The declining specificity of exercise radionuclide ventriculography.

Area of Science:

• Cardiology
• Diagnostic Imaging

Background:

• Exercise radionuclide ventriculography (ERNV) was initially considered highly specific for coronary artery disease (CAD).
• Later studies indicated a high false-positive rate for ERNV, prompting re-evaluation of its diagnostic accuracy.

Purpose of the Study:

• To investigate the reasons for the reported decline in specificity of ERNV for CAD.
• To analyze temporal changes in ERNV test responses and patient characteristics.

Main Methods:

• Retrospective analysis of 77 angiographically normal patients divided into an early (1978-1979) and recent (1980-1982) study period.
• Comparison of ejection fraction and wall motion responses between the two periods.
• Evaluation of pre-test probability of CAD and referral patterns (pre- and post-angiography).

Main Results:

• Normal ERNV responses decreased significantly in the recent period (ejection fraction: 49% vs. 94%; wall motion: 36% vs. 84%).
• Patients in the recent period had a higher pre-test probability of CAD (38% vs. 7%).
• Increased prior ERNV testing before angiography in the recent period led to more abnormal results (55% vs. 6%).

Conclusions:

• The decline in ERNV specificity for CAD is attributed to pre-test referral bias (testing sicker patients) and post-test referral bias (selecting positive tests for angiography).
• These biases significantly impact the interpretation of ERNV results over time.