A nationwide overview of 1-year mortality in cardiac arrest patients admitted to intensive care units in the Netherlands between 2010 and 2016

Neurologic Recovery at Discharge and Long-Term Survival After Cardiac Arrest

Importance

Brain injury is the leading cause of death following cardiac arrest and is associated with severe neurologic disabilities among survivors, with profound implications for patients and their families, as well as broader societal impacts. How these disabilities affect long-term survival is largely unknown.

Objective

To investigate whether complete neurologic recovery at hospital discharge after cardiac arrest is associated with better long-term survival compared with moderate or severe neurologic disabilities.

Design, Setting, and Participants

This cohort study used data from 4 mandatory national registers with structured and predefined data collection and nationwide coverage during a 10-year period in Sweden. Participants included adults who survived in-hospital cardiac arrest (IHCA) or out-of-hospital cardiac arrest (OHCA) beyond 30 days and who underwent predefined neurologic assessment conducted by health care professionals at hospital discharge using the Cerebral Performance Category (CPC) scale between January 2010 and December 2019. Patients were divided into 3 categories: complete recovery (CPC 1), moderate disabilities (CPC 2), and severe disabilities (CPC 3-4). Statistical analyses were performed in December 2023.

Exposure

CPC score at hospital discharge.

Main Outcomes and Measures

The primary outcome was long-term survival among patients with CPC 1 compared with those with CPC 2 or CPC 3 or 4.

Results

A total of 9390 cardiac arrest survivors (median [IQR] age, 69 .0 [58.0-77.0] years; 6544 [69.7%] male) were included. The distribution of functional neurologic outcomes at discharge was 7374 patients (78.5%) with CPC 1, 1358 patients (14.5%) with CPC 2, and 658 patients (7.0%) with CPC 3 or 4. Survival proportions at 5 years were 73.8% (95% CI, 72.5%-75.0%) for patients with CPC 1, compared with 64.7% (95% CI, 62.4%-67.0%) for patients with CPC 2 and 54.2% (95% CI, 50.6%-57.8%) for patients with CPC 3 or 4. Compared with patients with CPC 1, there was significantly higher hazard of death for patients with CPC 2 (adjusted hazard ratio [aHR], 1.57 [95% CI, 1.40-1.75]) or CPC 3 or 4 (aHR, 2.46 [95% CI, 2.13-2.85]). Similar associations were seen in the OHCA and IHCA groups.

Conclusions and Relevance

In this cohort study of patients with cardiac arrest who survived beyond 30 days, complete neurologic recovery, defined as CPC 1 at discharge, was associated with better long-term survival compared with neurologic disabilities at the same time point.

All-cause mortality risks among participants in mass-participation sporting events

OBJECTIVES

Exercise transiently increases the risk for sudden death, whereas long-term exercise promotes longevity. This study assessed acute and intermediate-term mortality risks of participants in mass-participation sporting events.

METHODS

Data of participants in Dutch running, cycling and walking events were collected between 1995 and 2017. Survival status was obtained from the Dutch Population Register. A time-stratified, case-crossover design examined if deceased participants more frequently participated in mass-participation sporting events 0-7 days before death compared with the reference period (14-21 days before death). Mortality risks during follow-up were compared between participants and non-participants from the general population using Cox regression.

RESULTS

546 876 participants (median (IQR) age 41 (31-50) years, 56% male, 72% runners) and 211 592 non-participants (41 (31-50) years, 67% male) were included. In total, 4625 participants died of which more participants had partaken in a sporting event 0-7 days before death (n=23) compared with the reference period (n=12), and the mortality risk associated with acute exercise was greater but did not reach statistical significance (OR 1.92; 95% CI 0.95 to 3.85). During 3.3 (1.1-7.4) years of follow-up, participants had a 30% lower risk of death (HR 0.70; 95% CI 0.67 to 0.74) compared with non-participants after adjustment for age and sex. Runners (HR 0.65; 95% CI 0.62 to 0.69) and cyclists (HR 0.70; 95% CI 0.64 to 0.77) had the best survival during follow-up followed by walkers (HR 0.88; 95% CI 0.80 to 0.94).

