Clinical outcome and concomitant injuries in operatively treated fractures of the lateral process of the talus

The talus is the second largest tarsal bone with a broad-based triangular lateral process on its lateral side, i.e. the lateral process of the talus (LPT). It articulates with the fibula superiorly and builds the anterior-lateral portion of the posterior subtalar joint [1].

Overall, LPT fractures (LPTF) are rare [2] but have been stated to account for up to 20% of all talar fractures [3]. They are traditionally attributed to a high energy trauma mechanism [4–6]. More recent studies have shown a 17-fold fracture risk increase in snowboarders. Consequently, LPTF are also referred to as “Snowboarders Ankle” or “Snowboarders Fracture” [7–10]. The actual trauma mechanism is a matter of debate. Based on the data available, forced dorsiflexion, inversion, eversion, and potentially external rotation have been postulated [11–13].

Diagnosis of LPTF is challenging, as almost 50% of fractures are missed on plain radiographs [6, 14]. Consequently, computed tomography (CT) must be performed in case of suspicion [14, 15]. Various foot and ankle classification systems, such as the AO-classification or ICI classification, do included LPTF. Still, Hawkins [6] as well as McCrory and Bladin [16] are the only two classifications, specifically focusing on LPTF. The Hawkins classification, published in 1965, thereby is still the most commonly used. It classifies LPTF as simple fractures (Type I), comminuted fractures (Type II), or chip fracture (Type III) [6]. The McCrory and Bladin classification [16] describes similar fracture patterns just in a different order. Both classifications are based on plain radiographs and therefore might underestimate the complexity of these fractures. Studies including CT or MRI imaging reported concomitant injuries in 46 to 88% of patients suffering a LPTF [15, 17]. Sadly, up to date we are missing a comprehensive classification for LPTF including concomitant injuries. In order to present the data available in literature more uniformly, all fractures were re-classified per the Hawkins classification [6].

The best treatment for LPTF remains a matter of debate as these fractures are rare, the literature available is limited, and concomitant injuries are frequent. The vast majority of LPTF studies are descriptive, retrospective case series based on plain radiographs missing a standardized treatment concept [4–6, 10, 17, 18]. Only two studies validated a predefined treatment regimen [7, 15]. Valderrabano et al. [7] treated fractures with a displacement of more than 2 mm surgically. They presented the outcome of six conservatively (Type I: 2, Type II: 1, Type III: 3) and 14 operatively (Type I: 14) treated patients. In this cohort, operative treatment resulted in superior AOFAS scores compared to the conservative treatment (AOFAS: 97 ± 1) vs. 85 ± 5). Von Knoch et al. [15] recommended surgery in case of displacement of more than 1 mm. Seven patients were treated conservatively (Type I: 4, Type II: 2, Type III: 1), 16 operatively (Type I: 11, Type II: 2). In contrast to Valderrabano et al. [7], this study reported superior results for the conservative treatment (AOFAS: 98 (87–100) vs. 93 (82–100)). Not only do those two studies report contradictory results, but they are also limited because of their group sizes and the chosen outcome measure (AOFAS). Although the AOFAS is one of the most frequently applied scores, it has been questioned for its validity, responsiveness and applicability [19–22]. Consequently, we are missing an evidence-based treatment algorithm for LPTF.

At the authors’ orthopedic reference center, patients presenting with a displaced Hawkins Type I or Type II fracture as well as patients suffering relevant concomitant injuries, such as subtalar osteochondral lesions, loose bodies, or peroneal tendon dislocation, are treated surgically. Undisplaced fractures (< 2 mm) and Hawkins Type III fractures without relevant concomitant injuries are treated conservatively. In case of conservative treatment, the further treatment is conducted by the referring physician. Injuries necessitating operative treatment remain in the authors’ reference center.

The way outcome in orthopedic trauma is being assessed is changing, shifting to patient focused outcome measures such as quality of life (QOL) and patient-reported outcome measures (PROMs) [23–25]. From 1985 to 2015, the number of studies on injured patients including PROMs increased almost 30-fold [23]. PROMs, per the recommendations of the COSMIN group, should be valid, reliable, and responsive [26–28]. Today, various PROMs have been published, but we are still missing a universally applied gold standard [21]. In this study, we applied the VAS-FA, Karlsson Score and SF-12, which were the standard PROMs used in our department at that time.

The primary aim of this study was to review the PROM of all surgically treatment patients suffering a LPTF with a follow-up of at least one year. The secondary aim was to identify factors affecting the PROM.

Overall, 75 patients were identified from the database, 23 were eligible and 22 (96%) includ.ed in the final analysis (Fig. 1). Their mean age was 32 ± 9 (18 to 49) years, 73% were male with a mean BMI of 23 ± 3 (19 to 28) kg/m². No relevant preexisting medical conditions, including smoking or diabetes, were reported.

