Author + information
- Published online April 1, 2019.
- Varsha K. Tanguturi, MD,
- Vijeta Bhambhani, MS, MPH,
- Michael H. Picard, MD,
- Katrina Armstrong, MD, MSCE and
- Jason H. Wasfy, MD, MPhil∗ ()
- ↵∗Cardiology Division, GRB 8-843, Massachusetts General Hospital, 55 Fruit Street, Boston, Massachusetts 02114
Substantial disparities in clinical outcomes exist for patients with valvular heart disease. Black patients with aortic stenosis receive aortic valve replacements (AVRs) less often than white patients, and women have greater mortality following valve surgery (1,2). Although little is known about the etiology of these disparities, understanding the mechanisms underlying them is essential to improving health equity.
Chronic valvular lesions such as aortic stenosis (AS), aortic insufficiency (AI), and mitral regurgitation (MR) are followed with serial transthoracic echocardiograms (TTEs) (3). Regular TTE surveillance for a diseased valve has been recommended in guidelines to prevent delayed repair and poor cardiovascular outcomes (3).
To elucidate a mechanism for disparities in clinical outcomes, we investigated the relationship between sociodemographic factors and the receipt of TTEs within the guideline-recommended timeframes for AS, AI, and MR. We hypothesized that patients of specific races, genders, incomes, or insurance status were less likely to receive surveillance imaging within these intervals.
The Massachusetts General Hospital (MGH) maintains a database of all echocardiograms performed. We linked these data with administrative data to obtain patient demographic characteristics and included all patients older than 18 years undergoing inpatient or outpatient TTEs between 2001 and March 2016. We sought to include TTEs performed for surveillance of primary AS, AI, and MR and thus excluded TTEs following valvular intervention and those patients with left ventricular ejection fraction (LVEF) <50%.
The primary outcome was whether a patient received a TTE within the guideline-recommended timeframe for the disease type and severity. The variables of interest were patient gender, self-reported race/ethnicity, and insurance status. Clinical covariates included the severity of MR, AS, or AI, and change in left ventricular end systolic dimension (LVESD).
Time intervals between TTEs were assessed sequentially for a given patient. The interval between 2 TTEs was deemed appropriate if it fell within guideline recommendations for imaging based on the disease severity in the index TTE. In patients with multiple valvular diseases, the most severe lesion was used to define the appropriate interval.
Logistic regression was used to assess the association of appropriate TTE surveillance with sociodemographic factors after adjustment. A generalized estimating equation with a repeated statement was used to account for patients with multiple TTE intervals. Income was collinear with race/ethnicity and was thus not included in the model. Regression results are reported as odds ratios (ORs) with 95% confidence intervals (CI). Sensitivity analyses were performed.
The initial database included 302,869 TTEs, representing 161,043 unique patients. After application of inclusion/exclusion criteria, 130,725 TTEs representing 42,289 unique patients remained. In our analysis, 48,366 (37%) echocardiograms were assessed as MR, 70,765 (54%) as AS, and 11,594 (9%) as AI. The mean age was 64 years, and 51% of the study population was male. Of the study population, 85% was white, with 4% self-identified as non-Hispanic black and 3% as Hispanic; 56% had Medicare, 38% had commercial insurance, and 5% had Medicaid.
The results of our model are presented in Figure 1. Female gender was associated with decreased likelihood of appropriate TTE surveillance compared with men (OR: 0.90 [95% CI: 0.86 to 0.95]; p < 0.0001). Non-Hispanic black patients were less likely to receive appropriate TTE surveillance than white patients, with an OR of 0.74 (95% CI: 0.66 to 0.83; p < 0.0001). Patients with Medicaid had a lower likelihood of receiving appropriate TTE surveillance (OR: 0.85 [95% CI: 0.75 to 0.96; p = 0.0095]) compared with those on Medicare, and increased age correlated with a significantly lower likelihood of receiving appropriate surveillance TTEs for each decade of life after 60 years. A sensitivity analysis in a validated loyalty cohort of MGH patients confirmed our main findings.
Our study shows that black patients, women, older patients, and Medicaid patients are less likely to receive appropriate TTE surveillance for valvular disease (Figure 1). Although we are unable to account for studies performed outside our system, we are reassured that our main findings persisted in a cohort cared for largely at MGH. These results should encourage interventions to improve adherence to consensus guidelines and suggest the need for further investigation into clinical consequences of delayed surveillance.
Please note: Dr. Armstrong is a consultant for GlaxoSmithKline. Dr. Wasfy is supported by a career development award from the National Institutes of Health through Harvard Catalyst (KL2 TR001100). All other authors have no conflicts of interest to disclose nor do they receive any funding sources relevant to the contents of this paper.
- 2019 American College of Cardiology Foundation