Surgical Intensive Care Unit Patients Research Articles

Noninvasive cardiovascular monitoring based on electrocardiography and ballistocardiography: a feasibility study on patients in the surgical intensive care unit.

The time interval between the peaks in the electroccardiogram (ECG) and ballistocardiogram (BCG) waveforms, TEB, has been associated with the pre-ejection period (PEP), which is an important marker of ventricular contractility. However, the applicability of BCG-related markers in clinical practice is limited by the difficulty to obtain a replicable and consistent signal on patients. In this study, we test the feasibility of BCG measurements within a complex clinical setting, by means of an accelerometer under the head pillow of patients admitted to the Surgical Intensive Care Unit (SICU). The proposed technique proved capable of capturing TEB based on the R peaks in the ECG and the BCG in its head-to-toe and dorso- ventral directions. TEB detection was found to be consistent and repeatable both in healthy individuals and SICU patients over multiple data acquisition sessions. This work provides a promising starting point to investigate how TEB changes may relate to the patients' complex health conditions and give additional clinical insight into their care needs.

Inferior Vena Cava Collapsibility Index and Central Venous Pressure for Fluid Assessment in the Critically Ill Patient.

To determine the correlation of sonographic evaluation of inferior vena cava diameter and its collapsibility index with central venous pressure in both spontaneously breathing and mechanically ventilated patients in surgical ICU. Study Design: Cross-sectional study. Surgical ICU, Lahore General Hospital from November 2020 to May 2021. All patients above 18 years of age, who had central venous catheter placed, were included in the study. Patients with raised abdominal pressure, pregnancy, morbid obesity, heart diseases, and those unable to lie in supine position, were excluded. For inferior vena cava (IVC) measurement, IVC was visualised using curvilinear probe of ultrasound machine. Minimum and maximum diameters of inferior vena cava were calculated in every respiratory phase. IVC collapsibility index was expressed in percentage. Central venous pressure (CVP) was recorded soon after IVC measurement, using manometer. Total number of patients was 126. A significant correlation was seen between IVC measurements (inferior vena cava diameters and the collapsibility index) and CVP, (p<0.001), but the regression coefficients were less in patients who were mechanically ventilated (r=0.779 for IVC maximum diameter and -0.725 for collapsibility index) than the patients who were breathing spontaneously (r=0.850 for IVC maximum diameter and -0.899 for collapsibility index) Conclusion: Evaluation of IVC diameter and its collapsibility index is an easy and non-invasive method to evaluate intravascular volume status of critically ill patients. Its use is more helpful in patients who are spontaneously breathing than those who are mechanically ventilated. Key Words: Central venous pressure, Fluid status, Inferior vena cava diameter.

Ultrasound-Guided Pleural Effusion Drainage: Effect on Oxygenation, Respiratory Mechanics, and Liberation from Mechanical Ventilation in Surgical Intensive Care Unit Patients.

The question as to whether an aggressive management of post-operative pleural effusion may improve clinical outcomes after major surgery remains unanswered. The aim of this study was to investigate the effect of ultrasound-guided pleural effusion drainage on oxygenation, respiratory mechanics, and liberation from mechanical ventilation in surgical intensive care unit patients. Oxygenation and respiratory mechanics were measured before and after drainage. Over an 18-month period, a total of 62 patients were analyzed. The mean drainage volume during the first 24 h was 864 ± 493 mL, and there were no procedural complications. Both the mean PaO2/FiO2 ratio and lung compliance improved after drainage. Additionally, 41.9% (n = 26) of patients were ventilator-free within 72 h after drainage. Multivariable logistic regression analysis revealed that non-cardiovascular or thoracic surgery (odds ratio [OR] = 4.968, p = 0.046), a longer time interval from operation to the onset of pleural effusion (OR = 1.165, p = 0.005), and a higher peak airway pressure (OR = 1.303, p = 0.009) were independent adverse predictors for being free from mechanical ventilation within 72 h after drainage. Specifically, patients with a time from surgery to the onset of pleural effusion ≤6 days—but not those with an interval >6 days—showed a significant post-procedural improvement in terms of PaO2/FiO2 ratio, PaCO2, peak airway pressure, and dynamic lung compliance. In summary, ultrasound-guided pleural effusion drainage resulted in significant clinical benefits in mechanically ventilated ICU patients after major surgery—especially in those with early-onset effusion who received thoracic surgery.

