Unnecessary Hospitalization and Related Harm for Patients With Low-Risk Syncope

Testing in patients admitted for syncope rarely identifies an underlying cause.^1,2 The San Francisco Syncope Rule (SFSR) was developed to identify patients with syncope at low risk for short-term serious outcomes who were unlikely to benefit from hospital admission.³ During hospitalization, many patients experience adverse events.⁴ Admission and testing can also lead to incidental findings of unclear clinical significance (“incidentalomas”) that can trigger clinical cascades of further testing.⁵ We performed a retrospective cohort study examining outcomes of patients who presented with low-risk syncope and were unnecessarily admitted to the hospital.

From January 1, 2010, to December 31, 2012, all patients with an International Classification of Disease, Ninth Revision (ICD-9) diagnosis of syncope (codes 780.2, 780.0, 458.0, or 780.4) were reviewed by internal medicine physicians (J.V.C., E.A., H.H., and C.K.). Hospitalizations were excluded if there was an obvious alternative reason for admission on presentation; the admission involved drug intoxication, alcohol intoxication, or trauma; patients were directly admitted or transferred; or determinants of SFSR were not available.

We focused on patients with SFSR ratings of 0 (systolic blood pressure >90 mm Hg on triage, hematocrit >30% [to convert to proportion of 1.0, multiply by 0.01], no history of congestive heart failure, no shortness of breath, and no nonsinus rhythm or new electrocardiographic changes) because they are at low risk for bad outcomes and likely did not require admission.^5,6 Data abstraction included length of stay, laboratory testing, imaging testing, procedures, and consultations (Table 1). Adverse events were determined using the Institute for Healthcare Improvement tool.⁴ Reviewers also noted unexpected incidental findings during patient evaluation.

Descriptive statistics were computed using Microsoft Excel (Microsoft Corp). This study was approved by the institutional review board of the University of Maryland, Baltimore. The need for informed consent was waived by the institutional review board. Data were not deidentified because it would not have been possible to access medical records without the information.

A total of 507 admissions were identified by ICD-9 codes, of which 213 met inclusion criteria. Of 213 admissions with syncope, 72 of the admissions (34%) were for low-risk (SFSR score, 0). These patients has a mean 1.73-day length of stay and 10.8 tests. Clinical data are reported in (Table 1). Eleven adverse events were identified in 9 admissions for low-risk syncope (13% [95% CI, 0.06-0.23]). Four of these adverse events were classified as serious and included delirium, transfusion error, hypoglycemia, and fall. Other adverse events included missed medication errors and complications from intravenous and urinary catheter placement.

Twenty-three patients (32% [95% CI, 0.22-0.44]) had incidental findings of unclear significance (Table 2). Of these, 3 patients (13%) had additional evaluation conducted during that inpatient visit, and an additional 11 patients (48%) were advised to have further evaluation performed as an outpatient.

Five patients (7% [95% CI, 0.01-0.13]) had potentially beneficial incidental findings resulting in treatment changes: dysautonomia, paroxysmal atrial fibrillation, inadequately treated hypothyroidism, obstructive hypertrophic cardiomyopathy, and urinary tract infection.

Approximately one-third of patients admitted with syncope to a tertiary care center likely represented unnecessary admissions. Many patients underwent testing advised against by Choosing Wisely, including 88% with computed tomography of the head, 20% with magnetic resonance imaging of the head, and 25% with carotid Doppler ultrasonography.⁶ Of patients with low-risk syncope, 13% experienced an adverse event related to being in the hospital and 32% had an incidental finding of unclear significance. One-half of the patients with incidental findings required follow-up testing or evaluation, which is a burden for patients and potentially triggers other events. In contrast, 5 patients (7%) had clinically relevant unexpected findings identified that were treated.

This study is limited by its use of retrospective data from a single center and relying on the SFSR, some patients with an SFSR rating of 0 may require admission for nonmedical reasons. Strengths of this study include using the current criterion standard for adverse event identification.⁴

Although previous studies^1,2 of syncope have identified the lack of benefit of testing for low-risk syncope, to our knowledge, this is the first study demonstrating patient harm associated with hospitalization for low-risk syncope. Risk for adverse events and incidental findings should be considered before admitting patients with low-risk syncope.

Corresponding Author: Jenna VanLiere Canzoniero, MD, MS, Department of Internal Medicine, School of Medicine, Nelson 207, The Johns Hopkins University, 600 N Wolfe St, Baltimore, MD 21287 (Jcanzon1@jhmi.edu).

Conflict of Interest Disclosures: None reported.

Published Online: April 20, 2015. doi:10.1001/jamainternmed.2015.0945.

Author Contributions: Dr Canzoniero had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Canzoniero, Morgan.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Canzoniero, Morgan.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Canzoniero, Morgan.

Administrative, technical, or material support: Morgan.

Study supervision: Morgan.