The future of interventional cardiology

Although mechanical reperfusion has been shown to achieve epicardial recanalization in almost all acutely occluded arteries, the optimal myocardial reperfusion still remains a major issue, and it is only achieved in barely 50% to 70% of the patients with ST- segment elevation myocardial infraction (STEMI).1 Several factors have been demonstrated to have an impact on myocardial reperfusion including preoperative Thrombolysis in Myocardial Infarction (TIMI) flow, ischemia time, ageing, diabetes, thrombus burden, and vessel size.1-3 Therefore, over the last few decades, several studies have been conducted on adjunctive therapies and devices to improve reperfusion such as antithrombotic therapies,4-5 and thrombectomy.6

The use of coronary stents, in particular drug-eluting stents, currently represents the standard of care,7 and considerable attention has been paid over the last decade on stenting techniques,8 and their impact on procedural results and outcomes.

In a paper recently published in REC: Interventional Cardiology, Vega et al.9 conducted a randomized trial to address the impact of the delivery system speed deflation on myocardial perfusion and the outcomes of patients with STEMI treated with direct stenting.

In fact, fast balloon deflation has been suggested to cause abrupt changes in coronary flow that may trigger the detachment of thrombotic material, and plaque fragments, disrupted by the stent strut coverage.10 Also, variations of intravascular pressure may increase the wall shear stress, which has been shown to promote plaque destabilization, endothelial dysfunction,11 and also the hydrostatic pressure inside the interstitial space favoring myocardial oedema, and cellular damage.12

Indeed, in post-conditioning strategies, balloon inflation has been proposed as a mechanism of protection against ischemic damage by inducing repeated sequences of ischemia-reperfusion that have been proven to reduce the infarct size.13

In a previous randomized study that recruited 211 patients, Gu et al.14 reported an improvement of coronary flow with the stent delivery system slow deflation strategy, yet with a not significant reduction of the no-reflow phenomenon, and null effects on the long-term outcomes.

However, this study primary endpoint was the corrected TIMI frame count, an index of coronary flow, whereas Vega et al.9 assessed the myocardial blush grade (MBG) and the ST-resolution, both parameters of myocardial perfusion that may be conditioned by several other factors.

In fact, the Spanish study9 was consistent with previous larger reports,8 and diabetes, hypertension, kidney disease, hemodynamic parameters, and lesion location emerged as independent predictors of MBG. Therefore, it may be argued that slow balloon deflation may have facilitated a successful epicardial reperfusion, although this did not translate into microcirculation differences or myocardial salvage.

Furthermore, although the extensive use of thrombectomy and glycoprotein IIb/IIIa inhibitors in the overall cohort of patients, as the authors very well pointed out, is not representative of the current guidelines-indicated strategies regarding primary percutaneous coronary intervention (PCI), it may have prevented such complications and minimized any potential benefits with the delivery system slow deflation strategy. Indeed, former studies and meta-analyses have demonstrated that the administration of these potent antiplatelet agents during a primary PCI, mainly as a downstream strategy,15 could translate into better myocardial perfusion, and reduce mortality.

In addition, the recruitment restriction to those lesions eligible for direct stenting in the study conducted by Vega et al.9 may have led to the selection of a very low-risk population where the occurrence of an impaired MBG was extremely low (observed in about 25% of the study population).

Finally, since the study was stopped after the recruitment of 50% of the predefined sample size for futility, we cannot discard that, with a larger population and different endpoints, any differences would have emerged. Future larger studies with a more heterogeneous and higher-risk population of patients with STEMI, more complex lesions, a higher rate of comorbidities, less extensive use of glycoprotein IIb/IIIa inhibitors, are certainly justified to better define the potential role of slow balloon deflation during primary PCI in terms of periprocedural complications, myocardial reperfusion, short- and long-term outcomes.

FUNDING

None whatsoever.

CONFLICTS OF INTEREST

None declared.

