Machine learning approaches were introduced for better or comparable predictive ability than statistical analysis to predict postoperative outcomes. We sought to compare the performance of machine learning approaches with that of logistic regression analysis to predict acute kidney injury after cardiac surgery. We retrospectively reviewed 2010 patients who underwent open heart surgery and thoracic aortic surgery. Baseline medical condition, intraoperative anesthesia, and surgery-related data were obtained. The primary outcome was postoperative acute kidney injury (AKI) defined according to the Kidney Disease Improving Global Outcomes criteria. The following machine learning techniques were used: decision tree, random forest, extreme gradient boosting, support vector machine, neural network classifier, and deep learning. The performance of these techniques was compared with that of logistic regression analysis regarding the area under the receiver-operating characteristic curve (AUC). During the first postoperative week, AKI occurred in 770 patients (38.3%). The best performance regarding AUC was achieved by the gradient boosting machine to predict the AKI of all stages (0.78, 95% confidence interval (CI) 0.75–0.80) or stage 2 or 3 AKI. The AUC of logistic regression analysis was 0.69 (95% CI 0.66–0.72). Decision tree, random forest, and support vector machine showed similar performance to logistic regression. In our comprehensive comparison of machine learning approaches with logistic regression analysis, gradient boosting technique showed the best performance with the highest AUC and lower error rate. We developed an Internet–based risk estimator which could be used for real-time processing of patient data to estimate the risk of AKI at the end of surgery.

Introduction: Microvascular reactivity is decreased in patients with septic shock; this is associated with worse clinical outcomes. The objectives of the present study were to investigate microvascular reactivity in cardiac surgery patients and to assess any association with clinical outcomes. Methods: We retrospectively analyzed a prospectively collected registry. In total, 254 consecutive adult patients undergoing cardiac and thoracic aortic surgeries from January 2013 through May 2014 were analyzed. We performed a vascular occlusion test (VOT) by using near-infrared spectroscopy to measure microvascular reactivity. VOT was performed three times per patient: prior to the induction of anesthesia, at the end of surgery, and on postoperative day 1. The primary endpoint was a composite of major adverse complications, including death, myocardial infarction, acute kidney injury, acute respiratory distress syndrome, and persistent cardiogenic shock. Results: VOT recovery slope decreased during the surgery. VOT recovery slope on postoperative day 1 was significantly lower in patients with composite complications than those without (3.1 ± 1.6 versus 4.0 ± 1.5 %/s, P = 0.001), although conventional hemodynamic values, such as cardiac output and blood pressure, did not differ between the groups. On multivariable regression and linear analyses, low VOT recovery slope on postoperative day 1 was associated with increases of composite complications (odds ratio 0.742; 95 % confidence interval (CI) 0.584 to 0.943; P = 0.015) and hospital length of stay (regression coefficient (B) −1.276; 95 % CI −2.440 to −0.112; P = 0.032). Conclusion: Microvascular reactivity largely recovered on postoperative day 1 in the patients without composite complications, but this restoration was attenuated in patients with composite complications. Trial registration: ClinicalTrials.gov NCT01713192. Registered 22 October 2012.

The effects of remote ischemic preconditioning (RIPC) in cardiac surgery have been inconsistent. We investigated whether anesthesia or beta-blockers interfere with RIPC cardioprotection. Fifty patients undergoing cardiac surgery were randomized to receive limb RIPC (four cycles of 5-min of upper arm cuff inflation/deflation) in the awake state (no-anesthesia; n = 17), or under sevoflurane (n = 17) or propofol (n = 16) anesthesia. In a separate crossover study, 11 healthy volunteers received either carvedilol or no medication prior to RIPC. Plasma dialysates were obtained and perfused through an isolated male Sprague–Dawley rat heart subjected to 30-min ischemia/60-min reperfusion, following which myocardial infarct (MI) size was determined. In the cardiac surgery study, pre-RIPC MI sizes were similar among the groups (39.7 ± 4.5% no-anesthesia, 38.9 ± 5.3% sevoflurane, and 38.6 ± 3.6% propofol). However, post-RIPC MI size was reduced in the no-anesthesia group (27.5 ± 8.0%; p < 0.001), but not in the anesthesia groups (35.7 ± 6.9% sevoflurane and 35.8 ± 5.8% propofol). In the healthy volunteer study, there was a reduction in MI size with RIPC in the no-carvedilol group (41.7 ± 4.3% to 30.6 ± 8.5%; p < 0.0001), but not in the carvedilol group (41.0 ± 4.0% to 39.6 ± 5.6%; p = 0.452). We found that the cardioprotective effects of limb RIPC were abolished under propofol or sevoflurane anesthesia and in the presence of carvedilol therapy.

