Coronary Pressure Measurement After Stenting Predicts Adverse Events at Follow-Up
A Multicenter Registry
Background— Coronary stenting is associated with a restenosis rate of 15% to 20% at 6-month follow-up, despite optimum angiographic stent implantation. In this multicenter registry, we investigated the relation between optimum physiological stent implantation as assessed by poststent fractional flow reserve (FFR) and outcome at 6 months.
Methods and Results— In 750 patients, coronary pressure measurement at maximum hyperemia was performed after angiographically apparently satisfactory stent implantation. Poststenting FFR was calculated and related to major adverse events (including need for repeat target vessel revascularization) at 6 months. In 76 patients (10.2%), at least 1 adverse event occurred. Five patients died, 19 experienced myocardial infarction, and 52 underwent at least 1 repeat target vessel revascularization. By multivariate analysis, FFR immediately after stenting was the most significant independent variable related to all types of events. In 36% of the patients, FFR normalized (>0.95), and event rate was 4.9% in that group. In 32% of the patients, poststent FFR was between 0.90 and 0.95, and event rate was 6.2%. In 32% of patients, poststent FFR was <0.90, and event rate was 20.3%. In 6% of the patients, FFR was <0.80, and event rate was 29.5% (P<0.001).
Conclusions— FFR after stenting is a strong independent predictor of outcome at 6 months.
Received April 1, 2002; revision received April 22, 2002; accepted April 22, 2002.
Although coronary stenting has significantly reduced the rate of restenosis after percutaneous coronary intervention (PCI), ≈15% to 20% of stented patients require repeat target vessel revascularization (TVR) within 6 months after intervention.1–3⇓⇓
Excessive intimal hyperplasia is a major contributor to restenosis, but inadequate stent deployment or plaque dislocation to adjacent coronary segments resulting in abnormal shear stress also most likely contributes to in-stent restenosis and is often not detected by angiography.4–8⇓⇓⇓⇓ Therefore, additional methods beyond angiography are necessary to assess immediate stent result and evaluate adjacent vessel segments.
Coronary pressure measurement and determination of fractional flow reserve (FFR) have been proposed as relatively new techniques for optimizing PCI.6–9⇓⇓⇓ Abnormal FFR identifies both residual hyperemic pressure gradient and abnormal resistance across the stented segment and in the adjacent parts of the vessel.
Normal epicardial coronary arteries provide no resistance to blood flow, not even during maximum hyperemia.10 Therefore, optimum coronary stenting should at least result in the disappearance of any hyperemic pressure drop within the respective coronary segment.
Indeed, the elimination of any transstenotic hyperemic gradient after stenting (ie, restoring normal coronary conductance) correlates well with optimal stent deployment on the basis of intravascular ultrasound (IVUS) criteria.6,9⇓ However, before routinely incorporating coronary pressure measurement as an adjunctive to angiography after stent implantation, it is necessary to demonstrate that such methodology has prognostic significance for the patient in predicting outcome during follow-up. Accordingly, this registry was designed to investigate the correlation of FFR measured after stent implantation, with major adverse cardiac events, including the need for TVR, during the 6 months after implantation.
The registry was performed in 750 patients in 15 hospitals in 8 countries (5 centers in the United States, 5 centers in Europe, and 5 centers in Asia).
Between January 2000 and April 2001, all patients undergoing coronary stenting were enrolled in this registry, in which a pressure wire was used as a guidewire for any reason (Pressure Wire, Radi Medical Systems). There were no exclusion criteria. The reason for the use of the pressure wire was either the presence of an intermediate lesion (whether or not the target lesion) or multiple stenoses (of which it was not clear beforehand which one should be dilated) or routine use of the pressure wire.
Interventional Procedure and Measurement of FFR
The interventional procedures were performed completely according to local routine, and the operator was free to use any interventional technique, provided a stent was placed at the end of the procedure.
Immediately after angiographically apparently successful stent implantation (defined as delivery in the target lesion with a residual diameter stenosis <10% by visual estimation), coronary pressure measurement at maximum hyperemia was performed and FFR was determined by the ratio Pd/Pa, where Pd represents mean hyperemic coronary pressure distal to the stented segment measured by the pressure wire, and Pa represents mean aortic pressure measured by the guiding catheter.11,12⇓ Maximum hyperemia was induced by either intracoronary adenosine or ATP (≥30 μg for right coronary artery [RCA]; ≥40 μg for left coronary artery [LCA]), by intravenous adenosine or ATP infusion at a rate of 140 μg/kg per min, or by intracoronary papaverine administration (15 mg for RCA, 20 mg for LCA), as described previously.13
For the purpose of this registry, only the final value of FFR was taken into account and correlated to adverse events and repeat TVR at follow-up. Medical treatment during the procedure itself and during follow-up, including antithrombotic treatment, was according to the local routine in all participating centers. Quantitative coronary angiography was performed offline.
