Cardiac Surgical Intensive Care Unit Research Articles

Organization and structure of intensive care management of adult congenital heart disease in the United States

BackgroundThe growing number of adults with congenital heart disease has led to the formation of dedicated centers to provide specialized care. There is a paucity of published data on the organization, staffing, and resources available when adult congenital heart disease (ACHD) patients are hospitalized in intensive care settings. This study aims to describe the delivery of intensive care for ACHD patients in the United States and identify key differences between accredited centers or those pending accreditation and non-accredited Adult Congenital Heart Association (ACHA) centers. MethodsA web-based cross-sectional survey targeting centers with specialized ACHD programs was conducted. ResultsData were obtained from 48 of 96 centers (50%). Twenty-seven (56%) were accredited by the ACHA and 8 (17%) were pending accreditation. Surgical ACHD patients were primarily co-managed by cardiologists and intensivists (n = 11, 23%) and treated within an adult cardiac surgical intensive care unit (n = 34, 72%). Compared to non-accredited centers, ACHA accredited centers or those pending accreditation were more likely to offer daily rounds with ACHD providers (30, 86% vs. 6, 46%, p=<0.01), were more likely to support an ACHD fellowship training program (16, 46% vs. 0, 0%, p=<0.01); and were more likely to have durable mechanical support, such as ventricular assist devices (35, 100% vs. 8, 62%, p < 0.01). ConclusionThis survey reports current trends in organization, staffing, and resources available to ACHD patients requiring intensive care. ACHA accredited centers or those pending accreditation were more likely to have dedicated ACHD rounds, fellowship training programs, and access to durable mechanical support.

A machine learning-based prediction model for postoperative delirium in cardiac valve surgery using electronic health records

BackgroundPrevious models for predicting delirium after cardiac surgery remained inadequate. This study aimed to develop and validate a machine learning-based prediction model for postoperative delirium (POD) in cardiac valve surgery patients.MethodsThe electronic medical information of the cardiac surgical intensive care unit (CSICU) was extracted from a tertiary and major referral hospital in southern China over 1 year, from June 2019 to June 2020. A total of 507 patients admitted to the CSICU after cardiac valve surgery were included in this study. Seven classical machine learning algorithms (Random Forest Classifier, Logistic Regression, Support Vector Machine Classifier, K-nearest Neighbors Classifier, Gaussian Naive Bayes, Gradient Boosting Decision Tree, and Perceptron.) were used to develop delirium prediction models under full (q = 31) and selected (q = 19) feature sets, respectively.ResultThe Random Forest classifier performs exceptionally well in both feature datasets, with an Area Under the Curve (AUC) of 0.92 for the full feature dataset and an AUC of 0.86 for the selected feature dataset. Additionally, it achieves a relatively lower Expected Calibration Error (ECE) and the highest Average Precision (AP), with an AP of 0.80 for the full feature dataset and an AP of 0.73 for the selected feature dataset. To further evaluate the best-performing Random Forest classifier, SHAP (Shapley Additive Explanations) was used, and the importance matrix plot, scatter plots, and summary plots were generated.ConclusionsWe established machine learning-based prediction models to predict POD in patients undergoing cardiac valve surgery. The random forest model has the best predictive performance in prediction and can help improve the prognosis of patients with POD.

Effect of Eye Masks and Ear Plugs on Quality of Sleep of Patients in Intensive Care Unit

