DNR Status Research Articles

Establishment and validation of a risk prediction model for delayed neurocognitive recovery associated with cerebral oxygen saturation monitoring

BackgroundDelayed neurocognitive recovery (DNR) is a common complication in patients undergoing laparoscopic surgery, and there are currently no effective therapies. It is vital to provide a reliable basis for clinical prediction. This study tried to analyse the risk factors for DNR in patients undergoing laparoscopic colorectal surgery and to establish a risk prediction model. MethodsA retrospective analysis of the clinical data and DNR status of patients undergoing laparoscopic colorectal surgery at Xiangya Hospital of Central South University from March 2018 to July 2020 was conducted. Logistic regression was performed to analyse the related risk factors for DNR post-operatively, and the predictive model of DNR post-operatively was constructed and validated internally. Patients who underwent laparoscopic colorectal surgery between January and July 2021 were also selected for external validation of the predictive model, to ultimately investigate the risk factors for DNR in patients undergoing laparoscopic colorectal surgery.ResultsThe incidence of DNR in patients undergoing laparoscopic colorectal surgery was 15.2% (31/204). The maximum variability of cerebral oxygen, age, education, and pre-existing diabetes was related to the incidence of DNR (p < 0.05). The risk prediction model of DNR after laparoscopic colorectal surgery was established. The internal and external validation showed that the discrimination was good (the AUCs were 0.751 and 0.694, respectively).ConclusionsThe risk prediction model of DNR related to cerebral oxygen saturation monitoring shows good predictive performance and clinical value, providing a basis for postoperative DNR prevention.

Disparities in end-of-life care for minoritized racial and ethnic patients during terminal hospitalizations in New York State.

Racial and ethnic minorities often receive care at different hospitals than non-Hispanic white patients, but how hospital characteristics influence the occurrence of disparities at the end of life is unknown. The aim of this study was to determine if disparities in end-of-life care were present among minoritized patients during terminal hospitalizations, and if these disparities varied with hospital characteristics. We identified hospitalizations where a patient died in New York State, 2016-2018. Using multilevel logistic regression, we examined whether documented end-of-life care (do-not-resuscitate status (DNR), palliative care (PC) encounter) differed by race and ethnicity, and whether these disparities differed based on receiving care in hospitals with varying characteristics (Black or Hispanic-serving hospital; teaching status; bed size; and availability of specialty palliative care). We identified 143,713 terminal hospitalizations in 188 hospitals. Across all hospitals, only Black patients were less likely to have a PC encounter (adjusted odds ratio (aOR) 0.83 [0.80-0.87]) or DNR status (aOR 0.91 [0.87-0.95]) when compared with non-Hispanic White patients, while Hispanic patients were more likely to have DNR status (aOR 1.07 [1.01-1.13]). In non-teaching hospitals, all minoritized groups had decreased odds of PC (aOR 0.80 [0.76-0.85] for Black, aOR 0.91 [0.85-0.98] for Hispanic, aOR 0.93 [0.88-0.98] for Others), while in teaching hospitals, only Black patients had a decreased likelihood of a PC encounter (aOR 0.88 [0.82-0.93]). Also, Black patients in a Black-serving hospitals were less likely to have DNR status (aOR 0.80 [0.73-0.87]). Disparities did not differ based on whether specialty PC was available (p = 0.27 for PC encounter, p = 0.59 for DNR status). During terminal hospitalizations, Black patients were less likely than non-Hispanic White patients to have documented end-of-life care. This disparity appears to be more pronounced in non-teaching hospitals than in teaching hospitals.

Retrospective review of the code status of individuals with Down syndrome during the COVID-19 era.

Code status is a label in the medical record indicating a patient's wishes for end-of-life (EOL) care in the event of a cardiopulmonary arrest. People with intellectual disabilities had a higher risk of both diagnosis and mortality from coronavirus infections (COVID-19)than the general population. Clinicians and disability advocates raised concerns that bias, diagnostic overshadowing, and ableism could impact the allocation of code status and treatment options, for patients with intellectual disabilities, including Down syndrome (DS). To study this, retrospective claims data from the Vizient® Clinical Data Base (used with permission of Vizient, all rights reserved.) of inpatient encounters with pneumonia (PNA) and/or COVID-19 at 825 hospitals from January 2019 to June 2022 were included. Claims data was analyzed for risk of mortality and risk of "Do Not Resuscitate" (DNR) status upon admission, considering patient age, admission source, Elixhauser comorbidities (excluding behavioral health), and DS. Logistic regression models with backward selection were created. In total, 1,739,549 inpatient encounters with diagnoses of COVID-19, PNA, or both were included. After controlling for other risk factors, a person with a diagnosis of DS and a diagnosis of COVID-19 PNA had 6.321 odds ratio of having a DNR status ordered at admission to the hospital compared with those with COVID-19 PNA without DS. The diagnosis of DS had the strongest association with DNR status after controlling for other risk factors. Open and honest discussions among healthcare professionals to foster equitable approaches to EOL care and code status are needed.

