You are here

Article Text

Article menu
Valvular heart disease
Original research
Coronary microvascular dysfunction in patients undergoing transcatheter aortic valve implantation
  1. Roberto Scarsini1,2,
  2. Leonardo Portolan1,
  3. Francesco Della Mora1,
  4. Margherita Fabroni1,
  5. Stefano Andreaggi1,
  6. Andrea Mainardi1,
  7. Paolo Springhetti1,
  8. Alberto Dotto1,
  9. Paolo Alberto Del Sole1,
  10. Simone Fezzi1,
  11. Sara Pazzi1,
  12. Domenico Tavella1,2,
  13. Concetta Mammone1,2,
  14. Mattia Lunardi1,2,
  15. Gabriele Pesarini1,2,
  16. Giovanni Benfari1,
  17. Flavio Luciano Ribichini1,2
  1. 1Department of Medicine, Division of Cardiology, University of Verona, Verona, Italy
  2. 2Interventional Cardiology Unit, Azienda Ospedaliera Universitaria Integrata Verona, Verona, Italy
  1. Correspondence to Dr Roberto Scarsini; scarsini.roberto{at}gmail.com

Abstract

Objectives This study aimed to evaluate the prognostic value of coronary microvascular dysfunction (CMD) at long term after transcatheter aortic valve implantation (TAVI) and to explore its relationship with extravalvular cardiac damage (EVCD). Moreover, we sought to test the correlation between angiography-derived index of microcirculatory resistance (IMRangio) and invasive IMR in patients with aortic stenosis (AS).

Methods This was a retrospective analysis of the Verona Valvular Heart Disease Registry (Italy) including 250 patients (83 (80–86) years, 53% female) with severe AS who underwent TAVI between 2019 and 2021. IMRangio was calculated offline using a computational flow model applied to coronary angiography obtained during the TAVI workup. CMD was defined as IMRangio ≥30 units.

The primary endpoint was the composite of cardiovascular death and rehospitalisation for heart failure (HF). Advanced EVCD was defined as pulmonary circulation impairment, severe tricuspid regurgitation or right ventricular dysfunction.

The correlation between IMR and IMRangio was prospectively assessed in 31 patients undergoing TAVI.

Results The primary endpoint occurred in 28 (11.2%) patients at a median follow-up of 22 (IQR 12–30) months. Patients with CMD met the primary endpoint more frequently than those without CMD (22.9% vs 2.8%, p<0.0001). Patients with CMD were more frequently characterised by advanced EVCD (33 (31.4%) vs 27 (18.6%), p=0.024). CMD was an independent predictor of adverse outcomes (adjusted HR 6.672 (2.251 to 19.778), p=0.001) and provided incremental prognostic value compared with conventional clinical and imaging variables. IMRangio demonstrated fair correlation with IMR.

Conclusions CMD is an independent predictor of cardiovascular mortality and HF after TAVI.

  • Aortic Valve Stenosis
  • Coronary Artery Disease
  • Coronary Angiography
  • Transcatheter Aortic Valve Replacement

Data availability statement

Data are available upon reasonable request. Data are available upon reasonable request to the corresponding author.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/heartjnl-2023-323461

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • Coronary microvascular dysfunction was previously described in aortic stenosis but its relationship with extravalvular cardiac damage and its prognostic impact remain unknown.

    WHAT THIS STUDY ADDS

    • Coronary microvascular dysfunction is associated with advanced extravalvular cardiac damage, and it is a strong predictor of cardiovascular mortality and heart failure at long-term follow-up after transcatheter aortic valve implantation (TAVI).

    • Angiography-derived index of microcirculatory resistance is a valuable alternative to invasive diagnostic tools to identify coronary microvascular dysfunction in patients with aortic stenosis.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • If confirmed by prospective dedicated multicentric studies, patients with coronary microvascular dysfunction and aortic stenosis undergoing TAVI may be considered for close monitoring for the high risk of heart failure.  

    Introduction

    Aortic stenosis (AS) leads to left ventricle pressure overload with compensatory changes (hypertrophy, fibrosis, reduced capillary density) that1–4 adversely affects blood flow in the coronary epicardial arteries and microcirculation. Indeed, coronary microvascular dysfunction (CMD) was described in AS2–11 and it was associated with adverse cardiac remodelling, left atrial dysfunction and myocardial fibrosis.6 10

    Recently, advanced extravalvular cardiac damage (EVCD), characterised by right ventricular dysfunction, pulmonary vasculature impairment and severe low-flow phenotype, was associated with adverse clinical outcomes.12 13 Moreover, preliminary data have suggested an association between low-flow phenotype and CMD.6 However, whether CMD is associated with EVCD and worse clinical outcome in patients with severe AS undergoing transcatheter aortic valve implantation (TAVI) remains undetermined.

