Debate: MitraClip. The heart failure expert perspective

QUESTION: After the negative results of the SYMPLICITY HTN-3 (Renal Denervation in Patients With Uncontrolled Hypertension) study, what new studies are dealing with renal denervation (RD) as a possible therapeutic option?

ANSWER: The results of the SYMPLICITY HTN-3 study1 were totally unexpected, particularly because those of us who had some sort of experience with this technique had lived a completely dif-ferent reality. The limitations of the study have to do with the arguable selection of patients (with non-optimized pharmacological treatment that was adjusted during follow-up and generated an unexpectedly positive response in the control group), the operators’ lack of experience (most of them conducted their first procedures within the study), and the lack of knowledge on how to conduct this procedure in order to optimize the results.

After the detailed analysis of the SYMPLICITY HTN-3 study,1 two different studies were designed with an improved device to conduct the procedure and their results have recently been made public.

The first study, the SPYRAL HTN-OFF MED2 was a randomized sham-controlled clinical trial (with a sham control group) as the proof-of-concept on the efficacy of RD to reduce arterial blood pressure (BP) in patients without concomitant pharmacological treatment. Patients with mild-to-moderate arterial hypertension (HT) (office systolic BP 150 mmHg to 180 mmHg and diastolic BP > 90 mmHg, and 24-hour ambulatory systolic BP 140 mmHg to 170 mmHg) were included in the study. Patients did not receive any prior treatment or had been without any pharmacological treatment for 3 to 4 weeks. The protocol included a drug screening of serum and plasma to confirm the absence of drugs. Patients were randomized on a 1:1 basis to RD using the multi-electrode radiofrequency catheter designed by Symplicity Spyral (Medtronic Inc., Minneapolis, United States) (n = 38) or to sham control (n = 42). The primary endpoint included changes in the 24-hour ambulatory BP, much more sensitive and specific to detect changes in the BP that measuring BP at the doctor’s office. The analysis of the first 80 patients showed a significant reduction of the 24-hour ambulatory systolic BP and office systolic BP at 3 months in favor of RD. We should mention here that during follow-up no relevant adverse events were reported in any of the two treatment arms and, as former studies had already shown, the procedure turned out to be safe and had an extremely low rate of complications.1

The endpoint of the second study, the SPYRAL HTN-ON MED,3 was to assess the efficacy of RD in a different context. The study population were not patients with resistant HT or HT naïve to drug therapy but non-severe hypertensive patients on drug therapy. Same as it happened with the SPYRAL HTN-OFF MED2 study, the primary endpoint included changes in the 24-hour BP instead of office-recorded BP changes. The patients included in the study had mild-to-moderate HT (office systolic BP between ≥ 150 mmHg and < 180 mmHg, office diastolic BP ≥ 90 mmHg, and 24-hour ambulatory systolic BP between 140 mmHg and 170 mmHg). RD reduced the 24-hour ambulatory systolic BP and the office systolic BP at 6-month follow up compared to the control arm.2

Also, a third study with a totally different device based on ultrasounds and not radiofrequency, the RADIANCE-HTN SOLO,4 confirmed the good results shown by the SPYRAL HTN-OFF MED2 in patients with moderate untreated HT with similar results in the monitoring of BP figures at follow-up, which reinforces the idea of RD for the management of HT.

Q.: What kind of technical advances have led to these positive results and what are the limitations of this kind of therapy?

A.: The critical analysis of the results shown by the SYMPLICITY HTN-31 taught us how to improve the results obtained with this procedure. It confirmed that the patients’ response was significantly better when the four quadrants of the renal artery were treated. Also, further studies showed that RD was more effective not only when the main stem of the renal artery was treated but also when the secondary branches were treated as well. A look at the anatomy of the sympathetic innervation revealed that although there were more nervous fibers at a proximal level (where treatment was formerly recommended), these fibers were at a distance from the vascular lumen that made it difficult for the radiofrequency energy released inside the blood vessel to actually reach them. On the contrary, at the most distal portion, nervous fibers are closer to the lumen and they are affected by the radiofrequency lesion. Finally, the number of radiofrequency applications somehow had something to do with efficacy in such a way that today it is adviseable to perform the maximum number of applications.