CONCLUSION

Participating in mass-participation sporting events was associated with a non-significant increased odds (1.92) of mortality and a low absolute event rate (4.2/100 000 participants) within 7 days post-event, whereas a 30% lower risk of death was observed compared with non-participants during 3.3 years of follow-up. These results suggest that the health benefits of mass sporting event participation outweigh potential risks.

National ICU Registries as Enablers of Clinical Research and Quality Improvement

OBJECTIVES

Clinical quality registries (CQRs) have been implemented worldwide by several medical specialties aiming to generate a better characterization of epidemiology, treatments, and outcomes of patients. National ICU registries were created almost 3 decades ago to improve the understanding of case-mix, resource use, and outcomes of critically ill patients. This narrative review describes the challenges, proposed solutions, and evidence generated by National ICU registries as facilitators for research and quality improvement.

DATA SOURCES

English language articles were identified in PubMed using phrases related to ICU registries, CQRs, outcomes, and case-mix.

STUDY SELECTION

Original research, review articles, letters, and commentaries, were considered.

DATA EXTRACTION

Data from relevant literature were identified, reviewed, and integrated into a concise narrative review.

DATA SYNTHESIS

CQRs have been implemented worldwide by several medical specialties aiming to generate a better characterization of epidemiology, treatments, and outcomes of patients. National ICU registries were created almost 3 decades ago to improve the understanding of case-mix, resource use, and outcomes of critically ill patients. The initial experience in European countries and in Oceania ensured that through locally generated data, ICUs could assess their performances by using risk-adjusted measures and compare their results through fair and validated benchmarking metrics with other ICUs contributing to the CQR. The accomplishment of these initiatives, coupled with the increasing adoption of information technology, resulted in a broad geographic expansion of CQRs as well as their use in quality improvement studies, clinical trials as well as international comparisons, and benchmarking for ICUs.

CONCLUSIONS

ICU registries have provided increased knowledge of case-mix and outcomes of ICU patients based on real-world data and contributed to improve care delivery through quality improvement initiatives and trials. Recent increases in adoption of new technologies (i.e., cloud-based structures, artificial intelligence, machine learning) will ensure a broader and better use of data for epidemiology, healthcare policies, quality improvement, and clinical trials.

Plasma neurofilament light is a predictor of neurological outcome 12 h after cardiac arrest

BACKGROUND

Previous studies have reported high prognostic accuracy of circulating neurofilament light (NfL) at 24-72 h after out-of-hospital cardiac arrest (OHCA), but performance at earlier time points and after in-hospital cardiac arrest (IHCA) is less investigated. We aimed to assess plasma NfL during the first 48 h after OHCA and IHCA to predict long-term outcomes.

METHODS

Observational multicentre cohort study in adults admitted to intensive care after cardiac arrest. NfL was retrospectively analysed in plasma collected on admission to intensive care, 12 and 48 h after cardiac arrest. The outcome was assessed at two to six months using the Cerebral Performance Category (CPC) scale, where CPC 1-2 was considered a good outcome and CPC 3-5 a poor outcome. Predictive performance was measured with the area under the receiver operating characteristic curve (AUROC).

RESULTS

Of 428 patients, 328 (77%) suffered OHCA and 100 (23%) IHCA. Poor outcome was found in 68% of OHCA and 55% of IHCA patients. The overall prognostic performance of NfL was excellent at 12 and 48 h after OHCA, with AUROCs of 0.93 and 0.97, respectively. The predictive ability was lower after IHCA than OHCA at 12 and 48 h, with AUROCs of 0.81 and 0.86 (p ≤ 0.03). AUROCs on admission were 0.77 and 0.67 after OHCA and IHCA, respectively. At 12 and 48 h after OHCA, high NfL levels predicted poor outcome at 95% specificity with 70 and 89% sensitivity, while low NfL levels predicted good outcome at 95% sensitivity with 71 and 74% specificity and negative predictive values of 86 and 88%.

