Resting Pd/Pa Measured with Intracoronary Pressure Wire Strongly Predicts Fractional Flow Reserve

   ABSTRACT: Objective. To investigate the relationship between resting distal coronary pressure to aortic pressure ratio (Pd/Pa) and fractional flow reserve (FFR) obtained during maximal hyperemia. Background. FFR is an invasive index of the functional severity of a coronary artery stenosis determined from coronary pressure measurements. It is generally believed that there is little correlation between resting Pd/Pa and FFR obtained during maximal hyperemia. We have therefore studied this relationship in a large cohort of patients who had undergone pressure-wire assessments. Methods. 528 consecutive pressure-wire studies performed in 483 patients over a 2-year period were retrospectively analyzed. Results. A linear correlation between resting Pd/Pa and FFR post-pharmacological hyperemia was observed (rho = 0.74; p Conclusions. We demonstrate a strong correlation between resting Pd/Pa and FFR. Resting values of Pd/Pa can be used to predict a positive FFR result with relatively high PPV and NPV. This may potentially obviate the need for adenosine infusion in a proportion of pressure-wire studies. J INVASIVE CARDIOL 2010;22:260–265 Key words: Pressure Wire, PCI, fractional flow reserve    Fractional flow reserve (FFR) is an invasive index of the functional severity of a stenosis within a coronary artery. It is measured using a pressure sensor-tipped angioplasty guidewire and is expressed as the ratio of distal coronary pressure (Pd) to aortic pressure (Pa) during maximal hyperemia induced with pharmacological agents. FFR expresses the maximum achievable blood flow to the myocardium supplied by a stenotic artery as a fraction of normal maximum flow.^1,2 In the absence of coronary disease, FFR equals 1, while a value of ≤ 0.75 identifies coronary stenosis associated with inducible ischemia with a similar accuracy to noninvasive stress testing.^2,3,4 Furthermore, the information obtained from a pressure wire provides superior spatial resolution compared to myocardial perfusion studies because each stenotic lesion is analyzed separately, hence avoiding the masking of one ischemic zone by another more severely ischemic zone.⁵ The utility of FFR in guiding percutaneous coronary intervention (PCI) has been demonstrated in many patient subsets including those with intermediate coronary stenoses,⁵ prior myocardial infarction (MI),⁶ multivessel disease,⁷ left main disease⁸ and coronary artery bypass graft (CABG) interventions.⁹    Apart from lesion severity, numerous factors such as lesion diameter, lesion length and left ventricular function influence FFR.¹⁰ Among these factors, a major role is played by microvascular integrity, microvascular perfusion, microvascular endothelial function and autoregulation. Conventionally, it is thought that meaningful measurement of coronary pressure can only be derived in the maximal hyperemic state, which would obliterate the contribution of microvascular resistance to flow impediment. Consequently, it is believed that the resting Pd-to-Pa ratio (Pd/Pa) bears little relation to FFR measured during maximal hyperemia with pharmacological stress.¹¹ Therefore, the aim of this study was to determine the relationship between resting Pd/Pa and FFR in a contemporary cohort of patients undergoing pressure-wire assessments during cardiac catheterization and determine whether baseline values of Pd/Pa can accurately predict a positive or negative FFR result.

Methods

   We retrospectively analyzed 528 consecutive pressure-wire studies performed in 483 patients at the Manchester Heart Centre, U.K. over a 2-year period using a Radi pressure-wire system (Radi Medical Systems AB, Uppsala, Sweden). FFR methodology was employed in a standard fashion and intravenous administration of adenosine at a dose of 140 μg/kg/min was chosen to achieve maximum hyperemia, since increasing the dose beyond this has not been shown to have any further effect on microvascular resistance.¹² Patient demographic and procedural data were obtained from the Manchester Heart Centre Patient Database (CARDEX) in which procedural and clinical demographic data are entered for each patient undergoing PCI prospectively. Angiographic degree of stenosis was assessed by the operator and recorded prospectively into the CARDEX system. Data quality entered into the CARDEX system is cross-checked and validated by an independent Clinical Information Assistant using the PCI procedural reports generated by the operator and information obtained from the medical notes.    Descriptive statistics were reported as mean and standard deviation (SD) for normally distributed continuous variables. ROC curves were constructed using Med Calc software (Belgium).