The association between sarcopenia and functional outcomes after hospital discharge among critically ill surgical patients

To determine the association between sarcopenia in surgical intensive care unit (SICU) patients and long-term functional outcomes. This prospective, cohort study enrolled patients aged >65 years admitted to SICUs at the tertiary care hospital. Their muscle mass and strength were measured by bioelectrical impedance vector analysis (BIVA) and handgrip-strength or manual-muscle-strength tests. The functional outcomes were evaluated with the Thai version of the Barthel index for activities of daily living (ADL). 120 patients were enrolled. A multivariate analysis identified 3 independent predictors associated with poor functional outcomes (ADL scores ≤70) at one month after hospital discharge including sarcopenia (adjusted odds ratio [aOR]: 3.33; 95% confidence interval [CI]: 1.25-8.87); duration of mechanical ventilation (aOR: 1.19; 95% CI: 1.02-1.38); and length of hospital stay (aOR: 1.05; 95% CI: 1.01-1.10). Cox proportional-hazards regression models found that sarcopenia (adjusted hazard ratio [aHR]: 2.07; 95% CI: 1.02-4.22) and admission severity (aHR: 1.13; 95% CI: 1.04-1.23) were predictors of 120-day mortality. Sarcopenia was an independent predictor of poor functional outcomes at one month after hospital discharge.

Evaluation of MRSA Surveillance Nasal Swabs for Predicting MRSA Infection in Surgical Intensive Care Unit Patients

We aimed to examine the clinical value of serial MRSA surveillance cultures to rule out a MRSA diagnosis on subsequent cultures during a patient's surgical intensive care unit (SICU) admission. We performed a retrospective cohort study to evaluate patients who received a MRSA surveillance culture at admission to the SICU (n=6,915) and collected and assessed all patient cultures for MRSA positivity during their admission. The primary objective was to evaluate the transition from a MRSA negative surveillance on admission to MRSA positive on any subsequent culture during a patient's SICU stay. Percent of MRSA positive cultures by type following MRSA negative surveillance cultures was further analyzed. 6,303 patients received MRSA nasal surveillance cultures at admission with 21,597 clinical cultures and 7,269 MRSA surveillance cultures. Of the 6,163 patients with an initial negative, 53 patients (0.87%) transitioned to MRSA positive. Of the 139 patients with an initial positive, 30 (21.6%) had subsequent MRSA positive cultures. Individuals who had an initial MRSA surveillance positive status on admission predicted MRSA positivity rates for cultures in qualitative lower respiratory cultures (64.3% versus. 3.1%), superficial wound (60.0% versus 1.6%), deep wound (39.0% versus 0.8%), tissue culture (26.3% versus 0.6%), and body fluid (20.8% versus 0.7%) cultures when compared to MRSA negative patients on admission. Following MRSA negative nasal surveillance cultures patients showed low likelihood of MRSA infection suggesting empiric anti-MRSA treatment is unnecessary for specific patient populations. SICU patient's MRSA status at admission should guide empiric anti-MRSA therapy.

