377 Developing a Nomogram to Enhance Point-of-Care Ultrasound Utility in Predicting Small Bowel Obstruction Diagnosis and Outcome

Abstract

Predicting the diagnosis and a need for surgical intervention in patients with small bowel obstruction (SBO) can assist with emergency physicians’ decisions in selecting appropriate management strategy in the ED. We developed and validated a unique nomogram for the prediction of SBO diagnosis based on clinical features and point-of-care ultrasound (POCUS) findings. We used data from a prospective cohort of patients in a single-institution database that consisted of 125 patients with suspected SBO who underwent POCUS evaluation in the ED of a tertiary academic center. Patients’ demographic and clinical data were collected, including age, sex, history of SBO, comorbidities, prior abdominal surgery, and clinician’s pre-test probability of SBO was surveyed prior to POCUS or CT scan results. Data from this cohort were analyzed to develop a nomogram integrating age, sex, comorbidities, prior abdominal surgery, physician’s pre-test probability, and POCUS findings to determine post-test risk of SBO. The associations of relevant clinical and demographical variables and POCUS findings with SBO diagnosis were assessed using Cox proportional hazards regression models. The discriminative ability of the nomogram was tested using C statistics, calibration plots, and Kaplan-Meier curves. The derivation cohort included 125 patients with a median age of 54 years who underwent POCUS for a suspected SBO. One-fourth of the patients (25.6% [32/125]) had SBO, and (10% [12/125]) underwent a surgical intervention. Using a retrospective stepwise selection of clinically important variables with the POCUS results, the final nomogram incorporated four relevant factors for the prediction of SBO: dilated loop of small bowel of at least 2.5 cm (hazard ratio [HR], 0.10; 95% CI, 0.04-0.16; P < .001), positive inter-luminal fluid (HR, 2.10; 95% CI, 0.96-3.24; P < .001), clinician’s high pretest probability (HR, 2.42; 95% CI, 0.36-4.48; P=0.021), and patient age (HR, 0.03; 95% CI, -0.00 to 0.07; P = 0.08). The discriminative ability and calibration of the nomogram revealed good predictive ability as indicated by the C statistic of 0.89 for the SBO diagnosis. A unique nomogram was developed to utilize the function of patient age, POCUS findings of small bowel > 2.5 cm in diameter, inter-luminal fluid, and pretest probability before POCUS to accurately predict the diagnosis of SBO. The nomogram was able to stratify patients into prognostic groups and performed well on internal validation. The nomogram should be prospectively validated in a novel cohort of patients at risk for SBO.