Results

   Retrospective analysis was performed on 528 consecutive pressure-wire studies performed in 483 patients at the Manchester Heart Centre over a 2-year period (June 2006–June 2008) using the Radi PressureWire system (Radi Medical Systems/St. Jude Medical, St. Paul, Minnesota). Baseline characteristics of the patient cohort are presented in Table 1. Briefly, the average age of the cohort was 61.5 ± 1.1 years (mean ± SD), 70.0% were male and 35.6% underwent cardiac catheterization following an index presentation with acute coronary syndrome (ACS). Lesion characteristics are presented in Table 2.    Resting Pd/Pa varied between 0.73 and 1.0 (Table 3) and the calculated FFR following maximal hyperemia varied between 0.41 and 1.0. Figure 1 illustrates the almost linear relationship between resting Pd/Pa and the corresponding FFR measured during maximal hyperemia with adenosine. A statistically significant correlation between resting Pd/Pa and FFR post pharmacological hyperemia was calculated using Spearman's rank correlation with a coefficient (rho) of 0.74 (p 13) and 169 studies (32.0%) were positive if a FFR cut-off value of 0.80 (as per the FAME study¹⁴) was used. Figures 2A and 2B illustrate the positive and negative predictive value plotted against resting Pd/Pa for FFR values defined as positive as per DEFER13 and FAME¹⁴ criteria, respectively. Similarly Receiver-Operator curves (ROC curves) for resting Pd/Pa are presented for FFR values defined as positive at cut-off values of ≤ 0.75 (as per DEFER;¹³ Figure 3A) and ≤ 0.80 (as per FAME;¹⁴ Figure 3B), with areas under the curve of 0.87 and 0.86, respectively.    From the data presented in Figures 2 and 3, optimal values of Pd/Pa were calculated for predicting a positive/negative FFR result as per DEFER¹³ or FAME¹⁴ criteria. When a FFR value of ≤ 0.75 is defined as a positive result as per DEFER,¹³ a resting Pd/Pa of ≤ 0.85 has a positive predictive value of 95%, while a resting Pd/Pa of ≥ 0.93 has a negative predictive value of 95.7%. Using these parameters, a total of 347 out of 528 (65.7%) pressure-wire procedures performed had a resting Pd/Pa of ≤ 0.85 or ≥ 0.93 (Figure 4). Consequently, resting Pd/Pa would not be able to accurately predict a positive/negative FFR result in only 34.3% cases if the DEFER¹³ criteria were used.    Similarly, when defining a FFR value ≤ 0.80 as a positive result, as in the FAME study,¹⁴ a resting Pd/Pa of ≤ 0.87 has a positive predictive value of 94.6%, while a resting Pd/Pa of ≥ 0.96 has a negative predictive value of 93%. Using these parameters, a total of 250 out of 528 (47.3%) pressure-wire measurements performed had a resting Pd/Pa of ≤ 0.87 or ≥ 0.96 (Figure 5). Similarly, resting Pd/Pa would not be able to accurately predict a positive/negative FFR result in only 52.7% cases if FAME¹⁴ criteria were used. These cut-off values calculated for either DEFER or FAME criteria were then applied only to lesions of intermediate severity (defined as angiographic degree of stenosis between 50–74%) for validation since this may be the main application of FFR in clinical practice. When a FFR value of ≤ 0.75 is defined as a positive result according to DEFER,¹³ a resting Pd/Pa of ≤ 0.85 has a positive predictive value of 100%, while a resting Pd/Pa of ≥ 0.93 has a negative predictive value of 98.5%. Similarly, when defining a FFR value ≤ 0.80 as a positive result, as in the FAME study,¹⁴ a resting Pd/Pa of ≤ 0.87 has a positive predictive value of 100%, while a resting Pd/Pa of ≥ 0.96 has a negative predictive value of 93.9%.    This would suggest that with positive and negative predictive values close to 95% for lesions of all angiographic severity, calculation of the resting Pd/Pa parameters would have been sufficient since infusion of adenosine to enable calculation of FFR would only have been required in 34% and 53% of cases as per DEFER¹³ or FAME¹⁴ criteria, respectively. The greatest utility of the resting Pd/Pa appears to be in predicting a negative FFR post adenosine since resting values of Pd/Pa, which predict a positive FFR, constitute a minority of cases (3.8% and 7.0% when DEFER¹³ and FAME¹⁴ criteria are used, respectively).