In order to simplify the procedure and taking all these premises into consideration, a new RD catheter was developed, the Symplicity Spyral, whose main characteristics with respect to the Symplicity Flex catheter (Medtronic Inc., Minneapolis, United States) with which the SYMPLICITY HTN-31 study was conducted were that it was a tetrapolar catheter (compared to the former one that was monopolar) meaning that up to 4 simultaneous radiofrequency applications could be performed; also, it cut down the duration of the application from 120 to 60 seconds. Also, the spiral-shaped catheter allowed the radiofrequency application to cover the 4 quadrants of the renal artery. Finally, the optimized caliber of the new device facilitated treating arteries of up to 3 mm in diameter vs 4 mm with the former device.

Q.: What would the actual indications of this technique be, if any?

A.: The clinical practice guidelines established by the European Society of Cardiology and the European Society of Hypertension that have been published recently5 are older compared to the knowledge acquired from the last studies we mentioned before. In these clinical guidelines, recommendations were based on the SYMPLICITY HTN-31 and its use was recommended in the setting of clinical trials only and outside the routine clinical practice.

The results of the new studies consistently show that RD is effective when it comes to improving the monitoring of the BP. Similarly, different observational registries have shown significant improvements in a large number of patients with resistant HT. In Spain we conducted a registry6 that included 125 of these patients and saw a good response in over 80% of these patients, not only when it comes to the office BP but also, and most important, when it comes to monitoring ambulatory BP. Also, RD significantly reduced pharmacological treatment, a finding that opens the door to future studies. With the evidence available today, in my opinion, patients with maintained non-monitored HT on multi-drug therapy, including aldosterone antagonists, can benefit from this procedure. We know that a reduction of 20 mmHg in systolic BP or 10 mmHg in diastolic BP cuts in half mortality risk due to cardiovascular causes. This improvement is not difficult to achieve in many of these patients after RD.

On the other hand, there are many gaps of knowledge still to be filled in in the field of RD. It is essential to identify those patients who may respond better to this procedure since the pathophysiology of HT is complex and is not always due to alterations in the regulation of the sympathetic nervous system. With regard to the procedure itself, the lack of markers to determine whether RD has been successful or not puts us on a holding pattern to see how the BP figures have evolved before determining its efficacy. The development of a non-invasive test to obtain this information should be the goal of future research. Also, the arrival of new technologies to perform RD procedures requires assessing its safety and efficacy profile in the long run.

Q.: What studies do we need so that clinical practice guidelines can recommend RD as a therapeutic alternative for the management of HT?

A.: Yet despite the raising awareness on the risks of HT and the development of new and better drugs over the last 70 years, data from 2010 in developed countries showed that one third of those who had the disease did not know about it, a little over half of them received pharmacological treatment, and less than a third had an adequate blood pressure monitoring. In this sense, the road ahead of us is a long one.

With the new Symplicity Spyral catheter the long-term safety profile is an issue we should take into consideration to give more robust guarantees that the treatment algorithm—substantially more aggressive than the algorithm used in the SIMPLICITY studies—does not cause complications.

The number of patients included in the studies is not large enough to give us evidence that the reduced BP levels observed after RD actually reduce the rate of cardiovascular events at follow-up. Improving BP is but a surrogate primary endpoint, although it is accepted that there is a correlation between lower BP levels and less cardiovascular events. Also, similar BP reductions to the ones obtained in these studies led to less events in pharmacological studies. A study that showed clinical benefits beyond the monitoring of BP would actually be conclusive in this context. Unfortunately, the high number of patients who should be included in this study probably makes such a study unfeasible. However, we should not forget that damage caused by sympathetic hyperactivity goes beyond hypertension itself and is cause for the worse glucose metabolism seen in diabetic patients, the arterial stiffness of atherosclerosis, the poor prognosis of heart failure and the impaired renal function seen in renal failure, to mention but a few.