Resuscitation of people with a do-not-resuscitate order: When does it happen and what are the outcomes?

BackgroundDo-not-resuscitate (DNR) orders allow competent individuals to state their wishes to not receive resuscitation. Despite the existence of a DNR order, resuscitative efforts may still occur. MethodsThe statewide out-of-hospital cardiac arrest (OHCA) registry in Victoria, Australia (population 6.5 million) identified rates of DNR orders and resuscitative efforts in ≥60-year-olds from 2006-2021. ResultsOf 47,322 patients ≥60 years old with OHCA, 2,688 (5.7%) had a DNR order. DNR order was communicated to emergency call-takers in only 332 (12.4%) cases. Bystander resuscitation was commenced in 1,074 (40.0%) patients, with EMS terminating efforts in 346 (32.2%) patients on arrival but commencing resuscitation in 189 others. Strongest predictors of bystander resuscitation were shockable rhythm (OR 7.75, 95% CI 5.04–11.9) and OHCA in public location (OR 4.8, 95% CI 2.0–11.3). Predictors of EMS terminating resuscitation efforts were OHCA in residential aged care facility (OR 1.93, 95% CI 1.47–2.5) and older age (OR 1.02, 95% CI 1.00–1.04). Transfer to hospital was undertaken in 107 patients, with strongest predictors being return of spontaneous circulation (OR 91.1, 95% CI 51.7–160.7) and witnessed OHCA (OR 7.5, 95% CI 3.9–14.5). Only 7 (0.3%) transported patients survived to discharge, with 5 deceased at 12-month follow-up. ConclusionApproximately 5% of EMS-attended OHCAs in ≥60-year-olds involve a patient with a DNR order, with low rates of awareness of DNR orders. Resuscitation is commenced in 47% of patients, and 4% are transported to hospital with very poor outcomes. Strategies to communicate DNR status and respect patients’ autonomy are extremely important.

Black Americans have worse stroke outcome compared with non-Hispanic whites

IntroductionWe studied racial differences in post-stroke outcomes using a prospective, population-based cohort of stroke survivors as part of the Brain Attack Surveillance in Corpus Christi (BASIC) project. MethodsNeurologic (NIHSS, range of 0–42, higher scores are worse), functional (ADLs/IADLs, range 1–4, higher scores are worse), and cognitive (3MSE, range 0–100, higher scores are better) outcomes were measured 90 days after stroke. Cox proportional hazards and negative binomial linear regression models were used to examine the associations between race and 90-day all-cause mortality and NIHSS, respectively, whereas linear regression was used for ADLs/IADLs and 3MSE scores. Covariates included demographics, initial NIHSS, comorbidities, prior stroke history, tPA treatment status, pre-stroke disability, and pre-stroke cognition. The mortality model was also adjusted for DNR status. ResultsAt 90 days post-stroke, Black American individuals (BAs) (n = 122) had a median (IQR) NIHSS of 2 (1,6) compared to NIHSS of 1 (0,3) in non-Hispanic White American individuals (NHWs) (n = 795). BAs had a median (IQR) ADL/IADL score of 2.41 (1.50, 3.39) compared to 2.00 (1.27, 2.95) in NHWs. BAs scored a median of 84 (75, 92) on the 3MSE compared to NHWs’ score of 91.5 (83, 96). Death occurred in 23 (8%) of BAs and 268 (15%) of NHWs within 90 days among those who participated in baseline. After adjustment for covariates, functional outcomes at 90 days were worse in BAs compared to NHWs, with 15.8% (95% CI=5.2, 26.4) greater limitations in ADLs/IADLs and 43.9% (95% CI=12.0, 84.9) greater severity of stroke symptoms. Cognition at 90 days was 6.5% (95% CI=2.4, 10.6) lower in BAs compared to NHWs. BAs had a 35.4% lower (95% CI=-9.8, 61.9) hazard rate of mortality than NHWs. ConclusionsIn this prospective, population-based community sample, BAs had worse neurologic, functional and cognitive outcomes at 90 days compared to NHWs. Future research should investigate how social determinants of health including structural racism, neighborhood factors and access to preventive and recovery services influences these racial disparities.