CONCLUSIONS

The prognostic accuracy of NfL for predicting good and poor outcomes is excellent as early as 12 h after OHCA. NfL is less reliable for the prediction of outcome after IHCA.

Explainable and efficient deep early warning system for cardiac arrest prediction from electronic health records

Cardiac arrest (CA) is a fatal acute event. The development of new CA early warning system based on time series of vital signs from electronic health records (EHR) has great potential to reduce CA damage. In this process, recursive architecture-based deep learning, as a powerful tool for time series data processing, enables automatically extract features from various monitoring clinical parameters and to further improve the performance for acute critical illness prediction. However, the unexplainable nature and excessive time caused by black box structure with poor parallelism are the limitations of its development, especially in the CA clinical application with strict requirement of emergency treatment and low hidden dangers. In this study, we present an explainable and efficient deep early warning system for CA prediction, which features are captured by an efficient temporal convolutional network (TCN) on EHR clinical parameters sequence and explained by deep Taylor decomposition (DTD) theoretical framework. To demonstrate the feasibility of our method and further evaluate its performance, prediction and explanation experiments were performed. Experimental results show that our method achieves superior CA prediction accuracy compared with standard national early warning score (NEWS), in terms of overall AUROC (0.850 Vs. 0.476) and F1-Score (0.750 Vs. 0.450). Furthermore, our method improves the interpretability and efficiency of deep learning-based CA early warning system. It provides the relevance of prediction results for each clinical parameter and about 1.7 times speed enhancement for system calculation compared with the long short-term memory network.

In-hospital versus out-of-hospital cardiac arrest: characteristics and outcomes in patients admitted to intensive care after return of spontaneous circulation

INTRODUCTION

Cardiac arrest is characterized depending on location as in-hospital cardiac arrest (IHCA) or out-of-hospital cardiac arrest (OHCA). Strategies for Post Cardiac Arrest Care were developed based on evidence from OHCA. The aim of this study was to compare characteristics and outcomes in patients admitted to intensive care after IHCA and OHCA.

METHODS

A retrospective multicenter observational study of adult survivors of cardiac arrest admitted to intensive care in southern Sweden between 2014-2018. Data was collected from registries and medical notes. The primary outcome was neurological outcome according to the Cerebral Performance Category (CPC) scale at 2-6 months.

RESULTS

799 patients were included, 245 IHCA and 554 OHCA. IHCA patients were older, less frequently male and less frequently without comorbidity. In IHCA the first recorded rhythm was more often non-shockable, all delay-times (ROSC, no-flow, low-flow, time to advanced life support) were shorter and a cardiac cause of the arrest was less common. Good long-term neurological outcome was more common after IHCA than OHCA. In multivariable analysis, witnessed arrest, age, shorter arrest duration (no-flow and low-flow times), low lactate, shockable rhythm, and a cardiac cause were all independent predictors of good long-term neurological outcome whereas location of arrest (IHCA vs OHCA) was not.

CONCLUSION

In patients admitted to intensive care after cardiac arrest, patients who suffered IHCA vs OHCA differed in demographics, co-morbidities, cardiac arrest characteristics and outcomes. In multivariable analyses, cardiac arrest characteristics were independent predictors of outcome, whereas location of arrest (IHCA vs OHCA) was not.

Can we reliably automate clinical prognostic modelling? A retrospective cohort study for ICU triage prediction of in-hospital mortality of COVID-19 patients in the Netherlands

BACKGROUND

Building Machine Learning (ML) models in healthcare may suffer from time-consuming and potentially biased pre-selection of predictors by hand that can result in limited or trivial selection of suitable models. We aimed to assess the predictive performance of automating the process of building ML models (AutoML) in-hospital mortality prediction modelling of triage COVID-19 patients at ICU admission versus expert-based predictor pre-selection followed by logistic regression.

