Abstract
-
The aim of this study was to update the scientific evidence for ankle fracture prognosis by addressing radiographic osteoarthritis, time course and prognostic factors.
-
A systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Studies were included if they were randomized controlled trials, controlled trials or observational studies, including case series and case-control studies investigating radiologically confirmed osteoarthritis in adults with a classified ankle fracture, treated with or without surgery, with a minimum follow-up of 1 year. Also included were studies examining prognostic factors predicting radiologically confirmed osteoarthritis. Tibial plafond and talus fractures were excluded.
-
Thirty-four studies were included examining 3447 patients. Extracted data included study type, inclusion and exclusion criteria, age, number of patients, number of fractures according to the author-reported classification method, radiological osteoarthritis, follow-up period, prognostic factors, and treatment.
-
Severe heterogeneity was visible in the analyses (I2 > 90%), reflecting clinical heterogeneity possibly arising from the presence of osteoarthritis at baseline, the classifications used for the fractures and for osteoarthritis.
-
The incidence of osteoarthritis was 25% (95% CI: 18–32) and 34% (95% CI: 23–45) for more severe fractures with involvement of the posterior malleolus.
-
The severity of the trauma, as reflected by the fracture classification, was the most important prognostic factor for the development of radiographic osteoarthritis, but there is also a risk with simpler injuries.
-
The period within which osteoarthritis develops or becomes symptomatic with an indication for treatment could not be specified.
Introduction
Risks for the future are a source of concern for any trauma patient and are included in the settlement of personal injury claims. In the case of ankle fractures, this mainly concerns the risk of osteoarthritis and the resulting treatments in the years after the injury. Medical advisers and independent experts often consult Medicolegal Reporting in Orthopedic Trauma by Foy & Fagg (1). In this reference text, the incidence of osteoarthritis after ankle fractures is determined at 20–40%, depending on the extent of the fracture. In addition, it is stated that medical assessments are preferably postponed until 18–24 months after the trauma: if osteoarthritis starts to develop, it should already be visible radiologically. The scientific evidence for this statement is however only based on older publications by Willenegger (1961) and Lindsjø (1981) which allegedly found that most osteoarthritis occurs in the first 12–18 months (2, 3). In fact, the last author found no (statistical) difference in the prevalence of osteoarthritis in 161 patients between their examination at 18 months and at 4 years since the accident. Furthermore, Foy & Fagg (1) refer to a study by Horisberger et al. (2009) who found a latent period between accident and end-stage osteoarthritis (arthrodesis or prosthesis) of 1–52 years (4). Because of this confusing lack of clear and recent evidence for the prognosis of ankle fractures, we performed a systematic review addressing the following questions:
-
What is the incidence of radiographic osteoarthritis in adults after an ankle fracture?
-
How long does the risk of radiological osteoarthritis persist since the accident?
-
Which prognostic factors are related to the development of radiological osteoarthritis after an ankle fracture?
Methods
This report has been prepared in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (5). The protocol of this systematic review was prospectively registered in the International Prospective Registry of Systematic Reviews (PROSPERO) under number CRD42021255718.
Selection criteria
Studies were included if they were randomized controlled trials, controlled trials or observational studies, including case series and case-control studies investigating radiologically confirmed osteoarthritis in adults with an ankle fracture. All types of fractures were included, except for pilon fractures of the tibial plafond and talus fractures. The study had to describe which fracture was involved by a classification such as Weber, Lauge Hansen and AO/OTA (6, 7, 8) or otherwise radiological features of the fracture from which a classification could be derived. Also included were studies examining prognostic factors for radiologically confirmed osteoarthritis.
Other criteria for inclusion were a minimum mean follow-up of 1 year, osteoarthritis had to be radiologically confirmed and assessed by a systematic method, such as Van Dijk et al. or Kellgren and Lawrence (9, 10). The treatment the patient received (surgical or conservative) did not affect inclusion.
Case reports or studies on cadavers were excluded. Also excluded were studies with a combination of fractures including pilon fractures, the results of which were not reported separately.
Outcome measures
Osteoarthritis has been defined as the radiological narrowing of the joint space. The results were derived from the method used in the article.
Search method
Cochrane Central Register of Controlled Trials (CENTRAL), Medline, Embase and Cumulative Index to Nursing & Allied Health Literature (CINAHL) were searched on August 16, 2020. The search strategy for the topics of ankle fractures and osteoarthritis consisted of a combination of free text words, so-called keywords and MeSH terms combined with a filter for systematic review, randomized controlled trials and observational studies. The systematic review search was added to rule out the possibility of previous research similar to ours.
Study selection was performed by two investigators (B S, A v E) independently of each other. The selection was made in Covidence (9). Title and summary were first assessed. Potentially relevant references were searched for as a full article and reviewed again. Discrepancies were discussed at consensus meetings.
Data extraction
Predefined data were extracted by a first investigator (A v E) in Covidence and checked by a second investigator (B S). The following descriptive data were collected: study type, inclusion and exclusion criteria, age, number of patients, follow-up period, prognostic factors, treatment and number of fractures according to the author-reported classification method.
