Risk factors including preoperative echocardiographic parameters for post-induction hypotension in general anesthesia

Machine learning-based prediction of post-induction hypotension: identifying risk factors and enhancing anesthesia management

BACKGROUND

Post-induction hypotension (PIH) increases surgical complications including myocardial injury, acute kidney injury, delirium, stroke, prolonged hospitalization, and endangerment of the patient's life. Machine learning is an effective tool to analyze large amounts of data and identify perioperative complication factors. This study aims to identify risk factors for PIH and develop predictive models to support anesthesia management.

METHODS

A dataset of 5406 patients was analyzed using machine learning methods. Logistic regression, random forest, XGBoost, and neural network models were compared. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC), calibration curves, and decision curve analysis (DCA).

RESULTS

The logistic regression model achieved an AUROC of 0.74 (95% CI: 0.71-0.77), outperforming the random forest (AUROC: 0.71), XGBoost (AUROC: 0.72), and neural network (AUROC: 0.72) models. In terms of calibration, logistic regression demonstrated superior performance, as reflected by Brier Scores and calibration curves, followed by XGBoost, random forest, and neural network. Decision curve analysis indicated that the logistic regression model provided the greatest clinical utility among all models. Baseline blood pressure, age, sex, type of surgery, platelet count, and certain anesthesia-inducing drugs were identified as important features.

CONCLUSIONS

This study provides a valuable tool for personalized preoperative risk assessment and customized anesthesia management, allowing for early intervention and improved patient outcomes. Integration of machine learning models into electronic medical record systems can facilitate real-time risk assessment and prediction.

Defining Postinduction Hemodynamic Instability With an Automated Classification Model

BACKGROUND

Postinduction hypotension (PIH) may be associated with increased morbidity and mortality. In earlier studies, the definition of PIH is solely based on different absolute or relative thresholds. However, the time-course (eg, how fast blood pressure drops during induction) is rarely incorporated, whereas it might represent the hemodynamic instability of a patient. We propose a comprehensive model to distinguish hemodynamically unstable from stable patients by combining blood pressure thresholds with the magnitude and speed of decline.

METHODS

This prospective study included 375 adult elective noncardiac surgery patients. Noninvasive blood pressure was continuously measured between 5 minutes before up to 15 minutes after the first induction agent had been administered. An expert panel rated whether the patient experienced clinically relevant hemodynamic instability or not. Interrater correlation coefficient and intraclass correlation were computed to check for consistency between experts. Next, an automated classification model for clinically relevant hemodynamic instability was developed using mean, maximum, minimum systolic, mean, diastolic arterial blood pressure (SAP, MAP, and DAP, respectively) and their corresponding time course of decline. The model was trained and tested based on the hemodynamic instability labels provided by the experts.

RESULTS

In total 78 patients were classified as having experienced hemodynamic instability and 279 as not. The hemodynamically unstable patients were significantly older (7 years, 95% confidence interval (CI), 4-11, P < .001), with a higher prevalence of chronic obstructive pulmonary disease (COPD) (3% higher, 95% CI, 1-8, P = .036). Before induction, hemodynamically unstable patients had a higher SAP (median (first-third quartile): 161 (145-175) mm Hg vs 150 (134-166) mm Hg, P < .001) compared to hemodynamic stable patients. Interrater agreement between experts was 0.92 (95% CI, 0.89-0.94). The random forest classifier model showed excellent performance with an area under the receiver operating curve (AUROC) of 0.96, a sensitivity of 0.84, and specificity of 0.94.

CONCLUSIONS

Based on the high sensitivity and specificity, the developed model is able to differentiate between clinically relevant hemodynamic instability and hemodynamic stable patients. This classification model will pave the way for future research concerning hemodynamic instability and its prevention.

Preoperative echocardiography and anesthetic drugs as predictors of post-induction hypotension during general anesthesia: a prospective observational study

Post-induction hypotension (PIH) during general anesthesia is a serious concern due to its high incidence and association with adverse postoperative outcomes and prognosis. We prospectively enrolled consecutive 1,603 patients scheduled for non-cardiac and cardiac surgery under general anesthesia. The primary outcome was defined as the lowest mean blood pressure from induction of general anesthesia to the start of surgery: post-induction blood pressure (PIB). Multivariable regression analysis was used to investigate the relationships between PIB and preoperative echocardiographic measurements, types and doses of anesthetic agents, and preoperative comorbidities and oral medication. Left ventricle regional wall motion abnormality (RWMA) and lower estimated glomerular filtration rate significantly associated lower PIB. Factors significantly associated with higher PIB were high BMI, beta blockers, high pre-induction mean blood pressure, and starting induction in the afternoon. Sevoflurane and prophylactic norepinephrine use were associated with higher PIB as anesthetic interventions. Significant interactions were found between sevoflurane and RWMA (P for interaction: 0.01). This is the first study to demonstrate the feasibility of intervening in the selection of anesthetic agents based on echocardiographic findings and to recommend the use of sevoflurane and prophylactic norepinephrine for the prevention of PIH, especially in patients with RWMA.