Analysis of Data
Baseline, angiographic, and functional data of all patients were related to occurrence of adverse events at 6-month follow-up. Adverse events were defined as death from any cause, acute myocardial infarction (AMI), or repeat TVR by bypass surgery or PCI. In case of more events in 1 patient, only the first event was counted for analysis. If more than 1 stent was used within 1 coronary segment, the analysis was performed for the complete segment. The analysis was performed in 2 steps. First, the crude FFR-specific risks for any adverse event were determined with 95% CIs. Second, stepwise multiple logistic regression analysis was performed to identify additional independent prognostic factors and to adjust the FFR-outcome relation for potential confounding. FFR was used as a variable with 5 categories (0.75 to 0.80, 0.81 to 0.85, 0.86 to 0.90, 0.91 to 0.95, and 0.96 to 1.00). In addition, sensitivity analyses were performed with FFR as a dichotomous variable (>0.95 versus ≤0.95 and >0.90 versus ≤0.90). Student’s t test, Mann-Whitney U test, and χ2 test were used for univariate tests, where appropriate. For the 5-category tests, ANOVA was used. Crude and adjusted ORs from logistic regression were used to compare univariate and multivariate results. For all analyses, a P<0.05 was considered statistically significant.
Baseline Characteristics and Procedural Results
Baseline characteristics are listed in Table 1. Age, sex, risk factors, and extent of anatomic disease were not different from the average percutaneous transluminal coronary angioplasty population in other large studies or from the total percutaneous transluminal coronary angioplasty population in the participating centers. An average of 1.18 stents were used per patient. In 5 patients (0.7%), FFR could not be measured after stenting because of technical failures. No complications attributable to the pressure measurement occurred in any of the patients.
A FFR value >0.95 was achieved in 266 patients (36%), and a value of >0.90 in 507 patients (68%). In 11 patients (1.5%), FFR was <0.75, which indicates that the procedure, in spite of a satisfactory angiographic result, was definitely unsuccessful from a functional point of view.11,12⇓ These patients were included in the 0.75 to 0.80 group in the analysis.
Events at Follow-Up
Complete follow-up was obtained in 744 patients (99.2%). A total of 90 events occurred in 76 patients, as follows: 5 deaths (0.7%), 19 myocardial infarctions (2.6%), 12 bypass surgeries (1.6%), and 54 repeated PCIs (7.3%) (Table 2). In 14 patients, >1 TVR was necessary. At univariate analysis, no differences were present between patients with or without an adverse event except FFR after stenting (P<0.001), stent diameter (P=0.023), and length of stent (P=0.032). Angiographic results (percentage residual stenosis and minimal luminal diameter) were not different between patients with or without an event (Table 1). The distribution of events for each of the 5 different FFR groups is presented in Table 2. A significant inverse correlation was found between final FFR and occurrence of adverse events during follow-up (P<0.001). This correlation was not only significant for all adverse events and for repeat TVR (P<0.001) but also for the end point death or AMI (P<0.01). In patients with FFR >0.95 (normal reference value of FFR in earlier studies),9,10⇓ the event rate was 4.9%, whereas in the lowest FFR group (FFR ≤0.80), event rate was 29.5% (Figure). This was in contrast to the angiographic residual stenosis percentage, which was not different between the 5 FFR categories except for patients with FFR <0.80, who had a slightly worse final angiographic result (percent residual stenosis and minimal luminal diameter) compared with the other 4 groups (P<0.01, Figure). Similarly, none of the other baseline or procedural characteristics was significantly associated with FFR category.
At subsequent multivariate analysis, the only independent variables predicting adverse events were FFR category (P<0.001) and length of stent (P<0.01). Table 3 compares the crude OR (2.83; CI 1.53 to 5.26) and the stent length–adjusted OR (2.78; CI 1.49 to 5.17) of adverse events for FFR >0.95. Crude and adjusted ORs differ by <2%, indicating that length of stent is not a confounding factor for the association of FFR and adverse event.14 Similarly, Table 4 demonstrates that the crude and adjusted ORs for the 5 FFR categories are similar; the mean change in the FFR-specific OR is <2%.