REFERENCES

Blood pressure indices like fractional flow reserve (FFR) and the instantaneous wave-free ratio (iFR) measure the distal-to-aortic pressure ratio of a lesion in a similar way using a pressure guidewire. FFR requires the induction of maximal hyperemia while the iFR is an index based on the assessment of pressure curves at rest. There is solid confirmation of the validity and efficacy of such indices from multiple studies while their use is widely backed by the clinical practice guidelines.1

However, the efficacy of blood pressure indices is based on 2 fundamental objectives. The first one is to prove that they are useful to indicate the revascularization of a lesion when the resulting value is below the cut-off point (≤ 0.80 for the FFR, and ≤ 0.89 for the iFR). The second one is to demonstrate that lesions with values above the cut-off point (> 0.80 or > 0.89, respectively) have a low risk of requiring revascularization, and lack of ischemic events at follow-up. Conceptually, the first indication is clear: a lesion with pathological values is an obstructive lesion that causes ischemia now. Therefore, if clinically indicated (and technically feasible) the best thing to do is to revascularize such a lesion. The second indication has a higher degree of uncertainty because a lesion that does not cause ischemia (now) does not necessarily mean that it will not cause it at some point in the future. This is so because the capacity of blood pressure indices (with normal results) to detect vulnerable plaques is limited, and they show a greater tendency towards progression or destabilization. Therefore, the negative predictive value of pressure indices is supposed to be worse when used in patients with a higher risk of progression into atherosclerotic cardiovascular disease.

Proof of this is that recent studies conducted in patients with acute myocardial infarction (AMI) and multivessel disease have demonstrated that the FFR of non-culprit lesions does not provide any clinical benefits regarding the revascularization of such lesions estimated visually and, therefore, under angiographic guidance.2,3 The prevalence of vulnerable plaques causing visual stenosis between 50% and 69% in non-culprit vessels of patients with AMI is estimated at around 30%. Actually this rate is probably higher when lesions with a greater degree of angiographic obstruction are studied.4 The COMBINE OCT-FFR trial also reported a similar rate of vulnerable plaques (25%) on the optical coherence tomography performed in diabetic patients with angiographic stenoses between 40% and 80% and normal FFR values.5 In this study, angiographic lesions with normal FFR values associated with vulnerable plaques had more adverse events (cardiac death, target vessel myocardial infarction, and clinically driven target lesion revascularization) compared to those not associated with vulnerable plaques (13% vs 3%; P < .001).

The article by Castro-Mejía et al.6 recently published in REC: Interventional Cardiology provides the mid-term follow-up (3.5 years) of a large single-center registry of patients with intermediate angiographic lesions assessed using blood pressure indices and with normal results. This study compared the adverse events between diabetic and nondiabetic patients with the necessary statistical adjustments to compensate for the inherent differences of the baseline characteristics of both groups. The authors conclude that blood pressure indices (FFR and iFR) had a similar efficacy in diabetic and nondiabetic patients for not predicting future target vessel myocardial infarctions or the need for target lesion revascularization. However, we should mention that diabetic patients had higher rates of all-cause mortality, AMI, and need for revascularization compared to nondiabetic patients.6 We should specifically mention that there was a 2.6-fold higher risk of infarction in any vessels at the follow-up in diabetic compared to nondiabetic patients (P < .063 after adjustments). However, only 13% of the AMIs registered were adjudicated to the study vessel in diabetic vs 38% in nondiabetic patients.6 This difference can be due to chance alone or to the fact that the lesions that cause AMIs at the follow-up are scarcely susceptible to be studied with a pressure guidewire at the index procedure in diabetic patients. It is well known that these patients have more diffuse atherosclerotic cardiovascular disease, which complicates the correct angiographic assessment of the lesions (and, also, prevents the use of a pressure guidewire in such vessel for assessment purposes). It is highly likely that a strategy based on intravascular imaging modalities alone or in combination like in the COMBINE OCT-FFR trial can be of greater utility to detect this type of lesions.

Three recent studies have reported contradictory outcomes regarding the negative predictive value of blood pressure indices in diabetic patients. Kennedy et al.7 found more AMIs and a greater need for target lesion revascularization with normal FFR values in diabetic patients at the 3-year follow-up. On the contrary, Van Belle et al.8 found no differences between diabetic and nondiabetic patients who were deferred (for having normal FFR values) in 2 large registries conducted in Portugal, and France at the 1-year follow-up. No significant differences were seen either in the DEFINE-FLAIR substudy that analyzed adverse events between diabetic and nondiabetic patients in deferred patients at 1-year follow-up.9 It should be interesting to assess the long-term follow-up of these last 2 studies to see if the negative predictive value of blood pressure indices stays the same in diabetic and nondiabetic patients.

