General Information. We conducted a retrospective review of 109 patients (67 in the ELA group and 47 in the STA group) who had been diagnosed as type II/III calcaneus fracture and had undergone ORIF surgery via ELA or STA from January 2013 to October 2018. Basic demographic and clinical characteristic of the patients were obtained including age, gender, type of fracture, anesthesia, the length of hospital stay and follow-up time.
Inclusion and Exclusion Criteria. Participants were enrolled if they met the following eligibility criteria: (1) DIACFs classified as sanders type II/III; (2) closed and fresh fractures; (3) fractures that underwent procedures via ELA or STA; (4) follow-up period greater than 1 year. The exclusion criteria were as follows: (1) fractures classified as sander I or IV; (2) malunion, nonunion, open, and bilateral fractures; (3) associated or multiple trauma; (4) severe systemic diseases and commodity.
Clinical Assessment. The pain was measured using Visual Analogue Scale (VAS) with the score ranged from 0 to 10 points indicating no pain to worst pain. The VAS scores were collected at pre-operation, postoperative 3 days, 5 days and 7 days. Functional outcome was measured by American Orthopaedic Foot and Ankle Society (AOFAS) ankle-hindfoot scores at pre-operation, postoperative 3 months and at final follow up with the scores ranging from 0-100 points of pain, function and alignment. Radiographic including Bohler’s angle and Gissane angle were measured on fluoroscopic images at pre-operation, postoperative 3 months and at final follow-up time. The specific radiographic views including the anteroposterior (AP), lateral, axial projections and CT-scans were collected preoperatively. Based on the anatomic characteristic of calcaneus fractures shown in preoperative radiographs and CT, surgeons could attain the information of the type, level, locations of the fractures and measure height, width, Bohler’s and Gissane angle of the calcaneus.
Cost and Cost-utility Analysis. The medical and financial records of patients including the expense of total costs, laboratory and radiographic evaluation, surgery, anesthesia, analgesia, internal fixation materials and antibiotic were obtained from our hospital’s medical and financial information center. However, indirect costs including miss time from work or decreased productivity, rehabilitation, further consultation, transportation were not enumerated because each patients’ situation and the employment status varied. All costs were converted into U.S. dollars ($) at their value during January 2020.
Clinical outcome was monitored by recording health-related quality of life (HRQoL) scores before and after operation with a score ranging from − 0.11 to 1.0 and 1.0 indicates full health. Validation of HRQoL scores were converted by EQ-5D-3L including five dimensions: mobility, self-care, activities of daily life, pain and anxiety/depression and each depicted by three level from no, mild or moderate to severe problems. Thus, this depictive system includes 243 combinations, or health states. Combining HRQoL index and time were used to estimate QALYs, which are enumerated for the area under the curve by trapezoidal method.
The CUA was conducted from the healthcare perspective and was presented by the cost-effectiveness ratio (CER) and incremental cost-utility ratio (ICER). An annual discount of 3% was adjusted to QALYs and mean total costs.
Different discount rates (0% and 5%) were used to mean total costs and QALYs for sensitivity analysis.
The decisions including the surgical procedures, type of the incision: ELA or STA and type of the implants were made by senior surgeons based on the anatomic characteristic of the fractures and their preference.
Extensile Lateral Approach
Patients with a tourniquet at the thigh were under epidural, local, general or subarachnoid block anesthesia in the lateral position. The incision of ELA, which originated perpendicularly from 5 cm over lateral malleolus or the midpoint between the fibula and Achilles tendon and stopped at the base of 5th metatarsal, was performed by experienced surgeon as depicted by Benirshcke and Sangeorzan. With the flap held by several 2.0-mm Kirschner wires, this incision allowed the visualization of the lateral wall of calcaneus and subtalar joint and anatomic reduction was achieved directly under the guidance of C-arm fluoroscopy. Consequently, the length, height and width of the calcaneus were restored (Fig. 1).
Sinus Tarsi Approach
Patients with a tourniquet at the thigh were under epidural, local, general or subarachnoid block anesthesia in the lateral position. The incision was performed along a line from the tip of the lateral malleolus to the base of the fourth metatarsal and its length was approximately 3-5-cm. This incision allowed the visualization of the posterior facet, anterior process and even the calcaneocuboid joint with the calcaneofibular ligament dissected and the extensor brevis muscle elevated. Two 2.5-mm Schanz screw were inserted percutaneously into the calcaneal tuberosity and talus from medial to lateral, and tibia distraction device (Johnson & Johnson, USA) was applied to correct the deformity and restore the length, height, and width of the calcaneus with the guidance of C-arm fluoroscopy (Fig. 2).
The Kolmogorov-Smirnov test was applied to examine the normality of all variables. The between-group differences were compared by Student’s t-test for normal distribution variables (mean ± standard deviation) or a Mann-Whitney U test for abnormal distribution variables (median (25% quartile-75% quartile)) or a chi-square test for categorical variables (n (%)). The differences in the longitudinal changes of the Bohler’s, Gissane angle and VAS scores between ELA and STA group were compared by applying non-parametric test. All analyses were performed using SPSS 20.0 software (SPSS, IL, USA). Differences were considered to be statistically significant when P value was less than .05.