Different fracture classifications were used in the selected studies, most often the Weber classification (6). To facilitate analysis, we translated other fracture classifications such as Lauge Hansen (7) and AO/OTA (8) as much as possible to a Weber classification. Adjustments were made to the Weber B and Weber C categories to reflect the involvement of the posterior malleolus, and thus the severity of the trauma. This resulted in the Weber A, Weber B, Weber B+, Weber C and Weber C+ categories (+ meaning posterior malleolar fracture). For each study, the number of patients with radiological osteoarthritis (i.e. joint space narrowing) was reported.
For the third research question, prognostic factors were extracted that were significantly associated with radiological osteoarthritis.
Risk of bias
Assessment of the risk of bias is an integral part of any systematic review. However, it was decided not to formally assess the risk of bias per study in this review. The instrument described by Hoy et al. 2012 is suitable for systematic reviews examining incidence data (10). Items from this instrument are either not applicable to this review or are part of the selection criteria that results in the item being scored as low in each study (Table 1).
Risk of bias instrument according to Hoy et al. 2012 (10).
| Item | Explanation |
|---|---|
| 1. Was the study target population a good representation of the national population in relation to relevant variables? | Not applicable |
| 2. Was the sample a true or accurate representation of the target population? | Not applicable |
| 3. Was some form of random selection used to select the sample OR was a count performed? | Not applicable |
| 4. Was the probability of non-response bias minimal? | Not applicable |
| 5. Was the data collected directly from the subjects (as opposed to a proxy)? | Part of selection criteria |
| 6. Was an acceptable case definition used in the study? | Part of selection criteria |
| 7. Was the research instrument that measured the parameter in question valid and reliable? | Part of selection criteria |
| 8. Was the same data collection method used for all subjects? | Part of selection criteria |
| 9. Was the length of the shortest prevalence period appropriate for the parameter in question? | Not applicable |
| 10. Were the numerator(s) and denominator(s) for the parameter of interest appropriate? | Part of selection criteria |
Statistical analysis
The incidence of osteoarthritis was assessed for every study (number of patients with osteoarthritis/total study population). A meta-analysis of incidence data of all fractures was performed according to a random effects model. Heterogeneity was examined by visual inspection of the forest plots and calculated with inconsistency statistics (I2).
The relationship between incidence and mean follow-up duration was investigated using meta-regression. When no mean follow-up period was indicated in the study, but a range, the arithmetic mean was arbitrarily taken. When only the median was presented, this was used as the mean follow-up.
In addition to the meta-regression, subgroup analyses were performed in which studies with a follow-up of up to 2 years were analyzed separately from the studies with a follow-up longer than 2 years.
A meta-regression would also have been performed to investigate the prognosis per fracture type. However, insufficient data were available to perform an analysis per Weber classification. Therefore, Weber A, B and/or C fractures not involving the posterior malleolus were combined in comparison with fractures involving the posterior malleolus; Weber B+ and C+. A subgroup analysis was also performed for this. Finally, a subgroup analysis was performed, in which the fracture types were compared depending on the follow-up. All analyses were performed in Stata (11). No sensitivity analysis was planned.
Quality of evidence
The evidence of incidence data and prognostic factors (follow-up and fracture type) was examined according to Grading Recommendations Assessment, Development and Evaluation (GRADE) methodology (12).
Results
Study selection
The search identified 2224 references, of which 129 were assessed as a full-text article and 34 studies were included examining 3447 patients (Fig. 1) (13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47). The reasons for exclusion of 95 studies have been recorded. Most studies have been excluded because osteoarthritis had not been systematically studied (wrong outcome).
Characteristics of the included studies
The included studies were published between 1975 and 2020. Three studies had a randomized controlled study design. Other studies had an observational retrospective design. Follow-up ranged from 12 to 348 months with a mean of 81 (s.d. 75) months (Table 2).
Characteristics of the included studies.
