When to Say When? Timing of DNR After Cardiac Arrest Varies Widely Among US Hospitals

Hospitals most aggressive with early DNRs are least likely to achieve meaningful survival in patients successfully resuscitated from cardiac arrest.

When to Say When? Timing of DNR After Cardiac Arrest Varies Widely Among US Hospitals

Despite current guidelines that recommend waiting 48-72 hours after in-hospital cardiac arrest to assess patient prognosis, registry data show marked variation among US hospitals as to when survivors of cardiac arrest are placed on Do Not Resuscitate (DNR) orders, with up to 40% of patients being made DNR within 12 hours. Moreover, early implementation of a DNR order—potentially a marker of less aggressive efforts—was linked with worse neurological outcomes.

“The patient who survives an in-hospital cardiac arrest is in a real complex state of being after survival,” lead study author Timothy Fendler, MD, MS (Saint Luke’s Mid-America Heart Institute, Kansas City, MO), told TCTMD. “There are a lot of variables that may be affecting their prognoses, and there are a lot of things that change in that early after period for patients after surviving in-hospital cardiac arrest. . . . Clinicians in that clinical setting should really think about whether or not they ever are too quick to make recommendations or to advise patients and/or their families and loved ones.”

For their study—published online August 7, 2017, ahead of print in the American Heart Journal—Fendler and colleagues identified 24,899 patients from the Get With the Guidelines-Resuscitation registry who survived cardiac arrest at 236 hospitals between 2006 and 2012. Almost one-quarter (22.4%) of patients were made DNR within 12 hours of achieving return of spontaneous circulation, and 24.3% survived to discharge with a favorable neurological outcome.

After adjustment for patient demographics, comorbidities, and event characteristics, early DNR status adoption ranged from 7.1% to 40.5% among hospitals—a “striking amount of variability,” according to Fendler. Also, patients who reported a favorable neurological status also ranged between 2.5% and 44.8% among hospitals.

Interestingly, hospital rates of early DNR for this patient cohort were strongly and inversely tied with risk-standardized rates of favorable neurological survival (P = 0.006), suggesting that hospitals most aggressive with early DNRs were least likely to achieve meaningful survival in their patients successfully resuscitated from cardiac arrest.

Lessons for Physicians, Administrators

However, since qualitative data on the discussions among the healthcare team and patients and their families was not available in this analysis, “further studies are needed to determine which factors may be contributing and if quality improvement efforts can reduce” the variability in early DNR orders, Gregg Fonarow, MD (University of California, Los Angeles), told TCTMD in an email. Additionally, “further studies are needed to determine if there is a subgroup of patients where early DNR could be appropriate,” he added.

As for why a DNR order might be activated early, “we can only theorize,” Fendler said. He guessed that it could be related to a belief that prognosis could be assessed sooner, unmeasured variables related to counseling, or patients’ social or spiritual beliefs.

Also, the variation in favorable neurological survival observed in this study is wide enough that it cannot all be attributed to DNR code status, he observed. It’s possible that patients made DNR are inappropriately being treated less aggressively in general. “This suboptimal care may be due to clinicians misinterpreting DNR preferences and thus not providing other appropriate therapeutic interventions (eg, intensive care unit transfer or blood transfusion) or unwarranted pessimism about prognosis in patients made DNR thus leading to a ‘self-fulfilling prophecy’ of poor outcomes,” the authors write.

“A DNR order should only change one aspect of care, which is whether or not a patient receives CPR if they have a cardiac arrest,” Fendler emphasized.

A DNR order should only change one aspect of care, which is whether or not a patient receives CPR if they have a cardiac arrest. Timothy Fendler

Although the study points out inconsistencies in practice compared with current guideline recommendations, “guidelines are not rules of law,” he noted. “They are a guidepost for us, and they're based on best evidence. There are always extenuating circumstances and patients for whom the guidelines maybe don’t merit the best therapeutic option going forward, but certainly we would hope to see a little bit more homogeneity in the way that we as a community of doctors across the country follow the guidelines.”

He called on hospital administrators specifically to examine how well their institutions are adhering to DNR-related guidelines. “If we find that our hospital or our health system really deviates a lot from a guideline practices or evidence-based medicine, it’s an area that’s ripe for quality improvement,” Fendler concluded

  • Fendler is supported by a grant from the National Heart, Lung, and Blood Institute.
  • Fonarow reports serving on the Get With the Guidelines steering committee.