Stephen Edward Rees

Objective: Analysis delay and gas contamination can affect the accuracy of blood measurements. This study uses a mathematical model of blood acid-base chemistry and gas in the sample tubes to calculate values of pH, partial pressures of carbon dioxide (pCO₂), partial pressure of oxygen (pO₂), oxygen saturation in whole blood (SO₂), glucose, and lactate at sample time from measurements with delayed analysis and gas contamination.Methods: Data were analyzed from two published studies. Study 1: Samples were obtained from 30 critically ill patients in standard blood gas syringes and analyzed after 0, 36, 54, 72, 90, 108, 126, 144, 162, and 180 min. Study 2: Samples were taken from 20 healthy participants in standard blood gas syringes and vacuum tubes (2 mL and 4 mL) and analyzed after 0, 20, and 90 min. Calculated values from the mathematical model were compared to measured values at sample time.Results: For delays of up to 90 min, the accuracy (mean) and precision (standard deviation (SD)) values calculated at the sample time using syringes and 4-mL vacuum tubes remained within clinically acceptable limits when compared to measured values, with the exception of SO₂ in vacuum tubes. Values represent the mean difference ± standard deviation between calculated and measured values. For syringes, the results were as follows: pH = −0.004 ± 0.011, pCO₂ = 0.08 ± 0.18 kPa, pO₂ = 0.05 ± 0.34 kPa, SO₂ = 0.39 ± 2.21%, glucose = 0.07 ± 0.35 mmol/L, and lactate = 0.13 ± 0.22 mmol/L. For 4-mL vacuum tubes, the results were as follows: pH = 0.006 ± 0.007, pCO₂ = −0.07 ± 0.11 kPa, pO₂ = −0.37 ± 0.34 kPa, SO₂ = −7.79 ± 4.95%, glucose = 0.01 ± 0.11 mmol/L, and lactate = −0.00 ± 0.20 mmol/L. In addition, 2-mL vacuum tubes had poorer accuracy and precision values than syringes and 4-mL vacuum tubes in a subset of cases.Conclusion: This study has shown that a mathematical model can accurately and precisely calculate blood values at sample time, even following delayed analysis, using both standard blood gas syringes and selected vacuum tubes. This method may have clinical applications in improving the logistics of blood sampling and analysis.

Objectives: Measurements of acid-base status are performed quickly after blood sampling avoiding errors. This necessitates rapid sample transport which can be problematic. This study measures blood sampled in critically ill patients over 180 min and proposes a mathematical physio-chemical model to simulate changes. Methods: Eleven blood samples were taken from 30 critically ill patients and measured at baseline (2 samples) and 36, 54, 72, 90, 108, 126, 144, 162, and 180 min. A mathematical model was proposed including red blood cell metabolism, carbon dioxide diffusion, electrolyte distribution and water transport. This model was used to simulate values of plasma pH, pCO₂, pO₂, SO₂, glucose, lactate, Na⁺ and Cl⁻ during analysis delay. Simulated and measured values were compared using Bland-Altman and correlation analysis, and goodness of model fits evaluated with chi-squared. Results: The mathematical model provided a good fit to data in 29 of 30 patients with no significant differences (p>0.1) between simulated and measured plasma values. Differences were (bias±SD): pH 0.000 ± 0.012, pCO₂ 0.00 ± 0.24 kPa, lactate −0.10 ± 0.23 mmol/L, glucose 0.00 ± 0.34 mmol/L, Cl⁻ −0.2 ± 1.21 mmol/L, Na⁺ 0.0 ± 1.0 mmol/L, pO₂ 0.0 ± 0.44 kPa, SO₂ −0.6 ± 5.5 %, with these values close to manufacturers’ measurement errors. All linear correlations had R²>0.86. Simulations of pH, PCO₂, glucose and lactate could be performed from baseline values without patient specific parameters. Conclusions: This paper illustrates that analysis delay can be accurately simulated with a mathematical model of physio-chemistry. While further evaluation is necessary, this may indicate a role for this model in clinical practice to simulate analysis delay.

