A recent critical review by Rammelt and Bartoníček of the management of ankle fractures involving the posterior malleolus establishes the following treatment recommendations :
“CT scanning is essential for detecting the exact 3-dimensional fracture anatomy, classification, and treatment planning in posterior malleolar fractures
Operative treatment should be considered for posterior malleolar fragments with displacement(≥ 2mm), involvement of the fibular notch of the distal part of the tibia (incisura), the presence of an intra-articular step-off (≥ 2 mm), or a depressed intercalary fragment
Fixation of a displaced posterior malleolar fragment restores the integrity of the posterior inferior tibiofibular syndesmosis and significantly reduces the need for additional syndesmotic stabilization
Anatomic reduction of posterior malleolar fragments restores the tibial incisura and thus facilitates reduction of the distal part of the fibula
Direct open reduction and fixation of posterior malleolar fragments via posterolateral and posteromedial approaches is biomechanically more stable and provides a more accurate reduction than indirect reduction and anterior-to-posterior screw fixation
Screw fixation of the posterior malleolus should precede intramedullary nailing in concomitant posterior malleolar and distal tibial shaft fractures
Relevant malalignment should be corrected as early as possible upon detection in order to avoid joint incongruity, chronic syndesmotic instability, and posttraumatic ankle arthritis” 
The first six recommendations are all described as grade B recommendations according to Wright . The last recommendation is classified as grade C. If guidelines based on these recommendations are introduced, the incidence of direct fixation of PM fragments will certainly increase. The trend toward direct PM fragment fixation without peer-reviewed level I evidence of efficacy, safety, and generalizability has been criticized . In our setting, the rate of severe AEs in this subgroup was considered high based on registry data and observational studies.
Was the incidence of adverse events that required revision unacceptably high?
Severe AEs requiring revision within the first twelve months after surgery were experienced by 34 patients (8.1% (95%CI [5,9–11,1%])) in this cohort; 30 of these cases (88%) were due to AEs that could arguably be classified as potentially avoidable, for example, deep infection, malpositioning of implants requiring revision, and postoperative loss of reduction. Overall, 50 patients (11.9%) experienced AEs that, according to the OrthoSAVES system, were classified as AEs with adverse effect on outcome (graded III or higher). A revision rate within the first year after surgery of approximately 8% and an overall severe AEs rate of more than 10% seems unacceptably high.
To our knowledge, the present study is one of the largest cohorts assessing AEs after surgical treatment of ankle fractures involving the posterior malleolus. The results need to be interpreted in light of the inherent limitations of the retrospective design. We aimed to address misclassification bias and interobserver inconsistencies by including only well-defined variables and by having multiple observers complete data capture, perform radiographic assessments and classify AEs. We believe the use of standardized AE reporting with the OrthoSAVES system is a strength of this study. The sample size of 421 patients is adequate to estimate the incidence of AEs with acceptable external validity and the inclusion of consecutive patients decreases selection bias. Of the reviewed patients, 29 were treated with initial external fixation and were excluded from the study. Almost all these patients had ankle fractures with PM fragments, which could pose a selection bias, as these were either very severely unstable injuries and/or open fractures. If included, patients treated with initial external fixation would probably further increase the incidence of AEs.
The AE reporting terminology in orthopaedic clinical studies is highly variable and inconsistent [8,17]. AEs reported in the different clinical studies of ankle fractures cannot be analysed and compared adequately . Reported incidences of AEs vary substantially in the literature. A systematic review of outcome after surgical treatment of ankle fractures with PM fragments reported deep infection in less than 1% of 768 included patients, but secondary surgery was required in 13% of cases . McHale et al. retrospectively reviewed 75 patients with posterior malleolar fractures and reported AEs in 25.3% of cases. PM fragment fixation was performed in 33% of included cases, predominantly using indirect techniques . Tornetta et al. followed 72 patients treated with direct PM fragment fixation and found no deep infections or loss of reduction requiring revision . Shi et al. assigned 116 patients to either direct or indirect PM fragment fixation and reported no severe AEs requiring revision in either group . Little et al. reported deep infection in 4.4% of 112 patients treated with direct PM fragment fixation via a posterolateral approach and a revision rate of 12.5%. However, this revision rate included implant removal procedures .
Implications for the future management of ankle fractures involving posterior malleolar fragments
If open direct fixation through a posterior approach can decrease the rate of potentially avoidable AEs, such as deep infection and implant-related AEs requiring revision, then a change of guidelines based on the recommendations described above seems justified.
In this historical cohort, 27% of patients were assessed with a preoperative CT scan and thus PM fragment morphology could not be ascertained in most cases. Previous CT studies have reported most PM fragments to extend into the fibular incisura . Bartoníček reported 92% of PM fragments to be intraincisural in a consecutive CT series . The critical review, discussed above, recommends anatomic reduction of intraincisural PM fragments to restore articular anatomy in the incisura . Additionally, fragmentation is frequently seen when assessing PM fragments with preoperative CT scans [24–26]. Sultan et al. reported intercalary fragments in 43% of 247 patients with posterior malleolar fractures . Based on these findings, we hypothesize that the frequency of PM fragment fixation found in this study (17.8%) would significantly increase if the aforementioned recommendations were implemented as standard care in clinical practice.
One of the most prevalent indications for early revision in ankle fractures is malreduction of the syndesmosis after trans-syndesmotic screw fixation . As direct PM fragment fixation reinserts the posterior inferior tibio-fibular ligament, it has been shown that the need for additional trans-syndesmotic fixation can be decreased [29–31]. One in 4 patients in this cohort (117 (27.8%)) were treated with trans-syndesmotic fixation, 12 of whom were revised in the first year after surgery. In 7 cases (7/117 (1.7% of the total cohort)) revision was required due to malpositioning of the trans-syndesmotic fixation or loss of tibiotalar alignment. The author of “In defence of the posterior malleolus” argues that it is only a “minority of fractures” that require syndesmotic fixation when PM fragment fixation is not readily performed . In this study, syndesmotic fixation was performed in 27.8% of cases. It seems to be a considerable proportion of patients with ankle fractures involving the posterior malleolus that require syndesmotic fixation when PM fragment fixation is performed only in selected cases. It remains to be established whether syndesmotic stabilization through direct PM fragment fixation represents a clinical advantage over trans-syndesmotic fixation in patients who present with this common fracture.