| Reference | Country | Follow-up⁑ | Follow-up in months‡ | Fracture classification according to Weber | Patients, n | Method to measure osteoarthritis | Osteoarthritis patients, n (%) |
|---|---|---|---|---|---|---|---|
| Abarquero-Diezhandino et al. (18) | Spain | 33 (12–73) mo | 33 | Weber B+ and C+ | 45 | Takakura | 8 (18) |
| Van Dijk | 7 (16) | ||||||
| Bauer et al. (20) | Sweden | 29 years | 348 | Weber A | 15 | Magnusson | 1 (7) |
| Weber B | 61 | Magnusson | 4 (7) | ||||
| Weber C | 11 | Magnusson | 1 (9) | ||||
| Weber B+ and C+ | 56 | Magnusson | 20 (36) | ||||
| Beris et al. (17) | Greece | 8.6 (2–12) years | 103.2 | Weber B and C | 64 | Unspecified* | 5 (8) |
| Weber B+ and C+ | 66 | Unspecified | 12 (18) | ||||
| Bois et al. (19) | Canada | 9.4 years | 112.8 | Weber B+ and C+ | 17 | Kellgren and Lawrence | 8 (67) |
| Day et al. (21) | Australia | 10–14 | 120 | Weber B, C, B+ and C+ | 153 | Philips | 19 (76) |
| Donken et al. (16) | Netherlands | 20 (17–24) years† | 240 | Isolated posterior malleolus fracture | 23 | Unspecified | 1 (5) |
| van Hooff et al. (45) | Netherlands | 6.9 (2.5–15.9) y | 82.8 | Weber B+ and C+ | 131 | Kellgren and Lawrence | 25 (19) |
| Federici et al. (39) | Italy | 6.2 years | 74.4 | Weber A, C, B+ and C+ | 145 | Magnusson | 22 (28) |
| Heim et al. (35) | Switzerland | 7.3 (3.9 –10.9) years | 87.6 | Weber B, C, B+ and C+ | 56 | Modified Knirk and Jupiter | 3 (7) |
| Hoelsbrekken et al. (36) | Norway | 44 (24–72) mo; 41 (24–67) months | 39 | Weber B, C, B+ and C+ | 82 | Magnusson | 4 (5) |
| Hoiness et al. (14) | Norway | 3.7 ± 0.6 years (2.8–5.2) | 44.4 | Weber A, B, C, B+ and C+ | 130 | Magnusson | 23 (26) |
| Kohake et al. (38) | Germany | 6.6 (2–12 ) years | 79.2 | Weber B and B+ | 61 | Kellgren and Lawrence | 16 (28) |
| Kortekangas et al. (43) | Finland | 4 years | 48 | Weber B and B+ | 48 | Kellgren and Lawrence | 3 (6) |
| Lambers et al. (40) | Netherlands | 21 (9–35) years | 252 | Weber C and C+ | 117 | Takakura | 17 (35) |
| Van Dijk | 9 (18) | ||||||
| Lange et al. (41) | Germany | 2–9 years | 24 | Unknown | 7 | Unspecified | 6 (86) |
| Weber A | 4 | Unspecified | 1 (25) | ||||
| Weber B | 14 | Unspecified | 8 (57) | ||||
| Weber C | 7 | Unspecified | 5 (71) | ||||
| Weber B+ and C+ | 66 | Unspecified | 49 (74) | ||||
| Lindsjø (22) | Sweden | 2–6 years | 24 | Weber A | 26 | Unspecified | 4 (15) |
| Weber B and B+ | 229 | Unspecified | 27 (12) | ||||
| Weber C and C+ | 66 | Unspecified | 22 (33) | ||||
| Lübbeke et al. (13) | Switzerland | 12–22 years | 144 | Weber A, B, C, B+ and C+ | 102 | Kellgren and Lawrence | 56 (55) |
| May et al. (42) | Turkey | 24.7 ± 12.0; 24.6 ± 9.1 | 24.6 | Weber B, C, B+ and C+ | 48 | Kellgren and Lawrence | 1 (2) |
| McDaniel & Wilson (44) | United States | 42 (12–180) months | 42 | Weber B+ and C+ | 51 | Unspecified | 32 (63) |
| Meijer et al. (33) | Netherlands | 12.5–39.4 years | 150 | Weber A, B, C, B+ and C+ | 104 | Van Dijk | 23 (22) |
| Olerud & Molander (15) | Sweden | 18–54 months | 18 | Weber B+ and C+ | 127 | Unspecified | 41 (32) |
| Regan et al. (34) | United States | 11.6 (8.4–14.6) years | 139.2 | Weber A, B, C, B+ and C+ | 141 | Van Dijk | 26 (32) |
| Shih et al. (29) | Taiwan | 1 year | 12 | Weber B+ and C+ | 72 | Modified Kellgren–Lawrence | 42 (58) |
| Stadler et al. (26) | Switzerland | 3–4 years | 36 | Weber A, B, C, B+ and C+ | 216 | Unspecified | 75 (45) |
| Stufkens et al. (46) | Switzerland | 12.9 years | 153 | Weber A, B, C, B+ and C+ | 109 | AOFAS score | 43 (39) |
| Sukur et al. (28) | Turkey | 17.1 (12–24) months | 17.1 | Weber B+ and C+ | 14 | Van Dijk | 4 (29) |
| Tosun et al. (23) | Turkey | 15 (12–51) months | 15 | Weber B+ and C+ | 64 | Van Dijk | 16 (33) |
| Veen & Zuurmond (31) | Netherlands | >6.5 years | 78 | Weber B, C, B+ and C+ | 59 | Van Dijk | 10 (17) |
| Verhage et al. (24) | Netherlands | 9.6 (5––17) years | 115.2 | Weber A, B, C | 164 | AOS | 5 (3) |
| Weber B+ and C+ | 79 | AOS | 13 (16) | ||||
| Verhage et al. (37) | Netherlands | 6.3 (2.4–15.9) years | 75.6 | Weber B+ and C+ | 327 | Domic | 26 (16) |
| Weigelt et al. (32) | Switzerland | 7.9 (3.1–12.3) years | 94.8 | Weber B+ and C+ | 32 | Van Dijk | 18 (56) |
| Xu et al. (30) | China | 55.7 (2.4–109) months | 55.7 | Weber A, B, C, B+ and C+ | 235 | Van Dyk | 6 (3) |
| Zhou et al. (27) | China | 32.75 (24–44) months | 33 | Weber B+ and C+ | 34 | Unspecified | 0 (3) |
| Zidek et al. (25) | Czech Republic | 1 year | 12 | Weber A, B, C, B+ and C+ | 38 | Kellgren and Lawrence | 11 (29) |
*Unspecified, authors focused on joint space narrowing but did not refer to a formal classification; †value is median; ⁑average value as indicated in study; ‡values as used in the analysis.
