Wide variability in the yield of transthoracic echocardiocardiography depending on the indication: Who should order it and in whom should it be ordered?

It is estimated that one-sixth of the approximate one trillion dollars spent on health care cost in the US is wasted on 'unnecessary' tests. One way to define the term 'unnecessary' is to determine the yield of the test for a particular indication. Accordingly, we prospectively determined the utilization patterns and yield of transthoracic echocardiography (TTE) in 2175 consecutive patients in whom the indication of the test was not protocol based. For the 1338 in-patients, the 5 most common indications for TTE and their positive yields were: 1) LV dysfunction (44% of all pts - yield 43%); 2) pericardial disease (11% of all pts - yield 31%); 3) cardiac source of embolism (11% of all pts - yield 3%) ; 4) segmental wall motion abnormalities (7% of all pts - yield 51%; and, 5) native valvular disease (6% of all pts - yield 39%). For the 837 outpatients, the 5 most common indications for TTE and their positive yields were: 1) LV dysfunction (41% of all pts - yield 32%); 2) native valvular disease (18% of all patients - yield 27%); 3) changes in known valve disease (6% of all pts yield 45%; 4) pericardial disease (5% of all pts yield 24%; and, 5) congenital heart disease (4% of all pts - yield 34%) . The yield was almost twice as high for cardiologists compared to noncardiologists for all indications in patients with similar baseline characteristics. These findings indicate that there is wide variability (3% to 51%) in the yield of TTE for various indications. The yield is influenced by the source of patient referral (inpatient versus outpatient and cardiologists versus non-cardiologists). For most effective utilization of TTE in this era of cost-consciousness and capitation, therefore, the following questions need to be addressed: Who should order a TTE and in whom should it be ordered?