Sleep has physiologic importance and adequate sleep is necessary for the maintenance of good health and well-being of all patients especially the critically ill. The study aimed to test the effect of eye masks and ear plugs on the quality of sleep among patients admitted to Intensive Care Units (ICU) and to identify the factors affecting the quality of sleep in ICU. An experimental cross-over design was adopted. Fourty subjects were selected using consecutive sampling from the Cardiac ICU, Neuro ICU, and Surgical ICU of the hospital. The tools used were the Richard Campbell Sleep Questionnaire (RCSQ) to assess the quality of sleep and the Visual Analogue Scale to measure factors causing sleep disturbance in the ICU. Results showed that eye mask and ear plug were effective in improving sleep among patients in ICU. While sleeping without interventions all of the subjects in ICU reported poor sleep. After using the eye mask and ear plugs 27.5 percent of subjects reported very good sleep 65 percent of subjects reported good sleep and only 7.5 percent had poor sleep. The mean score and standard deviation of quality of sleep of patients in ICU increased from 17.90 ±3.90 to 36.87±4.09 while sleeping with eye mask and ear plugs. This increase was significant at a 0.05 level of significance with a p value of 0.000. Eye masks and ear plugs were effective in improving sleep among patients in ICU. The most important factor affecting the quality of sleep of patients in ICU was light followed by anxiety, pain and alarms. The study concluded that Intensive care nurses have to use eye mask and ear plug for ICU patients to promote sleep which will help in reducing the use of sedatives. Nurses should take measures to reduce light and sound in ICU during night time and take measures to reduce anxiety and pain effectively for promoting sleep.

AI-Enhanced Predictive Modeling for Identifying Depression and Delirium in Cardiovascular Patients Scheduled for Cardiac Surgery.

Our study was conducted at a cardiac surgical intensive care unit. A total of 224 participants who were scheduled for elective coronary artery bypass graft surgery (CABG) were enrolled in the study. Prior to surgery, each patient underwent psychiatric evaluation to identify major depressive disorder (MDD) based on the DSM-5 criteria. An advanced data curation workflow was applied to eliminate outliers and inconsistencies and improve data quality. An explainable AI-empowered pipeline was developed, where sophisticated machine learning techniques, including the AdaBoost, random forest, and XGBoost algorithms, were trained and tested on the curated data based on a stratified cross-validation approach. Our findings identified a significant correlation between the biomarker "sRAGE" and depression (r = 0.32, p = 0.038). Among the applied models, the random forest classifier demonstrated superior accuracy in predicting depression, with notable scores in accuracy (0.62), sensitivity (0.71), specificity (0.53), and area under the curve (0.67). This study provides compelling evidence that depression in CVD patients, particularly those with elevated "sRAGE" levels, can be predicted with a 62% accuracy rate. Our AI-driven approach offers a promising way for early identification and intervention, potentially revolutionizing care strategies in this vulnerable population.

Melatonin and Melatonin Agonists for Prevention of Delirium in the Cardiac Surgical ICU: A Meta-analysis.

Delirium is highly prevalent in the immediate postoperative period following cardiac surgery and adversely impacts outcomes. Melatonin has been increasingly used in pharmacological prevention of delirium. We aimed to synthesize the available evidence concerning the role of melatonin and melatonin agonists in preventing delirium in patients after cardiac surgery. PubMed, Google Scholar, and Web of Science databases were searched for relevant randomized and non-randomized trials in adults undergoing cardiac surgery investigating melatonin agonists to prevent delirium. Studies incorporating transplants, preoperative organ support, prophylactic antipsychotics, or children were excluded. Risk-of-bias was assessed using Cochrane ROB 2.0 and ROBINS-I tools. A systematic review and meta-analysis were conducted, calculating pooled odds ratio (OR) for the incidence of postoperative delirium using a random effects model with the Mantel-Haenszel method with restricted maximum-likelihood estimator. Trial sequential analysis was also carried out for the primary outcome. Six randomized trials and one non-randomized trial involving 1,179 patients were included. Incidence of delirium was 16.7 and 29.6% in the intervention and comparator groups respectively, indicating a pooled OR of 0.44 [95% confidence interval (CI) 0.27 - 0.71, p = 0.04] favoring melatonin. Two studies had a high risk of bias, and I2 statistics indicated significant heterogeneity. However, publication bias was insignificant, and trial sequential analysis indicated the significance of the attained effect size. Based on available studies, perioperative melatonin use significantly decreases postoperative incidence of delirium after adult cardiac surgery. However, the available quality of evidence is low, and larger trials with standardization of nonpharmacological delirium prevention interventions, in high-risk cohorts, and exploring various dosages and regimens should be carried out. Niyogi SG, Naskar C, Singh A, Kumar B, Grover S. Melatonin and Melatonin Agonists for Prevention of Delirium in the Cardiac Surgical ICU: A Meta-analysis. Indian J Crit Care Med 2023;27(11):837-844.

