By Michael H. Crawford, MD, Editor
Synopsis: A single health system study of electronic provider notifications when severe aortic stenosis was discovered on echocardiography showed that referrals for aortic valve replacement significantly increased, especially in women and those > 80 years of age.
Source: Tanguturi VK, Abou-Karam R, Cheng F, et al. Electronic provider notification to facilitate the recognition and management of severe aortic stenosis: A randomized clinical trial. Circulation. 2025;151(21):1498-1507.
Observational data have shown that patients with echocardiographic evidence of severe aortic stenosis (sAS) often do not get referred for aortic valve replacement (AVR). This especially is an issue with women and older adults. Thus, this pragmatic, single-blind, cluster randomized controlled trial of electronic provider notification (EPN) of sAS (DETECT-AS) by transthoracic echocardiography (TTE) is of interest.
The aim of DETECT-AS was to evaluate the efficacy of EPN for improving rates of AVR within one year of the TTE diagnosis. It was conducted in the Massachusetts General Hospital health system, which shares a common electronic medical record (EMR). DETECT-AS was investigator-initiated but financed by Edwards Lifesciences. Providers whose patients had sAS were randomly assigned to EPN or usual care (UC) and capped at 50 patients.
The diagnosis of sAS was based on an aortic valve (AV) area of ≤ 1.0 cm2. EPN was delivered by the EMR and email. It was customized by sAS subtype based on the AV pressure gradient and left ventricular ejection fraction (EF) with the recommendations from the American Heart Association/American College of Cardiology 2020 guidelines for each subtype.
The primary endpoint was AVR within 12 months of the diagnostic TTE. Secondary endpoints included all-cause mortality and heart failure hospitalization (HFH). Various subgroups were evaluated (female sex, age > 80 years, and symptomatic patients).
After excluding one patient with a prior AVR and three patients because of the cap on their provider, 939 patients were randomized to EPN (144 patients, 496 providers) or UC (141 patients, 443 providers). Providers included interventional valve disease cardiologists (4%), other cardiologists (48%), primary care (33%), and others (15%). The mean age of the patients was 77 years (with 41% > 80 years), 47% women, and 96% self-identified as white.
The average AV area was 0.8 cm2, the average mean pressure gradient was 36 mmHg, the average EF was 62%, and 58% of patients were symptomatic. The primary outcome was achieved in 48% of the EPN group and 37% of the UC group (odds ratio [OR], 1.62; 95% confidence interval [CI], 1.13-2.32; P = 0.009). In symptomatic patients, the primary outcome also was more frequently observed in the EPN group (OR, 1.4; 95% CI, 1.06-1.85; P = 0.02).
The EPN treatment effect was highest for women (OR, 2.78; 95% CI, 1.69-4.57; P < 0.001) and those > 80 years of age (OR, 2.00; 95% CI, 1.17-3.41; P = 0.01). Also, EPN patients experienced less mortality (12% vs. 14%, P = NS), which was nearly significant in the symptomatic patients (15% vs. 21%, P = 0.06). HFH was not significantly different between the two groups. The authors concluded that in patients with TTE-identified sAS, EPNs increased the rate of AVR and reduced sex and age disparities in AVR referral.
Commentary
The echocardiographic evaluation of aortic stenosis is complex and can be confusing to those not intimately familiar with it. In DETECT-AS, the EPN discussed the management of four subgroups of sAS based on valve gradient and EF, with instructions for more testing (usually computed tomography), referral to a valve disease specialist, or directly to AVR. This EPN was successful in increasing the quality of care in sAS patients. In the UC group, less than half of the sAS patients got AVR within one year and only 19% received it by 90 days. Also, women and patients > 80 years of age were more likely to get AVR in the EPN group. In addition, the rate of AVR in asymptomatic patients did not increase with EPNs, suggesting that they do not lead to unnecessary AVRs. The authors maintain that these findings validate the need for EPNs and recommended that AVR within 90 days of TTE diagnosis of sAS should be a performance measure for hospital systems. Of course, this would have to come with caveats for asymptomatic patients, those at high risk for AVR, and those who refuse AVR.
The authors mention the potential problem of EPN fatigue. You can imagine that many specialties in medicine could come up with similar categories of historically undertreated patients. What they do not discuss is whether the TTE report could be the vehicle, since this should be reviewed by the ordering physician. In some institutions, TTE reports already give advice to providers, such as “transesophageal echocardiography would have a better sensitivity for X diagnosis.” Another potential impediment is the ordering provider’s perception that the EPN is to increase the business of the AVR providers or is a “taking over” of their patient. The industry sponsorship of the study may reinforce this perception.
There are limitations to DETECT-AS to be considered. Provider trial awareness could have influenced care in the no EPN group. This was a very homogenous population, and the results may not apply to other groups. The study was conducted in an integrated health system. It may be hard to implement EPNs in other care settings. sAS was defined only by valve area, so patients with pseudo sAS may have been included. Despite these limitations, DETECT-AS emphasizes that we need better systems to align significant test results with patient care, especially in patients with aortic stenosis.
Michael H. Crawford, MD, is Professor Emeritus of Medicine and Consulting Cardiologist, UCSF Health, San Francisco.
