Background Hospital readmission within 30-days of an index hospitalization is receiving increased scrutiny as a marker of poor quality patient care. This study identifies factors associated with 30-day readmission following General Surgery procedures. Study Design Using standard National Surgical Quality Improvement Project (NSQIP) protocol, preoperative, intraoperative, and postoperative outcomes were collected on patients undergoing inpatient General Surgery procedures at a single academic center between 2009 and 2011. Data were merged with our institutional clinical data warehouse to identify unplanned 30-day readmissions. Demographics, comorbidities, type of procedure, postoperative complications, and ICD-9 coding data were reviewed for patients who were readmitted. Univariate and multivariate analysis was utilized to identify risk factors associated with 30-day readmission. Results 1442 General Surgery patients were reviewed. 163 (11.3%) were readmitted within 30 days of discharge. The most common reasons for readmission were gastrointestinal complaint/complication (27.6%), surgical infection (22.1%), and failure to thrive/malnutrition (10.4%). Comorbidities associated with risk of readmission included disseminated cancer, dyspnea, and preoperative open wound (p<0.05 for all variables). Surgical procedures associated with higher rates of readmission included pancreatectomy, colectomy, and liver resection. Postoperative occurrences leading to increased risk of readmission were blood transfusion, postoperative pulmonary complication, wound complication, sepsis/shock, urinary tract infection, and vascular complications. Multivariable analysis demonstrates that the most significant independent risk factor for readmission is the occurrence of any postoperative complication (OR 4.20, 95% CI 2.89–6.13). Conclusions Risk factors for readmission after General Surgery procedures are multi-factorial; however, postoperative complications appear to drive readmissions in surgical patients. Taking appropriate steps to minimize postoperative complications will decrease postoperative readmissions.

Chest CT is emerging as a valuable diagnostic tool for clinical management of COVID-19 associated lung disease. Artificial intelligence (AI) has the potential to aid in rapid evaluation of CT scans for differentiation of COVID-19 findings from other clinical entities. Here we show that a series of deep learning algorithms, trained in a diverse multinational cohort of 1280 patients to localize parietal pleura/lung parenchyma followed by classification of COVID-19 pneumonia, can achieve up to 90.8% accuracy, with 84% sensitivity and 93% specificity, as evaluated in an independent test set (not included in training and validation) of 1337 patients. Normal controls included chest CTs from oncology, emergency, and pneumonia-related indications. The false positive rate in 140 patients with laboratory confirmed other (non COVID-19) pneumonias was 10%. AI-based algorithms can readily identify CT scans with COVID-19 associated pneumonia, as well as distinguish non-COVID related pneumonias with high specificity in diverse patient populations.

No abstract