    Angiography-derived index of microcirculatory resistance (IMRangio) has been recently developed and validated to assess CMD, but it was never tested in AS.14–16

    In this study, we aimed to assess if the presence of CMD was associated with EVCD and long-term adverse clinical outcomes in patients undergoing TAVI. Furthermore, we sought to test the correlation between IMRangio and invasive IMR in patients with severe AS undergoing TAVI.

    Methods

    Study design and population

    This single-centre observational study consisted of two phases. In the first phase, we tested the correlation between IMRangio and pressure/thermodilution wire-based IMR in a prospective cohort of patients with severe AS undergoing TAVI (derivation cohort). In the second phase, the long-term prognostic value of IMRangio-defined CMD was assessed in a larger cohort of patients treated with TAVI and prospectively enrolled in the Verona Valvular Heart Disease Registry (prognostic cohort) (online supplemental figures 1 and 2). The main exclusion criteria were previous coronary artery bypass graft, severe angiographic stenosis of the left anterior descending artery (LAD) (>70%), significant angiographic stenosis (>50%) of the left main coronary artery and recent (<30 days) acute coronary syndrome. Details about the enrolment and the exclusion criteria of the two cohorts are provided in the online supplemental methods 1.

    The clinical follow-up of patients included in the prognostic cohort was prospectively assessed with outpatient clinic visits or telephone contacts at 1, 6, 12, 24, 36 and 48 months after TAVI. Clinical events were adjudicated by dedicated research personnel blinded to the IMRangio data. Discordance was resolved by consensus.

    TAVI procedure and invasive assessment of coronary microvascular function

    TAVI was performed in all cases under conscious sedation and local anaesthesia with transfemoral access. Technical decisions were all left to the operator’s discretion. All patients underwent coronary angiography as part of the TAVI workup.

    In the derivation cohort, coronary microvascular function was invasively assessed as previously described.17 Briefly, a pressure and temperature-sensor guidewire (PressureWire X, Abbott, Santa Clara, California, USA) was advanced in the distal third of the LAD after careful equalisation of distal pressure and aortic pressure at the tip of the guiding catheter. Coronary pressures and mean transit times were measured and analysed with the Coroflow software (Coroventis, Uppsala, Sweden) during rest and steady-state hyperaemia obtained with the intravenous infusion of adenosine (140 µg/kg/min) (figure 1). IMR was calculated as:17

    Figure 1
    Figure 1

    Wire-based and angiography-derived CMD assessment in the derivation cohort. Clinical example of wire-based IMR and IMRangio assessment. IMRangio is obtained using standard coronary angiography as the product of the measured mean aortic pressure, QFR and the ratio between the number of frames required for the dye to reach the distal landmark of the coronary vessel and the acquisition frame rate. CMD, coronary microvascular dysfunction; IMR, index of microcirculatory resistance; IMRangio, angiography-derived index of microcirculatory resistance; QFR, quantitative flow ratio; TAVI, transcatheter aortic valve implantation.

    Embedded Image

    IMR >25 units was considered abnormal as previously described.17 18

    Angiography-derived index of microcirculatory resistance

    IMRangio was assessed offline by expert quantitative flow ratio (QFR)-certified operators blinded to imaging data and clinical outcome (FDM, SA, MF) using standard coronary angiographic views obtained during TAVI workup.14 15 Dedicated view angles according to the target coronary vessel were selected (online supplemental table 1). Coronary angiograms were acquired at 15 frames/s. Contrast media injection was performed in all patients using a standard contrast delivery system (ACIST CVi). Briefly, two angiographic views of the LAD were analysed using the QAngio XA three-dimensional (3D) software (Medis, Leiden, the Netherlands) to obtain the 3D vessel reconstruction and the QFR as a surrogate of the fractional flow reserve. The characterisation of coronary blood flow was based on the analysis of the number of frames required for contrast to reach the distal landmark of the vessel. IMRangio was subsequently derived using the following formula15:

    Embedded Image

    Where Pa is the mean aortic pressure at rest during the coronary angiography, and n frames are the number of angiographic frames for contrast dye to travel from the catheter to the distal part of the LAD at rest (figure 1 and online supplemental figure 3).

