Prediction of Prolonged Ventilation after Coronary Artery Bypass Grafting: Data from an Artificial Neural Network.

OBJECTIVES: The need for mechanical ventilation 24 hours after coronary artery bypass grafting (CABG) is considered a morbidity by the Society of Thoracic Surgeons. The purpose of this investigation was twofold: to identify simple preoperative patient factors independently associated with prolonged ventilation and to optimize prediction and early identification of patients prone to prolonged ventilation using an artificial neural network (ANN). METHODS: Using the institutional Adult Cardiac Database, 738 patients who underwent CABG since 2005 were reviewed for preoperative factors independently associated with prolonged postoperative ventilation. Prediction of prolonged ventilation from the identified variables was modeled using both "traditional" multiple logistic regression and an ANN. The two models were compared using Pearson r2 and area under the curve (AUC) parameters. RESULTS: Of 738 included patients, 14% (104/738) required mechanical ventilation ≥ 24 hours postoperatively. Upon multivariate analysis, higher body-mass index (BMI; odds ratio [OR] 1.10 per unit, P < 0.001), lower ejection fraction (OR 0.97 per %, P = 0.01) and use of cardiopulmonary bypass (OR 2.59, P = 0.02) were independently predictive of prolonged ventilation. The Pearson r2 and AUC of the multivariate nominal logistic regression model were 0.086 and 0.698 ± 0.05, respectively; analogous statistics of the ANN model were 0.159 and 0.732 ± 0.05, respectively.BMI, ejection fraction and cardiopulmonary bypass represent three simple factors that may predict prolonged ventilation after CABG. Early identification of these patients can be optimized using an ANN, an emerging paradigm for clinical outcomes modeling that may consider complex relationships among these variables.

Development and validation of a prediction model for diagnosing blood stream infections in febrile, non-neutropenic children with cancer.

BACKGROUND: Pediatric oncology patients are at increased risk for blood stream infections (BSI). Risk in the absence of severe neutropenia (absolute neutrophil count [ANC] ≥500/µl) is not well defined. PROCEDURE: In a retrospective cohort of febrile (temperature ≥38.0° for >1 hr or ≥38.3°) pediatric oncology patients with ANC ≥500/µl, a diagnostic prediction model for BSI was constructed using logistic regression modeling and the following candidate predictors: age, ANC, absolute monocyte count, body temperature, inpatient/outpatient presentation, sex, central venous catheter type, hypotension, chills, cancer diagnosis, stem cell transplant, upper respiratory symptoms, and exposure to cytarabine, anti-thymocyte globulin, or anti-GD2 antibody. The model was internally validated with bootstrapping methods. RESULTS: Among 932 febrile episodes in 463 patients, we identified 91 cases of BSI. Independently significant predictors for BSI were higher body temperature (Odds ratio [OR] 2.36 P < 0.001), tunneled external catheter (OR 13.79 P < 0.001), peripherally inserted central catheter (OR 3.95 P = 0.005), elevated ANC (OR 1.19 P = 0.024), chills (OR 2.09 P = 0.031), and hypotension (OR 3.08 P = 0.004). Acute lymphoblastic leukemia diagnosis (OR 0.34 P = 0.026), increased age (OR 0.70 P = 0.049), and drug exposure (OR 0.08 P < 0.001) were associated with decreased risk for BSI. The risk prediction model had a C-index of 0.898; after bootstrapping adjustment for optimism, corrected C-index 0.885. CONCLUSIONS: We developed a diagnostic prediction model for BSI in febrile pediatric oncology patients without severe neutropenia. External validation is warranted before use in clinical practice. Pediatr Blood Cancer 2015;62:262-268. © 2014 Wiley Periodicals, Inc.