Question 1. Incidence of osteoarthritis after an ankle fracture
In a meta-analysis of all fractures from the selected studies regardless of the fracture type, the incidence of osteoarthritis was 25% (95% CI: 18–32) (Table 3).
Results of the meta-analyses on the incidence of radiological osteoarthritis.
| % Osteoarthritis | 95% CI | I2 | Number of studies included in analysis | |
|---|---|---|---|---|
| All fractures | 25 | 18–32 | 94% | 32* |
| Weber ABC | 13 | 3–28 | 90% | 5 |
| Weber B+C+ | 34 | 23–45 | 92% | 15 |
*The number of included studies in the ‘all fractures’ category exceeds the sum of Weber ABC and Weber B+C+ because for this analysis we also included fractures that were not categorized in the original study.
The weight of evidence was lowered by two levels due to the severe heterogeneity and came out as low. Heterogeneity could not be sufficiently reduced by adjusting for follow-up and type of fracture. The GRADE profile is described in Table 4.
Quality of the evidence: GRADE Evidence profile. Incidence of radiologically confirmed osteoarthritis (follow up: range 12–348 months).
| Certainty assessment | |
| Participants, n | 2624 |
| Studies, n | 32 |
| Risk of bias | Not serious |
| Inconsistency | Very seriousa |
| Circumstantial evidence | Not serious |
| Inaccuracy | Not serious |
| Publication bias | Not found |
| Overall certainty of evidence | ⨁⨁ ◯◯: Low |
| Summary results | |
| Number of events, % (95% CI) | 25 (18–32) |
aThere was a very high degree of heterogeneity both clinically and statistically (I2 > 90%).
Question 2. Follow-up moment
When the results are sorted by follow-up time, both studies with a short and long follow-up showed a high incidence of osteoarthritis (Fig. 2). A regression analysis between the incidence of osteoarthritis and follow-up in months gave a nonstatistically significant regression coefficient of −0.0002 (95% CI: −0.0012 to 0.0; P value = 0.66). This means that no relationship could be demonstrated between the incidence of osteoarthritis and the duration of follow-up. Therefore, a subgroup analysis did not seem useful.
Forest plot of incidence of radiological osteoarthritis sorted from short follow-up (top) to long follow-up (bottom).
Citation: EFORT Open Reviews 7, 10; 10.1530/EOR-22-0065
Question 3. Prognostic factors
There was insufficient data to perform separate subgroup analyses for Weber A, Weber B and Weber C fractures only. Subgroup analyses indicated that the incidence of osteoarthritis for Weber B+C+ was higher compared to Weber ABC without posterior malleolus involvement (Table 3). However, a regression analysis did not confirm a statistically significant relationship (ß = 0.18; 95% CI: −0.06 to 0.42; P = 0.14).
In the studies selected by us, insufficient homogeneous data were available for a (multiple) regression analysis of prognostic factors for the development of radiological osteoarthritis. In some of the studies, however, the authors had done their own research (Table 5).
Overview of prognostic factors mentioned by the authors in the study.
| Reference | Prognostic factor |
|---|---|
| Abarquero-Diezhandino et al. (18) | Malleolus posterior fracture |
| Beris et al. (17) | Bi/trimalleolar , insufficient posterior malleolus reposition |
| Heim et al. (35) | Malleolus posterior fracture, cartilage injury |
| Lindsjø (22) | Sex (F), fracture type (posterior malleolus), reposition |
| McDaniel & Wilson (44) | Unrepaired posterior malleolus fracture > 25% |
| Regan et al. (34) | Luxation fracture after trauma |
| Verhage et al. (24) | Trimalleolar fracture (especially with medial malleolus fracture) |
| Verhage et al. (37) | Age, insufficient reposition of posterior malleolus fracture |
| Lübbeke et al. (13) | Age > 30 years, BMI, medial malleolus fracture, Weber C, time since trauma |
| Stufkens et al. (46) | Osteochondral lesions |
Discussion
Our research focused on three aspects of post-traumatic radiographic ankle osteoarthritis, namely incidence, time course and prognostic factors.
The incidence of osteoarthritis defined as joint space narrowing was 25% (95% CI: 18–32) for all fractures and 34% (95% CI: 23–45) for more severe fractures. Thus, trauma severity was associated with a higher incidence of radiographic osteoarthritis, which is not unexpected based on the pathogenesis, and is consistent with the prognostic factors found in some of our selected studies. However, osteoarthritis after a simpler ankle fracture cannot be ruled out. Stufkens et al. (2010) (46) pointed out the importance of osteochondral injuries such as can occur in simple Weber A fracture. These are usually not visible on the accident radiographs but are visible with arthroscopy, and in his research, they formed an independent predictive factor for the development of post-traumatic osteoarthritis.