Cerebral monitoring in surgical ICU patients

To give an overview of cerebral monitoring techniques for surgical ICU patients. As the burden of postsurgical neurological and neurocognitive complications becomes increasingly recognized, cerebral monitoring in the surgical ICU might gain a relevant role in detecting and possibly preventing adverse outcomes. However, identifying neurological alterations in surgical ICU patients, who are often sedated and mechanically ventilated, can be challenging. Various noninvasive and invasive techniques are available for cerebral monitoring, providing an assessment of cortical electrical activity, cerebral oxygenation, blood flow autoregulation, intracranial pressure, and cerebral metabolism. These techniques can be used for the diagnosis of subclinical seizures, the assessment of sedation depth and delirium, the detection of an impaired cerebral blood flow, and the diagnosis of neurosurgical complications. Cerebral monitoring can be a valuable tool in the early detection of adverse outcomes in surgical ICU patients, but the evidence is limited, and clear clinical indications are still lacking.

The association of macronutrient deficit with functional status at discharge from the intensive care unit: a retrospective study from a single-center critical illness registry.

Nutrition is often thought to influence outcomes in critically ill patients. However, the relationship between macronutrient delivery and functional status is not well characterized. Our goal was to investigate whether caloric or protein deficit over the course of critical illness is associated with functional status at the time of intensive care unit (ICU) discharge. We performed a retrospective analysis of surgical ICU patients at a teaching hospital in Boston, MA. To investigate the association of caloric or protein deficit with Functional Status Score for the ICU (FSS-ICU), we constructed linear regression models, controlling for age, sex, race, body mass index, Nutritional Risk in the Critically Ill score, and ICU length of stay. We then dichotomized caloric as well as protein deficit, and performed logistic regressions to investigate their association with functional status, controlling for the same variables. Linear regression models (n = 976) demonstrated a caloric deficit of 238 kcal (237.88; 95%CI 75.13-400.63) or a protein deficit of 14 g (14.23; 95%CI 4.46-24.00) was associated with each unit decrement in FSS-ICU. Logistic regression models demonstrated a 6% likelihood (1.06; 95%CI 1.01-1.14) of caloric deficit ≥6000 vs. <6000 kcal and an 8% likelihood (1.08; 95%CI 1.01-1.15) of protein deficit ≥300 vs. <300 g with each unit decrement in FSS-ICU. In our cohort of patients, macronutrient deficit over the course of critical illness was associated with worse functional status at discharge. Future studies are needed to determine whether optimized macronutrient delivery can improve outcomes in ICU survivors.

Predicting brain function status changes in critically ill patients via Machine learning.

In intensive care units (ICUs), a patient's brain function status can shift from a state of acute brain dysfunction (ABD) to one that is ABD-free and vice versa, which is challenging to forecast and, in turn, hampers the allocation of hospital resources. We aim to develop a machine learning model to predict next-day brain function status changes. Using multicenter prospective adult cohorts involving medical and surgical ICU patients from 2 civilian and 3 Veteran Affairs hospitals, we trained and externally validated a light gradient boosting machine to predict brain function status changes. We compared the performances of the boosting model against state-of-the-art models-an ABD predictive model and its variants. We applied Shapley additive explanations to identify influential factors to develop a compact model. There were 1026 critically ill patients without evidence of prior major dementia, or structural brain diseases, from whom 12 295 daily transitions (ABD: 5847 days; ABD-free: 6448 days) were observed. The boosting model achieved an area under the receiver-operating characteristic curve (AUROC) of 0.824 (95% confidence interval [CI], 0.821-0.827), compared with the state-of-the-art models of 0.697 (95% CI, 0.693-0.701) with P < .001. Using 13 identified top influential factors, the compact model achieved 99.4% of the boosting model on AUROC. The boosting and the compact models demonstrated high generalizability in external validation by achieving an AUROC of 0.812 (95% CI, 0.812-0.813). The inputs of the compact model are based on several simple questions that clinicians can quickly answer in practice, which demonstrates the model has direct prospective deployment potential into clinical practice, aiding in critical hospital resource allocation.

Lung Ultrasound-Guided Fluid Management versus Standard Care in Surgical ICU Patients: A Randomised Controlled Trial.

