RESUMEN
BACKGROUND/PURPOSE: To identify factors that dim the efficacy of ultrasound guided saline enema (USGSE) and to design a mathematical model for predicting the probability of success of USGSE. METHODS: Retrospective review of patients admitted with the diagnosis of ileocolic intussusception from 2009 to 2014. Demographics, clinical and sonographic data were reviewed. RESULTS: 116 first episodes of ileocolic intussusceptions. 109 USGSE attempts were analyzed. Composite reduction rate was 77%. A significant relationship was found between initial location of the intussusception, free peritoneal fluid (OR=0.329, 95% CI: 0.124-0.875), negative Doppler signal and sonographic signs of intestinal occlusion and unsuccessful USGSE. Initial location beyond the splenic angle was an independent risk factor for USGSE failure (OR=0.053, 95% CI: 0.005-0.534). A predictive model based on onset of symptoms, free peritoneal fluid and intussusception location was a reliable tool for prediction (AUC 0.63, 95% CI: 0.53-0.817). Assuming that a patient with less than 75.3% chance of USGSE success is going to fail, we would identify more than 80.9% of the real failures. CONCLUSIONS: This predictive model could be a filter selection for the patients at risk of USGSE failure and therefore candidates to further imaging investigations or referral to a surgical unit. LEVEL OF EVIDENCE: III.