Environmental and Economic Impact of Using a Higher Efficiency Ventilator and Vaporizer During Surgery Under General Anesthesia: A Randomized Controlled Prospective Cohort.

Background Compared to traditional breathing circuits, low-volume anesthesia machines utilize a lower-volume breathing circuit paired with needle injection vaporizers that supply volatile agents into the circuit mainly during inspiration. We aimed to assess whether or not low-volume anesthesia machines, such as the Maquet Flow-i C20 anesthesia workstation (MQ), deliver volatile anesthetics more efficiently than traditional anesthesia machines, such as the GE Aisys CS2 anesthesia machine (GE), and, secondarily, whether this was in a meaningful economic or environmentally conscious way. Methodology Participants enrolled in the study (Institutional Review Board Identifier: 2014-1248) met the following inclusion criteria: 18-65 years old, scheduled for surgery requiring general anesthesia at the University of California Irvine Health, and expected to receive sevoflurane for the duration of the procedure. Exclusion criteria included age <18 years old, a history of chronic obstructive pulmonary disorder, cardiovascular disease, sevoflurane sensitivity, body mass index >30 kg/m2, American Society of Anesthesiologists >2, pregnancy, or surgery scheduled <120 minutes. We calculated the total amount of sevoflurane delivered and consumption rates during induction and maintenance periods and compared the groups using one-sided parametric testing (Student's t-test). There was no suspicion that the low-volume circuit could use more sevoflurane and that the outcome did not answer our research question. One-sided testing allowed for more power to be more certain of smaller differences in our results. Results In total, 103 subjects (MQ: n = 52, GE: n = 51) were analyzed. Seven subjects were lost to attrition of different types. Overall, the MQ group consumed significantly less sevoflurane (95.5 ± 49.3 g) compared to the GE group (118.3 ± 62.4 g) (p = 0.043), corresponding to an approximately 20% efficiency improvement in overall agent delivery. When accounting for the fresh gas flow setting, agent concentration, and length of induction, the MQ delivered the volatile agent at a significantly lower rate compared to the GE (7.4 ± 3.2 L/minute vs. 9.1 ± 4.1 L/minute; p = 0.017). Based on these results, we estimate that the MQ can save an estimated average of $239,440 over the expected 10-year machine lifespan. This 20% decrease in CO2 equivalent emissions corresponds to 201 metric tons less greenhouse gas emissions over a decade compared to the GE, which is equivalent to 491,760 miles driven by an average passenger vehicle or 219,881 pounds of coal burned. Conclusions Overall, our results from this study suggest that the MQ delivers statistically significantly less (~20%) volatile agent during routine elective surgery using a standardized anesthetic protocol and inclusion/exclusion criteria designed to minimize any patient or provider heterogeneity effects on the results. The results demonstrate an opportunity for economic and environmental benefits.

Detection of arterial pressure waveform error using machine learning trained algorithms.

In critically ill and high-risk surgical room patients, an invasive arterial catheter is often inserted to continuously measure arterial pressure (AP). The arterial waveform pressure measurement, however, may be compromised by damping or inappropriate reference placement of the pressure transducer. Clinicians, decision support systems, or closed-loop applications that rely on such information would benefit from the ability to detect error from the waveform alone. In the present study we hypothesized that machine-learning trained algorithms could discriminate three types of transducer error from accurate monitoring with receiver operator characteristic (ROC) curve areas greater than 0.9. After obtaining written consent, patient arterial line waveform data was collected in the operating room in real-time during routine surgery requiring arterial pressure monitoring. Three deliberate error conditions were introduced during monitoring: Damping, Transducer High, and Transducer Low. The waveforms were split up into 10 s clips that were featurized. The data was also either calibrated against the patient's own baseline or left uncalibrated. The data was then split into training and validation sets, and machine-learning algorithms were run in a Monte-Carlo fashion on the training data with variable sized training sets and hyperparameters. The algorithms with the highest balanced accuracy were pruned, then the highest performing algorithm in the training set for each error state (High, Low, Damped) for both calibrated and uncalibrated data was finally tested against the validation set and the ROC and precision-recall curve area-under the curve (AUC) calculated. 38 patients were enrolled in the study with a mean age of 52 ± 15 years. A total of 40 h of monitoring time was recorded with approximately 120,000 heart beats featurized. For all error states, ROC AUCs for algorithm performance on classification of the state were greater than 0.9; when using patient-specific calibrated data AUCs were 0.94, 0.95, and 0.99 for the transducer low, transducer high, and damped conditions respectively. Machine-learning trained algorithms were able to discriminate arterial line transducer error states from the waveform alone with a high degree of accuracy.