The value of lung ultrasound (LU) in assessing extravascular lung water (EVLW) was demonstrated by comparing LU with gold-standard methods for EVLW assessment. However, few studies have analysed the value of B-Line score (BLS) in guiding fluid management during critical illness. The purpose of this trial was to evaluate if a BLS-guided fluid management strategy could improve fluid balance and short-term mortality in surgical intensive care unit (ICU) patients. We conducted a randomised, controlled trial within the ICUs of two university hospitals. Critically ill patients were randomised upon ICU admission in a 1:1 ratio to BLS-guided fluid management (active group) or standard care (control group). In the active group, BLS was monitored daily until ICU discharge or day 28 (whichever came first). On the basis of BLS, different targets for daily fluid balance were set with the aim of avoiding or correcting moderate/severe EVLW increase. The primary outcome was 28-day mortality. Over 24 months, 166 ICU patients were enrolled in the trial and included in the final analysis. Trial results showed that daily BLS monitoring did not lead to a different cumulative fluid balance in surgical ICU patients as compared to standard care. Consecutively, no difference in 28-day mortality between groups was found (10.5% vs. 15.6%, p = 0.34). However, at least 400 patients would have been necessary for conclusive results.

Clinical profile, antifungal susceptibility, and molecular characterization of Candida auris isolated from patients in a South Indian surgical ICU.

BackgroundCandida auris is an emerging multidrug resistant yeast which causes blood stream infection especially among critically ill patients. This yeast can also colonize patients and are isolated from hospital environment causing outbreaks in hospital settings. ObjectiveTo describe possible outbreak of C. auris infection in surgical ICU and characterize the isolates by molecular typing and azole resistance mechanism. MethodsAfter isolation of Candida auris from cluster of patients from surgical ICU, environment survey was done to identify the source in the hospital. The identity of the isolates was confirmed by Matrix Assisted Laser Desorption Ionisation Time of Flight mass spectroscopy and sequencing 26S and ITS region of rDNA. Molecular typing was done by fluorescent amplified fragment length polymorphism technique. Antifungal susceptibility testing was performed by CLSI broth dilution technique. ERG11 gene was sequenced to screen for mutations responsible for azole resistance. Results and conclusionA total of eight C. auris was isolated during the four months (December 2018–March 2019) suggesting possible of outbreak in surgical ICU of tertiary care center in South India. C. auris (n = 8) was isolated from urine (n = 4), blood (n = 3) and ear discharge (n = 1) samples. Based on 26S sequence analysis all our isolates belonged to South Asian clade. All the isolates had minimum inhibitory concentration (MIC) of ≥16 µg/ml to fluconazole. ERG11 sequence exhibited amino acid substitution Y132F in all the isolates. The two environmental isolates clustered closely with an isolate from urine sample. Adherence to strict infection control practices prevented further spread of infection.

Chronic Critical Illness Elicits a Unique Circulating Leukocyte Transcriptome in Sepsis Survivors.

Surgical sepsis has evolved into two major subpopulations: patients who rapidly recover, and those who develop chronic critical illness (CCI). Our primary aim was to determine whether CCI sepsis survivors manifest unique blood leukocyte transcriptomes in late sepsis that differ from transcriptomes among sepsis survivors with rapid recovery. In a prospective cohort study of surgical ICU patients, genome-wide expression analysis was conducted on total leukocytes in human whole blood collected on days 1 and 14 from sepsis survivors who rapidly recovered or developed CCI, defined as ICU length of stay ≥ 14 days with persistent organ dysfunction. Both sepsis patients who developed CCI and those who rapidly recovered exhibited marked changes in genome-wide expression at day 1 which remained abnormal through day 14. Although summary changes in gene expression were similar between CCI patients and subjects who rapidly recovered, CCI patients exhibited differential expression of 185 unique genes compared with rapid recovery patients at day 14 (p < 0.001). The transcriptomic patterns in sepsis survivors reveal an ongoing immune dyscrasia at the level of the blood leukocyte transcriptome, consistent with persistent inflammation and immune suppression. Furthermore, the findings highlight important genes that could compose a prognostic transcriptomic metric or serve as therapeutic targets among sepsis patients that develop CCI.

