Weigner and colleagues [1] found a low risk for thromboembolism in patients admitted to the hospital for cardioversion of atrial fibrillation that had lasted less than 48 hours. Although I am impressed by the authors' principal conclusion, I was more intrigued by an inadvertent finding that could have been elucidated further. In 71.5% of cases, patients spontaneously converted to sinus rhythm or cardioversion was unsuccessful in restoring sinus rhythm.

The average hospitalization was 4.6 days for a condition that spontaneously converts in 66.7% of patients or does not convert in 4.8% of patients; thus, an enormous amount of money could be saved if physicians in emergency departments, urgent care centers, or other outpatient sites could identify low-risk patients (those without congestive heart failure, angina, or hypotension) and discharge the patient from the emergency department or urgent care center after the ventricular response is adequately slowed. Stable patients could then either be observed or expeditiously undergo safe cardioversion in the hospital's ambulatory surgery suite or other monitored setting. From the data presented by Weigner and colleagues, however, we cannot ascertain how many of the hospitalized patients in the study were actually at low risk.

Future endeavors may lead to the identification of patient characteristics that would help physicians distinguish between patients who spontaneously convert or do not convert from those who would benefit from cardioversion, thereby allowing more judicious use of medical and economic resources.