Discussion

   We have demonstrated in this large retrospective study of over 480 patients undergoing pressure-wire assessment (of similar size to the FAME study, n = 509 patients and larger than the DEFER study, n = 325) that there is a strong correlation between resting Pd/Pa and FFR measured during maximal hyperemia with adenosine. We have also demonstrated that certain cut-off values of resting Pd/Pa can be used to predict a positive or negative FFR result with high positive and negative predictive accuracy (in the order of 95%), irrespective of whether FAME or DEFER criteria are used. Indeed, the area under the receiver operator curves (AUC) for resting Pd/Pa for FFR values defined as positive by DEFER (0.87) and FAME (0.86) compare favorably to standard cardiac troponin T for the diagnosis of MI (0.85–0.90),^15,16 BNP for diagnosis of heart failure (0.84–0.91)^17–19 and fasting plasma glucose for the detection of diabetes (0.81–0.89).^20,21 In fact, our data might suggest that adenosine infusion may only be required in 34% and 52% of pressure-wire cases if DEFER or FAME criteria are used, respectively, if our findings were to apply to and be validated by other cohorts studied. Since the proportion of patients in our cohort whose resting Pd/Pa accurately predicted a positive FFR (as per DEFER or FAME definition) is small (3.8% and 7.0%, respectively), the importance of our findings is primarily applicable to the cohort of patients whose resting Pd/Pa was predictive of a negative FFR outcome (39.8% of all patients studied) if FAME criteria are utilized.    FFR measurements are widely used in contemporary PCI practice for the assessment of the hemodynamic significance of coronary stenoses, guiding interventional strategy and providing prognostic information. FFR can be simply and rapidly determined just before the planned intervention, and a value of 2,5 In the recent multicenter FAME study utilizing an FFR cut-off value of 0.80, routine measurement of FFR in patients with multivessel coronary artery disease undergoing PCI significantly reduced the rate of the composite endpoint of death, nonfatal MI and repeat revascularization at 1 year.    Similarly, the DEFER study has shown that PCI of a functionally nonsignificant stenotic lesion, as assessed by an FFR value ≥ 0.75, does not improve event-free survival compared to medical management alone.¹³ Furthermore, the risk that such a hemodynamically nonsignificant stenosis will cause death or acute MI is 22    The fraction of maximal flow to the myocardium in the presence of a stenosis (Qs) to the theoretical maximal flow in the same myocardial bed without a stenosis (Qn) defines myocardial FFR, i.e., FFR = Qs/Qn. Since measurements of flow in vivo are quite difficult to obtain, pressure measurements can be used as a surrogate for flow. This approximation utilizes Ohm’s law which states that Q = P/R. Based on this, FFR = [Ps/Rs] / [Pn/Rn], where Ps is the pressure and Rs is the resistance across a stenosed myocardial bed, and Pn is the pressure and Rn is the resistance across a normal myocardial bed. The purpose of achieving maximal hyperemia with adenosine is to minimize the contribution of resistance across the myocardial bed so that FFR approximates Ps/Pn.²³    Values of resting Pd/Pa in our series that have the highest positive and negative predictive values are at the lower and higher end of the range of Pd/Pa values recorded. Very low resting Pd/Pa values indicate that the pressure drop across the lesion is substantial and thus the additional contribution of minimizing Rs/Rn with maximal hyperemia to influence FFR outcome is negligible. Conversely, where the resting Pd/Pa is very high (indicative of a minimal pressure drop across the lesion), minimizing the contribution of Rs/Rn with maximal hyperemia would add little to the FFR outcome. For intermediate values of resting Pd/Pa (i.e., 0.88–0.95 when a cut-off value of FFR is taken as 0.80 as per FAME), the pressure drop across the lesion is intermediate, and thus the contribution of Rs/Rn to FFR outcome would be far more important. Under these circumstances, resting Pd/Pa would be of little use since there would be a need to achieve maximal hyperemia to minimize Rs/Rn pharmacologically in order to gain accurate FFR measurements.    The relation between resting Pd/Pa and FFR might be even more complex in patients with previous MI where the amount of nonviable tissue within the infarcted territory influences perfusion by the stenotic infarct-related artery. McClish et al²⁴ studied FFR values in stenotic arteries in recently infarcted myocardium in comparison with FFR values in angiographically matched control vessels supplying noninfarcted beds. In their study, no significant differences were observed in either resting Pd/Pa or FFR between patients with recent MI and angiographically matched controls without acute MI.    