Clinical and Economic Implications of Inconclusive Noninvasive Test Results in Patients with Suspected Coronary Artery Disease

Author: Akash Goyal, MD, Additional Authors^**

See the poster presentation |  See the poster

Background: Suspected coronary artery disease (CAD) is commonly evaluated with noninvasive diagnostic tests (NIT) accounting for over 4 million tests and $500 million in annual healthcare costs. NIT results can sometimes be inconclusive, but the relative frequencies and medical costs of such inconclusive results remain unclear.

Objectives: We aimed to assess the prevalence of inconclusive NIT results across various testing modalities, quantify the frequency of subsequent testing following inconclusive results, and estimate the 24-month attributable medical costs of inconclusive results.

Methods: In the PROMISE trial, patients with suspected CAD were randomized to and received either stress testing (4533 patients) or anatomic testing with coronary computed tomographic angiography (CTA) (4677 patients) and followed for a median of 25 months. Site-determined inconclusive NIT results included those which were non-diagnostic, uninterpretable, or if <85% target heart rate was achieved on exercise or dobutamine. The frequency of inconclusive NIT results by test type, subsequent testing, and the effect on prospectively-collected, cumulative 24-month medical costs using logistic and linear regression models were analyzed. Cost estimates were derived from: 1) the Premier Research Database for index testing, 2) hospital billing data for hospital costs, and 3) Current Procedural Terminology or Healthcare Common Procedure Coding System codes in the 2014 Medicare national reimbursement schedule for physician costs. Total costs (fixed plus variable) were discounted at 3% annual rate and adjusted to 2014 U.S. dollars using the Producer Price Index for hospital care.

Results: The frequency of inconclusive tests was 23.7% for exercise ECG, 11.9% for stress echo, 6.9% for stress nuclear, and 6.4% for CTA (8.0% overall) (p<0.01 all stress versus CTA; p=0.56 nuclear versus CTA). The median calcium score for inconclusive CTAs was 402 versus 387 for conclusive positive and 9 for conclusive negative CTAs, p<0.01. Compared to negative NITs, inconclusive NITs were more referred to a 2^nd NIT (stress: 14.6% vs 8.5%, OR 1.9; CTA: 36.5% vs 8.4%, OR 6.0, p<0.01) and to catheterization (stress: 5.5% vs 2.4%, OR 2.4; CTA: 23.4% vs 4.1%, OR 6.5, p<0.01). 24-month costs were higher for inconclusive NITs than negative NITs by $4030 (CTA) and $2905 (stress).

Conclusions: In conclusion, among patients with stable angina undergoing NIT, inconclusive CTA tests were less common than inconclusive stress tests overall, while nuclear stress and CTA had similar inconclusive rates. Inconclusive NITs were associated with significantly more referral to a 2^nd NIT and to catheterization when compared to negative NITs, perhaps driving higher costs for both CTA and stress tests. Possible contributors to observed higher costs with inconclusive CTA tests include high calcification burden in inconclusive CTAs test, lack of accompanying functional information, and provider and site unfamiliarity with management of CTA results. These results may help guide NIT selection in patients with chest pain.

**Additional Authors:

Akash Goyal, MD, Ohio State University Wexner Medical Center, Columbus, Ohio

Neha Pagidipati, MD, Duke Clinical Research Institute, Durham, North Carolina

Daniel Mark, MD, Duke Clinical Research Institute, Durham, North Carolina

Larry Hill, MD, Duke Clinical Research Institute, Durham, North Carolina

Brooke Alhanti, MD, Duke Clinical Research Institute, Durham, North Carolina

James Udelson, MD, Tufts Medical Center, Boston, Massachusetts

Michael Picard, MD, Harvard Medical School, Boston, Massachusetts

Patricia Pellikka, MD, Mayo Clinic, Rochester, Minnesota

Udo Hoffmann, MD, Harvard Medical School, Boston, Massachusetts

Pamela Douglas, MD, Duke Clinical Research Institute, Durham, North Carolina