Elderly patients who sign a preoperative “do not resuscitate” order are more likely to die from complications following surgery than matched controls because they turn down aggressive management of their complications, according to a new study presented at the 132nd annual meeting of the American Surgical Association (ASA).
Investigators call this phenomenon a “failure to pursue rescue,” and say surgeons, caregivers and policymakers need to be aware of the trend.
“Failure to pursue rescue is a more accurate description of what happens here,” said lead author John E. Scarborough, MD, assistant professor of trauma and surgical critical care, Duke University Medical Center, in Durham, N.C. “It’s not that patients die after massive attempts to prevent their deaths; it’s that these patients or their families refuse to accept aggressive management of their complications.”
The study is based on an analysis of data collected by the American College of Surgeons National Surgical Quality Improvement Program between 2005 and 2010. In that period, 25,518 patients older than 65 years underwent an emergency operation for one of 10 common general surgical conditions. Of these patients, 4.2% (n=1,061) signed preoperative “do not resuscitate” (DNR) orders before their index operation.
Overall, 37.2% of elderly DNR patients died in the 30 days following surgery, the study showed, and 42% experienced major complications. Patients who signed DNR orders were older, had poorer functional status, higher ASA physical status scores and higher rates of cognitive dysfunction, malignancy, congestive heart failure and other comorbidities than non-DNR elderly surgical patients. The patients who signed DNR orders were also more likely to receive preoperative transfusions, undergo longer operations and have procedures associated with higher total work relative value units, used as a marker for complexity.
Investigators performed propensity score matching to compare DNR and non-DNR patients with similar patient- and operation-related variables. The 30-day mortality was significantly higher among patients who signed a DNR, at 36.9%, than among non-DNR patients of whom 22.3% died within the same period (odds ratio, 2.07; 95% confidence interval, 1.69-2.55). Despite the difference in mortality, the incidence of major postoperative complications was similar between the two groups, at 42.1% for DNR patients and 40.2% for non-DNR patients.
Analysis revealed that the response to the postoperative complications led to the difference in mortality rates. Of DNR patients who developed a major complication, 57% died within 30 days of surgery, which was 18 percentage points higher than the rate among patients without a DNR. DNR patients were also significantly less likely to undergo reoperation following the index operation; only 8% had a second operation, compared with 12% in the non-DNR group. The investigators believed that the lower reoperation rate among DNR patients was due to a refusal to consent to follow-up interventions.
The findings could help surgeons and policymakers who manage the care of geriatric patients. For general and acute care surgeons, the results provide objective data they can use when advising elderly DNR patients about the risks of emergent surgery, the investigators said.
Surgeons’ conversations with compromised elderly patients and their families are difficult and surgeons need to learn how to better counsel these patients, said Ronnie A. Rosenthal, MS, MD, professor of surgery at Yale University School of Medicine, in New Haven, Conn.
Even after those discussions have occurred, an ethical dilemma can arise when a patient has complications after surgery. “The ethical dilemma is that the surgeon may well feel that rescue from the complication is possible—which the study supports since mortality following a complication in the matched non-DNR patients is lower while the complication rate is the same—but patient autonomy demands respect for the patient’s right to stop further treatment,” Dr. Rosenthal wrote in an e-mail interview with General Surgery News.
The results send a strong message about misclassifying these cases as “failure to rescue.” Failure to rescue typically is used to describe a patient death following postoperative complications, and is used as an important patient safety indicator; however, the researchers argue that failure to rescue and failure to pursue rescue should not be lumped as one.
The investigators added that they were pleased to find no difference in pre- and perioperative treatment of patients related to their DNR status. “Our findings do not support the existence of an overt bias among physicians against aggressive preoperative or intraoperative management of elderly DNR patients,” said Dr. Scarborough.