Effect of a “Do Not Resuscitate” (DNR) Status on Patient Care: A Descriptive Survey on the Perceptions of ICU and Medical/Surgical Nurses

Background: Nursing research literature demonstrates a persistent concern regarding how nurses interpret “Do Not Resuscitate” (DNR) code status. Aim: The aim of this study was to assess nurses’ perceptions of the effect of a DNR code status on patient care. Methods: This study was a cross-sectional, descriptive survey offered to all clinical nurses in the Medical, Surgical, Telemetry, Critical Care, and Emergency Departments within one urban community hospital in Southern California, USA. Descriptive statistics were used for demographic data and individual question analysis, which included frequency and percentages for categorical variables in addition to means with standard deviation for continuous data. Open- ended questions received content analysis. Results: This study resulted in a convenience sample of 120 registered nurses (RNs). Greater than 95% of nurses who completed the survey agreed that nursing care should continue until patient death. Yet when asked about physiologic, demographic, chronic, or cognitive status, participants answered with varying degrees of certainty about the care that should be provided. Conclusions: Further research is needed to explore the inconsistency in nurses’ understanding of the care provided to a patient with DNR status. There may be opportunities to improve consistency in practice with education or protocols to outline optimal care of a patient with DNR orders.

Abstract WP90: Black Americans Have Worse Functional And Cognitive Outcomes Than White Americans After Stroke

Introduction: We studied racial differences in post-stroke outcomes using a prospective, population-based cohort of stroke survivors as part of the Brain Attack Surveillance in Corpus Christi (BASIC) project. Methods: Neurologic (NIHSS, range of 0-42, higher scores are worse), functional (ADLs/IADLs, range 1-4, higher scores are worse), and cognitive (3MSE, range 0-100, higher scores are better) outcomes were measured 90 days after stroke. Cox proportional hazards and negative binomial linear regression models were used to examine the associations between race and 90-day all-cause mortality and NIHSS, respectively, whereas linear regression was used for ADLs/IADLs and 3MSE scores. Covariates included demographics, initial NIHSS, comorbidities, prior stroke history, tPA, pre-stroke disability, and pre-stroke cognition. The mortality model also adjusted for DNR status. Results: At 90 days post-stroke, Black Americans (BAs) (n=130) had a median (IQR) NIHSS of 2 (1,6) compared to NIHSS of 1 (0,3) in non-Hispanic white Americans (NHWs) (n=868). BAs had a median (IQR) ADL/IADL score of 2.4 (1.5, 3.4) compared to 2 (1.3, 3.0) in NHWs. BAs scored a median of 85.0 (75.0, 92.5) on the 3MSE compared to NHWs’ score of 92.0 (83.0, 96.0). Death occurred in 23 (11%) of BAs and 268 (19%) of NHWs within 90 days. After adjustment for covariates, functional outcomes at 90 days were worse in BAs compared to NHWs, with 14% (95% CI=4.9, 23) greater limitations in ADLs/IADLs and 47.9% (95% CI=16.3, 88.1) greater severity of stroke symptoms. Cognition at 90 days was 7% (95% CI=2.7, 11.3) lower in BAs compared to NHWs. BAs had a 25.2% lower (95% CI=-19.9, 53.3) hazard rate of mortality than NHWs (Figure). Conclusions: In this prospective, population-based community sample, BAs had worse functional and cognitive outcomes at 90 days compared to NHWs. Future research should investigate the cause of these racial disparities.

278: PRACTICE OF DO-NOT-RESUSCITATE ORDERS IN A TERTIARY CARE HOSPITAL IN SAUDI ARABIA

Introduction: The objective of the study is to describe DNR practices in Saudi Arabia. Methods: This is a retrospective cohort study based on a prospectively collected intensive care unit (ICU) database between 2002-2017. We compared patients who had DNR order during the ICU stay with patients with “full code” admitted to a tertiary medical-surgical ICU in Riyadh, Saudi Arabia. Results: Among 24790 patients admitted to the ICU over the 16-year period, 3217 (13%) patients had DNR orders during the ICU stay. Patients with DNR orders, compared to patients with “full code”, were older (median 67 years (Q1,Q3: 55,76) vs. 57 years (Q1,Q3: 33,71), P < 0.0001), were more likely to be females 1374 (43%) vs. 8200 (38%), P < 0.0001) and had higher APACHE II score (median 28 (Q1,Q3: 23, 34) vs. 19 (Q1,Q3: 13, 25), P < 0.0001). Patients with DNR orders, compared to patients with “full code”, were more likely to be mechanically ventilated (2660 (83%) vs. 11871 (55%), P < 0.0001), to have comorbid conditions and had a lower functional status (modified Rankin scores 4-5: 608 (18.9%) vs. 1897 (8.8%), P < 0.0001). Patients with DNR orders, compared to patients with “full code”, were more likely to die in the ICU (2168/3217 (67.8%) vs. 1807/21573 (8.5%) P < 0.0001); and in the hospital (2650/3217 (82.4%) vs. 3908/21573 (18.1%) P < 0.0001). On multivariate logistic regression analysis, the following were associated with an increased likelihood of DNR status: increasing age (OR 1.01, 95% CI 1.009, 1.014, P < 0.0001), higher APACHE II score (OR 1.09, 95% CI 1.008, 1.095, P < 0.0001), and increasing modified Rankin score. Patients admitted in the recent years (2012-2017 vs. 2002-2005) were less likely to have DNR orders (OR 0.36, 95% CI 0.33, 0.40, P < 0.0001). Conclusions: In a tertiary care ICU in Saudi Arabia, about 13% of patients had DNR orders. The study identified several predictors that were associated with the likelihood of DNR orders, including severity of illness and poor baseline functional status.