Phenotypic and Genotypic Evaluation of Antibiotic Resistance of Acinetobacter baumannii Bacteria Isolated from Surgical Intensive Care Unit Patients in Pakistan

Background: Carbapenem-resistant Acinetobacter baumannii (CRAB) is a significant nosocomial pathogen, causing serious threats concerning community-wide outbreaks globally, as well as in Pakistan. Antimicrobial resistance in A. baumannii is increasing day by day. Objectives: The study aimed to find out the antibiotic resistance (AMR) patterns and evaluate the AMR genes in clinical isolates from patients admitted to the surgical Intensive Care units (ICUs) at different hospitals in Lahore, Pakistan. Methods: A total of 593 clinical specimens were collected from patients admitted to the surgical ICUs of three different local hospitals in Lahore, Pakistan. From these samples, a total of 90 A. baumannii isolates were identified and further investigated to observe phenotypic resistance patterns and detect carbapenemases resistance genes. Results: The results showed that phenotypic resistance against amikacin was 27.2%, ceftriaxone 100%, ceftazidime 27.2%, cefepime 63.3%, ciprofloxacin and co-trimoxazole 100%, gentamicin 40%, imipenem 22.2%, meropenem 21.1%, piperacillin-tazobactam 27.2%, tigecycline 27.2%, and tetracycline 63.3%. All A. baumannii isolates were found to be sensitive to colistin (CT), polymixin-B (PB), and tobramycin (TOB). The PCR amplification of carbapenemases genes revealed the prevalence of blaOXA-23, blaOXA-51, and blaOXA-40 in 73, 90, and 64.4% of the isolates, respectively, along with blaNDM1 (92.2%), blaVIM (40%), blaIMP (90%), ISAba1 (85.5%), sul1 (16.6%), sul2 (20%), armA (32.2%), and PER-1 (12%) while the blaOXA-24 and blaOXA-58 genes were not detected in the isolates. The sequence analysis of the blaOXA-23 and blaOXA-51 genes showed 98% and 95% similarity with previously reported sequences in the GenBank database. Conclusions: The present study indicated that the emergence of high carbapenem resistance in CRAB isolates has increased, which may pose serious limitations in the choice of drugs for nosocomial infections.

Hypoglycemia in a Surgical Intensive Care Unit.

Glycemic control is an important aspect of critical care because derangements are associated with morbidity and mortality. Patients at highest risk for hypoglycemia in the surgical intensive care unit (SICU) are incompletely described by existing literature. Our objective was to delineate this high-risk patient population in our SICU. In this single-center, retrospective, observational study, SICU patients admitted from June 1, 2019 to July 31, 2020 with ≥1 episode of hypoglycemia (blood glucose <60mg/dL) were included. There were 41 hypoglycemic events in 27 patients, comprising an incidence of 1.5% among SICU patients. The most common admission diagnoses were cirrhosis (n = 13, 48%), polytrauma (n = 12, 44%), multisystem organ failure (n = 11, 41%), diabetes mellitus (n = 9, 33%), and soft tissue infection (n = 8, 30%). Four high-risk populations were identified: patients in multisystem organ failure (MSOF) (n = 11, 41%); those who were nil per os (NPO) (n = 10, 37%); patients receiving long acting subcutaneous insulin, for example, Lantus (n = 3, 11%); and those on continuous intravenous insulin infusions (n = 3, 11%). After multi-disciplinary peer review, most hypoglycemic events (n = 16, 59%) were deemed iatrogenic. Hypoglycemia is rare in surgical critical care. When it does occur, patients are typically in MSOF, NPO, on long acting subcutaneous insulin or continuous insulin infusions, have soft tissue infections, or have acute or chronic liver failure. Increased vigilance with frequent blood glucose monitoring in these high-risk patients may reduce the risk of hypoglycemia in the SICU.