    The recommendations recently proposed by Mejía-Rentería et al were applied to maximise standardisation (online supplemental methods 2).19

    Conventional and speckle-tracking echocardiography

    During the TAVI workup, all patients underwent conventional transthoracic echocardiography using commercially available ultrasound systems (Epiq 7C, Philips), and parameters were collected accordingly to international guidelines.1 20

    Data were saved digitally and subsequently analysed offline using the TomTec software by two independent expert operators (PS, AD) blinded to the data on microvascular assessment and clinical outcome. Advanced speckle-tracking echocardiography was performed with TomTec Autostrain software for the assessment of global longitudinal strain (GLS) and peak atrial longitudinal strain (PALS) as recommended.21–23

    EVCD assessment

    The extent of EVCD was previously categorised into five stages according to a model described by Genereux et al.12 13 To evaluate the interaction between CMD and EVCD and increase the statistical power, cardiac damage was dichotomised into stages 0–2 (group 1) corresponding to isolated left heart dysfunction, and stage 3 or 4 (group 2), corresponding to damage extending to the pulmonary circulation and right heart (advanced EVCD).24 Details about the evaluation of EVCD are provided in online supplemental methods 3.

    Study endpoints

    The primary endpoint of the study was the composite of cardiovascular mortality and rehospitalisation for heart failure (HF) at long-term follow-up after TAVI.

    HF was defined accordingly to international guidelines25 as the evidence of central and/or peripheral congestion and/or peripheral hypoperfusion requiring hospitalisation for appropriate therapy. Cardiovascular death was defined as death attributable to HF, myocardial infarction, cerebrovascular accident or cardiac arrest because of other or unknown causes. Mortality was considered cardiovascular unless an alternative cause was documented.

    Secondary endpoints were the correlation between IMRangio and wire-based IMR and measures of EVCD.

    Statistical analysis

    Continuous variables were reported as median and IQR, while categorical variables were reported as numbers and percentages.

    Continuous variables were compared with the Mann-Whitney test, whereas the Fisher’s exact test was used to compare categorical variables.

    To assess interoperator reliability, intraclass correlation coefficients (ICC) estimates and their 95% CIs were calculated. Spearman correlation coefficients were calculated to assess the linear correlation between IMRangio and IMR, GLS and PALS using a non-parametric approach. The agreement between IMRangio and invasive IMR was assessed using Bland-Altman plot analysis.

    Areas under the curve (AUCs) of the receiver operating characteristic (ROC) curves were used to test the diagnostic accuracy of IMRangio in predicting CMD defined by IMR >25 units and in predicting the primary endpoint. The best cut-off was determined with the analysis of the ROC curve to maximise sensitivity and specificity.

    Survival analysis and endpoint comparison between groups were performed with the Cox regression analysis for the calculation of HR with 95% CI and the log-rank test. Survival analysis was adjusted for variables with p<0.10 at the univariable analysis. The test for proportional-hazards assumption was applied to confirm the validity of the model.26 Logistic regression analysis was performed to identify determinants of the primary endpoint at 36 months of follow-up. AUCs were calculated and compared with the DeLong’s method.27

    The continuous association of IMRangio with the primary endpoint was tested dividing the prognostic cohort in tertiles of IMRangio and in four different groups according to the presence or absence of IMRangio-defined CMD and/or advanced EVCD. Moreover, the analysis was conducted also in prespecified subgroups stratified by left ventricular ejection fraction (LVEF) <50%, presence of atrial fibrillation (AF) and older age (≥median age of the study cohort). Interaction analysis was conducted to assess the impact of given variables on the IMRangio-based prognostic stratification.

    All the tests were two tailed and a p<0.05 was considered significant. Statistical analysis was performed with SPSS V.26 (IBM) and Stata V.15.1 (StataCorp, College Station, Texas, USA).

    Patient and public involvement

    Patients or the public were not involved in the design, conduct, reporting or dissemination of our research.

    Results

    Correlation between invasive IMR and IMRangio

    Overall, paired invasive IMR and IMRangio measurements were available in 31 out of 42 patients with severe AS undergoing TAVI (online supplemental figures 1 and 2).

    Baseline clinical, echocardiographic and procedural characteristics of the derivation cohort are shown in online supplemental table 2. The correlation between IMRangio and IMR was moderate but significant (r=0.41, p=0.024). The agreement between the two indices is shown in figure 2. At ROC curve analysis, IMRangio showed an AUC of 0.81 (0.65–0.97, p=0.009) in predicting IMR >25 units.

    Figure 2
    Figure 2

    Correlation between IMRangio and wire-based IMR in patients with AS undergoing TAVI. Scatter plot and Bland-Altman analysis showing the correlation and agreement between IMRangio and invasive pressure wire-based IMR in our derivation cohort. AS, aortic stenosis; IMR, index of microcirculatory resistance; IMRangio, angiography-derived index of microcirculatory resistance; TAVI, transcatheter aortic valve implantation.