Remote ischemic conditioning has been investigated for cardioprotection to attenuate myocardial ischemia/reperfusion injury. In this review, we provide a comprehensive overview of the current knowledge of the signal transduction pathways of remote ischemic conditioning according to three stages: Remote stimulus from source organ; protective signal transfer through neuronal and humoral factors; and target organ response, including myocardial response and coronary vascular response. The neuronal and humoral factors interact on three levels, including stimulus, systemic, and target levels. Subsequently, we reviewed the clinical studies evaluating the cardioprotective effect of remote ischemic conditioning. While clinical studies of percutaneous coronary intervention showed relatively consistent protective effects, the majority of multicenter studies of cardiac surgery reported neutral results although there have been several promising initial trials. Failure to translate the protective effects of remote ischemic conditioning into cardiac surgery may be due to the multifactorial etiology of myocardial injury, potential confounding factors of patient age, comorbidities including diabetes, concomitant medications, and the coadministered cardioprotective general anesthetic agents. Given the complexity of signal transfer pathways and confounding factors, further studies should evaluate the multitarget strategies with optimal measures of composite outcomes.

One-lung ventilation during thoracic surgery frequently disturbs normal systemic oxygenation. However, the effect of anesthetics on arterial oxygenation during one-lung ventilation has not been well established in human study. In this clinical trial, we investigated whether a difference between desflurane-remifentanil and propofol-remifentanil anesthesia can be observed with regard to oxygenation during one-lung ventilation for thoracoscopic surgery. Adult patients with lung cancer, scheduled for video-assisted thoracoscopic lobectomy without preoperative oxygen support, were screened and randomized to receive desflurane or propofol, with remifentanil continuous infusion in both groups. Mechanical ventilation was performed with tidal volume of 8 ml/kg and FIO2 0.5 during two-lung ventilation, and 6 ml/kg and 1.0 during one-lung ventilation, both with positive end-expiratory pressure of 5 cmH2O. Arterial blood gas analysis was performed preoperatively, during two-lung ventilation, and after 15, 30, 45, and 60 min of one-lung ventilation. The primary endpoint was PaO2 at 30 min after initiating one-lung ventilation. Statistical analyses included the independent t-test for the primary endpoint and a mixed model with a post-hoc analysis to evaluate the serial changes in values. Patients were recruited between July 9 and December 2, 2014. In total, 103 patients were analyzed (n = 52 in desflurane group and n = 51 in propofol group). The primary endpoint, PaO2 at 30 min of one-lung ventilation was lower in the desflurane group than the propofol group (170 ± 72 vs. 202 ± 82 mmHg; p = 0.039). Serial changes in PaO2 during one-lung ventilation showed lower levels during desflurane anesthesia compared with propofol anesthesia (mean difference, 45 mmHg; 95% confidence interval, 16–75 mmHg; p = 0.003). In conclusion, desflurane-remifentanil anesthesia resulted in decreased arterial oxygenation compared with that of propofol-remifentanil anesthesia during one-lung ventilation for thoracoscopic surgery in patients with lung cancer. ClinicalTrials.gov identifier: NCT02191371 , registered on July 7, 2014