Withholding and withdrawing life-sustaining treatment: Experiences in limiting futile therapy from three Polish intensive care departments

Limitation of Non-Beneficial Interventions and their Impact on the Intensive Care Unit Costs

Introduction

Using a plan to limit non-beneficial life support interventions has significantly reduced harm and loss of dignity for patients at the end of life. The association of these limitations with patients' clinical characteristics and health care costs in the intensive care unit (ICU) needs further scientific evidence.

Aim of the study

To explore decisions to limit non-beneficial life support interventions, their correlation with patients' clinical data, and their effect on the cost of care in the ICU.

Material and Methods

We included all patients admitted to the general ICU of a hospital in Greece in a two-year (2019-2021) prospective study. Data collection included patient demographic and clinical variables, data related to decisions to limit (withholding, withdrawing) non-beneficial interventions (NBIs), and economic data. Comparisons were made between patients with and without limitation decisions.

Results

NBIs were limited in 164 of 454 patients (36.12%). Patients with limitation decisions were associated with older age (70y vs. 62y; p<0,001), greater disease severity score (APACHE IV, 71 vs. 50; p<0,001), longer length of stay (7d vs. 4.5d; p<0,001), and worse prognosis of death (APACHE IV PDR, 48.9 vs. 17.35; p<0,001). All cost categories and total cost per patient were also higher than the patient without limitation of NBIs (9247,79€ vs. 8029,46€, p<0,004). The mean daily cost has not differed between the groups (831,24€ vs. 832,59€; p<0,716). However, in the group of patients with limitations, all cost categories, including the average daily cost (767.31€ vs. 649.12€) after the limitation of NBIs, were reduced to a statistically significant degree (p<0.001).

Conclusions

Limiting NBIs in the ICU reduces healthcare costs and may lead to better management of ICU resource use.

Neurology ethics at the end of life

Ethical challenges in medical decision making are commonly encountered by clinicians caring for patients afflicted by neurological injury or disease at the end of life (EOL). In many of these cases, there are conflicting opinions as to what is right and wrong originating from multiple sources. There is a particularly high prevalence of impaired patient judgment and decision-making capacity in this population that may result in a misrepresentation of their premorbid values and goals. Conflict may originate from a discordance between what is legal or from stakeholders who view and value life and existence differently from the patient, at times due to religious or cultural influences. Promotion of life, rather than preservation of existence, is the goal of many patients and the foundation on which palliative care is built. Those who provide EOL care, while being respectful of potential cultural, religious, and legal stakeholder perspectives, must at the same time recognize that these perspectives may conflict with the optimal ethical course to follow. In this chapter, we will attempt to review some of the more notable ethical challenges that may arise in the neurologically afflicted at the EOL. We will identify what we believe to be the most compelling ethical arguments both in support of and opposition to specific EOL issues. At the same time, we will consider how ethical analysis may be influenced by these legal, cultural, and religious considerations that commonly arise.

[Ethics of resuscitation and end of life decisions]

These European Resuscitation Council Ethics guidelines provide evidence-based recommendations for the ethical, routine practice of resuscitation and end-of-life care of adults and children. The guideline primarily focus on major ethical practice interventions (i.e. advance directives, advance care planning, and shared decision making), decision making regarding resuscitation, education, and research. These areas are tightly related to the application of the principles of bioethics in the practice of resuscitation and end-of-life care.

European Resuscitation Council Guidelines 2021: Ethics of resuscitation and end of life decisions

These European Resuscitation Council Ethics guidelines provide evidence-based recommendations for the ethical, routine practice of resuscitation and end-of-life care of adults and children. The guideline primarily focus on major ethical practice interventions (i.e. advance directives, advance care planning, and shared decision making), decision making regarding resuscitation, education, and research. These areas are tightly related to the application of the principles of bioethics in the practice of resuscitation and end-of-life care.

Machine learning predicts mortality in septic patients using only routinely available ABG variables: a multi-centre evaluation

PURPOSE

To evaluate the application of machine learning methods, specifically Deep Neural Networks (DNN) models for intensive care (ICU) mortality prediction. The aim was to predict mortality within 96 hours after admission to mirror the clinical situation of patient evaluation after an ICU trial, which consists of 24-48 hours of ICU treatment and then "re-triage". The input variables were deliberately restricted to ABG values to maximise real-world practicability.

METHODS

We retrospectively evaluated septic patients in the multi-centre eICU dataset as well as single centre MIMIC-III dataset. Included were all patients alive after 48 hours with available data on ABG (n = 3979 and n = 9655 ICU stays for the multi-centre and single centre respectively). The primary endpoint was 96 -h-mortality.

RESULTS

The model was developed using long short-term memory (LSTM), a type of DNN designed to learn temporal dependencies between variables. Input variables were all ABG values within the first 48 hours. The SOFA score (AUC of 0.72) was moderately predictive. Logistic regression showed good performance (AUC of 0.82). The best performance was achieved by the LSTM-based model with AUC of 0.88 in the multi-centre study and AUC of 0.85 in the single centre study.

CONCLUSIONS

An LSTM-based model could help physicians with the "re-triage" and the decision to restrict treatment in patients with a poor prognosis.