Shock Severity Classification and Mortality in Adults With Cardiac, Medical, Surgical, and Neurological Critical Illness

ObjectiveTo evaluate whether the Society for Cardiovascular Angiography and Interventions (SCAI) Shock Classification could perform risk stratification in a mixed cohort of intensive care unit (ICU) patients, similar to its validation in patients with acute cardiac disease. MethodsWe included 21,461 adult Mayo Clinic ICU patient admissions from December 1, 2014, to February 28, 2018, including cardiac ICU (16.7%), medical ICU (37.4%), neurosciences ICU (27.7%), and surgical ICU (18.2%). The SCAI Shock Classification (a 5-stage classification from no shock [A] to refractory shock [E]) was assigned in each 4-hour period during the first 24 hours of ICU admission. ResultsThe median age was 65 years, and 43.2% were female. In-hospital mortality occurred in 1611 (7.5%) patients, with a stepwise increase in in-hospital mortality in each higher maximum SCAI Shock stage overall: A, 4.0%; B, 4.6%; C, 7.0%; D, 13.9%; and E, 40.2%. The SCAI Shock Classification provided incremental mortality risk stratification in each ICU, with the best performance in the cardiac ICU and the worse performance in the neurosciences ICU. The SCAI Shock Classification was associated with higher adjusted in-hospital mortality (adjusted odds ratio, 1.32 per each stage; 95% CI, 1.24 to 1.41; P<.001); this association was not observed in the neurosciences ICU when considered separately. ConclusionThe SCAI Shock Classification provided incremental mortality risk stratification beyond established prognostic markers across the spectrum of medical and surgical critical illness, proving utility outside its original intent.

The Level of Knowledge Among Nurses Regarding Care of Patients with Hepatic Encephalopathy at Najran Hospitals, Saudi Arabia.

Hepatic encephalopathy (HE) is a significant complication of acute and chronic liver disease. It is crucial for nurses to have knowledge of encephalopathy symptoms to enable prompt and effective responses when caring for such patients. Therefore, this study aimed to evaluate nurses' HE knowledge in Najran hospitals in Saudi Arabia. This cross-sectional analytical study was conducted with nurses at the Najran Hospital in Najran City, Saudi Arabia. The sample size was determined using Epi-Calc 2000, resulting in a final sample of 125 nurses from intensive care units (ICU), cardiac surgical ICU and pediatric ICU, and they were selected by convenience sampling technique. Data were collected using a self-administered questionnaire, and 125 nurses participated in knowledge assessment. The data were coded and entered into SPSS version 21.0 for descriptive and inferential statistics. Information letters and consent forms were obtained from all the participants. The majority of nurses (88.8%) were female, and 59.2% were married. The age range of participants was 20-30 years (50.4%). Approximately 36.8% had 4-6 years of work experience. The majority of the nurses (87.2%) held a bachelor's degree. More than half (56%) of the nurses demonstrated moderate HE knowledge, whereas over a quarter (28%) exhibited good knowledge. There was a highly statistically significant association between marital status, level of education, nationality, and level of nurses' knowledge regarding HE (P-value = 0.01, 0.02, and 0.02, respectively). The study concludes that nurses in Najran hospitals have an adequate knowledge levels, ranging from moderate to good. This highlights the importance of tailored educational programs to enhance nurses' understanding of HE symptoms and management. Furthermore, the marital status, education level, and nationality have significant association with nurses' knowledge.

Percutaneous Dilatational Tracheostomy in a Cardiac Surgical Intensive Care Unit: A Single-Center Experience.

