Abstract 14426: Accuracy and Usefulness of Finger Pulse Wave Analysis during Brief Deep Breathing Exercise (Respiratory Stress Response) as a Marker of Significant Coronary Artery Disease
Retraction, March 2011—For the 2010 American Heart Association Scientific Sessions abstract (14426) by Arthur Shiyovich, Amos Katz, Steven J Sushinsky, Yosi Blaer, Petros Okubagzi, Jamal Jafari, and Ron Waksman (Accuracy and usefulness of finger pulse wave analysis during brief deep breathing exercise [respiratory stress response] as a marker of significant coronary artery disease. Circulation. 2010;122:A14426), the authors have notified the editors that results reported in this article are significantly biased and not reliable and therefore retract the abstract.
Background: The novel Respiratory Stress Response (RSR) test which analyses finger-pulse wave oscillations during 70 seconds of deep, paced breathing has been shown to be a strong indicator of significant coronary artery disease (S-CAD).
Hypothesis: We assessed the hypotheses that dose response like association exists between RSR and severity of CAD and that RSR is accurate and useful in detecting and excluding S-CAD.
Methods: We retrospectively analyzed the RSR tests, performed prior to coronary angiography in Washington Hospital Center, USA, and Barzilai Medical Center, Israel, and compared them with quantitative coronary angiography (QCA) results. Receiver operating characteristic methodology determined the optimal cut-off points and accuracy of RSR in indicating CAD.
Results: 245 patients, age 59.9+11.4 years, 69% men were analyzed. RSRs of 10.2%, 10.2% and 8.8% were identified as optimal detecting >50%, >70% and >90% stenosis respectively. RSR significantly and inversely correlated with the degree of stenosis (r2=−0.46, p<0.001). Mean RSR decreased with increased stenosis with dose response like association (Fig 1). Sensitivity and NPV increased, while specificity and PPV decreased as the number of affected arteries or severity of stenosis in single-vessel CAD increased. Sensitivity and NPV increases while specificity and PPV decreases as the RSR cut-off increased (Fig 1). RSR>11.5% excludes S-CAD with >90% accuracy while RSR<7.5% rules-in CAD (>50% stenosis) with >90% accuracy. Multivariate analyses showed that RSR is a strong independent indicator of CAD, with increase in odds ratio as the degree of stenosis increase.
Conclusions: RSR is an accurate and useful test for detecting and excluding CAD. There is a dose-response like, association between RSR and QCA results: Low RSR is an effective, independent indicator of CAD; high RSR strongly rules out S-CAD
- © 2010 by American Heart Association, Inc.