Right atrial pressure: Can it be ignored when calculating fractional flow reserve and collateral flow index? [2]

TY - JOUR

T1 - Right atrial pressure

T2 - Can it be ignored when calculating fractional flow reserve and collateral flow index? [2]

AU - Perera, Divaka

AU - Biggart, Simon

AU - Postema, Pieter

AU - Patel, Sundip

AU - Lambiase, Pier

AU - Marber, Michael

AU - Redwood, Simon

N1 - Funding Information: Divaka Perera MBBChir, MRCP 1 Simon Biggart MBBS, MRCP Pieter Postema MSc Sundip Patel MBChB, MRCP Pier Lambiase PhD, MRCP Michael Marber PhD, FRCP, FACC Simon Redwood MD, FRCP, FACC Simon.Redwood@gstt.sthames.nhs.uk Department of Cardiology, St. Thomas' Hospital Campus, King's College, Lambeth Palace Road, London, England SE1 7EH, United Kingdom 1 Please note: Dr. Perera is funded by a fellowship from the Guy's and St. Thomas' Charitable Foundation. To the Editor: Intracoronary pressure measurements are increasingly used to assess the physiologic significance of epicardial coronary stenoses, particularly when the angiographic severity is considered equivocal. Myocardial fractional flow reserve (FFR) expresses the maximum myocardial flow achievable in the presence of an epicardial stenosis in relation to the theoretical maximum flow without the obstruction and is calculated by simultaneous measurement of mean aortic (P a ), right atrial (P ra ), and distal coronary pressure (P d ) ( 1 ): FFR = (P d − P ra )/(P a − P ra ). In patients with intermediate coronary lesions, without evidence of reversible ischemia, a FFR below 0.75 can be used to identify patients who would benefit from a percutaneous coronary intervention (PCI) ( 2 ). An analogous index is the pressure-derived collateral flow index (CFI), which can be used to quantify the relative contribution of collateral flow to myocardial perfusion. CFI = (P w − P ra )/(P a − P ra ), where P w is coronary wedge pressure measured during transient coronary occlusion by balloon inflation ( 3 ). A CFI <0.25 is usually associated with myocardial ischemia during coronary occlusion and predicts the clinical outcome after PCI ( 4 ). The CFI has become the reference standard for measuring collateral flow in humans, and patients are categorized as having an adequate or inadequate collateral supply on the basis of a threshold of 0.25. Despite the reliance of these indexes on aortic and right atrial pressure, in the interest of expediency, the latter is rarely measured in clinical practice and is either ignored or assigned a fixed value. When P ra is ignored, FFR and CFI become the simplified indexes P d /P a and P w /P a , respectively. Although convenient, the errors introduced by these simplifications have not been systematically evaluated. This study addressed the hypothesis that failure to measure P ra when calculating FFR or CFI can result in artifactual values that cross physiologic dichotomy limits, which determine clinical care. Sixty-six patients referred for physiologic evaluation of coronary stenoses were enrolled in the FFR study. All patients had a lesion of intermediate severity in a major coronary artery (50% to 75% diameter stenosis on visual estimation). Patients with angiographically significant stenoses (>75%) in two or more coronary arteries and those with unstable coronary syndromes were excluded. Sixty-two patients with single-vessel coronary disease undergoing elective PCI were enrolled in the CFI study. All patients had stable angina pectoris and/or evidence of reversible ischemia on noninvasive testing. Patients were excluded if they had undergone a previous PCI to any artery or had an acute coronary syndrome within the previous month. Coronary angiography and PCI were performed via the femoral route. Lesion severity was characterized by off-line quantitative coronary angiography. Right atrial pressure was measured with a 6-F multipurpose catheter inserted via the femoral vein. After calibration, a pressure-sensing guide wire (Pressurewire, Radi Medical Systems, Uppsala, Sweden) was advanced beyond the lesion to measure distal coronary pressure. All measurements were carried out under conditions of maximal hyperemia induced by intravenous adenosine. In the CFI study, P w was measured during balloon occlusion of the lesion. Both FFR and CFI were recalculated for each measurement of P a and P d , assuming P ra to be 0 or 5, 8, and 10 mm Hg. Absolute and relative errors introduced by these estimations were calculated by comparison with the actual values for FFR and CFI using a paired, two-tailed Student t test. Using a dichotomy threshold of 0.75 for FFR and 0.25 for CFI, the sensitivity and specificity were calculated for both indexes at each substituted value of P ra ; actual values of FFR and CFI (incorporating