Minimally invasive percutaneous osteosynthesis versus ORIF for Sanders type II and III calcaneal fractures: a prospective, randomized intervention trial

This was a prospective randomized controlled trial of patients with displaced intraarticular fractures of the calcaneus. The study was conducted according to the “CONSORT statement” guidelines for randomized controlled trials [17]. Patients were allocated in accordance with the random number table and divided into two groups, a MIPO group and an open reduction and internal fixation (ORIF) using a conventional approach group (conventional treatment group). A calcaneus plate (AO Synthes, Oberdorf, Switzerland) was used for internal fixation in both groups. The procedures were all performed by a team of two senior surgeons. All radiographic assessments were conducted by two independent senior radiologists. The final result was the average of the two testing values. The study protocol was approved by the Medical Ethics Committee of Shaoxing People’s Hospital (No. 2010024).

The inclusion criteria for this study were displaced intraarticular fractures of the calcaneum with standard radiographic and CT evidence of Sanders type II or III fractures. Patients with an open fracture, bilateral calcaneal fractures, peripheral vascular disease, skin infection, signs of compartment syndrome, neurologic deficit following head injury or spinal injury, other fractures of the ipsilateral or contralateral limb, and severe osteoporosis were excluded. From May 2011 to November 2013, 82 consecutive patients with unilateral calcaneal fractures that met these criteria were included in the study. All patients received the type of treatment to which they had been allocated (Fig. 1). Written informed consent was obtained from all participants. Demographic data including age, sex, injury mechanism, and follow-up time were recorded.

Following admission, the patients’ extremities were elevated and ice was applied in an effort to minimize swelling and avoid blisters. Surgery was performed either within 24 h or after the swelling had subsided and skin wrinkling had appeared. For antibiotic prophylaxis, 1.5 g of intravenous cefuroxime was administered 30 min preoperatively [18]. All patients underwent general anesthesia or epidural anesthesia.

In the MIPO group, the patient was placed in a lateral position on a radiolucent operating table and a tourniquet was applied. Landmarks for skin incision were identified, including the posterior border of the lateral malleolus and the anterolateral border of the Achilles tendon. An axially directed incision (3–4 cm) was made from the level of the tip of the lateral malleolus to midway between the posterior border of the fibula and the anterolateral border of the Achilles tendon (Fig. 2a). After dividing the subcutaneous tissue directly to the bone, a periosteum elevator was used to create a lateral calcaneal subfascial channel to fit the calcaneal plate (Fig. 2b). Afterward, a Steinmann pin was inserted into the calcaneal tuberosity from the lateral side and traction was applied along the long axis of the foot to correct the varus and length of calcaneus while the posterior articular surface was reduced by leverage (Fig. 2c). For cases of Sanders type III calcaneal fractures, the posterior articular surface was also reduced with the help of a periosteum elevator. After incompletely uncovering the fragment of the lateral wall, a periosteum elevator was inserted in the calcaneus and placed under the posterior articular surface (Fig. 2d). Under C-arm fluoroscopy, the posterior articular surface was directly reduced by leverage (Fig. 2e). Once satisfactory anatomical reduction of the calcaneus was achieved, a pre-bent calcaneus plate (AO Synthes, Oberdorf, Switzerland) (Fig. 2f) was inserted into the subcutaneous calcaneus lateral channel. After confirming the proper placement of the plate under C-arm fluoroscopy, an identical plate was placed on the implant area as a scope for selective stab incisions and screws were placed into the rear calcaneal body, the subtalar articular surface, and the anterior process of the calcaneus through the stab incisions (Fig. 2g). Next, the plate and screw were secured under C-arm fluoroscopy, rubber drains were inserted, and the incision was closed in a layered fashion followed by compression bandaging (Fig. 2h).

In the conventional treatment group, ORIF using an extended L-shaped lateral approach [2] was performed. An incision measuring 10–12 cm was made directly to the bone to create a full-thickness flap. In contrast to the MIPO group, the entire lateral wall of the calcaneus was exposed and the subtalar joints were directly visualized after overturning the fracture fragment of the lateral wall. Once anatomical reduction of the calcaneus was achieved, the allogeneic bone was used for the reconstruction of calcaneal defects. Rigid fixation with the calcaneus plate and screws (AO Synthes, Oberdorf, Switzerland) was also achieved under C-arm fluoroscopy.

Postoperatively, the affected limbs of the patients were elevated to minimize swelling. At 3 and 12 h postoperatively, 1.5 g of prophylactic intravenous cefuroxime was administered [18]. At 24 h postoperatively, it was possible to begin passive and active motion of the ankle and subtalar joint without weight-bearing. Rubber drainage was extracted after 48 h and the incisions were cleaned every 2 days until the sutures were removed approximately 2 weeks after surgery. All patients were treated with analgesic therapy for 1 week (celecoxib 200 mg, orally, twice daily). Partial weight-bearing was begun after 6 weeks post-operation, and full weight-bearing was initiated after 12 weeks.