Abstract 14519: End Stage Heart Failure Patients With Do-Not-Resuscitate Code Status Have Dramatically Reduced Hospitalization Cost and Length of Stay

Introduction: End-stage CHF (ES-CHF) poses a significant burden on healthcare system, DNR orders are often associated with comfort care measures, decreased pain and suffering, as well as healthcare resource utilization (cost and LOS). Objective: To quantify the discrepancy in cost and LOS between ES-CHF patients with DNR orders versus without DNR code status. Methodology: Nationwide Inpatient Sample (NIS) database years 2017-2018 was invoked for hospitalizations with primary or secondary diagnosis of ES-CHF, DNR status, comorbidities, and hospitalization outcome using ICD-10 codes. Characteristics of ES-CHF hospitalization with DNR versus no DNR code status were compared. Because of the nature of NIS database, timing of diagnoses and DNR code status could not be established, thus regression analysis for predictors of hospital charges, cost, and LOS was not feasible. Regression analyses were performed to explore predictors of mortality. Stata/MP 16.0 software was used. Results: 40,655 adult hospitalization had the diagnosis of ES-CHF in years 2017-2018 in the United States, among them, 12,300 (30%) had DNR code status, 35% were females, 83% of patients had &gt;=3 comorbid conditions (Charlson comorbidity index&gt;=3). Overall inpatient mortality rate was 12% (25% vs 7% in DNR vs no DNR group, respectively, p&lt;0.001), mean LOS was 10+-15 days (8 vs 11 days in DNR vs no DNR group, respectively, p&lt;0.001), mean total hospital charges was $159,122+-358,186 ($79,434 vs $193,873 in DNR vs no DNR group, respectively, p&lt;0.001). Significantly higher rates of CABG history, atrial fibrillation, COPD, peripheral vascular disease, hypothyroidism, CKD, liver disease, electrolyte disturbances, AKI, and sepsis were found among DNR group. In all patients with ES-CHF, independent predictors of inpatient mortality included older age, DNR status, history of CABG, peripheral vascular disease, liver disease, electrolyte disturbances, smoking, AKI, and sepsis. Conclusion: The presence of DNR code status in ES-CHF patients admitted to the hospital for any was associated with had dramatically reduced hospital charges and cost by more than the half, and with significantly shorter LOS.

Abstract 324: Out-of-hospital Cardiac Arrest Intervention and Outcome Variation in Detroit

Introduction: Significant variation exists in treatment and outcomes from out-of-hospital cardiac arrest (OHCA). Disparities between hospitals serving a similar population is less known. Hypothesis: OHCA post-resuscitation practices and outcomes differ significantly between 4 hospitals in Detroit, which serve a similar population. Methods: The Detroit-Cardiac Registry to Enhance Survival were analyzed between January 1st, 2014 to December 31st, 2019. Patients &lt;18 years age, had a pre-existing DNR, and those transferred to another facility were excluded. 4 tertiary care hospitals (median distance: 10.6 miles) were compared on utilization of 2 treatment decisions (coronary angiography, new DNR order) and 2 clinical outcomes (cerebral performance category {CPC} ≤ 2, in-hospital death) using logistic regression models adjusted for known confounders. Models for death and CPC outcome were tested with vs. without the addition of angiography and DNR initiation as additional confounders (sensitivity analysis). Results: 999-included patients at hospitals A - D differed (p&lt;0.05) before multivariable adjustment by age, race, witnessed arrest, dispatch-ED time, TTM, coronary angiography, DNR order, and in-hospital death, but not CPC ≤ 2. Rates of death and CPC ≤ 2 were worse in Hospital A (82.8%, 10%, respectively) compared to all others (69.1%, 14.1%). After multivariable adjustment, Hospital A performed angiography less compared to B (OR=0.17) and was more likely to initiate new DNR status than B (OR=2.9), C (OR=16.1), or D (OR=3.6). Multivariable-adjusted odds of CPC ≤ 2 were worse in Hospital A compared to B (OR=0.27) and D (OR=0.35). After sensitivity analysis for the different treatment propensities, CPC ≤ 2 odds did not differ for A versus B (OR=0.58, adjusted for angiography) or D (OR=0.65, adjusted for DNR order). Multivariable-adjusted odds of death, regardless of angiography and DNR differences, were worse in Hospital A compared to B (OR=1.87) and D (OR=1.81). Conclusion: OHCA treatment and outcome disparities were noted between nearby Detroit hospitals serving related patient populations. Rates of coronary angiography and DNR orders partially explain disparities in CPC &lt;2 but not death.