METHODS

We conducted an observational study of all COVID-19 patients admitted to Dutch ICUs between February and July 2020. We included 2,690 COVID-19 patients from 70 ICUs participating in the Dutch National Intensive Care Evaluation (NICE) registry. The main outcome measure was in-hospital mortality. We asessed model performance (at admission and after 24h, respectively) of AutoML compared to the more traditional approach of predictor pre-selection and logistic regression.

FINDINGS

Predictive performance of the autoML models with variables available at admission shows fair discrimination (average AUROC = 0·75-0·76 (sdev = 0·03), PPV = 0·70-0·76 (sdev = 0·1) at cut-off = 0·3 (the observed mortality rate), and good calibration. This performance is on par with a logistic regression model with selection of patient variables by three experts (average AUROC = 0·78 (sdev = 0·03) and PPV = 0·79 (sdev = 0·2)). Extending the models with variables that are available at 24h after admission resulted in models with higher predictive performance (average AUROC = 0·77-0·79 (sdev = 0·03) and PPV = 0·79-0·80 (sdev = 0·10-0·17)).

CONCLUSIONS

AutoML delivers prediction models with fair discriminatory performance, and good calibration and accuracy, which is as good as regression models with expert-based predictor pre-selection. In the context of the restricted availability of data in an ICU quality registry, extending the models with variables that are available at 24h after admission showed small (but significantly) performance increase.

Long-term outcomes after out-of-hospital cardiac arrest: a systematic review and meta-analysis

AIMS

Long term outcomes after out-of-hospital cardiac arrest (OHCA) are not well understood. This study aimed to evaluate the long-term (1-year and beyond) survival outcomes, including overall survival and survival with favorable neurological status and the quality-of-life (QOL) outcomes, among patients who survived the initial OHCA event (30 days or till hospital discharge).

METHODS

Embase, Medline and PubMed were searched for primary studies (randomized controlled trials, cohort and cross-sectional studies) which reported the long-term survival outcomes of OHCA patients. Data abstraction and quality assessment was conducted, and survival at predetermined timepoints were assessed via single-arm meta-analyses of proportions, using generalized linear mixed models. Comparative meta-analyses were conducted using the Mantel-Haenszel Risk Ratio (RR) estimates, using the DerSimonian and Laird model.

RESULTS

67 studies were included, and among patients that survived to hospital discharge or 30-days, 77.3% (CI=71.2-82.4), 69.6% (CI=54.5-70.3), 62.7% (CI=54.5-70.3), 46.5% (CI=32.0-61.6), and 20.8% (CI=7.8-44.9) survived to 1-, 3-, 5-, 10- and 15-years respectively. Compared to Asia, the probability of 1-year survival was greater in Europe (RR=2.1, CI=1.8-2.3), North America (RR=2.0, CI=1.7-2.2) and Oceania (RR=1.9,CI=1.6-2.1). Males had a higher 1-year survival (RR:1.41, CI=1.25-1.59), and patients with initial shockable rhythm had improved 1-year (RR=3.07, CI=1.78-5.30) and 3-year survival (RR=1.45, CI=1.19-1.77). OHCA occurring in residential locations had worse 1-year survival (RR=0.42, CI=0.25-0.73).

CONCLUSION

Our study found that up to 20.8% of OHCA patients survived to 15-years, and survival was lower in Asia compared to the other regions. Further analysis on the differences in survival between the regions are needed to direct future long-term treatment of OHCA patients.

