Alternative referent standards for cardiac normality. Implications for diagnostic testing

Area of Science:

• Cardiology
• Nuclear Cardiology
• Diagnostic Imaging

Background:

• Accurate reference standards are essential for evaluating cardiac diagnostic tests.
• Radionuclide ventriculography is used to assess cardiac function during exercise.
• Defining normality in patient populations is critical for test interpretation.

Purpose of the Study:

• To evaluate radionuclide ventriculographic exercise response in three distinct populations used as referent standards for cardiac normality.
• To determine if these populations (normal coronary arteriogram, healthy volunteers, low coronary artery disease probability) are equivalent.
• To assess the impact of different reference standards on test specificity.

Main Methods:

• Radionuclide ventriculography during exercise was performed.
• Patient populations included those with normal coronary arteriograms, healthy volunteers, and uncatheterized patients with low coronary artery disease probability.
• Disease probability was calculated using Bayesian analysis of clinical and diagnostic data.

Main Results:

• Catheterized normal patients showed a significant percentage of abnormal ejection fraction (34%) and wall motion (35%) responses.
• Healthy volunteers exhibited normal ejection fraction and wall motion responses.
• Patients with low coronary artery disease probability (<1%) rarely had abnormal responses (7% ejection fraction, 8% wall motion).

Conclusions:

• The three evaluated populations are not equivalent referent standards for cardiac normality.
• Using healthy volunteers or low-probability patients as a standard may overestimate test specificity.
• Employing catheterized normal patients as a standard may underestimate test specificity.