Trauma & Fracture characteristics

The left side was affected in 23% of the patients. A low velocity trauma mechanism (distorsion) was reported in 23%. High velocity trauma mechanisms included a fall during rock climbing (32%), or following a jump during snowboarding (14%). Each fracture was assessed by CT. An MRI was available in 46% of the patients. Table 1 summarized the assessed fracture characteristics. Neither displacement (p = 0.521), nor comminution (p = 0.060), age (p = 0.189.), gender (p = 0.160), or BMI (p = 0.754) differed significantly between the Hawkins classification (Type 1 and 2).

Hawkins classification	Type 1 n=16 (73%)	Type 2 n=5 (23%)	Type 3 n=1 (5%)
Displacement ≥2 mm	13 (81%)	5 (100%)	1
Comminuted	6* (38%)	5 (100%)	1
Concomitant injuries	13 (73%)	4 (80%)	1
Number concomitant injuries	1.7 ± 1.1 (0 to 4)	2.0 ± 1.6 (0 to 4)	3
Primary surgical procedure
ORIF	16 (100%)	0 (0%)	0
Resection	0 (0%)	5 (100%)	1
Additional surgical procedures
None	5 (31%)	1 (20%)	0
Resection i.a. loose bodies	7 (44%)	0	0
Subtalar microfracturing	2 (13%)	1 (20%)	1
Subtalar AMIC	2 (13%)	0	0
Reconstruction lateral ankle ligament complex	0	2 (40%)	0
Peroneal tendon repair	0	1 (20%)	0

* As assessed on CT imaging, one main fragment accompanied by smaller chip-off fragments

Six patients were classified as Hawkins Type 1 with comminution. This discrepancy to the traditional Hawkins classification is due to the different imaging modalities. Hawkins classified all fractures based on plain radiographs [6]. In this study, all fractures were classified based on CT images. Those patients classified as Hawkins Type 1 with comminution presented one large main fragment accompanied by few small comminuted fragments (Fig. 2), which most likely would not have been visible on plain radiographs.

Patient rated outcome

Patients returned to work after 63 ± 47 (7 to 200) days and to sports after 8 ± 5 (3 to 24) months after the initial surgery. Again, the Hawkins classification had no significant influence on the time to return to work (p = 0.813) or sports (p = 0.368).

The mean final follow-up was 44 ± 2 (12 to 97) months. At that time, the mean VAS FA overall was 77 ± 21 (20 to 100), the Karlsson Score 72 ± 21 (34 to 97), and for the SF 12 the PCS 53 ± 8 (36 to 64), the MCS 53 ± 7 (32 to 63), the PF 52 ± 8 (34 to 57), and the BP 53 ± 8 (31 to 58). The PROMs’ distribution per the Hawkins classification is illustrated in Fig. 3. Only 50% of patients have returned to their previous level (≥90%) of sports.

This study presents the largest cohort on surgically treated LPT fractures. Hawkins Type 1 fractures were treated by open reduction and screw fixation, Type 2 and 3 fractures by resection of the fracture fragments. Furthermore, all concomitant injuries were addressed. Despite this progressive treatment approach, after more than three years after the surgery, the patient rated outcome revealed only moderate to good scores. Twelve (55%) patients suffered from posttraumatic symptomatic subtalar osteoarthritis. Although Hawkins Type 1 fractures resulted in better VAS FA Overall scores than Type 2 fractures, the only independent factor associated to an impaired patient rated outcome was posttraumatic symptomatic subtalar osteoarthritis.

The authors are only aware of eight case series (> 6 patients) reporting on the outcome of LPTFs [4–7, 10, 17, 18]. Out of those, only two studies applied a standardized treatment algorithm and evaluated the surgical outcome using PROMs [7, 15]. Valderrabano et al. reported excellent results for 14 operatively treated Hawkins Type I fractures. Their mean AOFAS score was 97 (1) and a VAS of 0.2 (0.6) points [7]. However, no data was reported for operatively treated Type II fractures. Von Knoch et al. reported good to moderate AOFAS scores of 93 (82 to 100) points for operatively treated displaced Type I (n = 11) and Type II (n = 5) fractures [15]. Still, various studies have questioned the validity of the AOFAS [19, 20, 22]. As outlined in the introduction, the fractures included in the study by Valderrabano et al. [7] and von Knoch et al. [15] were transformed to the Hawkins classification to increase the comparability.

The herein presented 22 operatively treated LPTFs resulted in only moderate to good PROMs. The VAS FA and Karlsson Score revealed residual impairment. The VAS FA Overall Score (77 ± 21 (20 to 100)) was lower than the published reference values for healthy individuals (86 to 100) but comparable to patients with an isolated hallux valgus (45 to 83) [32]. Similar results were found for the Karlsson Score (72 ± 21 (34 to 97)) with scores above 80 points representing good to very good results [30]. The patients’ quality of life (SF-12: PCS 53 ± 8 (36 to 64), MCS 53 ± 7 (32 to 63)) was in the range of a healthy population. Overall, operatively treated displaced Type I or II fractures result in good to moderate results.

