, Crowther MJ, Grove A,
a study of the National Inpatient Sample. Clin Rheumatol 2016;35(1):239–45. et al. Setting benchmark revision rates for total hip replacement: analysis of
[16] Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The registry evidence. BMJ (Clinical research ed) 2015;350:h756.
PRISMA statement for reporting systematic reviews and meta-analyses of [40] Henricson A, Skoog A, Carlsson A. The Swedish Ankle Arthroplasty Register: an
studies that evaluate healthcare interventions: explanation and elaboration. analysis of 531 arthroplasties between 1993 and 2005. Acta Orthop 2007;78
BMJ (Clinical research ed) 2009;339:b2700. (5):569–74.
[17] Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, et al. Meta- [41] D’Ambrosi R, Banfi G, Usuelli FG. Total ankle arthroplasty and national
analysis of observational studies in epidemiology: a proposal for reporting. registers: what is the impact on scientific production? Foot Ankle Surg: Official
Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. Journal of the European Society of Foot and Ankle Surgeons 2018.
JAMA 2000;283(15):2008–12. [42] Skyttä ET, Koivu H, Eskelinen A, Ikävalko M, Paavolainen P, Remes V. Total ankle
[18] Caravaggi P, Lullini G, Leardini A, Berti L, Vannini F, Giannini S. Functional and replacement: a population-based study of 515 cases from the Finnish
clinical evaluation at 5-year follow-up of a three-component prosthesis and Arthroplasty Register. Acta Orthop 2010;81(1):114–8.
osteochondral allograft transplantation for total ankle replacement. Clin [43] National Joint Registry for England. Wales and Northern Ireland. NJR 15th
Biomech (Bristol, Avon) 2015;30(1):59–65. annual report [Internet]. 2018 Available from: [Link]
[19] Daniels TR, Mayich DJ, Penner MJ. Intermediate to long-term outcomes of total content/uploads/2018/11/NJR-15th-Annual-Report-pdf.
ankle replacement with the Scandinavian total ankle replacement (STAR). J [44] The New Zeland Joint Registry. Eighteen year report January 1999 to December
Bone Joint Surg Am Vol 2015;97(11):895–903. 2016 [Internet]. Available from: [Link]
[20] Di Iorio A, Viste A, Fessy MH, Besse JL. The AES total ankle arthroplasty analysis DH7827_NZJR__Report_v4_26Oct17.pdf.
of failures and survivorship at ten years. Int Orthop 2017;41(12):2525–33. [45] Henricson A, Nilsson JA, Carlsson A. 10-year survival of total ankle
[21] Frigg A, Germann U, Huber M, Horisberger M. Survival of the Scandinavian arthroplasties: a report on 780 cases from the Swedish Ankle Register. Acta
total ankle replacement (STAR): results of ten to nineteen years follow-up. Int Orthop. 2011;82(6):655–9.
Orthop 2017;41(10):2075–82.

Please cite this article in press as: J.R. Onggo, et al., Outcome after total ankle arthroplasty with a minimum of five years follow-up: A systematic
review and meta-analysis, Foot Ankle Surg (2019), [Link]