Preoperative Echocardiographic Assessment in Elective Surgery Patients: A Cross-Sectional Study From Duhok, Iraqi Kurdistan

Background and aim Echocardiography plays a pivotal role in the preoperative risk assessment of patients undergoing various surgical procedures. Therefore, we conducted this study to detect various cardiac abnormalities through a preoperative echocardiographic study for cases undergoing elective surgeries in Duhok, Iraqi Kurdistan. Methods This cross-sectional study assessed echocardiographic findings in preoperative patients at Azadi Teaching Hospital, Duhok, Kurdistan Region of Iraq, from 2023 to 2024. The study encompasses 468 adult patients. We gathered clinicodemographic characteristics, indications of referral to a preoperative echo study, and the echocardiographic findings, particularly regional wall motion abnormalities, valvular heart diseases, and heart failure (systolic or diastolic/left ventricular hypertrophy). Results In a cohort of 468 patients, 205 (43.80%) were aged 35-54 years, and 269 (57.48%) were female. Most patients (366; 78.21%) resided in Duhok. Nearly only half of the cases 219 (46.79%) had clear preoperative echostudy indications. A total of 289 (61.74%) had no remarkable echocardiographic findings. Diastolic heart failure was most prevalent at 65 (13.89%). Older cases had more prevalent echo findings in terms of valve dysfunctions and heart failure. Notable associations were found between echocardiographic abnormalities and surgical types, particularly higher regional wall motion abnormalities in genito-urinary system operations (4; 8.16%) and valvular heart disease in orthopedic surgeries (13; 15.85%). Conclusions Echocardiographic abnormalities were remarkably observed in patients aged 65 and older. Many cases had no clear indications for preoperative echo study and hence unremarkable echo findings. More scrutiny is indicated during referral, and focusing on older adults' preoperative cardiac screening is recommended.

Changes in blood catecholamines during induction of general anesthesia in patients with post-induction hypotension undergoing laparoscopic cholecystectomy: A single-center prospective cohort study

BACKGROUND

Post-induction hypotension (PIH) often occurs during general anesthesia induction. This study aimed to investigate blood catecholamine levels during induction of general anesthesia in patients with PIH undergoing laparoscopic cholecystectomy.

METHODS

This prospective study included 557 adult patients who underwent laparoscopic cholecystectomy under general anesthesia. PIH was defined as a greater than 20% decrease in systolic blood pressure from the pre-induction value, a systolic arterial pressure of less than 90 mmHg, or both. Plasma concentrations of epinephrine and norepinephrine during the induction of general anesthesia were determined using enzyme-linked immunosorbent assay. Multivariate logistic regression analysis evaluated the association between the clinical factors and PIH.

RESULTS

Of the 557 patients, 390 had PIH, and the remaining 167 were allocated to the non-PIH group. Changes in blood adrenaline, noradrenaline levels, or both were more pronounced in the PIH than in the non-PIH group (p<0.05). Age, body mass index, a history of hypertension, preoperative systolic blood pressure, and propofol or sufentanil dose were independent predictors of PIH.

CONCLUSION

The changes of blood catecholamines in patients with more stable hemodynamics during the induction of general anesthesia are smaller than that in patients with post-induction hypotension.

TRIAL REGISTRATION

ChiCTR2200055549, 12/01/2022.