Outcomes of COVID-19 Infection in Patients with Left Ventricular Assist Device

PurposeLeft ventricular assist device (LVAD) patients are presumed to be high risk for novel SARS-CoV2 virus complications given their multiple comorbidities. Data on COVID-19 outcomes in LVAD patients is limited. Our single-center, retrospective, cohort study aimed to study the characteristics and outcomes of LVAD patients with COVID-19 infection.MethodsInstitutional IRB approval was obtained. A total of 70 Heartmate 3 (HM3) LVAD patients at our center were screened from December 2019 - September 2021. Variables analyzed included mortality, need for hospitalization, respiratory failure, thromboembolic events, bleeding events, and International Normalized Ratio (INR).ResultsThe rate of COVID-19 infection in our cohort of 70 patients was 15.7% (n=11) similar to the rate observed in the state of Texas (14.1%). The average time interval between LVAD implantation and COVID-19 infection was 41.1 months. 63% (n=7) patients when to ER/hospitalized with a mortality rate of 9% (n=1). All patients had community-acquired infections except one patient who acquired from a visiting family member at the hospital. 81% (n=9) patients acquired the infection during a period of high-infectivity rate in the state of Texas (November 2020-February 2021). Cough was the most common symptom (63.3%, n=7). No patient required intubation. One patient with DNR status who was also unvaccinated died of acute respiratory failure. Three patients developed chronic hypoxic respiratory failure. There were no incidents of pump thrombosis, strokes, or pulmonary embolisms. 72% of patients had an INR >2.0 closest to testing COVID+ with an average INR of 3.5 (range: 1.3 - 9.7). As of September 2021, 63% are vaccinated.ConclusionCOVID-19 infection in LVAD patients can result in substantial morbidity requiring hospitalization and/or death. Interestingly, in our series, no patients suffered a thromboembolic event. Thus, the superior hemocompatibility of the Heartmate 3 device seems resistant to the reputed hypercoagulability of COVID infection.

Code status orders in patients admitted to the intensive care unit with COVID-19: A retrospective cohort study

PurposeCode status orders impact clinical outcomes as well as patients’ and surrogates’ experiences. This is the first multicenter cohort examining code status orders of ICU patients with COVID-19 reported to date. Materials and methodsThis is a retrospective cohort study including adult patients who tested positive for SARS-CoV-2 and were admitted to the ICU at three hospitals in Massachusetts from March 11, 2020 - May 31, 2020. We examined differences in code status orders at multiple timepoints and performed multivariable regression analysis to identify variables associated with code status at admission. ResultsAmong 459 ICU patients with COVID-19, 421 (91.7%) were Full Code at hospital admission. Age and admission from a facility were positively associated with DNR status (adjusted OR 1.10, 95% CI 1.05–1.15, p < 0.001 and adjusted OR 2.68, CI 1.23–5.71, p = 0.011, respectively) while non-English preferred language was negatively associated with DNR status (adjusted OR 0.29, 95% CI 0.10–0.74, p = 0.012). Among 147 patients who died during hospitalization, 95.2% (140) died with DNR code status; most (86.4%) died within two days of final code status change. ConclusionsThe association of non-English preferred language with Full Code status in critically ill COVID-19 patients highlights the importance of medical interpreters in the ICU. Patients who died were transitioned to DNR more than in previous studies, possibly reflecting changes in practice during a novel pandemic.

Abstract 11311: Geographic Variation and Temporal Trends in Management and Outcomes of Cardiac Arrest Complicating Acute Myocardial Infarction in the United States

Introduction: Regional variation in the uptake of newer techniques, procedures, and management strategies have been previously reported with acute cardiac care. Limited contemporary information exists on regional differences in the care of acute myocardial infarction (AMI) patients with cardiac arrest (CA). Methods: Using the National Inpatient Sample (2000-2017), we identified adults with AMI and concomitant CA admitted to the United States census regions of Northeast, Midwest, South, and West. Clinical outcomes of interest including in-hospital mortality, use of coronary angiography, percutaneous coronary intervention (PCI), mechanical circulatory support (MCS), hospitalization costs, and length of stay were compared across these four regions. Results: Of 9,680,257 admissions for AMI, 494,083 (5.1%) had concomitant CA. The prevalence of CA in AMI was highest in the West (6.0%) compared to the Northeast (4.4%), Midwest (5.0%), and South (5.1%), p&lt;0.001. Admissions in the West had higher rates of STEMI presentation, cardiogenic shock, multiorgan failure, mechanical ventilation, and hemodialysis. Northeast AMI-CA admissions had lower use of coronary angiography (52.0% vs. 67.9% vs. 60.9% vs. 61.5%), PCI (38.7% vs. 51.9% vs. 44.8% vs. 46.7%), and MCS (18.4% vs. 21.8% vs. 18.1%, vs. 20.0%) compared to the Midwest, West and South (all p&lt;0.001). Compared with the Northeast, adjusted in-hospital mortality for AMI-CA was higher in the Midwest (odds ratio [OR] 1.06 [95% confidence interval {CI} 1.03-1.08]; p&lt;0.001), South (OR 1.11 [95% CI 1.09-1.13]; p&lt;0.001), and highest in the West (OR 1.16 [95% CI 1.13-1.18]; p&lt;0.001). Temporal trends showed a decline in in-hospital mortality except in the West, which showed a slight increase. Compared to other regions, AMI-CA admissions from the Northeast had longer in-hospital stays and less frequent palliative care consultations (all p&lt;0.001). Admissions in the West had higher rates of palliative care consultations, DNR status, and highest hospitalization costs. Conclusions: We identified significant regional disparities in the management and outcomes of AMI-CA. Efforts to improve and provide uniform care for AMI-CA across regions may be essential.

