A descriptive analysis of obstacles to fulfilling the end of life care goals among cardiac arrest patients.

BACKGROUND AND OBJECTIVES: Performing cardiopulmonary resuscitation (CPR) on cardiac arrest patients with Do Not Resuscitate (DNR) orders does not respect patients' autonomy. We aimed to 1) determine the frequency of patients who wished to be DNR prior to cardiac arrest; 2) determine what proportion received CPR; and 3) explain why DNR patients received CPR.
METHODS: This was a retrospective chart review study of cardiac arrest patients at an emergency department. We reported the frequency and proportion of patients who 1) had valid DNR orders presented at the time of cardiac arrest; 2) had valid DNR orders that were unavailable at the time of arrest 3) had Advanced Directives but no DNR orders; 4) had their DNR orders rescinded; and 5) wished to be DNR but did not have any documents in place.
RESULTS: Of 419 patients, 65 (15.51%) wished to be DNR. Out of these 65 patients, 38 (58.46 %) patients were resuscitated. Among the 38 patients, 23 (60.52%) received CPR accordingly and 15 (39.47%) patients were inappropriately resuscitated.
CONCLUSION: In our sample of 419 patients, 65 (15.51%) did not want CPR in the event of cardiac arrest and 38 (9.06%) received CPR against their wishes. This was due to failure to document DNR orders, lack of recognition of documented valid DNR orders, and use of non-actionable Advanced Directives to relay patients' wishes.

Introduction

There is no international uniform approach to Do-Not-Resuscitate (DNR) as it may vary based on culture, religion, government policies, or the lack of them. Some patients opt to forego resuscitation and allow a natural death. This decision is often influenced by age, comorbidities, and functional status.1., 2. Advanced Directives serve as the primary legal tool for patients to communicate their end-of-life healthcare goals and/or to appoint a proxy decision maker in the event of incapacity.

The ideal method to communicate DNR status is to have a Medical or Physician Order for Life Sustaining Treatment stating DNR, signed by the patient and a physician. Other advanced directives, such as living wills can describe what therapy the patient would or would not want in the event the patient is unable to make an informed decision, but are not medical orders. Patients can also appoint a healthcare proxy to make decisions on their behalf in accordance with their wishes.

Unfortunately, some patients receive Cardiopulmonary Resuscitation (CPR) during a cardiac arrest despite their wishes to allow a natural death. The objectives of this study were to 1) determine the frequency of cardiac arrest patients who wished to be DNR prior to cardiac arrest, 2) determine what proportion of these DNR patients received CPR, and 3) explain why these patients received CPR.

Methods

This study was approved by our health system’s Institutional Review Board with exempt status. It was a retrospective medical record review study of all non-traumatic cardiac arrest patients, age ≥ 18 years at the Emergency Department (ED) of a quaternary care hospital from January 1, 2017 to September 30, 2020. All medical record reviews were conducted by trained research staff and emergency medicine residents and confirmed by the first author (GH).

We identified patients who suffered a cardiac arrest and did not wish to be resuscitated. This was determined based on a medical record review of DNR orders, living wills, and documentation of discussions with healthcare proxies. Among those patients who did not want resuscitation, we identified the ones who received CPR and classified them into two groups: 1) patients who were rightly resuscitated, and 2) patients who were not supposed to receive CPR.

Results

We identified 419 adult cardiac arrest patients during the study period. Overall, median age at time of cardiac arrest was 79 years, 57.04 % were male and 42.72 % were female. 308 (73.50 %) patients had out of hospital cardiac arrest and 110 (27.21 %) patients had the cardiac arrest while they were in the ED (Table 1). Based on documentation from Emergency Medical Services regarding initial cardiac rhythm, 114 (56.7 %) patients were in asystole, 41 (9.79 %) in PEA, 26 (6.21 %) in VF, 6 (1.43 %) in VT, and 14 (3.34 %) had an unknown rhythm or undocumented. The ED first arrest rhythm were as follow: 157 (37.47 %) patients were in asystole, 124 (29.59 %) patients were in PEA, 35 (8.35 %) patients were in VF, 13(3.10 %) were in VT and 47(11.22 %) patients had unknown or undocumented rhythm.