Finally, Castro-Mejía et al.6 propose that resting physiological indices with nonpathological findings may have a better negative predictive value compared to the FFR values of diabetic patients. This would be explained by the fact that one of the causes of discrepancy between the FFR and the iFR is microvascular dysfunction (that is more common in diabetic patients). Patients with microvascular dysfunction can have smaller drops of the pressure index during maximal hyperemia since not enough hyperemia is induced. However, resting physiological indices are not hyperemia-dependent, meaning that they should be more sensitive to detect significant epicardial lesions. However, this hypothesis was not confirmed by the DEFINE-FLAIR substudy on deferred diabetic patients.9

In conclusion, diabetic patients still remain as one of the challenges of interventional cardiology because they have more adverse events compared to nondiabetic patients (despite the proper optimal medical therapy). A more aggressive strategy to assess intermediate coronary lesions using pressure guidewires is advisable to guide complete revascularization in diabetic patients. Future studies should also assess whether a mixed strategy with intravascular imaging modalities of nonfunctionally significant arteries can be useful to prevent future events in this group of patients.

FUNDING

None whatsoever.

CONFLICTS OF INTEREST

J. Gómez-Lara has received speaking fees for presentations on behalf of Abbott Vascular, Boston Scientific, and Terumo. R. Romaguera declared no conflicts of interests whatsoever.

REFERENCES

The use of drug-eluting stents (DES) for the treatment of coronary artery stenosis substantially reduced the need for repeat revascularization compared to bare-metal stents.1 However, as many patients undergoing stenting have long life expectancy, and the incidence rate of stent failure increases with time since implantation, the number of patients presenting with DES restenosis is not insignificant and the treatment of these patients remains a challenge.2

Current clinical practice guidelines recommend treatment of restenosis associated with angina or ischemia by repeat revascularization with either repeat stenting with DES or angioplasty with drug coated balloon (DCB).3 Certain situations favour repeat stenting with DES, most notably loss of mechanical integrity of the restenosed stent. In general, however, although repeat stenting with DES may be more effective than angioplasty with DCB in the short-to-medium-term,4 avoidance of additional stent layers is an important consideration in the longer-term. Indeed, many centres prefer DCB angioplasty as a first-line approach for the treatment of restenosis in the absence of a compelling indication for repeat stenting.

The efficacy of DCB treatment relies on rapid transfer and subsequent tissue retention of the anti-proliferative agent, which is necessary for a persistent suppression of cell proliferation.5 Preclinical data suggest that micro-injuries to the vessel wall may enhance the ability of DCBs to inhibit neointimal growth.6 These micro-injuries can be achieved with a number of different types of modified balloon catheters, such as cutting or scoring balloons. Cutting balloon angioplasty is an attractive option thanks to its ability to effectively incise neointimal tissue and its ease of use.7 Scoring balloons are based on the same principle and may offer superior flexibility and deliverability at the expense of a somehow lower plaque disruption.

In a recent paper published in REC: Interventional Cardiology, Linares Vicente et al. reported on the 5-year results of cutting or scoring balloon angioplasty combined with DCB to treat in-stent restenosis.8 A total of 51 lesions (42 patients) were treated with cutting balloons plus DCBs, and 56 lesions (49 patients) with a standard DCB angioplasty. Both the SeQuent Please (B. Braun Melsungen AG, Germany), and the Pantera Lux (Biotronik, Switzerland) balloons were used. The primary endpoint was clinically driven target lesion revascularization at 5 years. It appears that, compared to the standard DCB strategy, the use of cutting or scoring balloons considerably reduced the 5-year rate of target lesion revascularization, although this difference was not statistically significant (9.8% vs 23.6%; odds ratio = 0.36; 95% confidence interval, 0.19-1.09; P = .05) (figure 1).

Figure 1. Target lesion revascularization (%): studies comparing cutting balloon/scoring balloon vs standard therapy in patients treated with percutaneous intervention for in-stent restenosis.