Comparison of Hemodynamic Parameters Based on the Administration of Remimazolam or Sevoflurane in Patients under General Anesthesia in the Beach Chair Position: A Single-Blinded Randomized Controlled Trial

Background: We aimed to evaluate whether the administration of remimazolam as a maintenance agent for general anesthesia affects the occurrence of hypotension compared with sevoflurane when switching to the beach chair position (BCP). Methods: We conducted a prospective randomized controlled trial from June 2023 to October 2023 in adult patients undergoing orthopedic surgery under general anesthesia in the BCP. A total of 78 participants were randomly allocated to the remimazolam (R) or sevoflurane (S) groups. The primary outcome was the incidence of hypotension that occurred immediately after switching to a BCP. The secondary outcomes included differences between the study groups in perioperative blood pressure (BP), heart rate (HR), endotracheal tube extubation time, postoperative complications, and hospital length of stay (LOS). Results: The incidence of hypotension immediately after switching to a BCP was significantly higher in the S group. The risk factors associated with hypotension included sevoflurane administration and a high baseline systolic BP. In the receiver operating characteristic curve analysis for the occurrence of hypotension after the transition to a BCP, the cutoff value for systolic BP was 142 mmHg. The perioperative BP and HR were higher in the R group at several timepoints. Postoperative endotracheal tube extubation time was shorter in the R group. There were no significant differences in the postoperative complications or hospital LOS between the two groups. Conclusions: Remimazolam should be considered as an anesthetic agent to prevent hypotension when switching to BCP, and hypotension may occur frequently in patients with high baseline BP.

Muscle-related parameters-based machine learning model for predicting postinduction hypotension in patients undergoing colorectal tumor resection surgery

Objectives

This study used machine learning algorithms to identify important variables and predict postinduction hypotension (PIH) in patients undergoing colorectal tumor resection surgery.

Methods

Data from 318 patients who underwent colorectal tumor resection under general anesthesia were analyzed. The training and test sets are divided based on the timeline. The Boruta algorithm was used to screen relevant basic characteristic variables and establish a model for the training set. Four models, regression tree, K-nearest neighbor, neural network, and random forest (RF), were built using repeated cross-validation and hyperparameter optimization. The best model was selected, and a sorting chart of the feature variables, a univariate partial dependency profile, and a breakdown profile were drawn. R2, mean absolute error (MAE), mean squared error (MSE), and root MSE (RMSE) were used to plot regression fitting curves for the training and test sets.

Results

The basic feature variables associated with the Boruta screening were age, sex, body mass index, L3 skeletal muscle index, and HUAC. In the optimal RF model, R2 was 0.7708 and 0.7591, MAE was 0.0483 and 0.0408, MSE was 0.0038 and 0.0028, and RMSE was 0.0623 and 0.0534 for the training and test sets, respectively.

Conclusion

A high-performance algorithm was established and validated to demonstrate the degree of change in blood pressure after induction to control important characteristic variables and reduce PIH occurrence.

Evaluation of pre-induction dynamic arterial elastance as an adjustable predictor of post-induction hypotension: A prospective observational study

STUDY OBJECTIVE

Dynamic arterial elastance (Eadyn) has been suggested as a functional measure of arterial load. We aimed to evaluate whether pre-induction Eadyn can predict post-induction hypotension.

DESIGN

Prospective observational study.

PATIENTS

Adult patients undergoing general anesthesia with invasive and non-invasive arterial pressure monitoring systems.

MEASUREMENTS

We collected invasive and non-invasive Eadyns (n = 38 in each), respectively. In both invasive and non-invasive Eadyns, pre-induction Eadyns were obtained during one-minute tidal and deep breathing in each patient before anesthetic induction. Post-induction hypotension was defined as a decrease of >30% in mean blood pressure from the baseline value or any absolute mean blood pressure value of <65 mmHg for 10 min after anesthetic induction. The predictabilities of Eadyns for the development of post-induction hypotension were tested using receiver-operating characteristic curve analysis.

MAIN RESULTS

Invasive Eadyn during deep breathing showed significant predictability with an area under the curve (AUC) of 0.78 (95% Confidence interval [CI], 0.61-0.90, P = 0.001). But non-invasive Eadyn during tidal breathing (AUC = 0.66, 95% CI, 0.49-0.81, P = 0.096) and deep breathing (AUC = 0.53, 95% CI, 0.36-0.70, P = 0.75), and invasive Eadyn during tidal breathing (AUC = 0.66, 95% CI, 0.41-0.74, P = 0.095) failed to predict post-induction hypotension.

CONCLUSION

In our study, invasive pre-induction Eadyn during deep breathing -could predict post-induction hypotension. Despite its invasiveness, future studies will be needed to evaluate the usefulness of Eadyn as a predictor of post-induction hypotension because it is an adjustable parameter.

