No Evidence for Blanket DNR Orders for US COVID-19 Patients

By Linda Carroll

May 27, 2020

(Reuters Health) - U.S. hospitals should not be applying blanket 'do not resuscitate' orders for COVID-19 patients on ventilators because there isn't enough data to justify that approach, a new study suggests.

Data from China showing poor cardiac arrest survival among critical COVID patients may not be applicable to patients treated in U.S. hospitals, the researchers assert in Circulation: Cardiovascular Quality and Outcomes. To make their point, they analyzed historic data on U.S. non-COVID patients with pneumonia or sepsis who were on ventilators when they arrested.

"Due to limited resources and risk of transmission of infection to healthcare workers, some hospitals have considered withholding CPR in all COVID-19 patients," said the study's lead author, Dr. Saket Girotra, an assistant professor of medicine at the University of Iowa Carver College of Medicine, in Iowa City.

"The main take-home from our study is that instead of a blanket DNR policy, it is reasonable to consider age, co-morbidities and illness severity to determine whether resuscitation would be appropriate and balance the likelihood of success of resuscitation with the risk to healthcare workers," Dr. Girotra said in an email.

Because there is little data on CPR and COVID-19 patients who've arrested in American hospitals, Dr. Girotra and his colleagues opted to look at a comparable group: 5,690 patients who received CPR between 2014 and 2018 for a cardiac arrest while being treated in an intensive care unit (ICU) for pneumonia or sepsis and who were receiving mechanical ventilation at the time their hearts stopped.

The researchers found an overall survival rate of 12.5%, but depending on certain patient-related factors, that number could be as high as 26%. The probability of survival without severe neurological disability ranged from 3% to more than 22%, while the probability of survival with mild to no disability ranged from 1% to 17%.

Survival rates were lower in older and sicker patients who had asystole or pulseless electrical activity, and much higher in younger patients with a shockable rhythm who were not being treated with vasopressor medications prior to the arrest.

"We created our cohort to resemble severely ill COVID-19 patients as closely as possible," Dr. Girotra said. "It is possible that these survival rates represent a best-case scenario and actual survival in COVID-19 patients who arrest may be lower because of the delay caused by the need to wear personal protective gear."

The new research is "very important," said Dr. Oscar Cingolani, an associate professor of medicine in the division of cardiology and associate director of the cardiac intensive care unit at the Johns Hopkins University Hospital, in Baltimore.

"These are not COVID-19 patients, but they are similar to those with COVID-19," Dr. Cingolani said. "Initially when the pandemic hit the U.S., physicians looked at the mortality in Wuhan and in Italy and talked about not resuscitating patients with COVID-19. But this is a classic example of why you need to be prudent. The policy in one country at one moment could be different in another country at a different moment."

The decision to resuscitate should be individualized, Dr. Cingolani said. And keep in mind that in the U.S., the population is younger due to the higher rates of obesity here, he added. "If we implemented a blanket policy here those younger patients would have been allowed to die," Dr. Cingolani said.

Dr. Kelly Michelson welcomed the new study. "More information is helpful," said Dr. Michelson, director of the Center for Bioethics and Medical Humanities at the Northwestern University Feinberg School of Medicine. "I'm grateful to the authors. This is a great topic. We need to understand outcomes."

"But it's also important to recognize the study's limitations," Dr. Michelson said. "For patients with COVID the outcomes might be significantly different in part because of the need to put on PPE before coming into the room."

The new study underscores the need for more data, said Dr. Robert Arnold, chief of the section of palliative care and medical ethics at the University of Pittsburgh School of Medicine.

"The problem is that early in a new disease, we don't know enough about the prognostic implications of the disease," Dr. Arnold said. "Good ethics depends on good data. We shouldn't make decisions based on fears rather than good data."

SOURCE: Circulation: Cardiovascular Quality and Outcomes, online May 22, 2020.