Table1

Sample demographics and cardiac arrest rhythm phenotypes.

Total sample	419
Median age	79
Sex
Male	239 (57.04 %)
Female	179 (42.72 %)
Missing	1 (0.23 %)
Cardiac arrest location
Out of hospital	308 (73.50 %)
In-hospital	110 (26.25 %)
Missing	1 (0.23 %)
EMS First pulse rhythm
Asystole	114 (27.21 %)
Pulseless Electrical Activity	41 (9.79 %)
Ventricular Fibrillation	26 (6.21 %)
Ventricular Tachycardia	6 (1.43 %)
Unknown/Undocumented	14 (3.34 %)
Missing	218 (52.03 %)
ED First pulse rhythm
Asystole	157 (37.47 %)
Pulseless Electrical Activity	124 (29.59 %)
Ventricular Fibrillation	35 (8.35 %)
Ventricular Tachycardia	13 (3.10 %)
Unknown/Undocumented	47 (11.22 %)
Missing	43 (10.26 %)

Out of 419 cardiac arrest patients, 65 (15.51 %) patients wished to be DNR yet, 38 (58.46 %) patients received CPR (Fig. 1). The 38 patients who wished to be DNR and received CPR, were categorized in 2 groups; 1) 23 patients received a justified CPR and 2) 15 patients were inappropriately resuscitated. Table 2 provides a more detailed description of circumstances in which patients received CPR.

Fig. 1

Different groups of DNR patients who received CPR DNR: Do Not Resuscitate; CPR: Cardiopulmonary Resuscitation.

Discussion

In the vast majority of cases, performing CPR on patients with DNR orders is unethical as it does not respect patient autonomy. We quantified the proportion of patients who received CPR against their wishes and identified barriers that prevented them from being honored. We found that 9.06 % (38 out of 419) patients received CPR, despite their wishes to not be resuscitated (Table 2).

Table 2

DNR status Determination, reasons for CPR and examples of clinical notes.

DNR status	CPR decision	Circumstances in which patients received CPR	DNR status determination	Examples of provider documentation
38 patients did not want CPR	15 inappropriately received CPR	1 patient had a signed DNR form but was not presented to EMS until the patient was intubated and receiving CPR.	DNR order visualized in patients chart and was available prior to cardiac arrest event
		14 patients had signed DNR form, but the paperwork was not handed over to EMS at the time of arrest. Resuscitation started but terminated when the DNR orders were presented in the ED.	DNR order added after cardiac arrest event and visualized in patient’s chart, or not in file but its existence confirmed by attending provider	“As per family, patient is DNR but paperwork was not available and EMS began CPR”
	23 patients received CPR accordingly	3 patients had signed DNR orders that were rescinded by a family member.	Rescinded DNR order visualized in patient’s chart or report by provider of the revoked order.	“Patient had a DNR order. Family wanted chest compressions and intubation at the scene”
		6 patients had a living will and healthcare proxy documents presented at time of arrest and resuscitation was appropriately initiated, however, healthcare proxy terminated ACLS efforts in the emergency department.	Documents either visualized in patient’s chart or not found in file but confirmed in the provider’s note.	“Living will was produced and as per document in ED, grand-son is a Healthcare Proxy “
		14 Patients relayed wishes to be DNR to family but did not prepare MOLST. Resuscitation started but family asked for comfort care.	Report in the provider’s note of conversation with family member stating patient’s wishes to be DNR.	“CPR in progress. Patient’s son on the phone stating that patient is DNR/DNI. Decision made to stop vasopressors and extubate the patient.”“Per family, patient did not want to be a vegetable. After lengthy discussion of prognosis, decision made to withdraw care.”

ROSC: Return Of Sustained Circulation; DNR: Do Not Resuscitate; ACLS: Advanced Cardiovascular Life Support; CPR: Cardio Pulmonary Resuscitation.

There were 15 patients who received CPR despite having the appropriate paperwork completed prior to their cardiac arrest. The majority of these events occurred because the patient’s DNR paperwork was not readily available. This is the most concerning failure of the current paradigm in cardiac arrest care since patients or their proxies went through the appropriate channels to document their wishes but in the end, their wishes were not honored. Completing an end-of-life planning document stating DNR is not enough, its accessibility in an emergency is critically important.