The strength of our study is that it is the first comprehensive systematic review of the incidence of post-traumatic radiologically confirmed ankle osteoarthritis.
However, the study also has several limitations. An important part of the current research was the classification of ankle fractures. Different fracture classifications were used in the selected studies. The interobserver variation in these classifications is considerable. Fonseca et al. (2018) (48) found a Kappa index of only 0.49 for the Danis–Weber classification, 0.32 for Lauge-Hansen and 0.38 for AO/OTA. An originally incorrect classified fracture might also incorrectly be classified in our translation to a Weber classification.
Various well-known and less common methods for staging radiological osteoarthritis were used in our selected studies. They lack consensus regarding the different radiological characteristics of osteoarthritis. For example, subchondral sclerosis is considered ‘normal’ in Van Dijk's classification (49), while according to Kellgren and Lawrence (50), this is classified as grade III osteoarthritis. In addition, low-to-moderate interobserver agreements have been found for the most used classifications of Van Dijk, Takakura and Kellgren and Lawrence (51, 52). Therefore, we have chosen radiographic joint space narrowing as the most uniform and useful outcome parameter in our study. But even this parameter could have been dependent on the technique of the radiological examination: in most studies, no details are given about weight-bearing or not, and certainly before 2000 it was not common to take weight-bearing radiographs.
Two studies reported pre-existing osteoarthritis on accident radiographs (29, 43). We have corrected for this. The question remains, of course, whether there was actually no pre-existing osteoarthritis in all those studies in which this was not reported.
An important part in the settlement of the medical aspects of personal injury claims is the prognosis: how long does the risk of developing radiological osteoarthritis and its consequences persist. Our study could not confirm the position of Foy & Fagg that possible post-traumatic osteoarthritis becomes visible within 2 years (1). This position is only based on the thesis of Lindsjø (1981) who found no statistically significant difference in the prevalence of radiological osteoarthritis between the study after 18 months (20.5%) and after 4 years (25.5%) since the accident (3). We have not been able to find a comparable study in the past 40 years. The number of studies with a follow-up of 2 years or less in our review was found to be limited to 4, and in addition, exceptionally high incidences were reported. There was even a study with 58% joint space narrowing after 12 months (29). That is unlikely and must be related to the radiological assessment.
Osteoarthritis can lead to complaints and medical treatment. Lübbeke et al. (2012) (13) found based on a hospital registry with a clinical and radiological follow-up of 12–22 years postoperatively that the time since the accident was a continuous risk factor for the development of symptomatic advanced radiological (Kellgren and Lawrence grades 3 and 4) osteoarthritis; longer follow-up was thus associated with more severe osteoarthritis. Ankle osteoarthritis can lead to an end-stage with an indication for an arthrodesis or prosthesis. Papa and Meyerson (1992) (53) reported a mean time between trauma and ankle arthrodesis of 6 years (range: 1–24 years). Horisberger et al. (2009) (4) found a latent period between accident and end-stage osteoarthritis (arthrodesis or prosthesis) of 21 years on average (range: 1–52 years) with negative prognostic factors including a more severe fracture type (in their study also pilon fractures), postoperative complications and older age. So, ankle fractures can have long-term consequences. Unfortunately, our study was unable to estimate the time at which radiological osteoarthritis develops after an ankle fracture. The data were insufficiently homogeneous to perform a regression analysis to establish a relationship between the onset of radiological osteoarthritis and the follow-up period since injury. This also implies that there is no time limit for insurers for the reservations they wish to make during the settlement procedure in injury claims for the development of osteoarthritis. If an insurance reservation applies to any treatment as a result of the ankle fracture, the reservation should be lifelong.
Very severe heterogeneity was visible in the analyses (I2 > 90%). This probably reflects clinical heterogeneity arising from the possible presence of osteoarthritis at baseline, the classifications used for the fractures and for osteoarthritis. As mentioned earlier, there was insufficient data to perform the analysis in more homogeneous subgroups. The usefulness of systematic reviews and meta-analyses of observational studies that are based on different study populations, different data collection procedures, different statistical analysis strategies, different underlying assumptions and with bias adjustments for different sets of confounding factors has been debated (54).
Conclusions
-
One in four ankle fractures is followed by radiological osteoarthritis, and after more severe fractures even one in three.
-
The severity of the trauma, as reflected by the fracture classification, is the most important prognostic factor for the development of radiographic osteoarthritis, but there is also a risk with simpler injuries.
-
The period within which osteoarthritis develops or becomes symptomatic with an indication for treatment cannot be specified.
ICMJE Conflict of Interest Statement
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding Statement
The authors received financial support from the Dutch Association of Medical Advisers in Insurance Matters (Nederlandse Vereniging van Geneeskundig Adviseurs in Particuliere Verzekeringszaken (GAV)). The funding organization had no influence whatsoever on any part of the research or the manuscript.
Author contribution statement
Both authors were equally involved in all parts of the study.
