Descriptive characteristics of the cohort
This cohort study was registered at ClinicalTrials.gov (NCT03769909) and approved by the National Committee on Health Research Ethics (J.No. S-20170139). The study is reported according to the STROBE guidelines [12]. Patients with acute intra-articular ankle fracture admitted to Odense University Hospital and Svendborg Hospital from October 2017 to Marts 2019 were enrolled in the study. Inclusion criteria were an acute intra-articular fracture involving the ankle joint, a need for internal or external fixation within 14 days, patient age between 18 and 65 years, ability to read and understand Danish, and written informed consent. Exclusion criteria were open fracture, associated arterial and nerve injury, multiple injury patients with an Injury Severity Score >15, primary or secondary infection, systemic inflammatory disease such as rheumatoid arthritis, anti-inflammatory medication, and injuries associated with a Charcot foot. Patients were also excluded if they had any sign of radiographic OA in the fractured or healthy contralateral ankle joint (Figure 1).
The following epidemiological parameters were collected: age, sex, body mass index (BMI), American Society of Anesthesiologists Classification (ASA), and fracture classification of injury according to Arbeitsgemeinschaft für Osteosynthesefragen (AO) standards (44A, 44B, and 44C, time of samples collection, and surgical managements).
Perioperative aspiration of synovial fluids for chemiluminescence analysis
Briefly, synovial fluid (SF) was collected prior to surgery from the healthy contralateral ankle joint and then the fractured joint by puncture using the antero-medial portal. As it can be difficult to obtain sufficient SF from the ankle joint [13], 5 ml saline was injected prior to aspiration. Within 2 hours of sampling, SF samples were centrifuged at 2,000 revolutions per minute (RPM) for 15 minutes, aliquoted, and stored at -80°C until chemiluminescence analysis.
Chemiluminescence analysis of the synovial fluids
SF from the fractured and healthy contralateral ankle joints were analyzed for 12 pro-inflammatory and 4 anti-inflammatory cytokines, 4 metabolic proteins, and 2 cartilage degradation proteins.
SF levels of interleukin (IL)-1α, IL-1 receptor antagonist (IL-1RA), IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-13, interferon gamma (IFN-y), tumor necrosis factor (TNF)-α, and TNF-β were measured by an electrochemiluminescence immunoassay using a human customized U-Plex (Mesoscale, Rockville, MD). Matrix metalloproteinase (MMP)-1, MMP-3, and MMP-9 were measured using a human MMP-3 Plex Ultrasensitive kit (Mesoscale, Rockville, MD); transforming growth factor (TGF)-β1, TGF-β2, and TGF-β3 using a human U-PLEX TGF-β Combo kit (Mesoscale, Rockville, MD); and basic fibroblast growth factor (bFGF) using a Human V-PLEX bFGF kit (Mesoscale, Rockville, MD). SF levels of C-terminal telopeptides of type 2 collagen (CTX-2) and Aggrecan (ACG) were analyzed by enzyme-linked immunosorbent assay (ELISA) (MyBiosource, VersaMax™). All samples were run in duplex and performed according to the manufacturer’s instructions. Protein values below lower limit of detection (LLOD) were replaced by a value of ½ LLOD for statistical analysis. The percentage of proteins below LLOD and the coefficient of variation (CV) are presented in Supplementary 1.
Clinical follow-up 3 and 12 months after ankle surgery
After 3 and 12 months, the patients were invited to clinical and radiographic examination with plain X-ray and low-dose weight-bearing cone beam CT (WBCT) [14]. The following parameters were recorded: ankle swelling, VAS pain score (at rest and on activity), time of return to work after surgery, degree of mobilization (4-point scale), EQ-5D-5L index score Kellgren Lawrence score for severity of osteoarthritis (Grades 0-4, where 0=healthy ankle) [15], the American Orthopedic Foot and Ankle Society (AOFAS) score (ranging 0-100, where 100 = healthy ankle) [16], and the Danish version of Foot Function Index (FFI-DK) (ranging 0-230, with a high score indicating poor outcome) [17]. All examinations were performed by the same surgeon (TMP) at both follow-ups.
Evaluation of post-operative fracture malreduction, Kellgren Lawrence score, and joint space narrowing
Fracture malreduction was based on 9 criteria on plain X-ray (medial step-off, posterior step-off, tibiotalar tilt, tibiofibular (TF) overlap, oblique medial clear space, dime sign, lag screw, distal fibula screws, and proximal fibular screws) (Figure 2) and 5 criteria on WBCT (medial step-off, posterior step-off, fibular rotation, fibular anteroposterior translation, and tibiofibular widening) (Figure 3). All criteria have previously been defined as important for fracture reduction during ankle surgery [18, 19]. Furthermore, peri-operative X-rays from all patients were approved by orthopedic specialists after surgery. The measurements were performed in GE Web PACS 3.0 based on 3 months follow-up and were performed twice by two independent investigators (EBK/TMP) with two weeks between each measurement. Inter-observer and intra-observer reliability were determined using the intraclass correlation coefficient (ICC).
The Kellgren Lawrence score was measured on X-ray 12 months after surgery and joint space narrowing was defined as the difference in joint space measured at 3 months and 12 months on WBCT (Figure 3).
Statistical methods
Quantile-quantile (q-q)-plot tests of nearly all cytokines, radiographic criteria, and clinical outcomes indicated a parametric pattern. The correlation of cytokine levels and radiographic criteria versus clinical outcomes was performed using mixed effects logistic regression models for longitudinal data adjusted for age, sex, BMI, and AO fracture classification. Correlation of fracture classification versus clinical outcome and protein levels was performed using Spearman’s rho correlation coefficient. Results are presented as means with standard deviations, and a p-value <0.05 was considered significant. All statistical analyses were performed using STATA MP 16.