A significant proportion of cardiac surgery intensive care unit (CSICU) patients require long-term ventilation, necessitating tracheostomy placement. The goal of this study was to evaluate the long-term postoperative outcomes and complications associated with percutaneous dilatational tracheostomy (PDT) in CSICU patients. All patients undergoing PDT after cardiac, thoracic, or vascular operations in the CSICU between January 1, 2013 and January 1, 2021 were identified. They were evaluated for mortality, decannulation time, and complications including bleeding, infection, and need for surgical intervention. Multivariable regression models were used to identify predictors of early decannulation and the complication rate. Ninety-three patients were identified for this study (70 [75.3%] male and 23 [24.7%] female). Furthermore, 18.3% of patients had chronic obstructive pulmonary disease (COPD), 21.5% had history of stroke, 7.5% had end-stage renal disease, 33.3% had diabetes, and 59.1% were current smokers. The mean time from PDT to decannulation was 39 days. Roughly one-fifth (20.4%) of patients were on dual antiplatelet therapy and 81.7% had anticoagulation restarted 8 hours post-tracheostomy. Eight complications were noted, including 5 instances of bleeding requiring packing and 1 case of mediastinitis. There were no significant predictors of decannulation prior to discharge. Only COPD was identified as a negative predictor of decannulation at any point in time (hazard ratio, 0.28; 95% confidence interval, 0.08-0.95; p=0.04). Percutaneous tracheostomy is a safe and viable alternative to surgical tracheostomy in cardiac surgery ICU patients. Patients who undergo PDT have a relatively short duration of tracheostomy and do not have major post-procedural complications.

Improved clinical outcomes following introduction of an attending intensivist for patients admitted to the cardiac surgical intensive care unit after valvular heart surgery: a single-center experience.

Although numerous studies have documented the improved clinical outcomes of patients undergoing cardiac surgery following introduction of attending intensivist, most of these studies included heterogeneous patient populations. We aimed to investigate the impact of an attending intensivist on the clinical outcomes of patients admitted to the cardiac surgical intensive care unit (CSICU) following valvular heart surgery. Patients who underwent valvular heart surgery between January 2007 and December 2012 (control group, n=337) were propensity matched (1:1) between January 2013 and June 2017 (intensivist group, n=407). During the propensity score matching analysis, 285 patients were extracted from each group. Patients in the intensivist group underwent mechanical ventilation for a significantly shorter time than those in the control group (21.8±69.8 vs. 39.2±115.3 hours, P=0.021). More patients were extubated within 6 hours in the intensivist group than in the control group (53.7% vs. 42.8%, P=0.015). The incidence of ventilator-associated pneumonia (1.4% vs. 4.9%, P=0.031), cardiac arrest due to cardiac tamponade associated with post-cardiotomy bleeding (0.4% vs. 3.9%, P=0.002), and acute kidney injury (2.8% vs. 7.7%, P=0.011) in the intensivist group was significantly lower than that in the control group. The 30-day mortality rate of the intensivist group was significantly lower than that of the control group (2.1% vs. 6.7%, P=0.015). Critical care provided in the CSICU staffed by an attending intensivist is associated with a lower 30-day mortality rate and reduced incidence of postoperative complications.

Agreement between cardiac output estimation with a wireless, wearable pulse decomposition analysis device and continuous thermodilution in post cardiac surgery intensive care unit patients.

Pulse Decomposition Analysis (PDA) uses integration of the systolic area of a distally transmitted aortic pulse as well as arterial stiffness estimates to compute cardiac output. We sought to assess agreement of cardiac output (CO) estimation between continuous pulmonary artery catheter (PAC) guided thermodilution (CO-CCO) and a wireless, wearable noninvasive device, (Vitalstream, Caretaker Medical, Charlottesville, VA), that utilizes the Pulse Decomposition Analysis (CO-PDA) method in postoperative cardiac surgery patients in the intensive care unit. CO-CCO measurements were compared with post processed CO-PDA measurements in prospectively enrolled adult cardiac surgical intensive care unit patients. Uncalibrated CO-PDA values were compared for accuracy with CO-CCO via a Bland-Altman analysis considering repeated measurements and a concordance analysis with a 10% exclusion zone. 259.7h of monitoring data from 41 patients matching 15,583 data points were analyzed. Mean CO-CCO was 5.55L/min, while mean values for the CO-PDA were 5.73L/min (mean of differences +- SD 0.79 ± 1.11L/min; limits of agreement - 1.43 to 3.01L/min), with a percentage error of 37.5%. CO-CCO correlation with CO-PDA was moderate (0.54) and concordance was 0.83. Compared with the CO-CCO Swan-Ganz, cardiac output measurements obtained using the CO-PDA were not interchangeable when using a 30% threshold. These preliminary results were within the 45% limits for minimally invasive devices, and pending further robust trials, the CO-PDA offers a noninvasive, wireless solution to complement and extend hemodynamic monitoring within and outside the ICU.

