Everyone Wants You to Discharge Syncope Patients, but Should You?

Figure: discharge, syncope, chest pain, ECG, troponins, tachycardia, asystole, low-yield, ROSE score, FAINT score, EGSYS, OESIL, San Francisco Syncope Rule, admissions, pacemaker, dialysis, administrative costs, P-R interval, Boston Syncope Rule, sensitivity, shortness of breathFigureI once saw a woman who had been having burning chest pain, usually after eating. She arrived asymptomatic, and her ECG and troponins were normal. I was going over discharge instructions when she put a hand to her chest, and said, “Oooh, there it is again! Just a burning!” Simultaneously, the telemetry alarms went ding, ding, ding! She was in ventricular tachycardia. She looked uncomfortable but not seriously so. She was rubbing her chest and saying, “Oooh!” and “Phew, that really burns!” “Seriously?” I said to myself. I was really annoyed! Who experiences ventricular tachycardia as burning chest pain that usually comes after eating? The nurse, wide-eyed, had been seconds from taking the patient off the monitor. Years later, I still wonder: Would I have gotten an ECG if that burning pain at discharge had come without the telemetry? Or would I have told her to pick up some famotidine on the way home—for her ventricular tachycardia? I saw a woman in the midst of a neurology workup for new-onset seizures around the same time. She had become briefly unresponsive reading a bedtime story to her grandkids. I had spoken to her neurologist, and the patient liked our plan to send her home on levetiracetam. Then her face went slack, and her cheeks started twitching. Her grandkids, in the room, said, “She's doing it again!” Simultaneously, the telemetry monitor alarms went ding, ding, ding! She was in asystole. Low-Yield Testing I think of humbling cases like these whenever someone complains about the money wasted on low-yield syncope admissions. Everybody with a syncope score wants me to send people home. There's the ROSE score, the FAINT score, and the San Francisco Syncope Rule. There's a Boston score, developed by the attendings who trained me and with whom I still work. There are two Italian scores (EGSYS and OESIL). And there's a Canadian score. (There's always a Canadian score.) Papers that validate or review these scores note with indignation that no one uses them. “Despite the development of tools for risk stratifying patients presenting with syncope, many patients continue to be admitted at an estimated cost of $5400 per admission,” according to a review article. (Med Clin North Am. 2016;100[5]:1019; https://tinyurl.com/tnh53z7n.) Every third syncope gets admitted, and this hasn't changed for years, even though the inpatient workup—telemetry, a cardiac echo, and perhaps more exotica like tilt-table testing—is usually low-yield. The syncope shamers all cite the same (20-year-old) statistic that we spend more than $2.5 billion a year on syncope admissions. Some of those admissions resulted in everything from a pacemaker to emergent dialysis. But others, we are told, were unnecessary. The thinking seems to be: They could have gotten an echo as an outpatient! Think of the savings to a grateful nation! It's not a very compelling argument. Syncope workups are a half-teaspoon of a fraction of a percent of the $4.4 trillion we spend annually on health care. (JAMA Network. April 25, 2024; https://tinyurl.com/4wy8th9t.) Meanwhile, we waste every fourth health care dollar, including by squandering $266 billion every year on excess administrative costs (compared with other nations). (Peter G. Peterson Foundation. April 3, 2023; https://tinyurl.com/fjbfkds7.) To put $266 billion in context: The Environmental Protection Agency's budget is about $9 billion, and the Food & Drug Administration's is around $7 billion. (S&P Global. March 11, 2024; https://tinyurl.com/6cyf7r26; FDA. 2024; https://tinyurl.com/2r9ww6jx.) We could fund 16 EPAs and 16 FDAs if we got rid of corporate parasitism and reduced health care bureaucracy just to the bloated levels seen in France or Germany by, for example, going to a single-payer system. But we could never have Medicare for All because “that's expensive,” so let's complain about syncope admits. Physicist Sold His Nobel In all seriousness, I do think about the finances of syncope (and other) admissions but from a physician's “first, do no harm” perspective. A single night's admission might cost my patient $1000 to $2000 out-of-pocket, and surveys show most Americans cannot absorb an unexpected $1000 cost. (JAMA Health Forum. 