Machine-Learned Algorithms to Predict the Risk of Pneumothorax Requiring Chest Tube Placement after Lung Biopsy.

PURPOSE: To develop a machine-learned algorithm to predict the risk of postlung biopsy pneumothorax requiring chest tube placement (CTP) to facilitate preprocedural decision making, optimize patient care, and improve resource allocation. MATERIALS AND METHODS: This retrospective study collected clinical and imaging features of biopsy samples obtained from patients with lung nodule biopsy and included information from 59 procedures resulting in pneumothorax requiring CTP and randomly selected 67 procedures without CTP (convenience sample). The data were divided into 70 and 30 as training and testing sets, respectively. Conventional machine-learned binary classifiers were explored with preprocedural imaging and clinical data as input features and CTP as the output. RESULTS: There was no single pathognomonic imaging or predictive clinical feature. For the independent test set under the high-specificity mode, a decision tree, logistic regression, and Naïve Bayes classifier achieved accuracies of identifying CTP at 0.79, 0.93, and 0.89 and area under receiver operating curves (AUROCs) of 0.68, 0.76, and 0.82, respectively. Under high-sensitivity mode, a decision tree, logistic regression, and Naïve Bayes achieved accuracies of identifying CTP of 0.60, 0.45, and 0.60 with AUROCs of 0.71, 0.81, and 0.82, respectively. High importance features included lesion character, chronic obstructive pulmonary disease, lesion depth, and age. A coarse decision tree requiring 4 inputs achieved comparable performance as other methods and previous machine learning prediction studies. CONCLUSIONS: The results support the possibility of predicting pneumothorax requiring CTP after biopsy based on an automated decision support, reliant on readily available preprocedural information.

Taking Control of Your Surgery: Impact of a Prehabilitation Program on Major Abdominal Surgery.

BACKGROUND: Surgery is a major physiologic stress comparable to intense exercise. Diminished cardiopulmonary reserve is a major predictor of poor outcomes. Current preoperative workup focuses mainly on identifying risk factors; however, little attention is devoted to improving cardiopulmonary reserve beyond counseling. We propose that patients could be optimized for a "surgical marathon" similar to the preparation of an athlete. STUDY DESIGN: The Michigan Surgical and Health Optimization Program (MSHOP) is a formal prehabilitation program that engages patients in 4 activities before surgery: physical activity, pulmonary rehabilitation, nutritional optimization, and stress reduction. We prospectively collected demographic, intraoperative (first hour), and postoperative data for patients enrolled in MSHOP undergoing major abdominal surgery. Statistical analysis was performed using 2:1 propensity score matching to compare the MSHOP group (n = 40) to emergency (n = 40) and elective, non-MSHOP (n = 76) patients. RESULTS: Overall, 70% of MSHOP patients complied with the program. Age, sex, American Society of Anesthesiologists (ASA) classification, and BMI did not differ significantly between groups. One hour intraoperatively, MSHOP patients showed improved systolic and diastolic blood pressures and lower heart rate (Figure). There was a significant reduction in Clavien-Dindo class 3 to 4 complications in the MSHOP group (30%) compared with the nonprehabilitation (38%) and emergency (48%) groups (p = 0.05). This translated to total hospital charges averaging $75,494 for the MSHOP group, $97,440 for the nonprehabilitation group, and $166,085 for the emergency group (p < 0.001). CONCLUSIONS: Patients undergoing prehabilitation before colectomy showed positive physiologic effects and experienced fewer complications. The average savings of $21,946 per patient represents a significant cost offset for a prehabilitation program, and should be considered for all patients undergoing surgery.

Electroporation-mediated delivery of FER gene enhances innate immune response and improves survival in a murine model of pneumonia.

Previously, we reported that electroporation-mediated (EP) delivery of the FER gene improved survival in a combined trauma-pneumonia model. The mechanism of this protective effect is unknown. In this paper, we performed a pneumonia model in C57/BL6 mice with 500 CFU of Klebsiella pneumoniae. After inoculation, a plasmid encoding human FER was delivered by EP into the lung (PNA/pFER-EP). Survival of FER-treated vs. controls (PNA; PNA/EP-pcDNA) was recorded. In parallel cohorts, bronchial alveolar lavage (BAL) and lung were harvested at 24 and 72 h with markers of infection measured. FER-EP-treated animals reduced bacterial counts and had better 5-day survival compared to controls (80 vs. 20 vs. 25%; p < 0.05). Pre-treatment resulted in 100% survival. With FER, inflammatory monocytes were quickly recruited into BAL. These cells had increased surface expression for Toll-receptor 2 and 4, and increased phagocytic and myeloperoxidase activity at 24 h. Samples from FER electroporated animals had increased phosphorylation of STAT transcription factors, varied gene expression of IL1ß, TNFα, Nrf2, Nlrp3, Cxcl2, HSP90 and increased cytokine production of TNF-α, CCL-2, KC, IFN-γ, and IL-1RA. In a follow-up experiment, using Methicillin-resistant Staphylococcus aureus (MRSA) similar bacterial reduction effects were obtained with FER gene delivery. We conclude that FER overexpression improves survival through STAT activation enhancing innate immunity and accelerating bacterial clearance in the lung. This constitutes a novel mechanism of inflammatory regulation with therapeutic potential in the setting of hospital-acquired pneumonia.

Molecular Characterization of Hypoxic Alveolar Epithelial Cells After Lung Contusion Indicates an Important Role for HIF-1α.

OBJECTIVE: To understand the fate and regulation of hypoxic type II alveolar epithelial cells (AECs) after lung contusion (LC). BACKGROUND: LC due to thoracic trauma is a major risk factor for the development of acute respiratory distress syndrome. AECs have recently been implicated as a primary driver of inflammation in LC. The main pathological consequence of LC is hypoxia, and a key mediator of adaptation to hypoxia is hypoxia-inducible factor (HIF)-1. We have recently published that HIF-1α is a major driver of acute inflammation after LC through type II AEC. METHODS: LC was induced in wild-type mice (C57BL/6), luciferase-based hypoxia reporter mice (ODD-Luc), and HIF-1α conditional knockout mice. The degree of hypoxia was assessed using hypoxyprobe and in vivo imaging system. The fate of hypoxic AEC was evaluated by luciferase dual staining with caspases-3 and Ki-67, terminal deoxynucleotidyl transferase dUTP nick end labeling, and flow cytometry with ApoStat. NLRP-3 expression was determined by western blot. Laser capture microdissection was used to isolate AECs in vivo, and collected RNA was analyzed by Q-PCR for HIF-related pathways. RESULTS: Global hypoxia was present after LC, but hypoxic foci were not uniform. Hypoxic AECs preferentially undergo apoptosis. There were significant reductions in NLRP-3 in HIF-1α conditional knockout mice. The expression of proteins involved in HIF-related pathways and inflammasome activation were significantly increased in hypoxic AECs. CONCLUSIONS: These are the first in vivo data to identify, isolate, and characterize hypoxic AECs. HIF-1α regulation through hypoxic AECs is critical to the initiation of acute inflammation after LC.