    Prognostic value of IMRangio in patients with AS undergoing TAVI

    Overall, 250 patients with severe AS undergoing TAVI were included in the prognostic cohort. Satisfactory ICC was observed for IMRangio assessment (0.86 (95% CI 0.69 to 0.94), p<0.0001). The primary endpoint occurred in 28 patients (11.2%) at a median follow-up time of 22 months (12–30 months) and it was mainly driven by rehospitalisation for HF (22 of 28 events, 78.6%). Patients who experienced the primary endpoint presented higher Society of Thoracic Surgeons (STS) score (2.941 (2.035–4.310) vs 2.335 (1.738–3.295), p=0.050), larger left atrial volume index (55 (49–65) vs 45 (38–55), p=0.001), higher rate of AF (67.9% vs 27.9%, p<0.0001) and advanced EVCD (15 (53.6%) vs 45 (20.3%), p<0.0001), with lower tricuspid annular plane systolic excursion (TAPSE), higher systolic pulmonary arterial pressure and left ventricular mass index, and higher rate of more than mild mitral and tricuspid regurgitation (table 1).

    View this table:
    Table 1

    Clinical, procedural data and CMD assessment in patients suffering or not the primary endpoint

    IMRangio was significantly higher in patients who met the primary endpoint (37 (30–44) vs 26 (20–36), p<0.0001), and overall, IMRangio demonstrated fair discriminatory power in predicting the composite endpoint (AUC 0.72 (0.63–0.80), p<0.0001) (figure 3). IMRangio ≥30 units demonstrated the best performance in predicting the primary endpoint (online supplemental table 3, figure 3 and online supplemental figure 4). CMD (defined as IMRangio ≥30 units) was present in 105 (42%) patients.

    Figure 3
    Figure 3

    IMRangio in patients who met the primary endpoint. IMRangio was significantly higher in patients who met the primary endpoint (left panel). IMRangio demonstrated fair diagnostic accuracy in predicting the primary endpoint at ROC curve analysis (right panel). The ROC-defined IMRangio cut-off ≥30 units (Youden index 0.49) demonstrated a sensitivity, specificity, NPV and PPV of 85.7%, 63.5%, 97.2% and 22.9%, respectively. IMRangio, angiography-derived index of microcirculatory resistance; NPV, negative predictive value; PPV, positive predictive value; ROC, receiver operating characteristics.

    Patients with CMD showed significantly worse clinical outcomes in terms of cardiovascular mortality and rehospitalisation for HF compared with patients with preserved microvascular function (22.9% vs 2.8%, p<0.0001) (figure 4).

    Figure 4
    Figure 4

    Survival analysis. Kaplan-Meier curves of patients with and without CMD defined by IMRangio ≥30 units (left panel) and of patients stratified by tertiles of IMRangio (right panel). CMD, coronary microvascular dysfunction; HF, heart failure; IMRangio, angiography-derived index of microcirculatory resistance.

    The continuous association of IMRangio with the primary endpoint was confirmed by dividing the prognostic cohort in tertiles of IMRangio. Notably, the number of adverse events was 2.4%, 10.6% and 20.5% in the first, second and third tertiles, respectively (p<0.0001) (figure 4).

    At multivariable Cox regression analysis, CMD (adjusted HR (aHR) 6.672 (2.251 to 19.778), p=0.001), AF (aHR 2.621 (1.105 to 6.217), p=0.029) and advanced EVCD (aHR 2.196 (0.999 to 4.828), p=0.050) were independent predictors of the primary endpoint (online supplemental tables 4 and 5).

    Sensitivity analyses excluding patients who underwent a percutaneous coronary intervention to the LAD (10.8% patients) (aHR 4.844 (1.614 to 14.537), p=0.005) and patients with QFR ≤0.80 (8.4% patients) (aHR 5.073 (1.693 to 15.202), p=0.004) confirmed the independent prognostic value of CMD (online supplemental figure 5).

    CMD and adverse cardiac remodelling in the prognostic cohort

    Clinical features of patients stratified by CMD are summarised in table 2. Briefly, patients with CMD showed a higher rate of AF (42.9% vs 24.8%, p=0.004), lower peak transvalvular velocity (4.0 (3.8–4.5) vs 4.2 (3.9–4.6), p=0.028), lower mean transvalvular gradient (40 (34–48) vs 44 (38–53), p=0.024), larger left atria (49 (40–58) vs 45 (37–55), p=0.042) and higher QFR (0.95 (0.90–0.99) vs 0.94 (0.87–0.97), p=0.002). Importantly, patients with CMD were more frequently characterised by an advanced EVCD (Genereux stages 3–4) (33 (31.4%) vs 27 (18.6%), p=0.024), with lower TAPSE and higher rates of more than mild mitral and tricuspid regurgitation (table 2).