ERCP-related adverse events: incidence, mechanisms, risk factors, prevention, and management

INTRODUCTION

Endoscopic retrograde cholangiopancreatography (ERCP) is a commonly performed procedure for pancreaticobiliary disease. While ERCP is highly effective, it is also associated with the highest adverse event (AE) rates of all commonly performed endoscopic procedures. Thus, it is critical that endoscopists and caregivers of patients undergoing ERCP have clear understandings of ERCP-related AEs.

AREAS COVERED

This narrative review provides a comprehensive overview of the available evidence on ERCP-related AEs. For the purposes of this review, we subdivide the presentation of each ERCP-related AE according to the following clinically relevant domains: definitions and incidence, proposed mechanisms, risk factors, prevention, and recognition and management. The evidence informing this review was derived in part from a search of the electronic databases PubMed, Embase, and Cochrane, performed on 1 May 20231 May 2023.

EXPERT OPINION

Knowledge of ERCP-related AEs is critical not only given potential improvements in peri-procedural quality and related care that can ensue but also given the importance of reviewing these considerations with patients during informed consent. The ERCP community and researchers should aim to apply standardized definitions of AEs. Evidence-based knowledge of ERCP risk factors should inform patient care decisions during training and beyond.

Post-induction hypotension and intraoperative hypotension as potential separate risk factors for the adverse outcome: a cohort study

PURPOSE

Intraoperative hypotension (IOH) is associated with organ hypoperfusion. There are different underlying causes of IOH depending on the phase of surgery. Post-induction hypotension (PIH) and early-intraoperative hypotension tend to be frequently differentiated. We aimed to explore further different phases of IOH and verify whether they are differently associated with postoperative complications.

METHODS

Patients undergoing abdominal surgery between October 2018 and July 2019 in a university hospital were screened. Post-induction hypotension was defined as MAP ≤ 65 mmHg between the induction of anaesthesia and the onset of surgery. Hypotension during surgery (IOH) was defined as MAP ≤ 65 mmHg occurring between the onset of surgery and its completion. Acute kidney injury, stroke or transient ischaemic attack, delirium, and myocardial infarction were considered as the outcome.

RESULTS

We enrolled 508 patients (219 males, median age 62 years). 158 subjects (31.1%) met PIH, 171 (33.7%) met IOH criteria, and 67 (13.2%) patients experienced both. PIH time accounted for 22.8% of the total hypotension time and 29.7% of the IOH time. The IOH time accounted for 5.17% of the total intraoperative time, while PIH for 8.91% of the pre-incision time. Female sex, lower height, body mass and lower pre-induction BP (SBP and MAP) were found to be associated with the incidence of PIH. The negative outcome was observed in 38 (7.5%) patients. Intraoperative MAP ≤ 65 mmHg, longer duration of the procedure (≥ 230 min), chronic arterial hypertension and age were associated with the presence of the outcome (p < 0.01 each).

CONCLUSIONS

The presence of IOH defined as MAP ≤ 65 mmHg is relevant to post-operative organ complications, the presence of PIH does not appear to be of such significance. Because cumulative duration of PIH and IOH differs significantly, especially in long-lasting procedures, direct comparison of the influence of PIH and IOH on outcome separately may be biased and should be taken into account in data interpretation. Further research is needed to deeply investigate this phenomenon.

Effect of pneumatic leg compression on post-induction hypotension in elderly patients undergoing robot-assisted laparoscopic prostatectomy: a double-blind randomised controlled trial

Post-induction hypotension is common and associated with postoperative complications. We hypothesised that pneumatic leg compression reduces post-induction hypotension in elderly patients undergoing robot-assisted laparoscopic prostatectomy. In this double-blind randomised study, patients were allocated randomly to the pneumatic leg compression group (n = 50) or control (n = 50). In the intervention group, pneumatic leg compression was initiated before induction of anaesthesia. In the control group, pneumatic leg compression was initiated 20 min after anaesthesia induction. The primary outcome was the incidence of post-induction hypotension in these groups. Post-induction hypotension was defined as systolic blood pressure < 90 mmHg during the first 20 min after induction. Haemodynamic variables and area under the curve of post-induction systolic blood pressure over time were assessed. Complications associated with pneumatic leg compression were recorded, including: peripheral neuropathy; compartment syndrome; extensive bullae beneath the leg sleeves; and pulmonary thromboembolism. The incidence of post-induction hypotension decreased in the pneumatic leg compression group compared with that in the control group; 5 (10%) vs. 29 (58%), respectively, p < 0.001. In the pneumatic leg compression group, the lowest systolic, diastolic and mean blood pressures 20 min after induction of anaesthesia were significantly greater than the control group. Pneumatic leg compression resulted in an increased area under the curve of systolic blood pressure in the first 20 min after induction, p = 0.001. There were no pneumatic leg compression-related complications. Pneumatic leg compression reduced post-induction hypotension in elderly patients undergoing robot-assisted laparoscopic prostatectomy, suggesting that it is an effective and safe intervention to prevent post-induction hypotension among elderly patients undergoing general anaesthesia.