measured P ra ) were considered the reference standard in each case. Data are expressed as the mean value ± standard deviation. In the FFR study, the mean diameter stenosis was 49 ± 16%. Using a dichotomy threshold of 0.75, 12 functionally significant stenoses (14% of all lesions) were misclassified as insignificant when P All misclassifications occurred in lesions with P ra was ignored. Fewer misclassifications occurred when P ra was estimated to be 5, 8, or 10 mm Hg, but several physiologically insignificant lesions were erroneously considered severe when right atrial pressure was assumed to have a fixed value ( Fig. 1 A). d /P ra values between 0.74 and 0.78. The sensitivity of FFR for detecting significant coronary lesions was reduced to 64% when P ra was ignored. Substituting a fixed P ra value when calculating FFR increased the sensitivity of the technique, but specificity decreased with higher assumed values of P ra ( Fig. 1 B). In the CFI study, the CFI was 0.20 ± 0.11 when measured P When P ra was included in the calculation and 0.26 ± 0.10 when P ra was ignored (p < 0.001). The mean error in CFI was 57 ± 89%, 21 ± 53%, −3 ± 52%, and −20 ± 64% when P ra was assumed to be 0, 5, 8, and 10 mm Hg, respectively ( Fig. 2 ). ra was ignored, 16% of individuals with insufficient collateral flow to prevent ischemia on coronary occlusion (CFI 0.25). Of the cases, 10% to 12% were misclassified when P ra was assigned a fixed value, with CFI being increasingly underestimated at higher substituted values of P ra . The sensitivity of pressure-derived CFI to detect inadequate collateral supply was reduced to 75% when P ra was ignored. Substituting values of 5, 8, and 10 mm Hg in the CFI calculation increased the sensitivity at the expense of specificity, and no fixed value of P ra significantly improved the accuracy of the test. The principal virtue of FFR is its independence of prevailing hemodynamic conditions, which in turn reflects the incorporation of mean aortic and right atrial pressures in its calculation. The original validation of FFR relied on calculations that included measurement of central venous pressure ( 1 ). However, in practice, right atrial pressure is often assumed to be negligible, and a simplified ratio of distal coronary to mean aortic pressure is used instead. Our study demonstrates that using this simplified index or assuming a fixed P ra value when calculating FFR could result in artifactual values that lie on opposite sides of the 0.75 dichotomy limit, which in turn may lead to inappropriate therapy. It is also noteworthy that these misclassifications occurred in a population where mean right atrial pressure was 5.5 mm Hg, which challenges the notion that right atrial pressure should only be measured in circumstances where it may be pathologically elevated. In view of the lower distal coronary pressures encountered during coronary occlusion, failure to include actual right atrial pressure would be expected to cause larger errors in calculation of CFI than FFR. These errors have not been formally quantified before, although it has been suggested that assuming a P ra value of 5 mm Hg results in insignificant differences in CFI ( 5 ). In contrast, our findings show that substantial errors occur when right atrial pressure is assumed to be a fixed value. It is interesting to note that Seiler et al. ( 6 ) found a better correlation between Doppler- and pressure-derived CFI when a measurement of central venous pressure was included, as compared with those calculations where venous pressure was assumed to be 5 mm Hg. Given that almost one-sixth of patients in our study were assigned to the wrong CFI category when P ra was assumed to be negligible or a fixed value, we believe that measurement of right atrial pressure is imperative when calculating CFI. The FFR and CFI were derived and validated with the inclusion of right atrial pressure. Ignoring P ra in everyday practice debases the fidelity of FFR and may lead to inappropriate therapy in some cases. Similarly, assuming a fixed arbitrary P ra value leads to substantial errors in CFI calculations. Therefore, right atrial pressure should always be measured when calculating these physiologic indexes. If the simplified index P d /P a is used, values between 0.70 and 0.80 mandate recalculation of true FFR, after measurement of P ra .

PY - 2004/11/16

Y1 - 2004/11/16

UR - http://www.scopus.com/inward/record.url?scp=8144221924&partnerID=8YFLogxK

U2 - https://doi.org/10.1016/j.jacc.2004.08.018

DO - https://doi.org/10.1016/j.jacc.2004.08.018

M3 - Article

C2 - 15542298

SN - 0735-1097

VL - 44

SP - 2089

EP - 2091

JO - Journal of the American College of Cardiology

JF - Journal of the American College of Cardiology

IS - 10

ER -