Search

Background Postoperative nausea and vomiting (PONV) is a still highly relevant problem and is known to be a distressing side effect in patients. The aim of this study was to develop a machine learning model to predict PONV up to 24 h with fentanyl-based intravenous patient-controlled analgesia (IV-PCA). Methods From July 2019 and July 2020, data from 2,149 patients who received fentanyl-based IV-PCA for analgesia after non-cardiac surgery under general anesthesia were applied to develop predictive models. The rates of PONV at 1 day after surgery were measured according to patient characteristics as well as anesthetic, surgical, or PCA-related factors. All statistical analyses and computations were performed using the R software. Results A total of 2,149 patients were enrolled in this study, 337 of whom (15.7%) experienced PONV. After applying the machine-learning algorithm and Apfel model to the test dataset to predict PONV, we found that the area under the receiver operating characteristic curve using logistic regression was 0.576 (95% confidence interval [CI], 0.520–0.633), k-nearest neighbor was 0.597 (95% CI, 0.537–0.656), decision tree was 0.561 (95% CI, 0.498–0.625), random forest was 0.610 (95% CI, 0.552–0.668), gradient boosting machine was 0.580 (95% CI, 0.520–0.639), support vector machine was 0.649 (95% CI, 0.592–0.707), artificial neural network was 0.686 (95% CI, 0.630–0.742), and Apfel model was 0.643 (95% CI, 0.596–0.690). Conclusions We developed and validated machine learning models for predicting PONV in the first 24 h. The machine learning model showed better performance than the Apfel model in predicting PONV.

Background & aims Fentanyl buccal tablets (FBTs) are a rapid-onset opioid indicated for breakthrough cancer pain (BTcP) and FBT titration is needed to optimize BTcP management. We aimed to predict which patients could tolerate a high dose of FBT (400 μg or more at a time). Methods A retrospective analysis was performed to assess the final FBT dose. The final FBT doses were compared according to the clinical features. The prediction accuracy of patients tolerant of 400 μg or higher FBT was compared using the area under the receiver operating characteristic (ROC) curves. A risk scoring model based on the odds ratio (OR) was developed from the final multivariable model, and patients were assigned into two groups: low tolerance (0–1 point) and high tolerance (2–3 points). Results Among 131 patients, the most frequently effective dose of FBT was 200 μg (54%), followed by 100 μg (30%). The median value of morphine equivalent daily doses (MEDD) was 60 mg/day, and the most common daily use was 3–4 times/day. In multivariable analysis, male sex, younger age, and use of FBTs three or more times per day were independently associated with high-dose FBT. According to the risk scoring model, the patients with a final FBT of 400 μg or higher were significantly more in the high tolerance group (17%) compared to the low tolerance group (3%; p = 0.023) Conclusions According to the dose relationship between the final FBT dose and the clinical features, three factors (sex, age, daily use of FBT) were independently associated with the final dose of FBT. Our risk score model could help predict tolerance to high-dose FBT and guide the titration plan for BTcP.

Objectives The occurrence of postoperative neurocognitive deficits(POND)after major cardiac surgery is associated with an increase in perioperative mortality and morbidity. Oxidative stress caused by oxygen can affect neuronal damage, which can lead to POND. Whether the intraoperative rSO2 value reflects oxidative stress and the associated incidence of POND is unknown. Methods Among 3482 patients undergoing cardiac surgery, 976 patients were allocated for this retrospective study. Of these, 230 patients (32.5%) were observed to have postoperative neurologic symptoms. After propensity score 1:2 ratio matching, a total of 690 patients were included in the analysis. Recorded data on the occurrence of POND from the postoperative period to predischarge were collected from the electronic records. Results The mean baseline rSO2 value was higher in the POND (–) group than in the POND (+) group. The mean overall minimum rSO2 value was lower in the POND (+) group (52.2 ± 8.3 vs 48.3 ± 10.5, P < 0.001). The mean overall maximum rSO2 values were not significantly different between the two groups (72.7 ± 8.3 vs 73.2 ± 9.2, P = 0.526). However, there was a greater increase in the overall maximum rSO2 values as compared with baseline in the POND (+) group (10.9 ± 8.2 vs 17.9 ± 10.2, P < 0.001). The degree of increase in the maximum rSO2 value was a risk factor affecting the occurrence of POND (adjusted odds ratio, 1.08; 95% confidence interval [CI], 1.04–1.11; P < 0.001). The areas under the receiver-operating characteristic curve for delta values of minimal and maximal compared with baseline values were 0.60 and 0.71, respectively. Conclusions Increased cerebral oximeter levels during cardiac surgery may also be a risk factor for POND. This is considered to reflect the possibility of oxidative neuronal damage, and further studies are needed in the future.

Objective To construct a prediction model for optimal tracheal tube depth in pediatric patients using machine learning. Methods Pediatric patients aged Results For each method, the percentage with optimal tracheal tube depth predictions in the test set was calculated as follows: 79.0 (95% confidence interval [CI], 73.5 to 83.6) for random forest, 77.4 (95% CI, 71.8 to 82.2; P = 0.719) for elastic net, 77.0 (95% CI, 71.4 to 81.8; P = 0.486) for support vector machine, 76.6 (95% CI, 71.0 to 81.5; P = 1.0) for artificial neural network, 66.9 (95% CI, 60.9 to 72.5; P < 0.001) for the age-based formula, 58.5 (95% CI, 52.3 to 64.4; P< 0.001) for the tube ID-based formula, and 44.4 (95% CI, 38.3 to 50.6; P < 0.001) for the height-based formula. Conclusions In this study, the machine learning models predicted the optimal tracheal tube tip location for pediatric patients more accurately than the formula-based methods. Machine learning models using biometric variables may help clinicians make decisions regarding optimal tracheal tube depth in pediatric patients.

Pigs are anatomically and physiologically closer to humans than other laboratory animals. Transgenic (TG) pigs are widely used as models of human diseases. The aim of this study was to produce pigs expressing a tetracycline (Tet)-inducible transgene. The Tet-on system was first tested in infected donor cells. Porcine fetal fibroblasts were infected with a universal doxycycline-inducible vector containing the target gene enhanced green fluorescent protein (eGFP). At 1 day after treatment with 1 µg/ml doxycycline, the fluorescence intensity of these cells was increased. Somatic cell nuclear transfer (SCNT) was then performed using these donor cells. The Tet-on system was then tested in the generated porcine SCNT-TG embryos. Of 4,951 porcine SCNT-TG embryos generated, 850 were cultured in the presence of 1 µg/ml doxycycline in vitro. All of these embryos expressed eGFP and 15 embryos developed to blastocyst stage. The remaining 4,101 embryos were transferred to thirty three surrogate pigs from which thirty eight cloned TG piglets were obtained. PCR analysis showed that the transgene was inserted into the genome of each of these piglets. Two TG fibroblast cell lines were established from these TG piglets, and these cells were used as donor cells for re-cloning. The re-cloned SCNT embryos expressed the eGFP transgene under the control of doxycycline. These data show that the expression of transgenes in cloned TG pigs can be regulated by the Tet-on/off systems.