Abstract 13412: Gender Based Differences in Duration of Cardiopulmonary Resuscitation in the Inpatient Setting

Introduction: The decision to terminate cardiopulmonary resuscitation (CPR) prior to return of spontaneous circulation (ROSC) may be impacted by resuscitation parameters or by patient demographics. Studies show that longer resuscitation efforts have a higher likelihood of ROSC and survival to discharge. It is unclear if gender of the code leader and the patient may be associated with duration of unsuccessful CPR attempt. Methods: Retrospective chart review of inpatient CA that occurred at the UC San Diego Health System between 2011-2019. All adult inpatients with an index cardiac arrest who expired within three days of admission were included. Exclusion criteria included patients who achieved ROSC, CA that occurred outside of the wards or ICU, those with active pre-arrest DNR status, and patients with CA &gt; 3 days after admission (done to minimize provider bias based on prior knowledge of the patient). Data were identified from an internal cardiac arrest quality improvement database. Patient gender was self-reported and code leader gender was identified. Primary outcome was duration of resuscitation efforts. To determine differences in duration of CPR between code leader and patient gender a one-sided ANOVA test was used. A p value &lt; 0.05 was considered statistically significant. Results: We identified 91 patients between 2011 and 2019 that met inclusion criteria. Thirty-eight patients (41.8%) identified as female, fifty-three identified as male (58.2%), the average age was 62.7 years old, and seventeen (18.7%) had initial shockable rhythm. Thirty (33%) of the code leaders identified as female and sixty-one (67%) identified as male. Average duration of CPR was 31.6 minutes (2 - 135 minutes) for male patients with male code leader, 32.2 minutes (9 - 71 minutes) for male patients with female code leader, 27.0 minutes (7 - 60 minutes) for female patients with female code leader, and 33.7 minutes (10 - 73 minutes) for female patients with male code leader. No significant difference in duration of unsuccessful CPR attempt associated with gender of code leader and gender of patient in ANOVA analysis was found (p = 0.512). Conclusions: We did not identify a significant association between code leader and patient gender on duration of CPR in the inpatient setting.

QOLP-35. PEDIATRIC POST-MORTEM TISSUE DONATION DURING A PANDEMIC: A MODEL OF COLLABORATION

BACKGROUNDObtaining post-mortem tissue from pediatric oncology patients is not only critical to research, but studies show that participating in the process can also help grieving families heal. Since 2019, the national Gift from a Child program — a multi-institutional effort to increase the rate of rapid autopsies for pediatric CNS tumor patients — has made significant progress. Collecting high-quality post-mortem tissue has advanced research through cell line generation and genomic analyses. Unfortunately, some autopsy programs temporarily shut down during the COVID-19 pandemic.METHODSWe retrospectively reviewed autopsies of four patients treated at Memorial Sloan Kettering (MSK) who underwent limited brain post-mortem examination at Weill Cornell Medicine College (WCMC) from June 2020 to June 2021. We collected patient demographics; DNR status; time of death and procedure; restrictions due to the COVID-19 pandemic; and results of the tissue analysis. Each case presented unique challenges and the timing of securing parental consent varied.RESULTSThree of four specimens were processed within 12 hours of the time of death. Two Spanish-speaking families required interpreters services to obtain consent. In all cases tumor aliquots were flash frozen for further study. All specimens contained viable tumor and cell line generation was successful in one case. All families/caregivers expressed gratitude for the opportunity to participate and for the handling of the procedures.DISCUSSIONDespite the sensitive nature of these cases, clinicians should offer the option of a rapid autopsy to caregivers of pediatric patients based on the scientific need and the positive effect it has on grieving families. This paper outlines the logistical efforts required for these donations to take place and provides a framework for providers to offer rapid autopsy as an option for families through this program.

Trends and Predictors of Palliative Care Consultation Among Patients Admitted for LVAD: A Retrospective Analysis From the Nationwide Inpatient Sample Database From 2006-2014.

