893. Occurrence’s Prediction of Surgical Site Infection in Limb Amputation Surgery

BACKGROUND: A research was conducted between July 2016 and June 2018 in five hospitals in Belo Horizonte, a city of 3,000,000 inhabitants, focused on surgical site infection (SSI) in patients undergoing limb amputation surgery procedure. The main objective is to statistically evaluate such incidences and enable a study of the prediction power of SSI through MLPs (Multilayer Perceptron), a pattern recognition algorithm. METHODS: Data were collected on SSI by the Hospital Infection Control Committees (CCIH) of the hospitals involved. The information was forwarded to the NOIS (Nosocomial Infection Study) Project. After data collection, three procedures were performed: a treatment of the database collected for the use of intact samples; a statistical analysis on the profile of the hospitals collected and; an assessment of the predictive power of five types of MLP (Backpropagation Standard, Momentum, Resilient Propagation, Weight Decay, and Quick Propagation) for SSI prediction. MLPs were tested with 3, 5, 7, and 10 hidden layer neurons and a database split for the resampling process (65% or 75% for testing, 35% or 25% for validation). They were compared by measuring AUC (Area Under the Curve - ranging from 0 to 1) presented for each of the configurations. RESULTS: From 969 data, only 507 were intact for analysis. Statistically: in 12.45% there was an incidence of global infection and that in 10.67% of the cases were SSI (among which, 94.6% had to be hospitalized for more than 10 days); patients were hospitalized on average 21 days (from 0 to 141 days); the average duration is 78 minutes (maximum 360 minutes); 53 deaths (a 16.98% death rate in case of SSI). A maximum prediction power of 0.688 was found. CONCLUSION: Despite the loss rate of almost 40% of the database samples due to the presence of noise, it was obtained a relevant sampling to evaluate the profile the hospitals. For the predictive process, although some configurations reached 0.688, which makes promising the use of the automated SSI monitoring framework for patients undergoing limb amputation surgery. To optimize data collection and enable other hospitals to use the SSI prediction tool (available in www.sacihweb.com), two mobile application were developed: one for monitoring the patient in the hospital and the other for post-hospital discharge monitoring. DISCLOSURES: All Authors: No reported disclosures