Correcting Coagulopathy With Fresh Frozen Plasma in the Surgical Intensive Care Unit: How Much Do We Need to Transfuse?

Thromboelastography (TEG) is an assay that assesses the coagulation status. Patients with prolonged reaction time (R) require fresh frozen plasma (FFP); however, the volume required to correct the R time is unknown. We sought to quantify the volume required to correct the R time and calculate the response ratio in our surgical intensive care unit (SICU) to allow for targeted resuscitation. Surgical intensive care unit patients between Aug 2017 and July 2019 with a prolonged initial R time and at least two TEG tests performed within 24hours were included. The response ratio was defined as the change in the R time divided by the number of FFP units. High responders (response ratio >5minutes/unit) were compared to low responders (response ratio ≤5minutes/unit). Forty-six patients were included. While the mean response ratio was 5minutes/unit, there was significant variation among patients. There were 28.0 (60.9%) low responders and 18.0 (39.1%) high responders. Low responders were more likely male (64.0% vs. 33.0%, P = .04), had a higher Acute Physiology and Chronic Health Evaluation (APACHE) IV score (42.0 vs. 27.0, P = .03), and a higher mortality rate (54.0% vs. 22.0%, P = .04). On average, one unit of FFP corrects the R time by 5minutes; however, there was significant variation between high and low responders. Male patients with higher APACHE IV score are expected to be low responders with a higher mortality rate. These findings can guide FFP transfusion and provide additional prognostication.

An examination of cardiovascular intensive care unit mortality based on admission day and time

BackgroundRecent interest in the ‘weekend effect’ has been expanded to cardiovascular intensive care units, yet the impact of off-hours admission on mortality and cardiovascular ICU (CICU) length of stay remains uncertain. ObjectivesWe examine the association between CICU admission day and time with mortality. Additionally, length-of-stay was also evaluated in relation to admission time. MethodsA single-center, retrospective cohort study was conducted including 10,638 adult patients admitted to a CICU in a tertiary-care academic medical center from July 1, 2012 to June 30, 2019. ICU mortality and length-of-stay were assessed by admission day and time adjusting for comorbid conditions and other clinical variables. We used logistic regression models to evaluate the factors associated with mortality and a generalized linear model (GLM) with log link function and gamma distribution was used to evaluate the factors associated with ICU length of stay. ResultsCompared to weekday-day admissions, we observed an increased mortality for weekend-day for all admissions (6.5 vs 9.6%, Adjusted OR: 1.32 (1.03–1.72)), and for medical CICU admissions (7.6 vs 9.9%, Adjusted OR: 1.35 (1.02–1.79)). Additionally, compared to weekday-day, weekday-night admission was associated with 7% longer ICU length of stay in surgical ICU patients, 7% shorter length of stay in medical ICU patients. ConclusionAdmission to this open-model CICU during weekend hours (Saturday 08:00-Sunday 17:59) versus nights or weekdays is associated with increased mortality. ICU staffing care models should not significantly change based on the day of the week.