One-year Mortality of Cancer Patients with an Unplanned ICU Admission: A Cohort Analysis Between 2008 and 2017 in the Netherlands

Introduction: A decrease in short-term mortality of critically ill cancer patients with an unplanned intensive care unit (ICU) admission has been described. Few studies describe a change over time of 1-year mortality. Therefore, we examined the 1-year mortality of cancer patients (hematological or solid) with an unplanned ICU admission and we described whether the mortality changed over time. Methods: We used the National Intensive Care Evaluation (NICE) registry and extracted all patients with an unplanned ICU admission in the Netherlands between 2008 and 2017. The primary outcome was 1-year mortality, analyzed with a mixed-effects Cox proportional hazard regression. We compared the 1-year mortality of cancer patients to that of patients without cancer. Furthermore, we examined changes in mortality over the study period. Results: We included 470,305 patients: 10,401 with hematological cancer, 35,920 with solid cancer, and 423,984 without cancer. The 1-year mortality rates were 60.1%, 46.2%, and 28.3% respectively (P< .01). Approximately 30% of the cancer patients surviving their hospital admission died within 1 year, this was 12% in patients without cancer. In hematological patients, 1-year mortality decreased between 2008 and 2011, after which it stabilized. In solid cancer patients, inspection showed neither an increasing nor decreasing trend over the inclusion period. For patients without cancer, 1-year mortality decreased between 2008 and 2013, after which it stabilized. A clear decrease in hospital mortality was seen within all three groups. Conclusion: The 1-year mortality of cancer patients with an unplanned ICU admission (hematological and solid) was higher than that of patients without cancer. About one-third of the cancer patients surviving their hospital admission died within 1 year after ICU admission. We found a decrease in 1-year mortality until 2011 in hematology patients and no decrease in solid cancer patients. Our results suggest that for many cancer patients, an unplanned ICU admission is still a way to recover from critical illness, and it does not necessarily lead to success in long-term survival. The underlying type of malignancy is an important factor for long-term outcomes in patients recovering from critical illness.

Documentation of neurological status in patients admitted to an intensive care unit after cardiac arrest: A 10-year cohort study

OBJECTIVE

The objective of this study was to describe the documented neurological assessment and investigations for neuroprognostication in patients after cardiac arrest.

DESIGN, SETTING, AND PARTICIPANTS

This was a retrospective cohort study of adult patients after cardiac arrest, admitted to a tertiary intensive care unit (ICU), between January 2009 and December 2018.

MAIN OUTCOME MEASURES

The main outcome measures were the proportion of patients with a documented Glasgow Coma Scale (GCS) score and investigations for neuroprognostication.

RESULTS

Four hundred twenty-seven patients formed the study cohort. The GCS score was documented for 267 (63%) patients at some time during their ICU stay. The proportion of patients with the GCS score documented decreased each day of ICU stay (59% at day 1, 20% at day 5). Pupil reflex to light was recorded in 352 (82%), corneal reflex in 155 (36%), and limb reflexes in 216 (51%) patients. Twenty-eight (6.6%) patients underwent brain magnetic resonance imaging, 10 (2.3%) an electroencephalogram, and two somatosensory evoked potentials. Withdrawal of life-sustaining treatments occurred in 166 (39%) patients, and 221 (52%) patients died in hospital.

CONCLUSIONS

In this single-centre study of patients admitted to the ICU after cardiac arrest, the GCS score was inconsistently documented, and investigations for neuroprognostication were infrequent.

Outcomes of Targeted Temperature Management for In-Hospital and Out-Of-Hospital Cardiac Arrest: A Matched Case-Control Study Using the National Database of Taiwan Network of Targeted Temperature Management for Cardiac Arrest (TIMECARD) Registry

BACKGROUND This study aimed to compare outcomes of targeted temperature management (TTM) for patients with in-hospital and out-of-hospital cardiac arrest using the national database of TaIwan network of targeted temperature ManagEment for CARDiac arrest (TIMECARD) registry. MATERIAL AND METHODS A retrospective, matched, case-control study was conducted. Patients with in-hospital cardiac arrest (IHCA) treated with TTM after the return of spontaneous circulation (ROSC) were selected as the case group and controls were defined as the same number of patients with out-of-hospital cardiac arrest (OHCA), matched for sex, age, Charlson comorbidity index, and cerebral performance category. Neurological outcome and survival at hospital discharge were the primary outcome measures. RESULTS Data of 103 patients with IHCA and matched controls with OHCA were analyzed. Patients with IHCA were more likely to experience witnessed arrest and bystander cardiopulmonary resuscitation (CPR). The duration from collapse to the beginning of CPR, CPR time, and the duration from ROSC to initiation of TTM were shorter in the IHCA group but their initial arterial blood pressure after ROSC was lower. Overall, 88% of patients survived to completion of TTM and 43% survived to hospital discharge. Hospital survival (42.7% vs 42.7%, P=1.00) and favorable neurological outcome at discharge (19.4% vs 12.7%, P=0.25) did not differ between the 2 groups. CONCLUSIONS The findings from the national TIMECARD registry showed that clinical outcomes following TTM for patients with IHCA were not significantly different from OHCA when baseline factors were matched.