This study investigated the clinical and physiological response to use of the BEACON Caresystem, a bedside open-loop decision support system providing advice to guide clinicians when weaning patients from invasive mechanical ventilation. Multicenter prospective study conducted in five adult intensive care units in the UK. Following screening and assent, intubated patients mechanically ventilated for > 24 h were randomized to intervention or usual care. Intervention consisted of application of the BEACON Caresystem’s advice on tidal volume/inspiratory pressure, inspired oxygen, respiratory rate and PEEP. Usual care was defined as local clinical practice. The primary outcome was duration of mechanical ventilation. Secondary outcomes quantified prolonged intubation and survival; adverse events; ventilator settings and physiological state; time spent in ventilator modes; links to other therapy; the frequency of advice utilization and time spent outside normal physiological limits. The study was terminated early with a total of 112 patients included. Fifty-four were randomised to the intervention arm and fifty-eight to usual care. The study was underpowered and no significant differences were seen in duration of mechanical ventilation (p = 0.773), prolonged intubation or survival. Intervention arm patients had lower rates of adverse events (p = 0.016), including fewer hypoxaemic events (p = 0.008) and lower values of PEEP (p = 0.030) and tidal volume (p = 0.042). Values of peak inspiratory pressure and pressure support were reduced but at the boarder of statistical significance (p = 0.104, p = 0.093, respectively). No differences were seen for time in ventilator mode or other therapy. Advice presented by the decision support system was applied at the beside an average of 88% of occasions, with a significantly increased number of changes only in inspired oxygen fraction. No significant differences were seen in time spent outside physiological limits. This study investigated the use of the BEACON Caresystem, an open loop clinical decision support system providing advice on ventilator settings. It was terminated early, with no significant difference shown in duration of mechanical ventilation, the primary outcome. Application of advice indicated potential for fewer adverse events and improved physiological status. (Trial registration ClinicalTrials.gov under NCT03249623. Registered 22nd June 2017)

Background: Acute respiratory distress syndrome (ARDS) is highly heterogeneous, both in its clinical presentation and in the patient’s physiological responses to changes in mechanical ventilator settings, such as PEEP. This study investigates the clinical efficacy of a physiological model-based ventilatory decision support system (DSS) to personalize ventilator therapy in ARDS patients.Methods: This international, multicenter, randomized, open-label study enrolled patients with ARDS during the COVID-19 pandemic. Patients were randomized to either receive active advice from the DSS (intervention) or standard care without DSS advice (control). The primary outcome was to detect a reduction in average driving pressure between groups. Secondary outcomes included several clinically relevant measures of respiratory physiology, ventilator-free days, time from control mode to support mode, number of changes in ventilator settings per day, percentage of time in control and support mode ventilation, ventilation- and device-related adverse events, and the number of times the advice was followed.Results: A total of 95 patients were randomized in this study. The DSS showed no significant effect on average driving pressure between groups. However, patients in the intervention arm had a statistically improved oxygenation index when in support mode ventilation (−1.41, 95% CI: −2.76, −0.08; p = 0.0370). Additionally, the ventilatory ratio significantly improved in the intervention arm for patients in control mode ventilation (−0.63, 95% CI: −1.08, −0.17, p = 0.0068). The application of the DSS led to a significantly increased number of ventilator changes for pressure settings and respiratory frequency.Conclusion: The use of a physiological model-based decision support system for providing advice on mechanical ventilation in patients with COVID-19 and non-COVID-19 ARDS showed no significant difference in driving pressure as a primary outcome measure. However, the application of approximately 60% of the DSS advice led to improvements in the patient’s physiological state.Clinical trial registration: clinicaltrials.gov, NCT04115709.

