Preoperative PROMIS Scores as a Predictive Method for Postoperative Success in Posterior Tibial Tendon Dysfunction Surgery

Abstract

Category: Hindfoot Introduction/Purpose: Patient Reported Outcomes Instrumentation System (PROMIS) pain interference (PI) and physical function (PF) scales were recently validated for foot & ankle surgery. Few studies have explored PROMIS in advanced posterior tibial tendon dysfunction (PTTD) surgery. We examined the change in PROMIS scores after PTTD reconstruction surgery and aimed to determine whether preoperative PROMIS scores, radiographic data, and demographic data can model and predict surgical improvement. Methods: PF and PI scores were prospectively obtained on 215 patients between November 2013 and May 2017. Of the 34 patients who had PTTD surgery, 25 patients were included with a minimum follow-up of 7-months (mean 21.8). Paired t-tests and multivariable linear regression models tested the difference in PROMIS scores pre- to postoperatively. Minimal clinically important differences (MCID) and receiver operator curve (ROC) analysis determined the accuracy of using preoperative scores in predicting postoperative change. The MCID for PROMIS PI and PF was defined as half of its respective standard deviation. Multivariable linear regressions with preoperative PROMIS scores, changes in radiographic imaging, and BMI were created to model change in PROMIS scores. Results: There was a significant difference in pre- to postoperative mean PI PROMIS scores (p = 0.0016, average change = -7.16). Mean PF scores improved postoperatively, but not significantly (p = .0595, average change = +4.17). Improvement in PI remained significant (p= 0.02) after adjusting for significant predictors (education level and diabetes status). Multivariable regression models utilizing preoperative PROMIS scores, radiographic variables, and BMI were unable to predict changes in PF and PI scores (p > 0.05). Using pre-determined MCID criteria, 68% achieved surgical success for PI (defined as change of at least -3.45) and 52% achieved surgical success for PF (defined as change of at least +3.00). However, preoperative PI (p=0.310) and PF (p=0.054) scores were unable to significantly predict MCID surgical success or failure. Conclusion: After major PTTD reconstruction, PI scores improve to significance but PF scores may not. After PTTD reaches an advanced stage, surgery may provide significant pain relief and improvement in physical function and radiographic parameters, but it may be difficult to predict clinical improvement based solely on preoperative PROMIS scores. Realistic expectations of improvements in pain and functionality should be maintained postoperatively in advanced PTTD surgery.