Likewise, in the cohort of patients with acute coronary syndromes in the present study, we have demonstrated a similar linear relationship between resting Pd/Pa and FFR, with a high correlation coefficient.    The findings of the current study are important for a number of reasons. Firstly, although we have not collected data related to significant complications/tolerability or symptoms associated with adenosine infusion, it is well documented that infusion of adenosine to assess FFR is frequently associated with significant patient discomfort that includes chest pain, dyspnea, flushing and bronchospasm resulting from activation of the A1, A2B, and A3 adenosine receptors in other vascular smooth muscle beds.^25,26 For example, in a large retrospective study in which 1,000 patients underwent pharmacological stress radionuclide perfusion imaging with adenosine, such adverse effects were documented in 78% of the study population.²⁵ In the prospective Adenoscan Multicenter Trial registry including 9,256 patients in whom adenosine was infused at the same rate as in our study (140 μg/kg/min), minor side effects were reported in 81% of the participants, with those over the age of 70 having a 9.5% risk of developing significant atrioventricular (AV) block.²⁶ Other studies have demonstrated that close to 20% of patients undergoing myocardial perfusion assessment describe symptoms associated with adenosine infusion as very uncomfortable or extremely uncomfortable, with 79% experiencing significant symptoms.²⁷ Even with intracoronary application of adenosine to minimize these effects, care must be taken in those with higher degrees of AV block, particularly when assessing the right coronary artery. Furthermore intracoronary application of adenosine to achieve maximal hyperemia limits the ability to perform pullback assessment of lesions studied. The potential for patient distress and adverse symptoms should not be underestimated following adenosine infusion, and our findings may be even more relevant in situations where adenosine is relatively contraindicated such as in patients with reversible airway limitation or conduction abnormalities.    Secondly, apart from potential benefits in terms of cost savings and reductions in radiation dose and procedural time, such a strategy would simplify and render more widely “acceptable” findings from the recent FAME study that mandated the assessment of all lesions of at least 50% diameter stenosis on angiography. This would be particularly relevant in multivessel coronary artery disease, where a simple “rule-out” test based on resting Pd/Pa values with high negative predictive values would shorten procedure times and reduce radiation doses significantly. Furthermore, multiple pressure-wire assessments would be made more acceptable to patients with multivessel coronary artery disease due to the avoidance of multiple episodes of chest pain, dyspnea and flushing associated with each adenosine infusion used to assess each coronary artery individually. This would particularly be the case in the 20% of patients who describe symptoms associated with adenosine infusion as very uncomfortable or extremely uncomfortable.²⁷    Thirdly, the ability to predict a result of an investigation is useful in itself; it alerts the operator to an unusual result that may prompt a repeat measurement. For example, pressure-wire drift is not uncommon during a pressure-wire study, and, particularly in the case of borderline FFRs, an unexpected result (FFR) in relation to an initial Pd/Pa reading may prompt the interventionist to repeat the FFR measurement (re-zero, re-equilibrate, etc).    Study limitations. Although our study cohort of 483 patients is large (e.g., similar in size to the FAME study, n = 509 patients and larger than the DEFER study, n = 325 patients), it is subject to all the recognized limitations of a retrospective study. For example, as a single-center study, our findings might not be applicable to other study populations, although our cohort is likely to be representative of a contemporary PCI patient population in the United Kingdom. Consequently, our findings would need to be applied in a prospective manner to other cohorts undergoing pressure-wire assessments to evaluate their applicability to other patient groups before we could recommend their use in clinical practice. Finally, even if our data were validated in a prospective study, adenosine infusion would still be required to assess FFR in 50% of pressure-wire cases if FAME criteria were used since there is a substantial gray zone in intermediate resting Pd/Pa values (0.88–0.95), which do not accurately predict FFR.

