BOSTON – U.S. physicians chose to withdraw life-sustaining care from critically ill ischemic stroke patients at much higher rates in 2011 than in 2006, according to results of a new study, although overall percentages remain low.
The trend was bolstered by big jumps in “withdrawal of care” among patients who underwent thrombolysis or both thrombolysis and endovascular treatment: Over the 5-year period, their likelihood of having care withdrawn increased fivefold.
The study authors don’t know whether the trends have continued over the past 6 years, and it’s not clear why the rates rose so much from 2006 to 2011. The researchers speculate that the increase could be linked to disease severity and the preferences of patients and their families.
Whatever the case, study lead author Malik Muhammad Adil, MD, a neurology resident at the Ochsner Clinic Foundation in New Orleans, cautioned that prematurely withdrawing care can throw off stroke prognostication estimates that are already fuzzy. In an interview, he said this can then lead to “significant consequences, including suboptimal outcomes and higher risk of short-term mortality.”
Dr. Adil defines withdrawal of care as “discontinuation of life-sustaining interventions from a patient who is expected to die without this support.” These interventions include such treatments as intubation, mechanical ventilation, feeding tubes, antibiotics, and brain surgery, he said.
For the new study, researchers examined the Nationwide Inpatient Survey database for the years 2006-2011. They reported their findings at the annual meeting of the American Academy of Neurology.
The study reports the following regarding withdrawal of care among ischemic stroke patients:
- The rate grew in those who received neither thrombolysis nor endovascular treatment from 0.8% in 2006 to 3.0% in 2011 (P less than or equal to .0001).
- In those who received thrombolysis alone, the rate rose from 0.9% to 5.5% (P less than or equal to .0001).
- In those who received endovascular treatment alone, the rate increased from 2.8% to 9.0% (P = .0006).
- In patients who received both thrombolysis and endovascular treatment, the rate grew from 2.0% to 10.3% (P = .0009).
Dr. Adil said several factors can affect rates of withdrawal of care in ischemic stroke patients, such as the level of illness (patients receiving aggressive treatment are sicker), advance directives, and the decisions of family members. Some institutions may be more likely to push for withdrawal of care, too, he said. “At my institution, we are not aggressive with withdrawal of care, and I have seen a few better outcomes than expected,” he said.
Also, the lack of useful data regarding prognosis for these patients may lead to premature decisions regarding withdrawal of care, he said.
“We have few prognostic models/scores that predict mortality, and these models are not very sensitive and specific,” Dr. Adil said. “Often, physicians make these decisions based on their previous experiences. All of this leads to premature withdrawal of care. On the other hand, because of premature withdrawal of care, we do not have the data on long-term outcomes on these patients, leading to errors in prognostication.”
In an interview, Adam G. Kelly, MD, a neurologist at the University of Rochester (N.Y.) and chief of neurology at Highland Hospital in Rochester, said the study is important but lacks crucial information such as the circumstances surrounding the care decisions. He hasn’t noticed trends in withdrawal of care among his patients.
“It’s possible that providers have become better at documenting discussions with patients and families which allowed this information to be better captured,” he said. “As the authors mention, it’s also possible that providers are consciously or unconsciously delivering prognoses that are biased towards negative outcomes, leading patients and families to be more apt to choose a palliative approach.”
Dr. Kelly added that neurologists need to objectively offer prognoses in stroke cases. “When the outcome is in doubt, I recommend time-limited trials of interventions of mechanical ventilation, artificial feeding, and other high-intensity interventions to allow patients and families to make the decision they feel is best,” he said.
What about trends in the years since 2011? Dr. Adil said information regarding the years since 2011 wasn’t available to him, but he hopes to analyze the period from 2012 to 2016.
Dr. Adil discussed his study and its implications in a video interview.
No funding is reported. Dr. Adil reports no relevant disclosures.
The video associated with this article is no longer available on this site. Please view all of our videos on the MDedge YouTube channel.