By Cara Pellegrini, MD
Assistant Professor of Medicine, UCSF; Cardiology Division, Electrophysiology Section, San Francisco VA Medical Center
Dr. Pellegrini reports no financial relationships relevant to this field of study.
SYNOPSIS: Typical cardiac symptoms frequently precede sudden cardiac arrest and are frequently unheeded, but when acted on are associated with decreased mortality.
SOURCE: Marijon E, et al. Warning symptoms are associated with survival from sudden cardiac arrest. Ann Intern Med 2016;164:23-29.
By definition, sudden cardiac death (SCD) is an unexpected loss of pulse without obvious cardiac cause. Affecting more than 550,000 Americans and accounting for more than half the cardiovascular deaths in the United States, SCD has traditionally confounded efforts to predict its victims beyond broad strokes. Few variables outside of depressed left ventricular ejection fraction really affect prediction meaningfully. Perhaps it is not surprising that despite continued research and clinical efforts, the survival after sudden cardiact arrest remains very low and stable around 7%.
Against this backdrop Marijon et al hypothesized that there might be a warning signal in the form of symptoms in the hours to days prior to an arrest that, if heeded, might improve outcomes. They performed a large prospective, community-based study of survivors and decedents of SCD in the Portland, OR, area. SCD cases were canvassed from the EMS system, the medical examiner’s office, and EDs of all local hospitals. Utilizing EMS reports, hospital records, and outpatient community physician charts, the authors collected data on 839 patients 35-65 years of age with symptoms reported during the 4 weeks prior to SCD. They compared patient characteristics and mortality among those who “acted on” these symptoms (called 911) and those who did not.
They found that 51% of patients had reported symptoms prior to their SCD. Of note, symptoms immediately prior to collapse were not included. Eighty percent of symptoms started more than 1 hour before SCD, including 34% of patients who had more than 1 day from symptom onset to ultimate SCD. (Ninety-three percent of these patients had recurrent symptom episodes during the 24 hours prior to the arrest.) The most common symptom was chest pain (46%) and this was mostly typical angina. Only 19% of those with symptoms called 911. Older patients and those with a history of heart disease or continuous chest pain were more likely to seek help. Survival was significantly higher among those who did call 911. While multiple known factors that predict survival — witnessed arrest, bystander cardiopulmonary resuscitation, and initially shockable rhythm — were all more common in those who called 911, after adjustment for these and other suspected confounders, those who called 911 retained a nearly five-fold survival benefit. The authors concluded that warning symptoms frequently occur before SCD but are unfortunately mostly ignored, suggesting a new target for awareness.
COMMENTARY
This study is provocative in suggesting that SCD occurrence could be specifically predicted and potentially prevented, albeit likely with only 1 day or perhaps 1 week of lead time. Still, in this era of rapid data sharing, near-instant automated electrogram interpretations, and geolocalization, even a brief warning period could prove sufficient to be life-saving. Rather than continuing to focus so much of our effects on defining risk predictors (with minimal success), perhaps we’d be better served educating the public on the importance of symptom recognition and preparing more rapid response capabilities to actual (near) events. The merits of this approach are both highlighted and detracted by the knowledge that 12% of patients in this study had consulted a physician within 30 days of their SCD and received a “systemic work-up.” Is recognition of the soon-to-be-afflicted still hazy only 1 month prior to the event? After all, symptoms such as chest pain and shortness of breath are not so unusual, and the denominator of those with such symptoms to view the numerator of those who ultimately had an arrest is not known. On the other hand, maybe this statistic simply underscores that the most important preventive action is that most proximate to the event, i.e., calling 911 and performing CPR.
In concert with data from myocardial infarction literature, women were less likely to experience chest pain and more likely to report shortness of breath prior to SCD. Notably, symptom prevalence was very similar between men and women. Also interesting was that overall symptom occurrence was similar between survivors and non-survivors, suggesting that the difference in outcome was not wholly related to disparate etiologies, such as stuttering ischemia leading to ventricular fibrillation and asystolic arrest, which might be expected to have a worse outcome.
About one-quarter of patients did not have symptom data, and the potential bias of this large proportion of missing data must be considered. The observational nature of the study obviously makes causal interpretation of the association between calling 911 and lower mortality fraught. The authors noted that this ongoing long-term study might have led to some recall and response bias on part of the EMS staff. Conversely, by including only those who actually did sustain a SCD, the positive effects of early symptom alert may be underestimated. Finally, it remains unproven that, even if there were greater recognition of symptoms and responsive action, lives would be saved. Nonetheless, I believe that these results warrant renewed public awareness efforts regarding the importance of careful attention to cardiac symptoms (including shortness of breath, particularly among women) and the need to seek emergent help when they occur.