Conclusion

   In summary, we have demonstrated a strong correlation between resting Pd/Pa and FFR in a contemporary PCI cohort undergoing pressure-wire measurements. We have also shown that resting values of Pd/Pa can be used to predict a positive FFR result with relatively high positive and negative predictive values (in the region of 95%). If our findings were proven to be applicable to other cohorts in future prospective studies, this may potentially obviate the need for adenosine infusion in at least 50% of patients undergoing pressure-wire assessments if FAME criteria are used. Our findings are of particular relevance in patients with a relative contraindication to adenosine or those with multivessel coronary artery disease.

_________________________________________________ From *Manchester Heart Centre, Manchester Royal Infirmary, Biomedical Research Centre, Oxford Road, Manchester, United Kingdom, and §Manchester Academic Health Science Centre, University of Manchester, Oxford Road, Man- chester, United Kingdom. The authors report no financial relationships or conflicts of interest regarding the content herein. Manuscript submitted November 30, 2009, provisional acceptance given January 4, 2010, final version accepted January 12, 2010. Address for correspondence: Dr. Magdi El-Omar, Manchester Heart Centre, Manchester Royal Infirmary, Manchester, M13 9PT, United Kingdom. E-mail: Magdi.El-Omar@cmft.nhs.uk

1. Pijls NH, Van Gelder B, Van der Voort P et al. Fractional flow reserve: A useful index to evaluate the influence of an epicardial coronary stenosis on myocardial blood flow. Circulation 1995;92:3183–3193.
2. Pijls NH, De Bruyne B, Peels K, et al. Measurement of fractional flow reserve to assess the functional severity of coronary artery stenoses. N Engl J Med 1996;334:2401–2406.
3. Caymaz O, Fak AS, Tezcan H, et al. Correlation of myocardial fractional flow reserve with thallium-201 SPECT imaging in intermediate-severity coronary artery lesions. J Invasive Cardiol 2000;12:345–350.
4. Leesar MA, Abdul-Baki T, Akkus NI, et al. Use of fractional flow reserve versus stress perfusion scintigraphy after unstable angina. Effect on duration of hospitalization, cost, procedural characteristics, and clinical outcome. J Am Coll Cardiol 2003;41:1115–1121.
5. Pijls NH. Optimum guidance of complex PCI by coronary pressure measurement. Heart 2004;90:1085–1093.
6. De Bruyne B, Pijls NHJ, Bartunek J, et al. Fractional flow reserve in patients with prior myocardial infarction. Circulation 2001;104:157–162.
7. Berger A, Botman KJ, MacCarthy PA, et al. Long-term clinical outcome after frac- tional flow reserve-guided percutaneous coronary intervention in patients with mul- tivessel disease. J Am Coll Cardiol 2005;46:438–442.
8. Jasti V, Ivan E, Yalamanchili V, et al. Correlations between fractional flow reserve and intravascular ultrasound in patients with an ambiguous left main coronary artery stenosis. Circulation 2004;110:2831–2836.
9. Aqel R, Zoghbi GJ, Hage F, et al. Hemodynamic evaluation of coronary artery bypass graft lesions using fractional flow reserve. Catheter Cardiovasc Interv 2008;72:479–485.
10. Blows LJ, Redwood SR. The pressure wire in practice. Heart 2007;93:419–422.
11. Pijls NH, De Bruyne B. In: Coronary Pressure, 2nd Edition. Fractional flow reserve from coronary pressure to coronary flow. The Netherlands: Kluwer Academic Pub- lishers. 2000, pp. 51–82.