Protocol Use in Surgical Intensive Care Units

BackgroundProtocols are common in intensive care, however the association between protocol prevalence and outcomes in surgical ICU patients is unclear. We hypothesized that ICUs in a multicenter database using more protocols had better outcomes. Material and MethodsThis is a retrospective analysis of prospectively collected data from a 2-d prevalence study with 30-d follow up, on surgical and trauma patients in ICUs at 42 trauma centers. Use of forty clinical protocols was queried. Protocol prevalence was categorized by quartile into Low (first), Moderate (second and third), or High (fourth) use ICUs. The primary outcome was in-hospital mortality; secondary outcomes were ventilator, ICU, and hospital days, mechanical ventilation, tracheostomy, renal replacement, transfusion, and hospital-acquired infections. ResultsData from 1044 surgical and trauma patients were analyzed. Protocol use was not different for “closed” (n = 20), “open” (n = 9), or “semi-open” (n = 13) ICUs (P= 0.20). Thirty-day in-hospital mortality was 8.4%, and not associated with number of protocols (OR 1.01 [95% CI 0.98-1.03], P= 0.65). There was no statistically significant difference between High and Low use ICUs for ventilator days (OR 0.86; 0.52-1.43), tracheostomy (OR 0.8; 0.47-1.38), renal replacement therapy (OR 0.66; 0.04-9.82), transfusion (OR 0.95; 0.58-1.57), or hospital-acquired infections (OR 1.07; 0.67-1.7). Higher mortality was seen in open (versusclosed; OR 1.74 [1.05-2.89], P= 0.033), and surgical/trauma (versustrauma; OR 1.86 [1.33-2.61]; P< 0.001). ConclusionsIn this multicenter observational study of surgical ICU patients, no association was found between the number of protocols used and patient outcomes.

The Impact of Earplugs and Eye Masks on Sleep Quality in Surgical ICU Patients at Risk for Frequent Awakenings.

Sleep disturbances may contribute to the development of delirium, prolonged ICU stay, and increased mortality. There is conflicting data on the effectiveness of earplugs and eye masks for sleep promotion in the ICU. This study evaluates the impact of earplugs and eye masks on sleep quality in postoperative surgical ICU patients at risk for frequent awakenings. Prospective randomized controlled trial. Surgical ICU within the University of Texas Southwestern Medical Center. Adult, female patients admitted to the surgical ICU requiring hourly postoperative assessments following breast free flap surgery between February 2018 and October 2019. Patients were randomized into an intervention group or a control group. The intervention group received earplugs and eye masks in addition to standard postoperative care, whereas the control group received standard postoperative care. The primary outcome was overall sleep quality assessed via the Richards-Campbell Sleep Questionnaire. Secondary outcomes of patient satisfaction and rates of ICU delirium were assessed with a modified version of the Family Satisfaction in the ICU survey and the Confusion Assessment Method for the ICU. After a planned interim analysis, the study was stopped early because prespecified criteria for significance were attained. Compared with the control group's average Richards-Campbell Sleep Questionnaire total score of 47.3 (95% CI, 40.8-53.8), the intervention group's average Richards-Campbell Sleep Questionnaire total score was significantly higher at 64.5 (95% CI, 58.3-70.7; p = 0.0007). There were no significant between-group differences for Confusion Assessment Method for the ICU scores or modified Family Satisfaction in the ICU survey scores. These results suggest that earplugs and eye masks are effective in improving sleep quality in ICU patients undergoing frequent assessments. The results strengthen the evidence for nonpharmacologic sleep-promoting adjuncts in the ICU.

600 Social Determinants and Their Impact on Burn Surgery Outcomes

Abstract Introduction To our knowledge, no studies have been conducted assessing the social determinants of health and the impact on the outcomes for burn patients. Such studies are needed considering burn injuries are associated with high costs, severe psychological impact, and a high burden placed on the healthcare systems. The burden is hypothesized to be aggravated by the increasing amount of diabetes and obesity seen in the general population which put patients at increased risk for developing chronic wounds. Studies have shown that several socioeconomic status (SES) factors are associated with increased risk of burns, but none have documented the outcomes of burn patients based on their social determinants of health. In our study, we will be comparing patients in the burn ICU (BICU) to patients in the surgical ICU (SICU). The purpose of this comparison is to evaluate whether the same social determinants of health have similar influences in both groups. Methods We performed a retrospective analysis of population group data from patients admitted to the BICU and SICU from January 1, 2016, to November 18, 2019. The primary outcomes were length-of-stay (LOS), mortality, 30-day-readmission, and hospital charges. Pearson’s chi-square test for categorical variables and t-test for continuous variables were used to compare population health groups. Results We analyzed a total of 487 burn and 510 surgical patients. When comparing BICU and SICU patients, we observed significantly higher mean hospital charges and LOS in burn patients with a history of mental health (mean difference: $42,756.04, p=0.013 and 7.12 days, p=0.0085), ESRD ($57,8124.7, p=0.0047 and 78.62 days, p=0.0104), sepsis ($168,825.19, p=&amp;lt; 0.001 and 20.68 days, p=0.0043), and VTE ($63,9924.1, p=&amp;lt; 0.001 and 72.9 days, p=0.002). Also, higher mortality was observed in burn patients with ESRD, STEMI, sepsis, VTE, and diabetes mellitus. Burn patients with a history of mental health, drug dependence, heart failure, and diabetes mellitus also had greater 30-day-readmissions rates. Conclusions This study sheds new knowledge on the considerable variability that exists between the different population health groups in terms of outcomes for each cohort of critically ill patients. It demonstrates the impacts of population health group on outcomes. These population groups and social determinants have different effects on BICU versus SICU patients and this study provides supporting evidence for the need to identify and develop new strategies to decrease overspending in healthcare. Further research to develop relevant and timely interventions that can improve these outcomes.

