Researchers devise effective prediction model for diagnosing thoracic aorta aneurysms: JAMA

To forecast ascending aortic diameter, a prediction model based on typical clinically accessible data was developed and validated. This study was published in the Journal of American Medical Association.

The majority of aortic aneurysms are discovered by accident, despite the fact that ascending thoracic aortic disease is a significant cause of sudden mortality in the US. In order to establish and verify a clinical score to gauge ascending aortic diameter, James Pirruccello and colleagues undertook this investigation.

Score derivation was carried out in 30 018 subjects and internal validation in an additional 6681 using an ongoing magnetic resonance imaging substudy of the UK Biobank cohort study, which has enrolled participants from 2006 through 2010. In the Community Care Cohort Project, a retrospective hospital-based cohort of longitudinal primary care patients in the Mass General Brigham (MGB) network between 2001 and 2018, external validation was carried out in 50 768 individuals who had undergone transthoracic echocardiography and in 1367 participants from the Framingham Heart Study (FHS) offspring cohort who had undergone computed tomography from 2002 to 2005. Hierarchical group least absolute shrinkage and selection operator (LASSO) regression was used to estimate the ascending aortic diameter. Performance for recognizing diameters of 4.0 cm or more and the correlation between estimated and measured diameter.

The key findings of this study were:

1. The median age of the 30 018 participants in the training cohort (52%) was 65.1 years.

2. For males and women, respectively, the mean (SD) ascending aortic diameters were 3.32 (0.34) cm and 3.04 (0.31) cm.

3. In the UK Biobank validation cohort, 30.8% in the FHS cohort, and 32.6% in the MGB cohort, a score to predict ascending aortic diameter explained 28.2%, 30.8%, and 32.6% of the variation in aortic diameter, respectively.

4. Although the model significantly overestimated or underestimated aortic diameter in external validation, the score had an area under the receiver operator characteristic curve of 0.770 for detecting individuals with an ascending aortic diameter of 4 cm or greater in the UK Biobank, 0.813 in the FHS, and 0.766 in the MGB cohorts.

5. With a fixed-score threshold of 3.537, imaging would be required for 9.7 individuals in the UK Biobank cohort, 1.8 individuals in the FHS cohort, and 4.6 individuals in the MGB cohort to confirm one person with an ascending aortic diameter of 4 cm or more.

6. With specificities of 98.1%, 99.2%, and 96.2%, respectively, the sensitivity at that threshold was 8.9% in the UK Biobank cohort, 11.3% in the FHS, and 18.8% in the MGB cohort.

In conclusion, checks and balances will need to be put in place to make sure that patients with small aortic aneurysms, which are unlikely to rupture, are not unnecessarily identified while searching the electronic health record for patients with aortic aneurysms, as this would undoubtedly cause patient anxiety, lead to over diagnosis, and result in low-value tests and procedures.

Reference: 

Pirruccello, J. P., Lin, H., Khurshid, S., Nekoui, M., Weng, L.-C., Ramachandran, V. S., Isselbacher, E. M., Benjamin, E. J., Lubitz, S. A., Lindsay, M. E., & Ellinor, P. T. (2022). Development of a Prediction Model for Ascending Aortic Diameter Among Asymptomatic Individuals. In JAMA (Vol. 328, Issue 19, p. 1935). American Medical Association (AMA). https://doi.org/10.1001/jama.2022.19701