Dynamic inflation prevents and standardized lung recruitment reverts volume loss associated with percutaneous tracheostomy during volume control ventilation: results from a Neuro-ICU population.

To determine how percutaneous tracheostomy (PT) impacts on respiratory system compliance (Crs) and end-expiratory lung volume (EELV) during volume control ventilation and to test whether a recruitment maneuver (RM) at the end of PT may reverse lung derecruitment. This is a single center, prospective, applied physiology study. 25 patients with acute brain injury who underwent PT were studied. Patients were ventilated in volume control ventilation. Electrical impedance tomography (EIT) monitoring and respiratory mechanics measurements were performed in three steps: (a) baseline, (b) after PT, and (c) after a standardized RM (10 sighs of 30 cmH2O lasting 3 s each within 1 min). End-expiratory lung impedance (EELI) was used as a surrogate of EELV. PT determined a significant EELI loss (mean reduction of 432 arbitrary units p = 0.049) leading to a reduction in Crs (55 ± 13 vs. 62 ± 13 mL/cmH2O; p < 0.001) as compared to baseline. RM was able to revert EELI loss and restore Crs (68 ± 15 vs. 55 ± 13 mL/cmH2O; p < 0.001). In a subgroup of patients (N = 8, 31%), we observed a gradual but progressive increase in EELI. In this subgroup, patients did not experience a decrease of Crs after PT as compared to patients without dynamic inflation. Dynamic inflation did not cause hemodynamic impairment nor raising of intracranial pressure. We propose a novel and explorative hyperinflation risk index (HRI) formula. Volume control ventilation did not prevent the PT-induced lung derecruitment. RM could restore the baseline lung volume and mechanics. Dynamic inflation is common during PT, it can be monitored real-time by EIT and anticipated by HRI. The presence of dynamic inflation during PT may prevent lung derecruitment.

Development of machine learning models to predict RT-PCR results for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in patients with influenza-like symptoms using only basic clinical data.

BACKGROUND: Reverse Transcription-Polymerase Chain Reaction (RT-PCR) for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) diagnosis currently requires quite a long time span. A quicker and more efficient diagnostic tool in emergency departments could improve management during this global crisis. Our main goal was assessing the accuracy of artificial intelligence in predicting the results of RT-PCR for SARS-COV-2, using basic information at hand in all emergency departments. METHODS: This is a retrospective study carried out between February 22, 2020 and March 16, 2020 in one of the main hospitals in Milan, Italy. We screened for eligibility all patients admitted with influenza-like symptoms tested for SARS-COV-2. Patients under 12 years old and patients in whom the leukocyte formula was not performed in the ED were excluded. Input data through artificial intelligence were made up of a combination of clinical, radiological and routine laboratory data upon hospital admission. Different Machine Learning algorithms available on WEKA data mining software and on Semeion Research Centre depository were trained using both the Training and Testing and the K-fold cross-validation protocol. RESULTS: Among 199 patients subject to study (median [interquartile range] age 65 [46-78] years; 127 [63.8%] men), 124 [62.3%] resulted positive to SARS-COV-2. The best Machine Learning System reached an accuracy of 91.4% with 94.1% sensitivity and 88.7% specificity. CONCLUSION: Our study suggests that properly trained artificial intelligence algorithms may be able to predict correct results in RT-PCR for SARS-COV-2, using basic clinical data. If confirmed, on a larger-scale study, this approach could have important clinical and organizational implications.