Background

Syncope accounts for a high number of hospital visits yearly and is associated with significant cost. Various syncope decision tools can help clinicians decide which patients are safe to discharge home when no serious cause is identified in the emergency department (ED). However, clinicians have been left without an equivalent of a HEART score (in the case of patients undergoing an ED workup for ACS) to help guide a disposition for patients presenting to the ED with syncope. There is a need to better understand the benefit of routine hospital admission for syncope in older adults.

Clinical Question

In older adults (>60 years old) admitted to the hospital or observation for syncope or near syncope when no serious cause was identified in the ED, when compared to patients who were discharged from the ED, is there a reduction in serious adverse events at 30 days?

Design

Retrospective secondary analysis of data from a multicenter, prospective, observational study of older adults who presented to an ED with syncope or near syncope

Propensity score matching using 43 covariates was used to adjust for confounders in this observational data set (comparing subjects with roughly equal probabilities of being admitted from the respective admitted and discharged groups, when comparing primary outcome)

Population

  • Inclusion: Patients presenting to the ED with a chief complaint of syncope or near syncope
  • Exclusion: Intoxication, seizure, stroke, head trauma, hypoglycemia, medical intervention to restore consciousness, new or worsening confusion, inability to obtain informed consent
  • Serious diagnosis identified in the ED: Death, significant cardiac arrhythmia, myocardial infarction, significant structural heart disease, stroke, pulmonary embolism, aortic dissection, hemorrhage or anemia requiring blood transfusion, acute pulmonary edema, pneumonia, sepsis, acute renal failure, intracranial bleeding, or acute surgical illness
  • Setting: 11 hospital EDs in academic institutions (US), nonprofit; 10/11 were teaching hospitals with a trauma center

Exposure

  • Admission, observation, or transfer, defined as hospitalization after no serious cause of syncope was identified in the ED

No Exposure

  • Discharge after no serious cause of syncope was identified in the ED

Outcomes

  • Primary Outcome: Serious adverse events within 30 days after ED disposition
  • Composite of: All cause mortality, significant cardiac arrhythmias, myocardial infarction, new diagnosis of structural heart disease, stroke, pulmonary embolism, aortic dissection, subarachnoid hemorrhage, cardiopulmonary resuscitation, internal hemorrhage requiring transfusion, recurrent syncope or fall leading to major trauma, cardiac intervention

Primary Results

N= 2,492 in final cohort, mean age 72.7 ±8.9 years, 50.8% women

Primary Outcome (serious adverse events within 30 days after ED disposition)

  • Most common SAE was serious cardiac arrhythmia (majority symptomatic SVT) 
  • Before propensity score matching (n = 2,492)
    • Admitted to hospital: 7.4% (138/1,866)
    • Discharged: 3.2% (20/626)
    • Absolute Risk Difference: 4.2%, 95% CI (2.38%, 6.02%), significant
  • After propensity score matching (n = 1,064 (532 pairs))
    • Admitted to hospital: 4.89% (24/532)
    • Discharged: 2.8% (15/523)
    • Absolute Risk Difference: 2.1%;95% CI (–0.24%, 4.38%), not significant
  • (ARD similar when SAE occurring during hospitalization were excluded)

Notable Secondary Outcomes (without propensity score matching)

  • Mortality
    • Admitted to hospital: 0.75% (14/1,866)
    • Discharged: 0.56% (3/626)
    • Absolute Risk Difference: 0.19%, 95% CI (-1.16, 0.78%) not significant
  • Length of time to serious adverse event from ED disposition
    • Admitted to hospital: 7.5 days (Hospital length of stay averaged 2.2 days)
    • Discharged: 13.8 days

Strengths

  • Asks a clinically meaningful question on a topic that needs to be better understood
  • Propensity score matched analysis to account for confounders
  • Good follow up >99%
  • Multicenter study using recent data likely making the study results generalizable to most academic emergency departments
  • Sensitivity analysis to account for higher likelihood or adverse events identified during hospitalization

Limitations

  • Secondary retrospective analysis
  • Unclear if powered to detect difference
  • Does not account for social factors affecting patient ability to follow-up
  • Propensity matching resulted in a decrease in sample size from 2,492 to 1,064 patients
  • Generalizability to community EDs is unclear

Author's Conclusions

“In our propensity-matched sample of older adults with syncope or near syncope and no serious diagnosis found on ED evaluation, and with clinical characteristics similar to those of the discharged cohort, hospitalization did not appear to be associated with a reduction in serious adverse events or mortality at 30 days post-ED visit.”

Our Conclusions

Routine admission for older adults presenting to the ED with syncope or near syncope without a serious cause identified the ED does not appear to be associated with a reduction in serious adverse events at 30 days, a clinically relevant outcome. Further prospective randomized studies would be helpful but may not be feasible. Moreover, length of time to an adverse event was ~7 days in hospitalized patients and up to ~13 days in discharged patients, far exceeding a standard “syncope observation” admission that is usually ~48 hours (2 days).

Potential Impact To Current Practice

Routine hospital admission or observation for older adults presenting with syncope or near syncope may not be associated with a meaningful reduction in 30 day adverse events. Discharge of these patients requires readily available follow up. Given that a prospective randomized controlled trial addressing our clinical question is not feasible, propensity score matching does a fine job, but is met with limitations. While this study is not practice-changing, in the right patient, the results may further support a clinician who feels comfortable discharging a patient presenting with syncope or near syncope.