Primary clinical outcomes including operative time, postoperative visual analog scale (VAS) score, and wound healing complications were recorded. VAS was used to track patients’ daily pain scores during the pre-operation and post-operation phases, as well as at 1 week postoperatively. The VAS was measured in centimeters (0–10.0 cm) in this study. The wound healing complications, including superficial infection, deep infection, and wound necrosis, were evaluated. According to the surgical site infection criteria [19], superficial infections could be treated non-operatively and infections that required surgical intervention, readmission, or intravenous antibiotics were classified as deep infections.

Secondary clinical outcomes including time to operation, length of incision, postoperative drainage, length of hospital stay, and medical expense were also recorded. The postoperative complications of peroneal tendon or sural nerve injury were also assessed. At the final follow-up visit, the American Orthopedic Foot and Ankle Society (AOFAS) Ankle-Hindfoot scale [20] and the Short Form-36 (SF-36) questionnaires [21] were used to quantify the functional outcomes.

Radiographic evaluation included measurement of the height and width of the calcaneus in ankle axial view, Bohler’s angle, Gissane’s angle, and length of the calcaneus on the lateral view during the pre-operation and post-operation period [9, 13‐15]. In addition, the surgical planning and classification of fractures using the Sanders et al. [22] classifications were determined by preoperative CT scans (Fig. 3).

Data were expressed as means ± standard deviation. Statistical evaluation including the Pearson χ ² test, contingency table χ ² test, and Student’s t test was conducted for comparisons between the MIPO group and the conventional treatment group. The tests were performed using SPSS 18.0 for Windows. The Shapiro-Wilk and Levene’s tests showed normal data distribution and variance. The level of significance was set at P < 0.05.

From May 2011 to November 2013, a total of 82 patients who fulfilled the study criteria were initially enrolled; 18 patients were lost to follow-up. The reasons for loss of follow-up included limited activity in 8 cases, loss of communication in 4 cases, and immigration to another area in 6 cases. Of the remaining 64 patients available for analysis, 29 patients had displaced intraarticular calcaneal fractures in the MIPO group and 35 in the conventional treatment group. All patients completed at least 1 year of follow-up. There were no statistically significant differences between the MIPO and conventional treatment groups in terms of age, sex, mechanism of injury, Sanders classification of calcaneal fractures, or follow-up time (Table 1).

The MIPO group was superior to the conventional group in terms of operative time and postoperative VAS values. The operative time of the MIPO group was significantly shorter than that of the conventional treatment group (52.5 ± 11.1 vs. 82.8 ± 16.2 min, respectively; P < 0.05). At 1 week postoperatively, the VAS values in the MIPO group (3.2 ± 1.4) were significantly lower than those in the conventional treatment group (3.9 ± 1.3, P < 0.05) (Table 2).

Overall, the incidence of wound healing complications in the MIPO group was significantly lower than that of the conventional treatment group. In the conventional treatment group, 13 of 35 fractures (37.1%) had issues with wound healing compared to only 2 of 29 fractures (6.7%) in the MIPO group; this difference was statistically significant (P < 0.05). Deep infection occurred in none of the 29 patients in the MIPO group and in 2 of the 35 (5.7%) patients in the conventional treatment group (P = 0.191). The incidence of superficial infection was 1 of 29 (3.4%) patients in the MIPO group and 4 of 35 (11.4%) patients in the conventional treatment group (P = 0.236). Wound necrosis occurred in 1 of 29 (3.4%) patients in the MIPO group and 7 of 35 (20%) patients in the conventional treatment group, and this difference was statistically significant (Table 2). Five patients with superficial infections in the two groups were managed using dressings and oral antibiotics. Two patients with deep infections were treated with debridement twice as well as with intravenous antibiotics, without success. Finally, implant removal was executed.

The length of hospital stay was 9.7 ± 2.8 days in the MIPO group and 11.7 ± 2.6 days in the conventional treatment group (P < 0.05). The length of incision was 4.2 ± 0.6 cm in the MIPO group and 10.9 ± 1.5 cm in the conventional treatment group (P < 0.05). Neither peroneal tendon nor sural nerve injury occurred in any patient in either group. The medical cost was 20.0 ± 2.3 × 10² USD in the MIPO group and 21.3 ± 4.4 × 10² USD in the conventional treatment group, with no statistical difference. In terms of time to operation and postoperative drainage, the data were comparable between the two groups (P > 0.05). At the time of the final follow-up, the SF-36 scores and AOFAS scores in the two groups were comparable (P > 0.05) (Table 2).

Similar radiographic results were achieved in both groups. Preoperatively, all the radiographic outcomes in the two groups were comparable (P > 0.05). Postoperative radiography showed that satisfactory calcaneal fracture reduction was achieved in both groups. The postoperative Bohler’s angle was 28.7 ± 4.6° in the MIPO group and 29.5 ± 4.1° in the conventional treatment group (P > 0.05). The postoperative Gissane’s angle was 123.0 ± 10.5° in the MIPO group, which was lower than that in the conventional treatment group, 124.3 ± 10.1°, but the difference was not statistically significant (P = 0.620). There were no significant differences in the calcaneal height, width, or length between the two groups postoperatively (P > 0.05) (Table 3).