Category: Ankle Introduction/Purpose: Treatment for patients with severe ankle instability or failed previous ankle stabilization is not well defined. Results after ankle stabilization techniques involving non-anatomic reconstruction have historically been suboptimal, and newer techniques have limited presence in the literature. The purpose of this study is to report clinical and radiographic outcomes after modified anatomic lateral ligament reconstruction using hamstring auto- or allograft in patients with severe ankle instability or failed previous ankle stabilization. A novel technique for ligament reconstruction is also presented that is hypothesized to restore functional and radiographic ankle stability. Methods: A retrospective chart review was performed on all patients that had undergone modified anatomic lateral ligament reconstruction by a single surgeon between 2011 and 2015 with at least 6 months follow-up. Indications for modified anatomic reconstruction included failure of previous ankle stabilization or severe laxity with greater than 20 degrees of talar tilt or anterior drawer greater than 15 mm on stress radiographs. Patients completed routine pre- and post-operative functional outcome scores including Foot and Ankle Outcome Score (FAOS), Short Form 12 Health Survey (SF-12), and Visual Analog Scale (VAS). Patients underwent pre- and post-operative stress radiographs using the Telos Stress Device (Hungen, Germany). Thirty-four patients (35 ankles) were included with average follow-up of 26.7 months. Average age was 34.2 years, and there were 29 female patients and 5 male patients. Hamstring autograft was utilized in 31 ankles and hamstring allograft in 4 ankles. Results: Indications for surgery included failure of previous ankle stabilization in 13 patients and severe ankle instability in 22 patients. All functional outcome scores improved; VAS increased from 5.3 to 1.0 points (p < 0.0001), SF-12 increased from 64 to 89 points (p < 0.0001), and FAOS scores increased in all categories (p < 0.05). Radiographic measurements of instability also improved; anterior drawer decreased by 3 mm (p = 0.0002) and talar tilt decreased by 11 degrees (p < 0.0001) (see Table 1). One patient (3%) returned to the operating room for removal of hardware after over 2 years. There were 4 patients (12%) with delayed wound healing, 2 patients (6%) with neurologic complications, and 2 patients (6%) with venous thromboembolic events. Conclusion: Patients demonstrated significant improvement in functional outcome scores as well as radiographic measures of ankle stability following modified anatomic lateral ligament reconstruction in a population with severe or recurrent instability. This is the largest series to date of ankle ligament reconstruction using autograft, and it is associated with high patient satisfaction, reduced pain, improved objective stability, and low morbidity. Further study is warranted to develop well-defined guidelines on the management of patients with severe or recurrent instability.