Ease of DNR identification needs to be improved. Patients and caregivers must be educated so that first responders can easily identify patients with known DNR status and provide the appropriate care. In addition, nursing facilities need to have easily accessible DNR forms, as well as education and training of employees on where to find DNR forms quickly. One potential modality to improve recognition and communication of DNR status is an electronic database of signed DNR forms, end-of-life conversations, verbal consents, and any substantiating signatures. In New York, the electronic Medical Order for Life-Sustaining Treatment (eMOLST) registry is accessible by any clinical site.6., 7. Some states have adopted and created medical alert bracelets for the documentation of DNR status. ILCOR received a proposal to discuss the possibility of implementing an international DNR symbol to be printed on patients’ wristbands. DNR bracelets or necklaces offer visual cues for healthcare providers to the presence of a valid DNR order in place. However, this creates concern for patient’s privacy and the disclosure of a personal decision, and for some families the concern of a constant reminder of patient’s prognosis or eventual mortality.

Another area of concern regarding patient autonomy is the act of rescinding valid DNR orders. In our study, 3 patients had forms indicating their preference for DNR. However, family members rescinded those orders, although only 2 of them were lawfully rescinded. In the United States, DNR orders should only be rescinded by the patient who completed the form or by the healthcare proxy if the patient lacks capacity and a major change in health condition occurred. While these issues vary by country and state, increasing medical providers’ awareness regarding the law and who may legally rescind a DNR order may spare patients from unwanted interventions.

Six patients had advanced directives such as living wills and healthcare proxies. DNR orders are the only legal medical orders in New York State,3., 10. allowing first responders to withhold resuscitation in the event of cardiac arrest. In the U.S., an advance planning care document or a medical order is agreed upon by both the patient and a physician. Alternatively, a living will is a document relaying the patient’s wishes but not medical recommendations, it is therefore not a medical order. Patients prepare a living will in case they lose their capacity to state their treatment preferences in the future. It can inform a healthcare proxy or can be extrapolated to the patient’s current situation to facilitate decision making. A healthcare proxy’s role is to follow the patient’s expressed wishes if the patient loses the capacity to make medical decisions. If these wishes are not known, the healthcare proxy should make decisions according to the patient’s best interest. A healthcare proxy has the authority to allow CPR or refuse it on behalf of the patient, unless the patient specifically stated otherwise. Educational institutions and teaching hospitals should provide clinicians training in advance care planning to help them navigate through the different documents of advanced directives and patients should be informed about the difference between an actionable medical order and the limits of a living will.

Fourteen patients had their wishes relayed by family members but did not have prior documents. This underscores a lack of communication between patients, families, and physicians about end-of-life wishes. It is important that conversations about end-of-life care happen early enough in the disease process, otherwise it would be a lost opportunity for the patient to exercise their autonomy. Care should be taken to involve patients, family members, healthcare proxies, and surrogates in these conversations, and they should be provided with information about advance planning to understand decisions and their implications. Finally, the importance of end-of-life planning could be addressed more prominently in the media, so people can begin these conversations outside the medical setting.

Limitations

The main limitation of this study was the lack of follow up with family members to confirm patient’s wishes, for patients who did not have any documentation in their medical records. This might not always reflect the patients’ true wishes, as that information might not have been available at the time of cardiac arrest.

Another limitation is that this was a single center study; however, these barriers to following DNR orders might still be the same for other centers.

Conclusions

Some patients receive resuscitation despite not wanting CPR. Reasons for this include not having DNR orders available at the time of cardiac arrest, as well as having improper documentation, or not having their wishes documented at all. Improved education and methods of making DNR wishes known are needed.

Disclosure Statement

Timmy Li receives research support from Nihon Kohden Corporation and United Therapeutics. Casey Owens and Ghania Haddad receive research support from United Therapeutics.

CRediT authorship contribution statement

Ghania Haddad: Methodology, Investigation, Data curation, Writing – original draft, Writing – review & editing. Timmy Li: Methodology, Writing – review & editing, Supervision, Project administration. Danielle Turrin: Writing – review & editing. Casey Owens: Writing – review & editing. Daniel Rolston: Conceptualization, Methodology, Writing – review & editing, Supervision.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.