Remimazolam: A New Ingress in Cardiac Surgical Intensive Care Unit

Remimazolam is a novel intravenous ultrashort acting benzodiazepine that has a potential for safe and effective sedative. This recent addition in anesthesia armamentarium has a great role in procedural sedation and general anesthesia which is well-approved. The main beauty of this drug is, if needed its effects can be reversed by flumazenil which allows prompt termination of sedation. Remimizolam has fast predictive effect and recovery time, high procedural success rate, and minor respiratory and hemodynamic fluctuation when used for procedural sedation and general anesthesia. Although has a great potential for sedation in patients admitted to intensive care unit (ICU), some randomized trials are necessary to prove its long-term efficacy and safety in patients admitted to ICU.

Cardiac surgery during the COVID-19 pandemic.

The COVID-19 pandemic had several specific as well as general implications on cardiac surgery. Acute respiratory distress made extracorporeal oxygenation necessary in asignificant number of patients and accordingly many patients were treated in anesthesiological and even more in cardiac surgical intensive care units, which left only alimited number of beds in the intensive care units available for elective surgery cases. Moreover, the necessary availability of intensive care beds for severely diseased COVID-19 patients in general posed afurther limit, as did the relevant number of diseased personnel. Specific emergency plans were developed for many heart surgery units, limiting the number of elective cases. Increasing waiting lists were of course stressful for numerous elective-surgery patients and the decreased number of heart operations also meant afinancial burden for many units.

Algorithm-based detection of acute kidney injury according to full KDIGO criteria including urine output following cardiac surgery: a descriptive analysis

BackgroundAutomated data analysis and processing has the potential to assist, improve and guide decision making in medical practice. However, by now it has not yet been fully integrated in a clinical setting. Herein we present the first results of applying algorithm-based detection to the diagnosis of postoperative acute kidney injury (AKI) comprising patient data from a cardiac surgical intensive care unit (ICU).MethodsFirst, we generated a well-defined study population of cardiac surgical ICU patients by implementing an application programming interface (API) to extract, clean and select relevant data from the archived digital patient management system. Health records of N = 21,045 adult patients admitted to the ICU following cardiac surgery between 2012 and 2022 were analyzed. Secondly, we developed a software functionality to detect the incidence of AKI according to Kidney Disease: Improving Global Outcomes (KDIGO) criteria, including urine output. Incidence, severity, and temporal evolution of AKI were assessed.ResultsWith the use of our automated data analyzing model the overall incidence of postoperative AKI was 65.4% (N = 13,755). Divided by stages, AKI 2 was the most frequent maximum disease stage with 30.5% of patients (stage 1 in 17.6%, stage 3 in 17.2%). We observed considerable temporal divergence between first detections and maximum AKI stages: 51% of patients developed AKI stage 2 or 3 after a previously identified lower stage. Length of ICU stay was significantly prolonged in AKI patients (8.8 vs. 6.6 days, p < 0.001) and increased for higher AKI stages up to 10.1 days on average. In terms of AKI criteria, urine output proved to be most relevant, contributing to detection in 87.3% (N = 12,004) of cases.ConclusionThe incidence of postoperative AKI following cardiac surgery is strikingly high with 65.4% when using full KDIGO-criteria including urine output. Automated data analysis demonstrated reliable early detection of AKI with progressive deterioration of renal function in the majority of patients, therefore allowing for potential earlier therapeutic intervention for preventing or lessening disease progression, reducing the length of ICU stay, and ultimately improving overall patient outcomes.Graphical

Survey on the Updated German S3 Guideline for Intensive Care in Cardiac Surgery Patients.