The observed residual impairment was also reflected in the return to sports rate. Only half of the patients reached a sport activity level ≥ 90% after on average 8.2 ± 4.9 (3 to 24) months. A surprisingly high rate (100%) of return to sports in operatively treated patients was found by Valderrabano et al. [7] Von Knoch et al. [15] and Klein et al. [17] reported rates of return to sports comparable to the herein presented study (63% / 59%).

The secondary aim of this study was to identify factors associated to a poor patient rated outcome. Although Hawkins Type 1 fractures resulted in superior results compared to Type 2 fractures (VAS FA Overall: p = 0.018 and -Other: p = 0.002) the strongest independent factor for impairment was posttraumatic symptomatic subtalar osteoarthritis (VAS FA and Karlsson Score). In the current study, 50% of patients suffered from symptomatic subtalar osteoarthritis, which compares well to literature with reported rates of 15 to 45% [6, 7, 10]. On the contrary, neither the occurrence nor the number of concomitant injuries affected the patient rated outcome. Only five studies have previously reported on concomitant injuries following LPTFs [6, 10, 15, 17, 18]. Von Knoch et al. reported a rate of concomitant injuries of 88% [15]. Klein et al. reported on peroneal tendon displacement in 46% of patients suffering a Type 2 LPTF [17]. Both studies did not further analyze the influence of these concomitant injuries. These figures are in the range of the herein observed rate of 82% of patients suffering concomitant injuries. This number might even underestimate the actual occurrence of concomitant injuries accompanying LPTFs, as several of these injuries are only detectable on MRI, and MRI was available for just 46% of patients in this study. On the contrary, the high number of concomitant injuries could have hindered sufficient statistical analysis.

The high incidence of symptomatic posttraumatic subtalar osteoarthritis and the concomitant injuries observed (osteochondral lesion to the posterolateral calcaneal facet, bony avulsion of the medial talocalcaneal ligament, lesion to the plantolateral aspect of the talar head), might have implications for the actual trauma mechanism (Fig. 4). Up to now, most authors discuss a combination of axial compression, dorsal extension, inversion and external rotation of the foot [4–6, 12, 13, 33]. Taking into consideration the above outlined combination of concomitant injuries, the mechanism of injury could also be a subluxation of the subtalar joint with external rotation and pronation (Fig. 4).

Several limitations of the herein presented study need to be discussed. First, no sample size calculation was conducted due to the observational study design and the low incidence of LPTF. Second, the study was retrospective, which again is in line with most other studies and again attributable to the low fracture incidence [4–6, 10, 15, 17, 18]. A further limitation could be the small number of patients (n = 22). Still, this is the largest published cohort on surgically treatment of LPTF [5–7, 10, 15]. The most pronounce limitation is a missing control group. Up to now, extensive comparative studies are missing completely. Although Valderrabano et al. and von Knoch et al. reported on the AOFAS of conservatively treated LPTF, they did not compare different treatment regimes in similar fracture patterns and reported conflicting results [7, 15]. It would be of great interest to prospectively asses the PROM of conservatively treated, displaced LPTF Type I and II according to Hawkins. Finally, the herein used PROMs are not fully validated per the COSMIN group recommendations [26–28] and the normative data available for the VAS-FA have not been validated in a foot and ankle trauma population [32]. Future studies should be even more aware of the PROMs chosen and consider scores such as the Self-reported Foot and Ankle Score (SEFAS) [21, 34], Manchester-Oxford foot questionnaire (MOXFQ) [35–37].

Despite the above outlined limitations, several strengths of this study are noteworthy. First the intermediate follow-up of more than three years. Second, a follow-up of 96% of patients. Third, the detailed fracture assessment based on CT and MRI imaging. The authors are not aware of any work that has investigated the concomitant injuries and fracture patterns in such detail. Finally, this is the first study to clearly show, that despite consequent treatment of all concomitant injuries, posttraumatic symptomatic subtalar osteoarthritis is the independent factor associated to an impaired patient rated outcome.

The authors present the largest cohort on surgically treated LPT fractures. Despite consequent treatment of all concomitant injuries, the patient rated outcome revealed only moderate results. Although Hawkins Type 1 fractures resulted in a better outcome than Type 2 fractures, the major factor affecting the outcome was symptomatic posttraumatic subtalar osteoarthritis. The authors have hypothesized that this is due to a subtalar subluxation. Future randomized studies have to compare the patient rated outcome in conservative to surgically treated displaced LPTF.