References
- 1.↑
Foy M, & Fagg P. Medicolegal Reporting in Orthopaedic Trauma, 4th ed ., pp. 1–560. Churchill Livingstone Elsevier, 2010.
- 2.↑
Willenegger H Treatment of luxation fractures of the tibiotarsal joint according to biomechanical viewpoints. Helvetica Chirurgica Acta 1961 28 225–239.
- 3.↑
Lindsjo U Operative treatment of ankle fractures. Acta Orthopaedica Scandinavica: Supplementum 1981 189 1–131. (https://doi.org/10.3109/ort.1981.52.suppl-189.01)
- 4.↑
Horisberger M, Valderrabano V, Hintermann B, Horisberger M, Valderrabano V, & Hintermann B. Posttraumatic ankle osteoarthritis after ankle-related fractures. Journal of Orthopaedic Trauma 2009 23 60–67. (https://doi.org/10.1097/BOT.0b013e31818915d9)
- 5.↑
Moher D, Liberati A, Tetzlaff J, Altman DG & PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Journal of Clinical Epidemiology 2009 62 1006–1012. (https://doi.org/10.1016/j.jclinepi.2009.06.005)
- 7.↑
Lauge-Hansen N Fractures of the ankle. II. Combined experimental-surgical and experimental-roentgenologic investigations. Archives of Surgery 1950 60 957–985.
- 8.↑
Meinberg EG, Agel J, Roberts CS, Karam MD, & Kellam JF. Fracture and dislocation classification compendium-2018. Journal of Orthopaedic Trauma 2018 32 (Supplement 1) S1–S170. (https://doi.org/10.1097/BOT.0000000000001063)
- 9.↑
Covidence Systematic Review Software. Melbourne, Australia: Veritas Health Innovation. (available at: www.covidence.org)
- 10.↑
Hoy D, Brooks P, Woolf A, Blyth F, March L, Bain C, Baker P, Smith E, & Buchbinder R. Assessing risk of bias in prevalence studies: modification of an existing tool and evidence of interrater agreement. Journal of Clinical Epidemiology 2012 65 934–939. (https://doi.org/10.1016/j.jclinepi.2011.11.014)
- 12.↑
Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, Schünemann HJ & GRADE Working Group. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ 2008 336 924–926. (https://doi.org/10.1136/bmj.39489.470347.AD)
- 13.↑
Lübbeke A, Salvo D, Stern R, Hoffmeyer P, Holzer N, & Assal M. Risk factors for post-traumatic osteoarthritis of the ankle: an eighteen year follow-up study. International Orthopaedics 2012 36 1403–1410. (https://doi.org/10.1007/s00264-011-1472-7)
- 14.↑
Hoiness P, Engebretsen L, & Strömsöe K. The influence of perioperative soft tissue complications on the clinical outcome in surgically treated ankle fractures. Foot and Ankle International 2001 22 642–648. (https://doi.org/10.1177/107110070102200805)
- 15.↑
Olerud C, & Molander H. Bi- and trimalleolar ankle fractures operated with nonrigid internal fixation. Clinical Orthopaedics and Related Research 1986 206 253–260. (https://doi.org/10.1097/00003086-198605000-00044)
- 16.↑
Donken CC, Verhofstad MH, Edwards MJ, & van Laarhoven CJ. Twenty-one-year follow-up of supination-external rotation type II–IV (OTA type B) ankle fractures: a retrospective cohort study. Journal of Orthopaedic Trauma 2012 26 e108–e114. (https://doi.org/10.1097/BOT.0b013e31822c4ea5)
- 17.↑
Beris AE, Kabbani KT, Xenakis TA, Mitsionis G, Soucacos PK, & Soucacos PN. Surgical treatment of malleolar fractures. A review of 144 patients. Clinical Orthopaedics and Related Research 1997 341 90–98. (https://doi.org/10.1097/00003086-199708000-00015)
- 18.↑
Abarquero-Diezhandino A, Luengo-Alonso G, Alonso-Tejero D, Sanchez-Morata EJ, Olaya-Gonzalez C, & Vilá Y Rico J. Study of the relation between the posterior malleolus fracture and the development of osteoarthritis. Revista Espanola de Cirugia Ortopedica y Traumatologia 2020 64 41–49. (https://doi.org/10.1016/j.recot.2019.09.002)
- 19.↑
Bois AJ, Dust W, Bois AJ, & Dust W. Posterior fracture dislocation of the ankle: technique and clinical experience using a posteromedial surgical approach. Journal of Orthopaedic Trauma 2008 22 629–636. (https://doi.org/10.1097/BOT.0b013e318184ba4e)
- 20.↑
Bauer M, Jonsson K, & Nilsson B. Thirty-year follow-up of ankle fractures. Acta Orthopaedica Scandinavica 1985 56 103–106. (https://doi.org/10.3109/17453678508994329)
- 21.↑
Day GA, Swanson CE, & Hulcombe BG. Operative treatment of ankle fractures: a minimum ten-year follow-up. Foot and Ankle International 2001 22 102–106. (https://doi.org/10.1177/107110070102200204)
- 22.↑
Lindsjø U Operative treatment of ankle fracture-dislocations. A follow-up study of 306/321 consecutive cases. Clinical Orthopaedics and Related Research 1985 199 28–38.
