Futility in surgical critical care
One in five Americans die during or shortly after an intensive care unit (ICU) stay,3 despite a strong desire to avoid aggressive care at the end of life, particularly if cognitive or functional disability is likely to result.4 Death in the ICU frequently follows a transition from disease or injury-directed therapy to end-of-life care, and the process is often prolonged, inefficient, and painful. The USA easily outpaces every other nation in terms of healthcare spending but yields inferior health outcomes,5 and a disproportionate share of US healthcare spending is dedicated to care provided at the end of life.6 Although admission to the ICU should be considered a therapeutic trial that is reconsidered in a timely manner if patient-centered results are not being achieved,7 intensive care interventions often prolong life in a manner that prolongs suffering. These interventions often lead to moral distress and overuse of scarce hospital resources.8 9 As such, the importance of optimizing goals of care discussions for critically ill patients and, in turn, reducing potentially ineffective care was deemed the number one research priority within the field of surgical critical care by the American Association for the Surgery of Trauma Surgical Critical Care Committee in 2020.10
The COVID-19 pandemic has had devastating consequences globally, and the brunt of its impact on the medical system fell on the ICU. The sheer volume of patients affected by the COVID-19 pandemic has drawn national attention to the (mis)usage of critical care resources. For the first time, many physicians have been confronted with the reality of resource limitation and the possibility of ‘rationing care’.11 The need to limit the allocation of intensive care resources such as ventilators and dialysis machines may contribute to a shift in the care offered by critical care providers to one focusing on the usage of limited resources on patients most likely to have a meaningful recovery. One small positive outcome from the pandemic is that it has brought heightened awareness to the need to improve discussion and documentation of healthcare goals, increase goal concordant care, reduce potentially ineffective care, and improve the quality of care at the end of life for all patients and their loved ones.
Advance care planning (ACP) is a key component of shared medical decision-making, an iterative process which allows patients to explore their medical care goals, identify a surrogate decision maker, complete advance directives (ADs), and translate their values into medical care plans. Thorough, early, and frequent ACP is the best way to avoid an undesirable outcome and preserve patient autonomy by allowing patients to express their goals and preferences for medical interventions. As such, ACP and ADs have been shown to increase quality of life for those at the end of life, and reduce potentially ineffective care and healthcare costs.12 13
Involvement of our palliative care colleagues and/or integration of palliative care concepts into our practice may help improve provider-patient communication, patient outcomes, and reduce costs. The American College of Surgeons Trauma Quality Improvement Program Palliative Care Best Practice Guidelines recommend that clinicians identify pre-existing ADs or ACP early and hold a structured family meeting addressing goals of care within 72 hours of admission.14 Integration of early palliative care interventions parallel to trauma care has been shown to improve patient satisfaction and increase the frequency of ACP from 4% to 36% patient days and decrease median ICU length of stay for patients that died from 3 days to 1 day, without a change in mortality rate.15 16 In addition, high-quality palliative care has been shown to increase the usage of hospice and reduce the utilization of inpatient units for patients at the end of life.17
Despite its many benefits, barriers to ACP remain, and unfortunately, documented ACP information is rarely in place prior to an ICU admission.18 Sometimes ADs are structured and suited more to a terminal disease like cancer instead of an acute process, making them less helpful for caregivers and families and not directly applicable to an acute trauma or emergency general surgery patient’s situation.19 Surgeon training in ACP is inadequate and many avoid discussing death and dying due to a perceived inability to adequately address the concerns of a frightened patient.20 Patients, on the other hand, demonstrate willingness to engage with providers about goals of care issues to maintain control over their healthcare. It is important to note that ACP is a complex combination of health values, care preferences, and prognostic awareness that evolves based on the patient’s current healthcare status—ACP is not a box one can check and be done with and must be rediscussed at frequent intervals.
In the context of medical decision-making in the ICU, prognostication is both essential and difficult, as the consequences are life-altering for both the patient and the family. Healthcare decision-making processes require providers to prognosticate about the likelihood of a patient’s meaningful survival so that surrogate decision makers can make an educated decision that is aligned with the patient’s goals and values. Resuscitative efforts may not always lead to death or survival, but to an in-between zone where the patient may be alive but in a state that is not aligned with their goals and preferences. An individual’s prognosis is a multifactorial estimation that depends on chronic comorbidities and type and severity of illness, among other things. ICU-specific prognostic tools such as the Acute Physiology and Chronic Health Evaluation, Simplified Acute Physiology Score and Mortality Probability Model scores exist that can help providers prognosticate in-hospital mortality and length of stay, but there is almost always some degree of uncertainty and the scores can be resource intensive to calculate.21 Evidence has shown that both scoring systems and provider ‘expert’ prognostication can be overly optimistic. When asked to decide whether a critical care patient was likely to survive 2 years, critical care physicians who answered ‘yes’ were only right 57.4% of the time.22 Better objective scoring systems and provider training are needed to improve on the ability to accurately prognosticate function and survival for critically ill patients.
Surgical critical care providers could reduce potentially ineffective care in the ICU in many ways. We have identified several potential barriers to quality end-of-life care including low rates of ACP and the paucity of research on prognostication tools that could help quantify futile care in the ICU. We must respond with research and clinical program development that helps combat these barriers. The focus should be on three main goals: assessing patients’ goals and preferences for their care early and often, improving communication about illness trajectory, and maximizing goal-concordant care.