Anant Mandawat, Aditya Mandawat, Gregory Curfman, L. Kristin Newby (2017). A “Shocking” New Code Status. American Heart Journal.
Division of Cardiology, Duke University Medical Center, Durham, NC
Harvard Medical School, Boston, MA
There are approximately 200,000 in-hospital cardiac arrests in the United States each year for which resuscitation is attempted.1 Although there are many components of resuscitation, it is often presented to patients as a binary option: resuscitate or do not attempt resuscitation (DNAR).
Since the seminal report by Zoll, et al., in 1956,2 defibrillation has been a success story of modern medicine. Early defibrillation for a shockable rhythm [pulseless ventricular tachycardia (VT) or ventricular fibrillation (VF)] can be a life-saving intervention that abruptly restores cardiac output. In fact, resuscitation of arrest patients who have shockable rhythms results in significantly better survival and neurological outcomes than for non-shockable rhythms (pulseless electrical activity or asystole).3 Due to the efficacy of defibrillation for shockable rhythms, automated external defibrillators (AEDs) have been placed in public places. Although used in an ambulatory population, AEDs are associated with a survival rate of 74% if a defibrillation is delivered within 3 minutes of collapse but only 49% if a defibrillation is delivered after 3 minutes.4 In the hospital setting, a similar graded inverse association has been shown between time to defibrillation and survival among pulseless VT/VF arrest patients.5
In an era of personalized medicine, should we also consider personalizing resuscitation? Resuscitation is a complex decision that synthesizes the chance of a successful resuscitation with patient preferences and goals of care. Patients may choose to forgo resuscitation due to concerns about trauma incurred during resuscitation, quality of life prior to and post resuscitation, and emotional and financial stress on family members. The SUPPORT Study,6 a multicenter prospective cohort study of 6802 hospitalized patients, sought to understand patient preferences regarding cardiopulmonary resuscitation. Factors associated with a higher preference for DNAR were advanced age, worsening functional status, and worsening quality of life, and these are factors that recur in the literature.7,8 But there are also some patients who forgo resuscitation because it is presented as an all-or-nothing choice. The current all-or-nothing paradigm may result in otherwise functional patients with a good quality of life choosing not to have resuscitation attempts because they must accept all components (e.g., chest compressions, shocks, medications, rescue breathing or intubation), and not receiving the benefits of early defibrillation for shockable rhythms. For example, over 40% of inpatients who rated their quality of life as “good” or better opted for DNAR, and 24% of inpatients with no functional impairment opted for DNAR.6
An alternative to do-not-attempt-resuscitation is shock-only resuscitation (SOR). With SOR, inpatients would not receive chest compressions, but could receive defibrillation for shockable rhythms. Capucci and colleagues provide intriguing evidence for SOR from a 2016 resuscitation study in Piacenza, Italy.9 They report the first use of resuscitation employing AEDs without CPR (i.e. SOR). They compared resuscitation in a small cohort of patients by community volunteers using an AED without CPR versus full resuscitation by EMS over a 13 year period. They found significantly higher rates of survival to hospital discharge with SOR (41.4%) compared with full resuscitation (5.9%). When comparing AED-only resuscitation events with the fastest EMS resuscitation events, the authors found that better outcomes with AED-only resuscitation persisted even though time to treatment was faster in the EMS group (1.7 +/− 0.6 versus 5.3 +/− 3.4 minutes). A meta-analysis of three RCTs also demonstrated that chest compressions prior to VF defibrillation resulted in no benefit for survival to hospital discharge compared with immediate defibrillation, irrespective of response time.10 Further supporting this, in animal models of cardiac arrest, chest compressions and epinephrine result in a significant improvement in survival compared with immediate defibrillation only after 7.5 minutes of VF.11 Therefore, the major determinant of outcomes following in-hospital cardiac arrest in which the rhythm is VT or VF appears to be time to initial defibrillation, with each additional minute of delay resulting in worse survival.5,12