- 23.↑
Tosun B, Selek O, Gok U, & Ceylan H. Posterior malleolus fractures in trimalleolar ankle fractures: malleolus versus transyndesmal fixation. Indian Journal of Orthopaedics 2018 52 309–314. (https://doi.org/10.4103/ortho.IJOrtho_308_16)
- 24.↑
Verhage SM, Schipper IB, & Hoogendoorn JM. Long-term functional and radiographic outcomes in 243 operated ankle fractures. Journal of Foot and Ankle Research 2015 8 45. (https://doi.org/10.1186/s13047-015-0098-1)
- 25.↑
Zidek T, Urban J, Holub K, Peml M, & Kloub M. Comparison of plaster fixation and Kirschner wire transfixation as temporary fixation of displaced ankle fractures – randomised prospective study. Acta Chirurgiae Orthopaedicae et Traumatologiae Cechoslovaca 2019 86 141–146.
- 26.↑
Stadler J, Gauer E, & Ruedi T. Surgically treated malleolar fractures – results after 3 to 4 years under special consideration of the talus profile (author’s transl). [German]. Archiv für Orthopadische und Unfall-Chirurgie 1975 82 311–323. (https://doi.org/10.1007/BF00418927)
- 27.↑
Zhou Q, Lu H, Wang Z, Yu S, & Zhang H. Posterolateral approach with buttress plates and cannulated screw fixation for large posterior malleolus fractures. Journal of Foot and Ankle Surgery 2017 56 1173–1179. (https://doi.org/10.1053/j.jfas.2017.05.028)
- 28.↑
Sukur E, Akman YE, Gokcen HB, Ozyurek EC, Senel A, & Ozturkmen Y. Open reduction in pilon variant posterior malleolar fractures: radiological and clinical evaluation. Orthopaedics and Traumatology, Surgery and Research 2017 103 703–707. (https://doi.org/10.1016/j.otsr.2017.05.012)
- 29.↑
Shih CA, Jou IM, Lee PY, Lu CL, Su WR, Yeh ML, & Wu PT. Treating AO/OTA 44B lateral malleolar fracture in patients over 50years of age: periarticular locking plate versus non-locking plate. Journal of Orthopaedic Surgery and Research 2020 15 112. (https://doi.org/10.1186/s13018-020-01622-9)
- 30.↑
Xu HL, Liu LM, Li X, Zhang DY, Fu ZG, Wang TB, Zhang PX, Jiang BG, Shen HL & Wang G et al.Multicenter follow-up study of ankle fracture surgery. Chinese Medical Journal 2012 125 574–578.
- 31.↑
Veen EJD, & Zuurmond RG. Mid-term results of ankle fractures with and without syndesmotic rupture. Foot and Ankle Surgery 2015 21 30–36. (https://doi.org/10.1016/j.fas.2014.09.001)
- 32.↑
Weigelt L, Hasler J, Flury A, Dimitriou D, & Helmy N. Clinical and radiological mid- to long-term results after direct fixation of posterior malleolar fractures through a posterolateral approach. Archives of Orthopaedic and Trauma Surgery 2020 140 1641–1647. (https://doi.org/10.1007/s00402-020-03353-2)
- 33.↑
Meijer DT, Gevers Deynoot BDJ, Stufkens SA, Sierevelt IN, Goslings JC, Kerkhoffs GMMJ, & Doornberg JN. What factors are associated with outcomes scores after surgical treatment of ankle fractures with a posterior malleolar fragment? Clinical Orthopaedics and Related Research 2019 477 863–869. (https://doi.org/10.1097/CORR.0000000000000623)
- 34.↑
Regan DK, Gould S, Manoli 3rd A, & Egol KA. Outcomes over a decade after surgery for unstable ankle fracture: functional recovery seen 1 year postoperatively does not decay with time. Journal of Orthopaedic Trauma 2016 30 e236–e241. (https://doi.org/10.1097/BOT.0000000000000571)
- 35.↑
Heim D, Niederhauser K, & Simbrey N. The Volkmann dogma: a retrospective, long-term, single-center study. European Journal of Trauma and Emergency Surgery 2010 36 515–519. (https://doi.org/10.1007/s00068-010-0061-6)
- 36.↑
Hoelsbrekken SE, Kaul-Jensen K, Mørch T, Vika H, Clementsen T, Paulsrud Ø, Petursson G, Stiris M, & Strømsøe K. Nonoperative treatment of the medial malleolus in bimalleolar and trimalleolar ankle fractures: a randomized controlled trial. Journal of Orthopaedic Trauma 2013 27 633–637. (https://doi.org/10.1097/BOT.0b013e31828e1bb7)
- 37.↑
Verhage SM, Krijnen P, Schipper IB, & Hoogendoorn JM. Persistent postoperative step-off of the posterior malleolus leads to higher incidence of post-traumatic osteoarthritis in trimalleolar fractures. Archives of Orthopaedic and Trauma Surgery 2019 139 323–329. (https://doi.org/10.1007/s00402-018-3056-0)
- 38.↑
Kohake MBJ, Wiebking U, O’Loughlin PF, Krettek C, & Gaulke R. Mid- to long-term outcomes after weber b-type ankle fractures with and without syndesmotic rupture. In Vivo 2019 33 255–261. (https://doi.org/10.21873/invivo.11469)
- 39.↑
Federici A, Sanguineti F, & Santolini F. The closed treatment of severe malleolar fractures. Acta Orthopaedica Belgica 1993 59 189–196.