2023;4[5]:e230784; https://tinyurl.com/2nznwbc2; USA Today. Jan. 25, 2024; https://tinyurl.com/3x8n7t6h.) Remember, this is a nation where medical bills contribute to the majority of personal bankruptcies and wreck the lives of hundreds of thousands each year and a place where a Nobel Prize-winning physicist paid his medical bills by selling his Nobel medal. Meanwhile, more than 200,000 others turned to GoFundMe campaigns. (Am J Public Health. 2019;109[3]:431; https://tinyurl.com/wpcxwymr; Vox. Oct. 4, 2018; https://tinyurl.com/yf8kadkj; J Philanthr Market. 2023;28[4]:e1777; https://tinyurl.com/26pa75aw.) I don't want to harm my patient with unnecessary medical debt. So, sure, tell me about your syncope rules. A Surplus of Scores An immediate red flag is how different all the syncope rules are. FAINT, one of the newest, looks at a history of congestive heart failure or dysrhythmia (F is for failure, A is for arrhythmia), and then bogs down in the ECG (I is for initial, as in the initial ECG). (Ann Emerg Med. 2020;75[2]:147; https://tinyurl.com/4b6d75ht.) This ECG can't have premature ventricular contractions (PVCs), left ventricular hypertrophy, an axis deviation, a first-degree heart block, a long P-R interval, a short P-R interval, or many other things. So, two pieces of past medical history, two cardiac biomarkers, and a meticulously scrutinized ECG. Got it. Italy's EGSYS (Evaluation of Guidelines in Syncope Score), another newer score, includes no labs, no past medical history, and has less precise ECG criteria. (Heart. 2008;94[12]:1620.) Instead, it asks just five questions about the syncope itself, including whether it happened with exertion, while supine, or after palpitations. The Boston Syncope Rule asks 25 questions, including whether anyone can hear a heart murmur. (J Emerg Med. 2007;33[3]:233; https://tinyurl.com/cbafzwz2.) The patient gets admitted if any answer is yes. ROSE, an older score out of Scotland, doesn't care about any of that, but among its seven criteria wants the oxygen saturation to be above 94% and would like a rectal exam, please. (J Am Coll Cardiol. 2010 23;55[8]:713; https://tinyurl.com/yhyu6kde.) San Francisco, one of the best validated, doesn't need any of that but the patient gets admitted as soon as he reports shortness of breath. (MDCalc. https://tinyurl.com/ha7unxaf.) Keeping Patients Safe The Canadian Syncope Risk Score, meanwhile, has some criteria to click through, but then asks if the emergency physician thinks this was a cardiac or vasovagal event. If the doctor thinks it was cardiac, the +2 points work out so the patient gets admitted; if the doctor thinks it was vagal, the -2 points usually gets the patient discharged. (MDCalc. https://tinyurl.com/3yuasbaa.) (This recognizes a key truth: None of these scores has ever been proven better than a physician's gestalt). I guess you could just pick one. A retrospective study out of China compared five of those scores in 221 older patients and found the sensitivities of Boston and FAINT were highest at 95% and 93%, with negative predictive values (NPVs) of 97% and 96%. (J Int Med Res. 2024;52[1]:300060523122089; https://tinyurl.com/mwhce2u2.) The Boston score has also been validated in a Jerusalem retrospective cohort, where it had a sensitivity of 95% and a NPV of 99%. (Isr Med Assoc J. 2021;23[7]:420; https://tinyurl.com/49jc776z.) So, Boston is perhaps “the best,” but good luck using it. Those 25 questions have multiple sub-questions, to the point that working through the checklist feels like a full admission anyway. Meanwhile, all of these scores have been studied by chart reviews, none by a randomized, prospective trial. Maybe we're afraid to do that? After all, fear is part and parcel of syncope. Patients and families are afraid. Someone just collapsed! Is it going to happen again? Is she about to drop dead? Sometimes, we're confident this was a benign event, says the doctor who diagnosed V-tach as GERD and tried to send asystole home on Keppra. But defaulting to 12-24 hours of telemetry is not wasting money when in doubt. It's therapeutic. It's reassurance. It tells patients and families we care; we're going to think about it some, and watch over them, at least keeping them safe for a little while. DR. BIVENS works at emergency departments in Massachusetts, including St. Luke's in New Bedford and Beth Israel Deaconess Medical Center in Boston. He is double-boarded in emergency medicine and addiction medicine. Follow him on X @matt_bivens. Read his past columns at http://tinyurl.com/EMN-Bivens.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.