The German guideline on intensive care treatment of cardiac surgical patients provides evidence-based recommendations on management and monitoring. It remains unclear if, respectively, to which degree the guidelines are implemented into the daily practice. Therefore, this study aims to characterize the implementation of guideline recommendations in German cardiac surgical intensive care units (ICUs). An internet-based online survey (42 questions, 9 topics) was sent to 158 German head physicians of cardiac surgical ICUs. To compare the effect over time, most questions were based on a previously performed survey (2013) after introduction of the last guideline update in 2008. A total of n = 65 (41.1%) questionnaires were included. Monitoring changed to increased provision of available transesophageal echocardiography specialists in 86% (2013: 72.6%), SvO2 measurement in 93.8% (2013: 55.1%), and electroencephalography in 58.5% (2013: 2.6%). The use of hydroxyethyl starch declined (9.4% vs. 2013: 38.7%), gelatin 4% presented the most administered colloid with 23.4% (2013: 17.4%). Low cardiac output syndrome was primarily treated with levosimendan (30.8%) and epinephrine (23.1%), while norepinephrine (44.6%) and dobutamine (16.9%) represented the most favored drug combination. The main way of distribution was web-based (50.9%), with increasing impact on therapy regimens (36.9% vs. 2013: 24%). Changes were found in all questioned sectors compared with the preceding survey, with persisting variability between ICUs. Recommendations of the updated guideline have increasingly entered clinical practice, with participants valuing the updated publication as clinically relevant.

Effect of Music Therapy in Relieving the Symptom Experiences and Improving Outcomes of Critical Care Patients: A Systematic Review

BACKGROUND: The critical care patients often have unpleasant experiences while facing critical care activities that may impact on physiological and psychological changes. Music therapy is a simple complementary therapy that can promote relaxation, potentially affecting some outcomes for critical care patients. However, a systematic review related to the application of music therapy and the results obtained from music therapy is still limited. AIM: This systematic review aimed to explore and evaluate the effectiveness of music therapy interventions in relieving symptom experiences and improving outcomes of critical care patients. METHODS: We conducted a systematic review of research articles focused on music therapy in critical care areas using three electronic databases, namely PubMed, CINAHL, and Science Direct. We selected English articles that had a Randomized Control Trial (RCT) design and published from 2005 to 2020. PICO format used to search literature; “Adult, Critical Care, Intensive Care Unit (ICU), Critically Ill Patients, Intensive Care, Surgical ICU, Cardiac Surgical ICU, Medical-Surgical ICU, Therapy Music, Music Therapy, Pain, Agitation, Stress. Anxiety, Delirium, Mobility, Sleep, Physiological, and Psychological. The risk of bias from 19 articles was evaluated by JBI. Furthermore, we assessed the manner and effect of music therapy on various symptoms. RESULTS: The researchers found 743 articles. After manual review, 19 RCTs (1711 participants) were identified and assessed. Based on the review, it was reported that music with a relaxed rhythm, soft tone, slow rhythm; duration from 15 to 60 min; 60–80 beats/min, and using headphones, has a significant effect on pain, anxiety, stress, delirium, mobilization, sleep quality, sedation levels, and physiological parameters. CONCLUSION: Music therapy has proved effective in reducing symptoms experiences of critical care patients. It is suggested to be used as a viable option in symptoms management of patients in ICU.

Agreement between cardiac output estimation by multi-beat analysis of arterial blood pressure waveforms and continuous thermodilution in post cardiac surgery intensive care unit patients

We sought to assess agreement of cardiac output estimation between continuous pulmonary artery catheter (PAC) guided thermodilution (CO-CTD) and a novel pulse wave analysis (PWA) method that performs an analysis of multiple beats of the arterial blood pressure waveform (CO-MBA) in post-operative cardiac surgery patients. PAC obtained CO-CTD measurements were compared with CO-MBA measurements from the Argos monitor (Retia Medical; Valhalla, NY, USA), in prospectively enrolled adult cardiac surgical intensive care unit patients. Agreement was assessed via Bland-Altman analysis. Subgroup analysis was performed on data segments identified as arrhythmia, or with low CO (less than 5 L/min). 927 hours of monitoring data from 79 patients was analyzed, of which 26 had arrhythmia. Mean CO-CTD was 5.29 ± 1.14 L/min (bias ± precision), whereas mean CO-MBA was 5.36 ± 1.33 L/min, (4.95 ± 0.80 L/min and 5.04 ± 1.07 L/min in the arrhythmia subgroup). Mean of differences was 0.04 ± 1.04 L/min with an error of 38.2%. In the arrhythmia subgroup, mean of differences was 0.14 ± 0.90 L/min with an error of 35.4%. In the low CO subgroup, mean of differences was 0.26 ± 0.89 L/min with an error of 40.4%. In adult patients after cardiac surgery, including those with low cardiac output and arrhythmia CO-MBA is not interchangeable with the continuous thermodilution method via a PAC, when using a 30% error threshold.