The study was retrospective and conducted at a single center with a small sample size of 91 patients. However, it is representative of real-world evidence, which may reflect clinical experiences across a broader and more diverse population of patients than those enrolled in randomized controlled trials. Regarding the baseline characteristics, almost 85% of patients were men with a mean age of 68.3 years. Patients had a high prevalence of diabetes mellitus and smoking (36% and 59%, respectively).

Interestingly, despite the current recommendations of the European Society of Cardiology,3 the use of intravascular ultrasound or optical coherence tomography was relatively low (5.9% in the cutting balloon group, and 8.9% in the standard group). Although this is consistent with rates observed in surveys of use in the clinical practice,11 it represents a missed opportunity for the mechanistic understanding of the disease etiology, and guidance for treatment optimization.12

The primary endpoint was the need for clinically driven target lesion revascularization at 5 years, which was 64% lower with cutting or scoring balloon angioplasty. Given the small sample size of this study and the relatively large treatment effect, it is a pity that insights from the angiographic follow-up were not available. Concordant data from systematic surveillance angiography would have given more confidence to the robustness of the observed treatment effect.

The results of this study should be interpreted in the context of earlier randomized controlled trials with cutting or scoring balloon angioplasty. In fact, evidence from such trials is scant including the RESCUT (Restenosis cutting balloon evaluation trial) trial9 published in 2003, and the more recent ISAR DESIRE 4 (Intracoronary stenting and angiographic results: optimizing treatment of drug-eluting stent in-stent restenosis 4) trial.10

In RESCUT, Albiero et al. randomized 428 patients with bare-metal stent in-stent restenosis across 23 European centers to receive either cutting balloon angioplasty or conventional balloon angioplasty.9 Overall, the trial showed neutral results: at late follow-up, the angiographic restenosis rate, minimal lumen diameter, and the rate of clinical events were similar in both arms (figure 1). Cutting balloon angioplasty, however, was associated with some important procedural advantages, such as use of fewer balloons, less requirement for additional stenting, and a significantly lower incidence of balloon slippage (6.5% vs 25%).

ISAR DESIRE 4 was a randomized, open-label, assessor-blinded trial that enrolled 252 patients with clinically significant DES restenosis undergoing DCB angioplasty at 4 different centres in Germany.10 This trial investigated scoring balloon rather than cutting balloon angioplasty. The primary endpoint–diameter stenosis at the 6-8-month follow-up angiography–was lower for the scoring balloon compared to the regular balloon angioplasty: 35% vs 40.4%; P = .047; in addition, target lesion revascularization was numerically lower (figure 1). Although, the size of treatment effect was modest, small incremental gains in efficacy in this challenging patient subset may translate into important clinical benefits.

Against this background, the observations made by Linares Vicente et al.8 are an important addition to the evidence supporting the clinical use of cutting or scoring balloons to treat stent failure. While repeat stenting with DES or angioplasty with DCB are the mainstay of in-stent restenosis procedures, the procedural efficiency and clinical efficacy of both approaches will likely improve with the adjunctive use of cutting or scoring balloons. The benefits of these devices are most likely mediated by a combination of factors: reduced balloon slippage (or watermelon-seeding), the mechanical advantage of increased disruption of restenotic tissue, and the potential for enhanced efficacy of the device-delivered drug. The management of patients with stent failure remains challenging and deserves the best treatment the operator can offer including the liberal use of cutting or scoring balloon lesion preparation. In this clinical setting, small margins can make a big difference.

FUNDING

L. McGovern is funded by the European Commission under the Horizons 2020 framework (grant agreement number 965246 CORE-MD).

CONFLICTS OF INTEREST

R.A. Byrne declared research or educational funding in the institution he currently works at from Abbott Vascular, Biosensors, Biotronik, and Boston Scientific. L. McGovern declared no conflicts of interest whatsoever.