The use of remimazolam versus propofol for induction and maintenance of general anesthesia: A systematic review and meta-analysis

Background: The primary objective of this study was to compare the risk of hypotension, as well as the induction and recovery characteristics between remimazolam and propofol in patients receiving surgery under general anesthesia. Methods: The Embase, Medline, Google scholar, and the Cochrane Library databases were searched from inception to March 2022 for randomized controlled trials The primary outcome was the risk of post-induction hypotension between the two agents, while the secondary outcomes included anesthetic depth, induction efficacy, time to loss of consciousness (LOC), hemodynamic profiles, time to eye opening, extubation time as well as the incidence of injection pain and postoperative nausea/vomiting (PONV). Results: Meta-analysis of eight studies published from 2020 to 2022 involving 738 patients revealed a significantly lower risk of post-induction hypotension with the use of remimazolam compared to that with propofol [risk ratio (RR) = 0.57, 95% confidence interval (CI): 0.43 to 0.75, p < 0.0001, I2 = 12%, five studies, 564 patients]. After anesthetic induction, the anesthetic depth measured by bispectral index (BIS) was lighter in the remimazolam group than that in the propofol group (MD = 9.26, 95% confidence interval: 3.06 to 15.47, p = 0.003, I2 = 94%, five studies, 490 patients). The time to loss of consciousness was also longer in the former compared to the latter (MD = 15.49 s, 95%CI: 6.53 to 24.46, p = 0.0007, I2 = 61%, three studies, 331 patients). However, the use of remimazolam correlated with a lower risk of injection pain (RR = 0.03, 95%CI: 0.01 to 0.16, p < 0.0001, I2 = 0%, three studies, 407 patients) despite comparable efficacy of anesthetic induction (RR = 0.98, 95%CI: 0.9 to 1.06, p = 0.57, I2 = 76%, two studies, 319 patients). Our results demonstrated no difference in time to eye opening, extubation time, and risk of PONV between the two groups. Conclusion: Remimazolam was associated with a lower risk of post-induction hypotension after anesthetic induction compared with propofol with similar recovery characteristics. Further studies are required to support our findings. Systematic Review Registration: https://www.crd.york.ac.uk/prospero/; Identifier: CRD42022320658.

Differential effects of remimazolam and propofol on heart rate variability during anesthesia induction

PURPOSE

The effects of remimazolam on autonomic nervous activity have not been elucidated. We investigated the differential effects of remimazolam and propofol on autonomic nervous activity during anesthesia induction.

METHODS

Thirty patients were randomly divided into a remimazolam group or a propofol group for anesthesia induction. Hemodynamics and indices of heart rate variability were recorded before and after anesthesia. Low frequency power (LF; 0.04-0.15 Hz, ms²) and high frequency power (HF; 0.15-0.4 Hz, ms²) were calculated from power spectral density of heart rate variability. LF reflects both sympathetic and parasympathetic activities and HF reflects parasympathetic activity. To investigate the balance between sympathetic and parasympathetic activities, the normalized unit (%) of LF (LF nu) and that of HF (HF nu) were calculated. Changes in LF nu (ΔLF nu) before and after anesthesia (ΔLF nu = LF nu at awake-LF nu after anesthesia) were compared between the groups.

RESULTS

Remimazolam and propofol decreased blood pressure and power spectral density of heart rate variability. Remimazolam did not change LF nu and HF nu, while propofol increased LF nu and decreased HF nu (P = 0.020). ΔLF nu in the remimazolam group (1.4 ± 23.6%) were less than that in the propofol group (19.3 ± 22.4%, P = 0.0415).

CONCLUSION

Remimazolam and propofol decreased autonomic nervous activity during anesthesia induction. Remimazolam preserved the balance of sympathetic and parasympathetic activities, while propofol modulated it to sympathetic dominance.