Higher 1-year mortality in women admitted to intensive care units after cardiac arrest: A nationwide overview from the Netherlands between 2010 and 2018

PURPOSE

We study sex differences in 1-year mortality of out-of-hospital cardiac arrest (OHCA) and in-hospital cardiac arrest (IHCA) patients admitted to the intensive care unit (ICU).

DATA

A retrospective cohort analysis of OHCA and IHCA patients registered in the NICE registry in the Netherlands. The primary and secondary outcomes were 1-year and hospital mortality, respectively.

RESULTS

We included 19,440 OHCA patients (5977 women, 30.7%) and 13,461 IHCA patients (4889 women, 36.3%). For OHCA, 1-year mortality was 63.9% in women and 52.6% in men (Hazard Ratio [HR] 1.28, 95% Confidence Interval [95% CI] 1.23-1.34). For IHCA, 1-year mortality was 60.0% in women and 57.0% in men (HR 1.09, 95% CI 1.04-1.14). In OHCA, hospital mortality was 57.4% in women and 46.5% in men (Odds Ratio [OR] 1.42, 95% CI 1.33-1.52). In IHCA, hospital mortality was 52.0% in women and 48.2% in men (OR 1.11, 95% CI 1.03-1.20).

CONCLUSION

Women admitted to the ICU after cardiac arrest have a higher mortality rate than men. After left-truncation, we found that this sex difference persisted for OHCA. For IHCA we found that the effect of sex was mainly present in the initial phase after the cardiac arrest.

Treatment Limitation Decisions in Critically Ill Patients With a Malignancy on the Intensive Care Unit

Background:

Treatment limitation decisions (TLDs) on the ICU can be challenging, especially in patients with a malignancy. Up-to-date literature regarding TLDs in critically ill patients with a malignancy admitted to the ICU is scarce. The aim was to compare the incidence of written TLDs between patients with an active malignancy, patients with a malignancy in their medical history (complete remission, CR) and patients without a malignancy admitted unplanned to the ICU.

Methods:

We conducted a retrospective cohort study in a large university hospital in the Netherlands. We identified all unplanned admissions to the ICU in 2017 and categorized the patients in 3 groups: patients with an active malignancy (study population), with CR and without a malignancy. A TLD was defined as a written instruction not to perform life-saving treatments, such as CPR in case of cardiac arrest. A multivariate binary logistic regression analysis was used to identify whether having a malignancy was associated with TLDs.

Results:

Of the 1046 unplanned admissions, 125 patients (12%) had an active malignancy and 76 (7.3%) patients had CR. The incidence of written TLDs in these subgroups were 37 (29.6%) and 20 (26.3%). Age (OR 1.03; 95% CI 1.01 -1.04), SOFA score at ICU admission (OR 1.11; 95% CI 1.05 -1.18) and having an active malignancy (OR 1.75; 95% CI 1.04-2.96) compared to no malignancy were independently associated with written TLDs. SOFA scores on the day of the TLD were not significantly different in patients with and without a malignancy.

Conclusions:

This study shows that the presence of an underlying malignancy is independently associated with written TLDs during ICU stay. Patients with CR were not at risk of more written TLDs. Whether this higher incidence of TLDs in patients with a malignancy is justified, is at least questionable and should be evaluated in future research.