Left ventricular assist devices (LVADs) are an essential part of advanced heart failure (HF) management, either as a bridge to transplantation or destination therapy. Patients with advanced HF have a poor prognosis and may benefit from palliative care consultation (PCC). However, there is scarce data regarding the trends and predictors of PCC among patients undergoing LVAD implantation. This study aims to assess the incidence, trends, and predictors of PCC in LVAD recipients using the United States Nationwide Inpatient Sample (NIS) database from 2006 until 2014. We conducted a weighted analysis on LVAD recipients during their index hospitalization. We compared those who had PCC with those who did not. We examined the trend in palliative care utilization and calculated adjusted odds ratios (aOR) to identify demographic, social, and hospital characteristics associated with PCC using multivariable logistic regression analysis. We identified 20,675 admissions who had LVAD implantation, and of them 4% had PCC. PCC yearly rate increased from 0.6% to 7.2% (P < 0.001). DNR status (aOR 28.30), female sex (aOR 1.41), metastatic cancer (aOR: 3.53), Midwest location (aOR 1.33), and small-sized hospitals (aOR 2.52) were positive predictors for PCC along with in-hospital complications. Differently, Black (aOR 0.43) and Hispanic patients (aOR 0.25) were less likely to receive PCC. There was an increasing trend for in-hospital PCC referral in LVAD admissions while the overall rate remained low. These findings suggest that integrative models to involve PCC early in advanced HF patients are needed to increase its generalized utilization.

Impact of Palliative Care Services on Treatment and Resource Utilization for Hepatorenal Syndrome in the United States.

Background: This study aimed to determine the rates of inpatient palliative care service use and assess the impact of palliative care service use on in-hospital treatments and resource utilization in hospital admissions for hepatorenal syndrome. Methods: Using the National Inpatient Sample, hospital admissions with a primary diagnosis of hepatorenal syndrome were identified from 2003 through 2014. The primary outcome of interest was the temporal trend and predictors of inpatient palliative care service use. Logistic and linear regression was performed to assess the impact of inpatient palliative care service on in-hospital treatments and resource use. Results: Of 5571 hospital admissions for hepatorenal syndrome, palliative care services were used in 748 (13.4%) admissions. There was an increasing trend in the rate of palliative care service use, from 3.3% in 2003 to 21.1% in 2014 (p < 0.001). Older age, more recent year of hospitalization, acute liver failure, alcoholic cirrhosis, and hepatocellular carcinoma were predictive of increased palliative care service use, whereas race other than Caucasian, African American, and Hispanic and chronic kidney disease were predictive of decreased palliative care service use. Although hospital admission with palliative care service use had higher mortality, palliative care service was associated with lower use of invasive mechanical ventilation, blood product transfusion, paracentesis, renal replacement, vasopressor but higher DNR status. Palliative care services reduced mean length of hospital stay and hospitalization cost. Conclusion: Although there was a substantial increase in the use of palliative care service in hospitalizations for hepatorenal syndrome, inpatient palliative care service was still underutilized. The use of palliative care service was associated with reduced resource use.

Abstract MP29: Effects of System-Level Factors on Race/Ethnic Differences in In-Hospital Mortality After Acute Ischemic Stroke

Introduction: Stroke mortality has declined, with differential changes by race; stroke is now the 5 th leading cause of death overall, but 2 nd leading cause of death in blacks. Little is known about recent race/ethnic trends in in-hospital mortality after acute ischemic stroke (AIS) and whether system-level factors contribute to possible differences. Methods: Using the National Inpatient Sample, adults (&gt;18 yrs) with a primary diagnosis of AIS from 2006 to 2017 (n=763,808) were identified. We assessed in-hospital mortality by race/ethnicity (white, black, Hispanic, Asian/Pacific Islander [API], other), sex, and age. Hospitals were categorized by proportion of minority patients served: &lt;25% minority (white hospitals); 25-50% (mixed hospitals), and &gt;50% (minority hospitals). Using survey adjusted logistic regression, the association between race/ethnicity and odds of mortality was assessed, adjusting for key sociodemographic, clinical, and hospital characteristics (e.g. age, comorbidities, stroke severity, DNR status, and palliative care). Results: Overall, mortality decreased from 5.0% in 2006 to 2.9% in 2017, p&lt;0.001. Comparing 2006-2011 to 2012-2017, there was a 66% reduction in mortality after adjustment for covariates, most prominent in whites (68%) and smallest in blacks (58%). Compared to whites, blacks and Hispanics had lower adjusted odds of mortality (AOR 0.82, 95% CI 0.78-0.86 and AOR 0.92, CI 0.86-0.98), primarily driven by those &gt;65 yrs (age x ethnicity interaction p = 0.003). Compared to white men, black, Hispanic, and API men and black women had lower odds of mortality. Adjusted mortality was lower in minorities vs. whites and most pronounced in white hospitals (white: AOR 0.78, 0.73-0.85; mixed: 0.85, 0.80-0.91; minority: 0.89, 0.82-0.95; interaction effect: p=0.018). These differences were present for both minority men and women in white and mixed hospitals, but not women in minority hospitals. Discussion: AIS mortality decreased dramatically in recent years. Overall, black and Hispanic AIS patients have lower mortality than whites, a difference that is most striking in white hospitals. Further study is needed to understand these differences and to what extent biological, sociocultural, and system-level factors play a role.