    View this table:
    Table 2

    Clinical, procedural data and outcomes in patients with and without CMD — prognostic cohort

    Pre-TAVI speckle-tracking echocardiography was available in 125 (50%) patients. GLS was not significantly different in patients with and without CMD (−14.4% (−9.0 to −17.2) vs −14.0% (−10.0 to −16.7), p=0.993). Conversely, PALS was significantly lower in patients with CMD with borderline significance (14.5% (9.0–26.0) vs 18.7% (13.0–26.7), p=0.050) (online supplemental figure 6). Moreover, GLS was not significantly correlated with IMRangio (r=0.014, p=0.874), while PALS showed a significant inverse correlation with IMRangio (r=−0.191, p=0.033) (online supplemental figure 6).

    Prognostic value of EVCD in patients with AS undergoing TAVI

    Patients with advanced EVCD met the primary endpoint more frequently compared with the rest of the study cohort (25.0% vs 6.8%, p<0.0001). Importantly, the coexistence of advanced EVCD and CMD portended significantly worse clinical outcomes compared with patients with EVCD or CMD in isolation (p<0.0001) (figure 5).

    Figure 5
    Figure 5

    Survival analysis stratified by CMD and advanced EVCD. Patients with advanced EVCD (Genereux stages 3–4) suffered more cardiovascular death and rehospitalisation for heart failure (HF) 36 months after TAVI compared with patients with Genereux stages 0–2 (left panel). Patients with CMD and EVCD were characterised by a significantly worse prognosis compared with patients with advanced EVCD and no CMD (right panel). Similarly, patients with CMD and no advanced EVCD showed worse clinical outcomes compared with patients without CMD and no advanced EVCD. CMD, coronary microvascular dysfunction; EVCD, extravalvular cardiac damage; TAVI, transcatheter aortic valve implantation.

    Incremental prognostic value of CMD in predicting cardiac mortality and HF after TAVI

    The logistic regression model including age, AF, STS score, LVEF <50%, advanced EVCD (Genereux stages 3–4) and CMD (IMRangio ≥30 units) showed an AUC of 0.86 (0.79–0.92) in predicting the primary endpoint, demonstrating a modest but significant prognostic gain compared with models including only clinical variables (age, AF and STS; p for AUC comparison=0.025) or clinical and echocardiographic parameters (age, AF, STS, LVEF and advanced EVCD; p for AUC comparison=0.042) (figure 6).

    Figure 6
    Figure 6

    Incremental prognostic value of CMD in patients undergoing TAVI. The logistic regression model including CMD on top of imaging and clinical variables demonstrated the best accuracy in predicting cardiovascular death and rehospitalisation for HF after TAVI. aOR, adjusted OR; AUC, area under the curve; CMD, coronary microvascular dysfunction; EVCD, extravalvular cardiac damage; HF, heart failure; LVEF, left ventricular ejection fraction; STS, Society of Thoracic Surgeons; TAVI, transcatheter aortic valve implantation.

    Subgroup analysis

    IMRangio was an effective risk-stratification tool in all the prespecified subgroups including patients with LVEF <50% vs LVEF ≥50% (p for interaction=0.426), patients with and without AF (p for interaction=0.428) and patients aged <83 years vs ≥83 years (p for interaction=0.328) (online supplemental figure 7).

    Discussion

    In this study, we observed that IMRangio is a valuable alternative to wire-based IMR for the evaluation of CMD in patients undergoing TAVI.

    CMD defined as high IMRangio is associated with EVCD and adverse cardiac remodelling at non-invasive imaging. More importantly, CMD is an independent predictor of cardiovascular death and rehospitalisation for HF at a median follow-up time of 22 months after TAVI.

    To the best of our knowledge, this is the first study to establish a direct link between CMD and long-term adverse clinical outcomes in patients with severe AS undergoing TAVI. In a previous study,6 we demonstrated that, in patients with AS, CMD was associated with impaired left atrial function and with a chronic low-flow phenotype. Importantly, this was likely mediated via an association between high microvascular resistance, reduced vasodilatory capacity and adverse cardiac remodelling. Furthermore, patients with CMD were more frequently characterised as low-flow low-gradient AS, a phenotype notoriously associated with poor clinical outcomes.6

    In AS, the left ventricular response to the valve narrowing is initially adaptive but it becomes soon maladaptive with excessive left ventricular hypertrophy and concentric remodelling.4 CMD contributes to myocardial ischaemia which ultimately leads to further adverse cardiac remodelling, EVCD and symptoms.28

    Although the association between CMD, adverse cardiac remodelling and adverse events was expected, whether CMD was associated with EVCD and the risk of cardiac mortality and HF in patients undergoing TAVI was not determined.9 We demonstrated that the risk of adverse clinical outcomes at long-term follow-up after TAVI was continuously associated with impaired coronary microvascular function. This was mediated via adverse cardiac remodelling as demonstrated by the association between CMD, advanced EVCD, impaired atrial function and low-flow phenotype.6 Consistently, previous histopathological studies suggested a relationship between left ventricular fibrosis and lower values of PALS in patients with AS undergoing surgical aortic valve replacement.29 In addition, CMD has been previously correlated with the degree of left ventricular function recovery after TAVI beyond the acute effect of left ventricular unloading induced by the aortic valve replacement.6