Every Deep-Drawn Breath: A Critical Care Doctor on Healing, Recovery, and Transforming Medicine in the ICU, by Wes Ely. Published by Scribner, New York, NY. 332 pages, $28.00. 2021. ISBN: 9781982171148.

1 Cardiac Surgical ICU, Cape Fear Valley Medical Center, Fayetteville, NC 2 Department of Anesthesiology, Perioperative, and Pain Medicine, Stanford University School of Medicine, VA Palo Alto Health Care System, Palo Alto, CA

Prediction of depression among patients with cardiovascular disease scheduled cardiac surgery using an AI-empowered pipeline

Abstract Introduction Several studies have demonstrated, that one in three patients with cardiovascular disease (CVD) suffers from depression, and depression increases the likelihood for cardiac morbidity and mortality in the CVD population by 2–3-fold, independently of traditional risk factors or gender. Purpose The aim of this work is to predict depression in patients with CVD. Methods The clinical study was conducted in the cardiac surgical intensive care unit of The Central Clinical Hospital, the Medical University of Lodz in Poland, among patients scheduled for elective coronary artery bypass graft surgery (CABG). 224 patients signed an informed consent form, met the inclusion criteria and were enrolled in the study. All the patients had chronic coronary syndromes. The inclusion criteria were: consecutive adult patients scheduled for CABG surgery or CABG surgery with cardiac valve repair or replacement (CVR). The study population was examined by a psychiatrist the day prior to the scheduled operation, a diagnosis of major depressive disorder (MDD) was established on the basis of DSM-5 criteria. A data curation pipeline was applied to automatically remove outliers and duplicated fields in the input dataset. An AI-empowered pipeline was developed to classify patients at higher risk for depression. Random downsampling with replacement was applied to deal with the increased class imbalance by taking into consideration two confound factors, namely the Mini Mental State Examination and Hemoglobin concentration &amp;lt;10mg/dl. The downsampling process was repeated K times. In each iteration, three bagging and boosting ML schemas were utilized for the classification task including the AdaBoost (adaptive boosting), Random Forests and Extreme Gradient boosting trees (XGBoost). To this end, a nested cross-validation process was applied for hyperparameter optimization and model validation, where: (i) a 3-fold cross-validation process was first applied to seek for the optimal set of hyperparameters based on the grid search approach by tuning core parameters, including the learning rate, number of estimators, and max depth, among others, and (ii) a stratified 5-fold cross-validation process was subsequently used to evaluate the performance of the best model from the previous stage by computing the classification accuracy, sensitivity, specificity, and area under the ROC curve (AUC). The performance evaluation results were averaged across the 5 folds and across the downsampling iterations. Shappley additive explanation analysis was finally applied to provide explainable risk factors for depression. Results sRAGE was significantly correlated with depression (r=0.32, p=0.038). The Random forests classifier presented the highest performance to predict depression (Accuracy: 0.71, Sensitivity: 0.71, Specificity: 0.75 and AUC: 0.74). Conclusions Depression can be predicted with 71% accuracy at patients with CVD and elevated sRAGE values. Funding Acknowledgement Type of funding sources: Public grant(s) – EU funding. Main funding source(s): HORIZON2020

INFRARED THERMOGRAPHY-BASED BODY-SURFACE THERMAL INHOMOGENEITY MONITORING TO ASSESS THE SEVERITY OF HYPOPERFUSION IN CRITICALLY ILL PATIENTS.