Prediction of Mortality in Surgical Intensive Care Unit Patients Using Machine Learning Algorithms.

Objective: Predicting prognosis of in-hospital patients is critical. However, it is challenging to accurately predict the life and death of certain patients at certain period. To determine whether machine learning algorithms could predict in-hospital death of critically ill patients with considerable accuracy and identify factors contributing to the prediction power.Materials and Methods: Using medical data of 1,384 patients admitted to the Surgical Intensive Care Unit (SICU) of our institution, we investigated whether machine learning algorithms could predict in-hospital death using demographic, laboratory, and other disease-related variables, and compared predictions using three different algorithmic methods. The outcome measurement was the incidence of unexpected postoperative mortality which was defined as mortality without pre-existing not-for-resuscitation order that occurred within 30 days of the surgery or within the same hospital stay as the surgery.Results: Machine learning algorithms trained with 43 variables successfully classified dead and live patients with very high accuracy. Most notably, the decision tree showed the higher classification results (Area Under the Receiver Operating Curve, AUC = 0.96) than the neural network classifier (AUC = 0.80). Further analysis provided the insight that serum albumin concentration, total prenatal nutritional intake, and peak dose of dopamine drug played an important role in predicting the mortality of SICU patients.Conclusion: Our results suggest that machine learning algorithms, especially the decision tree method, can provide information on structured and explainable decision flow and accurately predict hospital mortality in SICU hospitalized patients.

Association between Etomidate Use for Rapid Sequence Intubation and Adrenal Insufficiency in Sepsis.

BackgroundThe risk of adrenal insufficiency (AI) in using single-dose etomidate for intubation among patients with sepsis remains controversial. Our aim was to assess the prevalence of AI and characterize the risk factors in patients who received etomidate for rapid sequence intubation (RSI).MethodsThis is a retrospective study of prospectively-acquired data evaluating surgical intensive care unit (SICU) patients who developed respiratory failure undergoing RSI.ResultsOf the 44 adult SICU patients who developed respiratory failure, 34 patients received etomidate. The average age for the total cohort, for the patients that received etomidate and for those who did not, was 70.91 ± 14.92, 72.82 ± 13.61 years and 64.40 ± 15.93, respectively. Twenty-four patients of the total cohort (54.55%) developed AI; 26 had septic shock (59.09%), and 16 patients had AI and septic shock (36.36%). There was no statistical significance between the incidence of AI in patients who received etomidate (47%) and those who did not (80%). However, in the subset of patients who received etomidate for RSI, there was a non-significant trend toward increased incidence of AI in those who were septic compared to those who were not (p = 0.06).ConclusionA single dose of etomidate used for RSI in SICU patients is not associated with the development of AI or mortality. However, a trend was shown, although not statistically significant, towards the development of AI in septic patients. High-quality and adequately powered randomized control trials (RCTs) are warranted.