12. de Bruyne B, Pijls NH, Barbato E, et al. Intracoronary and intravenous adenosine 5’-triphosphate, adenosine, papaverine, and contrast medium to assess fractional flow reserve in humans. Circulation 2003;107:1877–1883.
13. Pijls NH, van Schaardenburgh P, Manoharan G et al. Percutaneous coronary intervention of functionally nonsignificant stenosis: 5-year follow-up of the DEFER Study. J Am Coll Cardiol. 2007;49:2105–2111.
14. Tonino PA, De Bruyne B, Pijls NH, et al; FAME Study Investigators. Fractional flow reserve versus angiography for guiding percutaneous coronary intervention. N Engl J Med 2009;360:213–224.
15. Keller T, Zeller T, Peetz D, et al. Sensitive troponin I assay in early diagnosis of acute myocardial infarction. N Engl J Med 2009;361:868–877.
16. Reichlin T, Hochholzer W, Bassetti S, et al. Early diagnosis of myocardial infarction with sensitive cardiac troponin assays. N Engl J Med 2009;361:858–867.
17. Mueller C, Scholer A, Laule-Kilian K, et al. Use of B-type natriuretic peptide in the evaluation and management of acute dyspnea. N Engl J Med 2004;350:647–654.
18. Ng LL, Loke IW, Davies JE, et al. Community screening for left ventricular systolic dysfunction using plasma and urinary natriuretic peptides. J Am Coll Cardiol 2005;45:1043–1050.
19. Hobbs FD, Davis RC, Roalfe AK, et al. Reliability of N-terminal proBNP assay in diagnosis of left ventricular systolic dysfunction within representative and high risk populations. Heart 2004;90:866–870.
20. Hu Y, Liu W, Chen Y et al. Combined use of fasting plasma glucose and glycated hemoglobin A1c in the screening of diabetes and impaired glucose tolerance. Acta Diabetol 2009, Sept. 17 [Epub ahead of print].
21. Sato KK, Hayashi T, Harita N et al. Combined measurement of fasting plasma glu- cose and A1C is effective for the prediction of type 2 diabetes: The Kansai Healthcare Study. Diabetes Care 2009;32:644–646.
22. Kushner FG, Hand M, Smith SC Jr, et al. 2009 Focused Updates: ACC/AHA Guide- lines for the Management of Patients With ST-Elevation Myocardial Infarction (Up- dating the 2004 Guideline and 2007 Focused Update) and ACC/AHA/SCAI Guidelines on Percutaneous Coronary Intervention (Updating the 2005 Guideline and 2007 Focused Update). A Report of the American College of Cardiology Foun- dation/American Heart Association Task Force on Practice Guidelines. Circulation 2009;120:2271–2306. Epub 2009 Nov 18.
23. Hau WK. Fractional flow reserve and complex coronary pathologic conditions. Eur Heart J 2004;25:723–727.
24. McClish JC, Ragosta M, Powers ER et al. Effect of acute myocardial infarction on the utility of fractional flow reserve for the physiologic assessment of the severity of coronary artery narrowing. Am J Cardiol 2004;93:1102–1106.
25. Johnston DL, Daley JR, Hodge DO, et al. Hemodynamic responses and adverse ef- fects associated with adenosine and dipyridamole pharmacologic stress testing: A comparison in 2,000 patients. Mayo Clin Proc 1995;70:331–336.
26. Cerqueira MD, Verani MS, Schwaiger M, et al. Safety profile of adenosine stress per- fusion imaging: Results from the Adenoscan Multicenter Trial Registry. J Am Coll Cardiol 1994;23:384–389.
27. Iskandrian AE, Bateman TM, Belardinelli L, et al. for the ADVANCE MPI Investi- gators. Adenosine versus regadenoson comparative evaluation in myocardial perfusion imaging: Results of the ADVANCE phase 3 multicenter international trial. J Nucl Cardiol 2007;14:645–658.