There are several potential benefits of SOR over DNAR. Because outcomes are significantly better for shockable versus non-shockable rhythms,3 patients who otherwise would be poor full resuscitation candidates and thus DNAR are given the best chance at survival with SOR. Interestingly, patients managed by attending cardiologists, who likely manage the highest number of patients with arrhythmias, including shockable rhythms, had the lowest rates of inpatient DNAR orders compared with other specialties.6 Further, SOR avoids patient fears and potential provider concerns associated with full resuscitation. If a shock is delivered, patients are typically unconscious or sedated, resulting in minimal emotional trauma. SOR forgoes CPR, avoiding the physical trauma of broken ribs, bleeding, or chest tubes post-resuscitation. Consistent with current resuscitation practices for patients who are full-code or DNAR, intravenous antiarrhythmics like amiodarone and lidocaine could be administered with SOR. Compared with DNAR, SOR may be lifesaving and within the goals of care of many functional elderly patients who are choosing DNAR to avoid the trauma of full resuscitation or concerns about poor post-resuscitation neurological function or quality of life after a prolonged full resuscitation effort. SOR may also be a reasonable alternative for patients who are uncomfortable with the concept of “no resuscitation” and reflexively choose to be full code. This may be a sizeable population as among inpatients (mean age 65 years) who would prefer to die rather than live with pain, on a ventilator, or in a coma, over 50% opted for CPR,13 suggesting an innate discomfort with “no resuscitation.” Ethically, SOR would be similar to allowing patients with implantable cardioverter-defibrillators (ICDs) to be DNAR with the defibrillation function in their ICD active, a paradigm accepted by Heart Rhythm Society guidelines.14
To better understand patient preferences regarding resuscitation and patient-focused outcomes from SOR, one could conduct a prospective observational study offering elderly patients, who are less likely to opt for full resuscitation,6 one of three options for resuscitation: (1) full resuscitation (2) SOR (3) DNAR. Consistent with the treatment of intensive care unit patients who are DNAR, use of ionotropes like epinephrine and antiarrythmics could be background therapy with any resuscitation choice. Patients would be enrolled into the study on admission to the hospital and prospectively followed. A discussion of the three options would be held by a multidisciplinary team including the attending physician of record, palliative care, and nursing staff to describe the differences between the three resuscitation options. The primary goals of the study would be to: (1) delineate patient preferences for full resuscitation versus SOR versus DNAR, and (2) compare outcomes between full resuscitation and SOR.
Several caveats exist. First, SOR, like full resuscitation, does not obviate the risk of brain damage post-resuscitation. Depending upon the interval between the onset of arrhythmia and the shock, there is the possibility that a shock could successfully restore a cardiac rhythm, but there could still be neurological damage, especially since time-to-defibrillation for SOR is likely longer than it is with an ICD. Comfortingly, the median time to defibrillation for an in-hospital pulseless VT/VF arrest is 1 minute.5 Even among the 30% of patients who underwent defibrillation more than 2 minutes after initial recognition of the arrest, more than 1 in 5 patients survived to hospital discharge and 51% of survivors had no major disability5 Second, SOR, like full resuscitation, will require clinical acumen during the resuscitation. For example if there is a persistently shockable arrhythmia, providers will have to balance additional defibrillations against relative survival and neurologic recovery. One option is to limit SOR to 3 defibrillations. Prior work in VF shows that 3 shocks terminated 96% of VF, with the first shock terminating 84% of VF, the second shock terminating 7.5% of VF, and the third shock terminating 4.8% of VF.15 Appropriately, decisions regarding multiple shocks will be time-limited. A shockable rhythm will often devolve rapidly into a non-shockable rhythm, removing the provider’s ability to further intervene and maintaining the patient’s wishes for limited resuscitation with only shocks. Situations may arise in which a limited duration of chest compressions could be beneficial, for instance a brief period of asystole or PEA due to stunned myocardium post-defibrillation. When formalizing a SOR protocol, discussion regarding a brief one-minute period of post-shock chest compressions will be reasonable. Although there may be added benefit of chest compressions in this scenario, brief compressions could cause SOR to transform into a de-facto full resuscitation with multiple rounds of shocks and chest compressions, limiting provider ability to maintain patient wishes. Lastly, outcomes supporting SOR are based on inferences from full resuscitation. Although it may be a logical choice for many patients, we need to collect prospective data on SOR.
A code status is one of the most critical choices made after a patient-physician discussion. Prior literature has highlighted a need for more nuanced resuscitation options.16,17 SOR is a tempered resuscitation option that leverages the benefits of shockable rhythm resuscitation, including improved outcomes, while avoiding the trauma associated with CPR or the short and long term consequences of a prolonged full resuscitation effort. Of note, our intention is not to challenge the value or outcomes of CPR in patients undergoing full-resuscitation. Rather, SOR is an important dialogue and may be a life-saving choice for some patients who do not conveniently fit into the box of “full-code” or “DNAR.” Ultimately, the decision to pursue SOR should be individualized to each patient based on risks, benefits, and patient preferences for resuscitation.