Background: Postoperative pulmonary complications can increase hospital length of stay, postoperative morbidity, and mortality. Although many factors can increase the risk of postoperative pulmonary complications, it is not known whether intraoperative ventilation/perfusion (V/Q) mismatch can be associated with an increased risk of postoperative pulmonary complications after major noncardiac surgery. Methods: This study enrolled patients undergoing general anesthesia for noncardiac surgery and evaluated intraoperative V/Q distribution using the automatic lung parameter estimator technique. The assessment was done after anesthesia induction, after 1 h from surgery start, and at the end of surgery. Demographic and procedural information were collected, and intraoperative ventilatory and hemodynamic parameters were measured at each timepoint. Patients were followed up for 7 days after surgery and assessed daily for postoperative pulmonary complication occurrence. Results: The study enrolled 101 patients with a median age of 71 [62 to 77] years, a body mass index of 25 [22.4 to 27.9] kg/m², and a preoperative Assess Respiratory Risk in Surgical Patients in Catalonia (ARISCAT) score of 41 [34 to 47]. Of these patients, 29 (29%) developed postoperative pulmonary complications, mainly acute respiratory failure (23%) and pleural effusion (11%). Patients with and without postoperative pulmonary complications did not differ in levels of shunt at T1 (postoperative pulmonary complications: 22.4% [10.4 to 35.9%] vs. no postoperative pulmonary complications:19.3% [9.4 to 24.1%]; P = 0.18) or during the protocol, whereas significantly different levels of high V/Q ratio were found during surgery (postoperative pulmonary complications: 13 [11 to 15] mmHg vs. no postoperative pulmonary complications: 10 [8 to 13.5] mmHg; P = 0.007) and before extubation (postoperative pulmonary complications: 13 [11 to 14] mmHg vs. no postoperative pulmonary complications: 10 [8 to 12] mmHg; P = 0.006). After adjusting for age, ARISCAT, body mass index, smoking, fluid balance, anesthesia type, laparoscopic procedure and surgery duration, high V/Q ratio before extubation was independently associated with the development of postoperative pulmonary complications (odds ratio, 1.147; 95% CI, 1.021 to 1.289; P = 0.02). The sensitivity analysis showed an E-value of 1.35 (CI, 1.11). Conclusions: In patients with intermediate or high risk of postoperative pulmonary complications undergoing major noncardiac surgery, intraoperative V/Q mismatch is associated with the development of postoperative pulmonary complications. Increased high V/Q ratio before extubation is independently associated with the occurrence of postoperative pulmonary complications in the first 7 days after surgery. Editor’s Perspective: What We Already Know about This Topic: What This Article Tells Us That Is New:

Pulse pressure variation (PPV) is a well-established method for predicting fluid responsiveness in mechanically ventilated patients. The predictive accuracy is, however, disputed for ventilation with low tidal volume (V_T) or low heart-rate-to-respiratory-rate ratio (HR/RR). We investigated the effects of V_T and RR on PPV and on PPV’s ability to predict fluid responsiveness. We included patients scheduled for open abdominal surgery. Prior to a 250 ml fluid bolus, we ventilated patients with combinations of V_T from 4 to 10 ml kg⁻¹ and RR from 10 to 31 min⁻¹. For each of 10 RR-V_T combinations, PPV was derived using both a classic approach and a generalized additive model (GAM) approach. The stroke volume (SV) response to fluid was evaluated using uncalibrated pulse contour analysis. An SV increase > 10% defined fluid responsiveness. Fifty of 52 included patients received a fluid bolus. Ten were fluid responders. For all ventilator settings, fluid responsiveness prediction with PPV was inconclusive with point estimates for the area under the receiver operating characteristics curve between 0.62 and 0.82. Both PPV measures were nearly proportional to V_T. Higher RR was associated with lower PPV. Classically derived PPV was affected more by RR than GAM-derived PPV. Correcting PPV for V_T could improve PPV’s predictive utility. Low HR/RR has limited effect on GAM-derived PPV, indicating that the low HR/RR limitation is related to how PPV is calculated. We did not demonstrate any benefit of GAM-derived PPV in predicting fluid responsiveness.Trial registration: ClinicalTrials.gov, reg. March 6, 2020, NCT04298931.

Rationale: The acute respiratory distress syndrome (ARDS) shows significant heterogeneity in responsiveness to changes in mechanical ventilation and lacks personalisation. Objectives: Investigate the clinical efficacy of a physiologic-based ventilatory decision support system (DSS) on ARDS patients. Methods: An international, multi-centre, randomized, open-label study enrolling patients with ARDS during the COVID-19 pandemic. The primary outcome was to detect a reduction in average driving pressure between groups. Secondary outcomes included several clinically relevant measures of respiratory physiology, ventilator free days; time from control mode to support mode; number of changes in ventilator settings per day; percentage of time in control and support mode ventilation; ventilation related and device related adverse events; and number of times the advice is followed. Measurements and Main Results. 95 patients were randomized to this study. The DSS showed was no effect in the average driving pressure between arms. Patients in the intervention arm had statistically improved oxygenation index when in support mode ventilation (−1.41, 95% CI: −2.76, −0.08; p=0.0370). Ventilatory ratio was also significantly improved in the intervention arm for patients in control mode ventilation (−0.63, 95% CI: −1.08, −0.17, p= 0.0068). The application of the DSS resulted in a significantly increased number of ventilator changes for pressure settings and respiratory frequency. Conclusions: The application of a physiological model-based decision support system for advice on mechanical ventilation in patients with COVID-19 and non-COVID-19 ARDS showed that application of about 60% of advice improved physiological state, despite no significant difference in driving pressure as a primary outcome measure.