    IMRangio-based risk stratification

    The assessment of CMD may contribute to refine the long-term risk stratification of patients undergoing TAVI. In particular, being based on standard coronary angiography performed during the workup for TAVI and obviating the need for coronary artery instrumentation, IMRangio may emerge as a practical tool to detect CMD overcoming most of the common limitations of pressure wire-based microvascular assessment.30

    Importantly, IMRangio demonstrated that a continuous risk stratification of the composite primary endpoint, as shown by the progressive higher rate of events of patients in the second and third tertiles of IMRangio, should be provided (figure 4). Moreover, although the Genereux staging retains its excellent prognostic value in patients with AS undergoing TAVI also in our cohort, IMRangio-defined CMD demonstrated further stratification of the risk of cardiovascular death and rehospitalisation for HF in patients characterised by advanced EVCD (Genereux stages 3–4) (figure 5). Furthermore, CMD demonstrated that the prognostic value in different subgroups of patients characterised by a baseline high risk of adverse events, such as low LVEF, AF and older age, should be maintained (online supplemental figure 7).

    Limitations

    Our study has limitations, including the need for external validation due to its retrospective, single-centre nature, making our initial findings hypothesis-generating. Moreover, the sample size of the derivation cohort was limited and it did not allow to define the best cut-off of IMRangio for detecting IMR >25 units. The absence of independent event adjudication is another limitation; however, the Verona Valvular Heart Disease Registry relies on trained researchers blinded to coronary physiology data for adjudication. In our series, a considerable proportion of patients did not demonstrate an adequate angiographic quality and were excluded from the analysis (online supplemental figure 2). Indeed, specific recommendations should be applied in order to increase the accuracy of IMRangio assessment, including a recommended acquisition rate of at least 15 frames/s.19

    Additionally, the diagnostic performance of IMRangio was compared with IMR but not with myocardial perfusion imaging which may be considered the gold standard for CMD evaluation in the whole left ventricle. In fact, we only assessed CMD in the LAD, which may limit generalisability, but given its role in myocardial perfusion and the absence of prior anterior myocardial infarction history, it is a reasonable proxy for the whole coronary microcirculation.

    Conclusions

    CMD, being associated with adverse cardiac remodelling and advanced EVCD, is a major determinant of cardiac mortality and rehospitalisation for HF after TAVI. IMRangio demonstrated a continuous association with adverse clinical outcomes after TAVI and showed incremental prognostic value compared with conventional clinical and echocardiographic risk stratification.

    Data availability statement

    Data are available upon reasonable request. Data are available upon reasonable request to the corresponding author.

    Ethics statements

    Patient consent for publication

    Ethics approval

    This study involves human participants and was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of the University of Verona (CESC 1701-1918). All patients provided their written informed consent.

