Single-FiO2 lung modelling with machine learning: a computer simulation incorporating volumetric capnography.

We investigated whether machine learning (ML) analysis of ICU monitoring data incorporating volumetric capnography measurements of mean alveolar PCO2 can partition venous admixture (VenAd) into its shunt and low V/Q components without manipulating the inspired oxygen fraction (FiO2). From a 21-compartment ventilation / perfusion (V/Q) model of pulmonary blood flow we generated blood gas and mean alveolar PCO2 data in simulated scenarios with shunt values from 7.3% to 36.5% and a range of FiO2 settings, indirect calorimetry and cardiac output measurements and acid- base and hemoglobin oxygen affinity conditions. A 'deep learning' ML application, trained and validated solely on single FiO2 bedside monitoring data from 14,736 scenarios, then recovered shunt values in 500 test scenarios with true shunt values 'held back'. ML shunt estimates versus true values (n = 500) produced a linear regression model with slope = 0.987, intercept = -0.001 and R2 = 0.999. Kernel density estimate and error plots confirmed close agreement. With corresponding VenAd values calculated from the same bedside data, low V/Q flow can be reported as VenAd-shunt. ML analysis of blood gas, indirect calorimetry, volumetric capnography and cardiac output measurements can quantify pulmonary oxygenation deficits as percentage shunt flow (V/Q = 0) versus percentage low V/Q flow (V/Q > 0). High fidelity reports are possible from analysis of data collected solely at the operating FiO2.

Hyperlactatemia in diabetic ketoacidosis is common and can be prolonged: lactate time-series from 25 intensive care admissions.

Hyperlactatemia is a documented complication of diabetic ketoacidosis (DKA). Lactate responses during DKA treatment have not been studied and were the focus of this investigation. Blood gas and electrolyte data from 25 DKA admissions to ICU were sequenced over 24 h from the first Emergency Department sample. Hyperlactatemia (> 2 mmol/L) was present in 22 of 25 DKA presentations [mean concentration = 3.2 mmol/L]. In 18 time-series (72%), all concentrations normalized in ≤ 2.6 h (aggregate decay t1/2 = 2.29 h). In the remaining 7 (28%), hyperlactatemia persisted > 12 h. These were females (P = 0.04) with relative anemia (hemoglobin concentrations 131 v 155 g/L; P = 0.004) and lower nadir glucose concentrations (5.2 v 8.0 mmol/L, P = 0.003). Their aggregate glucose decay curve commenced higher (42 mmol/L v 29 mmol/L), descending towards a lower asymptote (8 mmol/L v 11 mmol/L). Tonicity decay showed similar disparities. There was equivalent resolution of metabolic acidosis and similar lengths of stay in both groups. Hyperlactatemia is common in DKA. Resolution is often rapid, but high lactates can persist. Females with high glucose concentrations corrected aggressively are more at risk. Limiting initial hyperglycemia correction to ≥ 11 mmol/L may benefit.