This study shows that measurement of coronary pressure and FFR immediately after stent implantation has a strong predictive value with respect to death, myocardial infarction, or need for repeat revascularization of the target vessel within 6 months. The higher the FFR, the lower the event rate. This predictive value of FFR is independent of any other baseline or procedural parameter and is present for both death and AMI and for repeated TVR.
Earlier studies have shown that coronary pressure measurement has a high specificity compared with IVUS for adequate stent implantation.6,9⇓ Therefore, it is well accepted that the absence of a hyperemic residual pressure gradient is a prerequisite for optimum stent deployment.4 The present study extends these data by demonstrating that the presence and magnitude of a residual hyperemic pressure gradient across the stented segment are directly correlated to a higher event rate during follow-up, irrespective of other variables.
There are several potential explanations for this strong relation between FFR after stenting and 6-month outcome. First, it is well known from studies using IVUS that angiography is a rather inaccurate tool for evaluating the result of stenting.4–7⇓⇓⇓ Incomplete deployment, protruding struts, or dislocation of plaque at the entry or exit of the stent is often not recognized by angiography but may result in early in-stent restenosis.4–8⇓⇓⇓⇓ Such abnormalities can often be detected by the persistence of a hyperemic pressure gradient across the stented segment.6 Therefore, it is likely that several patients with restenosis in fact had suboptimal stent deployment not detected by angiography. Such patients most likely would also have been recognized by IVUS.4,5⇓
Second, there is increasing evidence that abnormal shear stress plays an important role in neointimal growth and restenosis.8 Presence of a persistent pressure gradient across the stented segment indicates an abnormal blood flow pattern with heterogeneous and abnormal low shear stress. Therefore, despite similar anatomic results, the pressure measurement might identify patients with increased risk of restenosis. Such patients will not be discovered by using IVUS.
Third, coronary pressure measurement not only indicates an abnormal conductance within the stented segment but also reveals other overt or occult disease in the remaining part of the coronary artery, contributing to decreased FFR.10,15⇓ Therefore, although not detected by angiography, diffuse disease may have been more common in the groups with lower FFR and have accounted in part for the higher adverse event rate.
In evaluating coronary stenting by pressure measurement, some important issues should be taken into account. First, because only small gradients are often present after stenting, it is of paramount importance to ensure that the FFR measurement is performed at true maximum hyperemia. In some earlier studies, inadequate doses of intracoronary adenosine were used, and FFR may have been overestimated.9 It is recommended to use dosages of at least 30 μg for the RCA and 40 μg for the LCA.13
Second, as mentioned above, it should be distinguished if a persistent hyperemic gradient after stenting is attributable to incomplete stent deployment or abnormalities within or adjacent to the stent or to disease more proximal in the vessel. Such disease is often not apparent before the procedure but may result in a significant hyperemic pressure drop after blood flow in the artery has increased by stenting the most severe stenosis.15 Therefore, for complete analysis of the coronary artery after the intervention, it is preferable to perform a pressure pull-back curve during sustained hyperemia, induced by intravenous adenosine or ATP. This pull-back pressure recording along the length of the epicardial coronary artery indicates the conductance of the entire artery as well as of every individual segment in mm Hg per unit of length. It is interesting, but as yet unknown, whether the in-stent restenosis rate is higher in the setting of optimum stenting but a persisting gradient within the remaining nonstented part of the vessel.
Finally, as shown in this study, a persistent hyperemic gradient across the stent itself indicates a suboptimal result but does not reveal its mechanism. Intravascular ultrasound may elucidate the cause of the persisting gradient within the stent in such cases.4–7⇓⇓⇓
The present study is a registry with all its inherent limitations. A selection bias could have been introduced, because the pressure wire is often used in patients with intermediate stenosis and less plaque burden, with a possibly better outcome after stenting. On the other hand, the average stenosis severity in this study was 75±14% and not different between patients with or without event (Table 1). Moreover, although the criterion for angiographic successful stent implantation in this study was a residual stenosis <10% by visual estimation (reflecting the common practice in most hospitals), in some patients a suboptimal FFR value might have triggered additional dilatations, leading to better results than in an average population. Finally, although it is clear that a persisting gradient is associated with adverse events and need for repeat revascularization, this study did not address whether additional action in the case of suboptimal FFR results in a decreased adverse event rate. Additional prospective studies are needed to address this question.
