Abstract 3235: Application of the Appropriateness Criteria for Echocardiography in an Academic Medical Center
Objectives: To examine the feasibility of application of the recent ACCF/ASE Appropriateness Criteria for Echocardiography.
Methods: Indications for transthoracic (TTE) and transesophageal echocardiography(TEE) were retrospectively determined by physician chart review in consecutive patients at a tertiary care center, and were classified according to the guidelines as appropriate (A), inappropriate (I), or uncertain (U), or, for situations not addressed in the document, as not classifiable (NC). To assess interobserver agreement, 63 studies were independently reviewed by 2 physicians.
Results: We included 493 consecutive pts who underwent echo: 176 (35%) outpatient TTE, 131 (26%) inpatient TTE, 95 (19%) outpatient TEE, and 91 (18%) inpatient TEE. Of all studies, 420 (85%) were A, 25 (5%) I, 1 (0.2%) U, and 47 (10%) NC. I and NC studies were more commonly TTE (7%, 13%, respectively) compared to TEE (2%, 5%) (p< 0.001). I studies were more common in outpatients than inpatients (9% vs 2%, p< 0.001). The most common I indications were routine evaluation of prosthetic valves (5, 20%), and evaluation of endocarditis with fever but no bacteremia or new murmur (5, 20%). NC cases included assessment after radiofrequency ablation (8, 17%) or mitral valve repair or valvuloplasty (6, 13%), moderate aortic stenosis (3, 6%), and periodic evaluation after cardiac transplant (3, 6%). The classification of the test as A, I, or U, and the proportion of studies that were NC varied with the specialty of the ordering physician (p < 0.05, p < 0.01). Of the tests ordered by primary care physicians, 18% were NC, compared to 10% by cardiologists, and none by neurologists. Of the tests classified as I, the largest percentage had been ordered by CV surgeons (13%), compared to 5 % by cardiologists, 7% by primary care physicians and 2 % by neurologists. Agreement between observers for classification as A, I, U or NC was present in 83% cases, but the same primary criterion for ordering the test was selected in only 65%.
Conclusion: Application of the criteria was challenging with limited interobserver agreement. In 9.5% of echocardiograms, appropriateness criteria could not be applied; NC studies were more common with TTE cases. Efforts are needed to refine the criteria.