    References

    1. ↵
      1. Vahanian A,
      2. Beyersdorf F,
      3. Praz F, et al
      . ESC/EACTS guidelines for the management of valvular heart disease. Eur Heart J 2022;43:561–632. doi:10.1093/eurheartj/ehab395
      OpenUrlCrossRefPubMed
    2. ↵
      1. McConkey HZR,
      2. Marber M,
      3. Chiribiri A, et al
      . Coronary Microcirculation in aortic stenosis. Circ Cardiovasc Interv 2019;12:e007547. doi:10.1161/CIRCINTERVENTIONS.118.007547
    3. ↵
      1. Michail M,
      2. Davies JE,
      3. Cameron JD, et al
      . Pathophysiological coronary and Microcirculatory flow alterations in aortic stenosis. Nat Rev Cardiol 2018;15:420–31. doi:10.1038/s41569-018-0011-2
      OpenUrlCrossRefPubMed
    4. ↵
      1. Treibel TA,
      2. Badiani S,
      3. Lloyd G, et al
      . Multimodality imaging markers of adverse myocardial remodeling in aortic stenosis. JACC Cardiovasc Imaging 2019;12(8 Pt 1):1532–48. doi:10.1016/j.jcmg.2019.02.034
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Sabbah M,
      2. Olsen NT,
      3. Holmvang L, et al
      . Long-term changes in coronary physiology after aortic valve replacement. EuroIntervention 2023;18:1156–64. doi:10.4244/EIJ-D-22-00621
      OpenUrlCrossRefPubMed
    6. ↵
      1. Scarsini R,
      2. Pighi M,
      3. Mainardi A, et al
      . Proof of concept study on coronary Microvascular function in low flow low gradient aortic stenosis. Heart 2022:heartjnl-2022-321907.
    7. ↵
      1. Wiegerinck EMA,
      2. van de Hoef TP,
      3. Rolandi MC, et al
      . Impact of aortic valve stenosis on coronary hemodynamics and the instantaneous effect of Transcatheter aortic valve implantation. Circ Cardiovasc Interv 2015;8:e002443. doi:10.1161/CIRCINTERVENTIONS.114.002443
    8. ↵
      1. Stoller M,
      2. Gloekler S,
      3. Zbinden R, et al
      . Left ventricular Afterload reduction by Transcatheter aortic valve implantation in severe aortic stenosis and its prompt effects on comprehensive coronary Haemodynamics. EuroIntervention 2018;14:166–73. doi:10.4244/EIJ-D-17-00719
      OpenUrlCrossRefPubMed
    9. ↵
      1. Singh A,
      2. Greenwood JP,
      3. Berry C, et al
      . Comparison of exercise testing and CMR measured myocardial perfusion reserve for predicting outcome in asymptomatic aortic stenosis: the Prognostic importance of Microvascular dysfunction in aortic stenosis (PRIMID AS) study. Eur Heart J 2017;38:1222–9. doi:10.1093/eurheartj/ehx001
      OpenUrlCrossRefPubMed
    10. ↵
      1. Steadman CD,
      2. Jerosch-Herold M,
      3. Grundy B, et al
      . Determinants and functional significance of myocardial perfusion Reserve in severe aortic stenosis. JACC Cardiovasc Imaging 2012;5:182–9. doi:10.1016/j.jcmg.2011.09.022
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Zhou W,
      2. Bajaj N,
      3. Gupta A, et al
      . Coronary Microvascular dysfunction, left ventricular remodeling, and clinical outcomes in aortic stenosis. J Nucl Cardiol 2021;28:579–88. doi:10.1007/s12350-019-01706-y
      OpenUrlPubMed
    12. ↵
      1. Généreux P,
      2. Pibarot P,
      3. Redfors B, et al
      . Staging classification of aortic stenosis based on the extent of cardiac damage. Eur Heart J 2017;38:3351–8. doi:10.1093/eurheartj/ehx381
      OpenUrlCrossRefPubMed
    13. ↵
      1. Tastet L,
      2. Tribouilloy C,
      3. Maréchaux S, et al
      . Staging cardiac damage in patients with asymptomatic aortic valve stenosis. J Am Coll Cardiol 2019;74:550–63. doi:10.1016/j.jacc.2019.04.065
      OpenUrlFREE Full Text
    14. ↵
      1. De Maria GL,
      2. Scarsini R,
      3. Shanmuganathan M, et al
      . Angiography-derived index of Microcirculatory resistance as a novel, pressure-wire-free tool to assess coronary Microcirculation in ST elevation myocardial infarction. Int J Cardiovasc Imaging 2020;36:1395–406. doi:10.1007/s10554-020-01831-7
      OpenUrlCrossRefPubMed
    15. ↵
      1. Scarsini R,
      2. Shanmuganathan M,
      3. Kotronias RA, et al
      . Angiography-derived index of Microcirculatory resistance (ImraNgio) as a novel pressure-wire-free tool to assess coronary Microvascular dysfunction in acute coronary syndromes and stable coronary artery disease. Int J Cardiovasc Imaging 2021;37:1801–13. doi:10.1007/s10554-021-02254-8
      OpenUrlCrossRefPubMed
    16. ↵
      1. Kotronias RA,
      2. Terentes-Printzios D,
      3. Shanmuganathan M, et al
      . Long-term clinical outcomes in patients with an acute ST-segment-elevation myocardial infarction stratified by angiography-derived index of Microcirculatory resistance. Front Cardiovasc Med 2021;8:717114. doi:10.3389/fcvm.2021.717114
    17. ↵
      1. Fearon WF,
      2. Kobayashi Y
      . Invasive assessment of the coronary Microvasculature: the index of Microcirculatory resistance. Circ Cardiovasc Interv 2017;10:e005361. doi:10.1161/CIRCINTERVENTIONS.117.005361
    18. ↵
      1. Ford TJ,
      2. Ong P,
      3. Sechtem U, et al