Background: Uneven body-surface thermal distribution is a manifestation of hypoperfusion and can be quantified by infrared thermography. Our aim was to investigate whether body-surface thermal inhomogeneity could accurately evaluate the severity of patients at risk of hypoperfusion. Methods: This was a prospective cohort study in which infrared thermography images were taken from unilateral legs of critically ill patients at high risk of hypoperfusion in a cardiac surgical intensive care unit. For each patient, five body-surface thermal inhomogeneity parameters, including standard deviation (SD), kurtosis, skewness, entropy, and low-temperature area rate (LTAR), were calculated. Demographic, clinical, and thermal characteristics of deceased and living patients were compared. The risk of mortality and capillary refill time (CRT) were chosen as the primary outcome and benchmarking parameter for hypoperfusion, respectively. The area under the receiver operating characteristic curve (AUROC) was used to evaluate predictive accuracy. Results: Three hundred seventy-three patients were included, and 55 (14.7%) died during hospital stay. Of inhomogeneity parameters, SD (0.738) and LTAR (0.768) had similar AUROC to CRT (0.757) for assessing mortality risk. Besides, there was a tendency for LTAR (1%-3%-7%) and SD (0.81°C-0.88°C-0.94°C) to increase in normotensive, hypotensive, and shock patients. These thermal parameters are associated with CRT, lactate, and blood pressure. The AUROC of a combined prediction incorporating three thermal inhomogeneity parameters (SD, kurtosis, and entropy) was considerably higher at 0.866. Conclusions: Body-surface thermal inhomogeneity provided a noninvasive and accurate assessment of the severity of critically ill patients at high risk of hypoperfusion.

A Contemporary Analysis of Phlebotomy and Iatrogenic Anemia Development Throughout Hospitalization in Critically Ill Adults.

Anemia is common in critically ill patients and may be exacerbated through phlebotomy-associated iatrogenic blood loss. Differences in phlebotomy practice across patient demographic characteristics, clinical features, and practice environments are unclear. This investigation provides a comprehensive description of contemporary phlebotomy practices for critically ill adults. This is an observational cohort study of adults ≥18 years of age requiring intensive care unit (ICU) admission between January 1, 2019, and December 31, 2019, at a large academic medical center. Descriptive statistics were utilized to summarize all phlebotomy episodes throughout hospitalization, with each phlebotomy episode defined by unique peripheral venous, central venous, or arterial accesses for laboratory draws, exclusive of finger sticks. Secondarily, financial costs of phlebotomy and the relationships between phlebotomy practices, hemoglobin concentrations, and red blood cell (RBC) transfusions were evaluated. A total of 6194 patients were included: 59% were men with a median (interquartile range) age of 66 (54-76) years and median ICU and hospital durations of 2.1 (1.4-3.9) and 7.1 (4.3-11.8) days, respectively. The median number of unique laboratory draws was 41 (18-88) throughout hospitalization, with a median volume of 232 (121-442) mL, corresponding to 5.2 (2.6-8.8) draws and 29 (19-43) mL per day. Waste (ie, discard) volume was responsible for 10.8% of total phlebotomy volume. Surgical patients had a higher number of phlebotomy episodes and greater total phlebotomy volumes compared to nonsurgical patients. Phlebotomy practices differed across ICU types, with the greatest frequency of laboratory draws in the cardiac surgical ICU and the greatest daily phlebotomy volume in the medical ICU. Across hospitalization, ICU environments had the greatest frequency and volumes of laboratory draws, with the least intensive phlebotomy practice observed in the general hospital wards. Patients in the highest quartile of cumulative blood drawn experienced the longest hospitalizations, lowest nadir hemoglobin concentrations, and greatest RBC transfusion utilization. Differences in phlebotomy practice were limited across patient age, gender, and race. Hemoglobin concentrations declined during hospitalization, congruent with intensity of phlebotomy practice. Each 100 mL of phlebotomy volume during hospitalization was associated with a 1.15 (95% confidence interval [CI], 1.14-1.17; P < .001) multiplicative increase in RBC units transfused in adjusted analyses. Estimated annual phlebotomy costs exceeded $15 million (approximately $2500 per patient admission). Phlebotomy continues to be a major source of blood loss in hospitalized patients with critical illness, and more intensive phlebotomy practices are associated with lower hemoglobin concentrations and greater transfusion utilization.