- 40.↑
Lambers KTA, van den Bekerom MPJ, Doornberg JN, Stufkens SAS, van Dijk CN, & Kloen P. Long-term outcome of pronation-external rotation ankle fractures treated with syndesmotic screws only. Journal of Bone and Joint Surgery: American Volume 2013 95 e1221–e1227. (https://doi.org/10.2106/JBJS.L.00426)
- 41.↑
Lange S, Mechsner K, & Langenscheidt P. Secondary arthrosis after surgical treatment of ankle fractures. RoFo: Fortschritte auf dem Gebiete Der Rontgenstrahlen und Der Nuklearmedizin 1984 140 69–74. (https://doi.org/10.1055/s-2008-1052924)
- 42.↑
May H, Alper Kati Y, Gumussuyu G, Yunus Emre T, Unal M, & Kose O. Bioabsorbable magnesium screw versus conventional titanium screw fixation for medial malleolar fractures. Journal of Orthopaedics and Traumatology 2020 21 9. (https://doi.org/10.1186/s10195-020-00547-7)
- 43.↑
Kortekangas T, Flinkkila T, Niinimaki J, Lepojarvi S, Ohtonen P, Savola O, & Pakarinen H. Effect of syndesmosis injury in ser IV (Weber B)-type ankle fractures on function and incidence of osteoarthritis. Foot and Ankle International 2015 36 180–187. (https://doi.org/10.1177/1071100714551788)
- 44.↑
McDaniel WJ, & Wilson FC. Trimalleolar fractures of the ankle. An end result study. Clinical Orthopaedics and Related Research 1977 122 37–45.
- 45.↑
Drijfhout van Hooff CC, Verhage SM, & Hoogendoorn JM. Influence of fragment size and postoperative joint congruency on long-term outcome of posterior malleolar fractures. Foot and Ankle International 2015 36 673–678. (https://doi.org/10.1177/1071100715570895)
- 46.↑
Stufkens SA, Knupp M, Horisberger M, Lampert C, & Hintermann B. Cartilage lesions and the development of osteoarthritis after internal fixation of ankle fractures: a prospective study. Journal of Bone and Joint Surgery: American Volume 2010 92 279–286. (https://doi.org/10.2106/JBJS.H.01635)
- 47.↑
Stufkens SA, Knupp M, Lampert C, van Dijk CN, & Hintermann B. Long-term outcome after supination-external rotation type-4 fractures of the ankle. Journal of Bone and Joint Surgery: British Volume 2009 91 1607–1611. (https://doi.org/10.1302/0301-620X.91B12.22553)
- 48.↑
Fonseca da LLD, Nunes IG, Nogueira RR, Martins GEV, Mesencio AC, & Kobata SI. Reproducibility of the Lauge-Hansen, Danis-Weber, and AO classifications for ankle fractures. Revista Brasileira de Ortopedia 2018 53 101–106. (https://doi.org/10.1016/j.rboe.2017.11.013)
- 49.↑
van Dijk CN, Verhagen RA, & Tol JL. Arthroscopy for problems after ankle fracture. Journal of Bone and Joint Surgery: British Volume 1997 79 280–284. (https://doi.org/10.1302/0301-620x.79b2.7153)
- 50.↑
Kellgren JH, & Lawrence JS. Radiological assessment of osteo-arthrosis. Annals of the Rheumatic Diseases 1957 16 494–502. (https://doi.org/10.1136/ard.16.4.494)
- 51.↑
Claessen FM, Meijer DT, Bekerom MP, Gevers Deynoot BD, Mallee WH, Doornberg JN, & van Dijk CN. Reliability of classification for post-traumatic ankle osteoarthritis. Knee Surgery, Sports Traumatology, Arthroscopy 2016 24 1332–1337. (https://doi.org/10.1007/s00167-015-3871-6)
- 52.↑
Moon JS, Shim JC, Suh JS, & Lee WC. Radiographic predictability of cartilage damage in medial ankle osteoarthritis. Clinical Orthopaedics and Related Research 2010 468 2188–2197. (https://doi.org/10.1007/s11999-010-1352-2)
- 53.↑
Papa JA, & Myerson MS. Pantalar and tibiotalocalcaneal arthrodesis for post-traumatic osteoarthrosis of the ankle and hindfoot. Journal of Bone and Joint Surgery. American Volume 1992 74 1042–1049. (https://doi.org/10.2106/00004623-199274070-00011)
- 54.↑
Ranstam J, & Wagner P. Systematic reviews, meta-analyses, randomized trials, and observational studies. Acta Orthopaedica 2022 93 1–2. (https://doi.org/10.1080/17453674.2021.1975398)