      . Assessment of vascular dysfunction in patients without obstructive coronary artery disease: Why, how, and when. JACC Cardiovasc Interv 2020;13:1847–64. doi:10.1016/j.jcin.2020.05.052
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Mejía-Rentería H,
      2. Wang L,
      3. Chipayo-Gonzales D, et al
      . Angiography-derived assessment of coronary Microcirculatory resistance in patients with suspected myocardial ischaemia and non-obstructive coronary arteries. EuroIntervention 2023;18:e1348–56. doi:10.4244/EIJ-D-22-00579
      OpenUrlCrossRefPubMed
    20. ↵
      1. Douglas PS,
      2. Carabello BA,
      3. Lang RM, et al
      . ACC/AHA/ASE key data elements and definitions for transthoracic echocardiography: A report of the American college of cardiology/American heart Association task force on clinical data standards (writing committee to develop cardiovascular endpoints data standards) and the American society of echocardiography. Circ Cardiovasc Imaging 2019;12:e000027. doi:10.1161/HCI.0000000000000027
    21. ↵
      1. Negishi K,
      2. Negishi T,
      3. Kurosawa K, et al
      . Practical guidance in echocardiographic assessment of global longitudinal strain. JACC Cardiovasc Imaging 2015;8:489–92. doi:10.1016/j.jcmg.2014.06.013
      OpenUrlFREE Full Text
    22. ↵
      1. Voigt J-U,
      2. Mălăescu G-G,
      3. Haugaa K, et al
      . How to do LA strain. Eur Heart J Cardiovasc Imaging 2020;21:715–7. doi:10.1093/ehjci/jeaa091
      OpenUrlCrossRefPubMed
    23. ↵
      1. Badano LP,
      2. Kolias TJ,
      3. Muraru D, et al
      . Standardization of left atrial, right ventricular, and right atrial deformation imaging using two-dimensional speckle tracking echocardiography: a consensus document of the EACVI/ASE/industry task force to standardize deformation imaging. Eur Heart J Cardiovasc Imaging 2018;19:591–600. doi:10.1093/ehjci/jey042
      OpenUrlCrossRefPubMed
    24. ↵
      1. Pighi M,
      2. Fezzi S,
      3. Pesarini G, et al
      . Extravalvular cardiac damage and renal function following Transcatheter aortic valve implantation for severe aortic stenosis. Can J Cardiol 2021;37:904–12. doi:10.1016/j.cjca.2020.12.021
      OpenUrlCrossRefPubMed
    25. ↵
      1. McDonagh TA,
      2. Metra M,
      3. Adamo M, et al
      . ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J 2021;42:3599–726. doi:10.1093/eurheartj/ehab368
      OpenUrlCrossRefPubMed
    26. ↵
      1. Grambsch PM,
      2. Therneau TM
      . Proportional hazards tests and diagnostics based on weighted residuals. Biometrika 1994;81:515–26. doi:10.1093/biomet/81.3.515
      OpenUrlCrossRefWeb of Science
    27. ↵
      1. Cox DR
      . Regression models and life-tables. Journal of the Royal Statistical Society: Series B (Methodological) 1972;34:187–202. doi:10.1111/j.2517-6161.1972.tb00899.x
      OpenUrlCrossRefWeb of Science
    28. ↵
      1. Zhou W,
      2. Sun Y-P,
      3. Divakaran S, et al
      . Association of myocardial blood flow Reserve with adverse left ventricular remodeling in patients with aortic stenosis: the Microvascular disease in aortic stenosis (MIDAS) study. JAMA Cardiol 2022;7:93–9. doi:10.1001/jamacardio.2021.3396
      OpenUrlCrossRefPubMed
    29. ↵
      1. Benfari G,
      2. Noni M,
      3. Onorati F, et al
      . Effects of aortic valve replacement on left ventricular diastolic function in patients with aortic valve stenosis. Am J Cardiol 2019;124:409–15. doi:10.1016/j.amjcard.2019.04.046
      OpenUrlCrossRefPubMed
    30. ↵
      1. Scarsini R,
      2. Portolan L,
      3. Della Mora F, et al
      . Angiography-derived and sensor-wire methods to assess coronary Microvascular dysfunction in patients with acute myocardial infarction. JACC Cardiovasc Imaging 2023;16:965–81. doi:10.1016/j.jcmg.2023.01.017
      OpenUrlCrossRefPubMed

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • RS and LP are joint first authors.

    • RS and LP contributed equally.

    • Correction notice This article has been corrected since it was first published. The OA license type has been updated to CC BY-NC. 18th November 2024.

    • Contributors RS and LP equally contributed to this work. All the authors contributed significantly to the study by planning (RS, GB, FLR), conducting (RS, LP, FDM), reporting (RS, LP, FDM), conceptualising (RS, GB, FLR) and designing (RS, GB, FLR), acquiring (RS, SA, MF, AM, PS, AD, PADS, SP) or analysing and interpreting the data (RS, LP, SF, DT, CM, ML, GP, GB). RS acts as guarantor of the study, accepting full responsibility for the work and/or the conduct of the study, had access to the data and controlled the decision to publish.

    • Funding This study was partially funded by a research grant from Abbott Vascular (n.1333-12/2020).

    • Competing interests RS reports research grant from Abbott Vascular and speakers fee from Abbott. FLR reports research grant from Abbott Vascular and Philips. The other authors report no conflict of interest.

    • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.