Using Cross-Checking Safety Tools, Decision Support, and Education Within the Electronic Medical Record to Increase Concordant Advance Care Planning Orders (QI728)

1.Understand the importance and appreciate the achievability of improving code status ordering accuracy.2.Gain initiative, insight and answers to specific questions on how to successfully replicate the project. Code status ordering lapses are a known cause of serious medical errors that are believed to be underreported, especially when unwanted resuscitation is performed. We describe a health system-wide iterative process of improving code status ordering in our EMR upon admission, to safely respect our patients’ wishes. At baseline, EMR code status ordering was mandatory upon admission (required field in the order sets and additional decision support alerts) and defined documentation processes existed including MOLST and eMOLST. The initiative had several phases:1.We synchronized our system EMR code status order options in both inpatient and ambulatory settings with the options on NY MOLST.2.We created enduring code status orders in the EMR that cross between settings and created an option of “affirming” the preexisting code status upon transitions of care rather than issuing a new order.3.We created modal decision support that monitors for unexplained discrepancy between the present code status order, the previous code status, and the discretely documented code status decisions in the chart (both in the provider note and the nurse flowsheet).4.We delivered a comprehensive educational program. Random chart reviews of 2-day span of admissions were conducted before the initiative and were repeated afterwards to measure the accuracy of admission code status orders by emergency and hospitalist providers for patients whose preexisting code status order or MOLST included DNR (27 of 229 patients). Post-intervention more patients had accurate continuation of DNR status (from 12.5% to 72.7%) and clearly preventable admission code status ordering errors were eliminated (from 75% to 0%). Some potentially preventable cases included missing order, external MOLST logged post-admission, or code status was erroneously changed prior to admission. A multifaceted EMR-based intervention may drastically reduce the risk of code status ordering errors. Further study is required in this field.

A rare and fatal case of Terrisporobacter glycolicus bacteremia: Case report and review of literature

Background: Terrisporobacter glycolicus (formerly known as Clostridium glycolicum) is a rare cause of infection and emerging anaerobic pathogen. It is a gram-positive, rod shaped obligate anaerobic organism, first identified in 1963 by Gaston and Stadman. However, the first human infection was reported only in 2007, more than 4 decades after its discovery. There are few published case reports reporting these gram-positive bacilli causing infection to human and mostly as part of a polymicrobial infection. Here, we described a case of septicaemia with mortality. Case description: We described a case of a 78-year-old gentleman with multiple co-morbidities and sacral sore admitted for altered sensorium due to cerebrovascular accident with uremia. During admission he developed urinary tract infection and was treated with intravenous cefuroxime, which he has responded well. However, he deteriorated later with poor oral intake and reduced in Glasgow coma scale of 8/15. Blood culture from anaerobic bottle isolated greyish colonies on anaerobic plate after 48 h of incubation, susceptible to metronidazole. By using Vitex 2.0 system, there was low discrimination between T. glycolicus and Clostridium sporogenes. The isolate was later identified by Matrix-assisted laser desorption/ionization – time of flight (MALDI-TOF) mass spectrometry as Terrisporobacter glycolicus. The result from the reference laboratory showed two anaerobic organisms were isolated (Gram-positive anaerobic bacilli and gram-negative anaerobic bacilli). The gram-positive anaerobic bacilli was identified as Terrisporobacter glycolicus which was susceptible to penicillin [minimum inhibitory concentration (MIC) of 0.380 ug/mL], amoxycillin-clavulanate (MIC of 0.750 mg/mL), imipenem (MIC of 0.500 ug/mL), clindamycin (MIC of 0.064 ug/mL) and metronidazole (MIC of 1.000 ug/mL). The patient however deteriorated further and was decided for DNR status. He succumbed to illness on day 15 of admission. Discussion: MALDI-TOF MS has become the method of choice for identification of anaerobic bacteria, replacing 16S rRNA gene sequencing and gas–liquid chromatography. The advent of MALDI-TOF MS has demonstrated increased specificity, cost-effectiveness, and decreased turnaround time. Conclusion: Infection by this organism could be fatal in elderly patient with co-morbidities. In this group of patients, the presentation is usually atypical. In addition, up-to-date laboratory identification such as MALDI-TOF plays important role in confirming the identity of this organism.