Recognizing these limitations, this study shows that coronary pressure measurement is an easy, rapid, and relatively cheap method to evaluate stent implantation and to predict occurrence of adverse events within 6 months of follow-up.
The following investigators and centers participated in the study: E. Powers, H. Samady, M. Ragosta, L. Gimple, I. Sarembock, L. Snyder, University of Virginia Charlottesville, Va; P. Yock, A. Yeung, P. Fitzgerald, W. Fearon, Stanford University, Stanford, Calif; S. Higano, B. Lehrman, Mayo Clinic, Rochester Minn; M. Kern, J. Muller, St Louis University, St Louis, Mo; N. Deeb, D. Jones, Arizona Heart Center, Phoenix, Ariz; G. Heyndrickx, W. Wijns, F. Staelens, B. De Bruyne, OLV Cardiovascular Center Aalst, Belgium; J. Bonnier, H. Michels, J. Koolen, C. Peels, C.J. Botman, N. Pijls, Catharina Hospital, Eindhoven, the Netherlands; V. Klauss, J. Rieber, P. Erdin, Universität Klinikum Innenstadt, München, Germany; J. Escaned, C. Macaya, A. Flores, St Carlos Hospital, Madrid, Spain; L. Marques, V. Matos, Coimbra, Portugal; K. Takazawa, N. Tanaka, Tokyo Medical University, Tokyo, Japan; Y. Tsurumi, Tokyo Women’s Medical University, Tokyo, Japan; T. Akasaka, Y. Neishi, Kawasaki University, Okayama, Japan; K. Yamabe, K. Tamita, Kobe General Hospital, Kobe, Japan; and Seung-jea Tahk, Ajou University School of Medicine, Korea.
This study was supported by an unrestricted grant from Radi Medical Systems, Uppsala, Sweden. The authors are indebted to Anne Hol for her assistance in preparing the manuscript.
This article originally appeared Online on May 28, 2002 (Circulation. 2002;105:r126–r130).
- ↵Koning R, Eltchaninoff H, Commeau P, et al. Stent placement compared with balloon angioplasty in small coronary arteries. Circulation. 2001; 104: 1604–1608.
- ↵Mudra H, Di Mario C, De Jaegere P, et al. Randomized comparison of coronary stent implantation under ultrasound or angiographic guidance to reduce stent restenosis (OPTICUS study). Circulation. 2001; 104: 1343–1349.
- ↵Fitzgerald PJ, Oshima A, Hayase M, et al. Final results of the Can Routine Ultrasound Influence Stent Expansion (CRUISE) study. Circulation. 2000; 102: 523–530.
- ↵Hanekamp C, Koolen JJ, Pijls NHJ, et al. Comparison of quantitative coronary angiography, intravascular ultrasound and pressure-derived fractional flow reserve to assess optimal stent deployment. Circulation. 1999; 99: 1015–1021.
- ↵Bermejo J, Botas J, Garaia E, et al. Mechanisms of residual lumen stenosis after high-pressure stent implantation. Circulation. 1998; 98: 112–118.
- ↵Wentzel J, Krams R, Schuurbiers JCH, et al. Relationship between neointimal thickness and shear stress after wallstent implantation in human coronary arteries. Circulation. 2001; 103: 1740–1745.
- ↵Fearon WF, Luna J, Samady H, et al. Fractional flow reserve compared with intravascular ultrasound guidance for optimizing stent deployment. Circulation. 2001; 104: 1917–1922.
- ↵De Bruyne B, Hersbach F, Pijls NHJ, et al. Abnormal epicardial coronary resistance in patients with diffuse atherosclerosis but “normal” coronary angiography. Circulation. 2001; 104: 2401–2406.
- ↵De Bruyne B, Pijls NHJ, Bartunek J, et al. Fractional flow reserve in patients with prior myocardial infarction. Circulation. 2001; 104: 157–162.
- ↵Pijls NHJ, De Bruyne B. Maximum hyperemic stimuli.In: Coronary Pressure, 2nd ed. Dordrecht, the Netherlands: Kluwer Academic Publishers; 2000: 97–107.
- ↵Pijls NHJ, De Bruyne B, Bech GJW, et al. Coronary pressure measurement to assess the hemodynamic significance of serial stenoses within one coronary artery: validation in humans. Circulation. 2000; 102: 2371–2377.