Preoperative Prediction of Postoperative Physical Function in Infected Diabetic Foot Ulcer Patients

Abstract

Category: Diabetes Introduction/Purpose: Diabetic foot ulcers (DFU) is a prevalent problem that can lead to devastating results such as limb loss if left untreated. Nevertheless, the prolonged treatment course can limit the patient’s overall function and quality of life. Utilization of Patient-Reported Outcomes Measurement Information System (PROMIS) in Orthopaedic practice has previously shown that preoperative PROMIS scores can predict postoperative outcomes in foot and ankle surgeries. However, PROMIS assessment has not been used to determine the impact of surgical treatment for DFU on patients’ physical function. We sought to investigate the impact of preoperative PROMIS scores (Physical Function (PF), Pain Interference (PI), Depression (D)), demographic and laboratory values on postoperative PF in this unique patient population. Methods: From an academic orthopaedic surgeon’s practice, we identified infected DFU patients who underwent surgical interventions between February 2015 and November 2018 using ICD-10 code E11.621 (n=240). Patients with at least 3 consecutive visits, 3 month minimum post-surgical follow up and completed PROMIS Computer Adaptive Testing (CAT) assessments for each visit were included (n=92). Demographic data, BMI, medical comorbidities, Hemoglobin A1C, procedure performed, and wound healing status were collected. Amputation level was categorized as the following: 0 = irrigation & debridement (I&D) (n=39), 1 = forefoot amputations (n=46), 2 = mid/hindfoot amputations (n=14), 3 = Syme or above amputations (n=12). Uni- and multivariate analysis were performed to identify factors affecting the post-operative PF within the cohort. Spearman’s rank correlation coefficient, Chi-Squared tests and multidimensional modelling were applied to all variables’ pre-operative and post-operative time points. Based on the results, we formulated a numeric equation to predict post-surgical PROMIS PF. Results: The mean age was 60.5 (33-96) and 4.7 (3-12) months follow up. Mean preoperative PF, PI, and D changed from 34.4, 58.7, 51.4 to postoperative 36.1, 58.8, 51.1, respectively (ΔPF = 1.7, ΔPI=0.1, ΔD = -0.3). Preoperative PF (p < 0.01), PI (p < 0.01), depression (p < 0.01), chronic renal failure (p < 0.02) and amputation level (p < 0.04) showed significant univariate correlation with post-operative PF. Multivariate model (r = 0.6) revealed postoperative PF is predicted by initial PF (p = 0.094), depression (p= 0.008), amputation level (p = 0.002), and wound healing status (p = 0.001). The model had greater prediction power than the best univariate association (Δr = +0.17). Follow up length was not significant (p = 0.08). Conclusion: This study demonstrates that preoperative PROMIS scores combined with clinical factors can predict postoperative PF in DFU patients. Postoperative PF is predicted by: PFlongest_FU = 45.4 +0.20 PFinitial -0.21 Dinitial -6.1 (Heal =1) -2.9 (Amputation Category, 1-3). Additional diseased states not captured in this study and psychosocial variables may improve prediction power of the multivariate model. 70% of the patients’ initial PF were 1 to 2 standard deviations below the US population (n = 49; 28). Therefore, the reported model may serve as a valuable tool for patient education, setting expectations and post-surgical PF prediction in infected DFU patients.