REFERENCES

In 1995, my colleagues and I at the Washington Hospital Center (Washington, DC, United States) published an intravascular ultrasound (IVUS) vs angiographic assessment of calcium in 1155 lesions targeted for percutaneous coronary intervention (figure 1)1. Angiography detected calcium in 440 lesions (38%), but IVUS detected lesion calcium in 841 lesions (73%). Among these 1155 lesions, 27% had no IVUS calcium, 26% had 1-quadrant IVUS calcium, 21% had 2 quadrants, 15% had 3 quadrants, and 11% had 4-quadrant IVUS calcium. When present, target lesion calcium was only superficial in 48%, only deep in 28%, and both superficial and deep in 24%. Therefore, some superficial calcium was present in 72% of the 841 calcium-containing lesions (1-quadrant superficial calcium in 35%, 2 quadrants in 31%, 3 quadrants in 18%, and 4-quadrant superficial calcium in 18%). The diagnostic ability of angiography to detect calcium was primarily dependent on the arc and length of calcium, but also on whether calcium was or not superficial (figure 1). However, there was also a curious 10% rate of angiographic false positives attributed to the difficulty differentiating perivascular or reference segment calcium from intralesional calcium. However, it was never clear whether there was a systematic problem with angiographic calcium detection or whether it was because, in the early 1990s, angiography was primitive compared to today and would improve over time.

Figure 1. A total of 3 studies (clockwise starting in the upper left-hand corner) comparing intravascular imaging to the angiography detection of coronary lesion calcium. A: the study conducted by Mintz et al.1 from 1995. B and C: the study by Wang et al.2 from 2017. D: the study by McGuire et al.3 from 2021. IVUS, intravascular ultrasound; OCT, optical coherence tomography.

This experiment was repeated more than 20 years later by Wang et al. in a smaller cohort of 440 lesions using state-of-the-art angiographic equipment and both IVUS and optical coherence tomography (OCT) imaging (figure 1).2 Any amount of calcium was detected by coronary angiography in 40.2% (177 of 440) of the lesions, by IVUS in 82.7% (364 of 440) of the lesions, and by OCT in 76.8% (338 of 440) of the lesions. Notably and compared to the 1995 study, almost all calcium was superficial, fewer lesions had no calcium, and more lesions had 1- or 2-quadrant calcium (figure 1). In 13.2% of the lesions with IVUS-detected calcium, calcium was not visible by OCT mostly because of attenuation due to superficial lipid plaque accumulation. In a recent paper published in REC: Interventional Cardiology, McGuire et al.3 compared angiographic vs OCT calcium detection in 75 lesions. OCT detected calcium in 69 lesions vs 30 lesions by angiography with no angiographic false positives (figure 1).3 Compared to IVUS, OCT can measure the thickness, area, and volume that affect the angiographic detection of calcium in addition to its arc and length.2,3

Other than a reduced rate of false positives in the 2 contemporary studies, which could be attributed to the improved resolution of modern x-ray equipment, the lower x-ray doses being used today vs 1995, and the clinical recognition of the existence perivascular calcium, the results were remarkably similar to those of 1995. Thus, there appears to be a fundamental limitation to x-ray that cannot be improved by technological advances.

Why is calcium detection so important? The primary cause of in-stent restenosis is stent underexpansion, the primary cause of stent underexpansion is calcium, and the natural history of in-stent restenosis is not benign with an annual mortality rate of 5% to 7% (associated with treatment and at the follow-up).4-7 There are calcium scores for both OCT and IVUS that reliably predict calcium-related stent underexpansion (figure 2);8,9 and there are technologies and approaches that can be used to modify calcium to promote a better stent expansion.4,10

Figure 2. Intravascular ultrasound and optical coherence tomography calcium scores to predict stent underexpansion.8,9 IVUS, intravascular ultrasound; MSA, minimum stent area; OCT, optical coherence tomography.

There are predictors of target lesion calcium at patient level (older age, non-insulin treated diabetes, stable angina rather than an acute coronary syndrome, chronic kidney disease—especially if a patient is on dialysis—, and calcium elsewhere in the coronary tree), and predictors at lesion level too (smaller vessels, more severe stenoses).11-15 However, for the most part, lesions behave independently with regards to calcium accumulation. Only intravascular imaging can reliably detect and quantify target lesion calcium and predict stent underexpansion in the severe target lesion calcium setting.

FUNDING

None.

CONFLICTS OF INTEREST

G.S. Mintz declares having received honoraria from BostonScientific, Philips, Medtronic, and Abiomed outside the present work.

REFERENCES