2.1

Following institutional review board approval, consecutive patients of a single, fellowship-trained orthopaedic foot and ankle surgeon were retrospectively reviewed. The criteria for inclusion in the current study included patients > 18 years of age diagnosed with rigid flatfoot deformity who underwent either double or triple arthrodesis for the treatment of rigid flatfoot deformity between October 2016 and November 2022, had failed at least six months of nonoperative management, and had a minimum of two years of postoperative follow-up available. Patients with a history of prior arthrodesis involving the subtalar, talonavicular, or calcaneocuboid joints; history of joint-preserving surgery for flatfoot deformity; incomplete follow-up; additional procedures unrelated to flatfoot correction performed during the index surgery; neuropathic arthropathy (e.g., Charcot neuroarthropathy); and neurodegenerative conditions (e.g., Parkinson’s disease, multiple sclerosis) affecting gait mechanics were excluded from the study.

After retrospective chart review, procedures were categorized as either open or MIS. All cases were performed utilizing an open technique from September 2016 to August 2020, after which the standard approach was transitioned to a percutaneous MIS technique starting in September 2020 through November 2022. Patients were included up to November 2022 to ensure a minimum of two years of follow-up.

2.2

In patients who underwent open double or triple arthrodesis, a dorsal incision was made over the talonavicular (TN) joint, along with a sinus tarsi incision to access the subtalar and calcaneocuboid (CC) joints. Fixation was achieved using nitinol staples or 4 mm headless compression screws for the CC joint, two 5–7 mm diameter headless compression screws for the subtalar joint, and a combination of nitinol staples and 4 mm fully threaded headless compression screws for the TN joint. Meanwhile, in patients who underwent an MIS arthrodesis procedure, fixation was performed using two 5–7 mm diameter headless compression screws for the subtalar joint and three 4 mm diameter fully threaded headless screws for the TN and CC joints.

2.3

Under fluoroscopic guidance, the posterolateral aspect of the subtalar joint is identified, and a 3-mm incision is made between just posterior to the peroneal tendons. Next, under fluoroscopy, the 3 × 30 cutting burr is advanced into the joint and debrided ( Fig. 1 ). After the debridement is completed, the joint surfaces are palpated with a curette to feel for any remaining areas of cartilage that can be removed with the curette itself ( Fig. 2 ). Next, a small bone rasp and a pituitary rongeur are used to remove debris from the joint and to further assist with cartilage debridement. The joint is then flushed copiously with a 50-ml syringe and an 18-gauge blunt needle. Finally, the joint surface was prepared; a 2 mm K-wire was used to perforate the subchondral bone, promote marrow bleeding, and facilitate bone fusion.

Under fluoroscopy, the 3 × 30 cutting burr is advanced into the subtalar joint and debrided.

Joint surfaces are palpated with a curette to feel for any remaining areas of cartilage that can be removed with the curette itself.

2.4

A single 3-mm stab incision is made at the lateral aspect of midline calcaneocuboid joint. A freer elevator is advanced to feel the joint. A 2×12 cutting burr is advanced into the joint and the debridement is performed. The debridement and preparation of the joint (curette, rasp, irrigation, 2-mm drill holes) is then completed as described for the subtalar joint.

2.5

The surgeon utilizes two 2-mm portals: one straight medial and one dorsolateral (between the anterior tibial and extensor hallucis longus tendons). Once the subcutaneous tissues are dissected with a mosquito clamp, a freer elevator is advanced into the joint to feel the anatomy and orientation ( Fig. 3 ). A 2 × 12 mm cutting burr is advanced through the portals and the joint is debrided as for the previously mentioned joints ( Fig. 4 ).

Once the subcutaneous tissues are dissected with a mosquito clamp, a freer elevator is advanced into the joint to feel the anatomy and orientation.

Final construct, A/P radiograph of the foot.

2.6

Following surgery, all patients were placed in a postoperative splint for two weeks. Afterward, patients were transitioned to a controlled ankle motion (CAM) boot, remaining non-weightbearing for a total of six weeks. Patients were allowed to weigh-bear as tolerated in a CAM boot for an additional six weeks before progressing to regular footwear. All patients underwent a computed tomography (CT) scan at three months postoperatively to assess healing. If fusion was not at least 50 % complete at this time, CT scans were repeated at six months postoperatively. Nonunion was defined as the absence of radiographic fusion at postoperative nine months, in accordance with current FDA guidelines.

2.7

Baseline demographic and clinical characteristics were collected via retrospective chart review. These included age, sex, BMI, smoking status, diabetes mellitus, peripheral neuropathy, peripheral vascular disease . The primary outcome of this study was defined as the change in patient-reported outcomes measured by Visual Analog Scale (VAS) and Foot Function Index (FFI) score from preoperative to final follow-up visits. Secondary outcomes included radiographic time to union, incidence of nonunion, and postoperative complication rates (including wound dehiscence). All statistical analyses were pre-specified as part of the study design. No post hoc or exploratory analyses were conducted beyond the stated primary and secondary outcomes.

VAS and FFI scores, were collected at preoperative and postoperative clinic appointments to measure patients’ functional limitations and pain presentation. Change in (∆) VAS and FFI scores were calculated by subtracting final postoperative follow-up scores from preoperative scores. VAS and all FFI scores were compared within each cohort by paired t -test. ∆VAS and ∆FFI scores were compared between cohorts on independent study t -test, and within each cohort against a value of 0 on one-sample t -test. In addition to clinical outcome measures, radiographic angular parameters—including Meary’s angle, calcaneal pitch, and Kite’s angle—were measured preoperatively and at final follow-up to assess the degree of deformity correction. These measurements were compared within cohorts using paired t -tests. All other continuous outcome measurements were compared between cohorts by independent study t -test. Noncontinuous data, including complications, were compared by Chi-squared analysis. All t -tests were two-tailed and all p < 0.05 were considered to be significant.

A priori power analysis was performed using a minimal clinically important difference (MCID) of 2.0 for VAS scores, a standard deviation of 2.5, an alpha level of 0.05, and a power of 80 %. The required sample size per group was 64 joints. Our final sample sizes exceeded this threshold, providing > 90 % power to detect clinically significant differences between cohorts. All analyses were each completed using SPSS 29 (IBM, New York).