The SPYRAL HTN-OFF MED2 and RADIANCE SOLO4 clinical trials have given us the first evidence, in a consistent way, on the possible clinical utility of RD for the management of hypertensive patients who may wish, or not, to use antihypertensive drugs. These preliminary results should be confirmed by the ongoing fundamental studies that intend to include a much larger number of patients. The pharmacological treatment of HT is a long-term option and, in most cases, for life. Even though drugs are usually well-tolerated, the noncompliance to pharmacological treatment is a common problem to the extent that almost one third of all hypertensive patients do not start a new prescription of antihypertensive medication and 50% of them become noncompliant during the first year after starting their antihypertensive medication.

In my opinion, should these results be confirmed, there will be a change of paradigm in the management of hypertension. Also, we will have to take every patient’s individual preference (shared decision-making process) into consideration on whether to follow pharmacological treatment for life or undergo a more individualized therapeutic approach through a catheter-based invasive procedure that has proven safe and with an active effect at all time.

Finally, in light of the preliminary results seen in other conditions, it will be essential to see if there is an added benefit to regulating the activity of the sympathetic nervous system beyond reducing the BP in other diseases where there is an increased sympathetic nervous system activation such as within cardiology, atrial fibrillation, and heart failure.

CONFLICTS OF INTEREST

None declared.

REFERENCES

QUESTION: After the negative results of the SYMPLICITY HTN-3 (Renal Denervation in Patients With Uncontrolled Hypertension) study, what new studies support renal denervation (RD) as a possible therapeutic option?

ANSWER: The introduction of RD as a therapeutic approach in patients with resistant arterial hypertension (RAHT) was very well received due to the difficulties in adequately reducing arterial blood pressure (BP) levels and, consequently, the cardiovascular and renal risk associated with the persistence of high BP. Also, this technique was backed by the extensive knowledge of the role renal sympathetic nerves play regulating BP. Some non-randomized studies revealed the virtues of this technique with significant reductions of BP levels. However, the first double-blind comparative study conducted, the SYMPLICITY HTN-3,1 did not confirm the superiority of RD over the pharmacological treatment of RAHT. This study showed that the response to RD had been lower in African-Americans and patients who had received anti-aldosterone drugs as antihypertensive therapy. The study was then heavily criticized because of the procedures used, the reduced number of ablations performed, the scarce experience of participant centers, etc.

After this first failure, several studies were designed, both in patients with and without RAHT to determine the capacity of RD to reduce the BP levels. The reason for this change can be found in the very nature of RAHT itself, the huge heterogeneity of patients, and the variability of BP values themselves. Among the studies in patients with RAHT, the DENER-HTN study (Renal Denervation in Patients With Resistant Hypertension)2 showed that RD was better compared to step-care treatment with antihypertensive drugs, although the degree of BP reduction may be biased because baseline values were much higher in the intervention group versus the control group. Among the studies in patients with no RAHT, the SPYRAL HTN-OFF MED (Global Clinical Study of Renal Denervation With the Symplicity Spyral Multi-electrode Renal Denervation System in Patients With Uncontrolled Hypertension in the Absence of Antihypertensive Medications)3 focused on patients without prior antihypertensive therapy and naïve to treatment during the study. The SPYRAL HTN-ON MED study (Global Clinical Study of Renal Denervation With the Symplicity Spyral Multi-electrode Renal Denervation System in Patients With Uncontrolled Hypertension on Standard Medical Therapy)4 included patients on antihypertensive treatment with 1 to 3 drugs and the RADIANCE-HTN SOLO5 study included controlled or non-controlled patients with low cardiovascular risk and on medication with 0 or up to 2 antihypertensive drugs. Whereas in the former, the antihypertensive treatment was kept, in the latter it was withdrawn. These studies confirmed reductions between 5 mm Hg and 7.4 mmHg in ambulatory systolic BP.6

Q.: What kind of technical advances have led to these positive results and what are the limitations of this kind of therapy?

A.: Together with the type of patients who are deemed eligible to receive RD, research has been made to develop devices capable of improving the ablation capabilities of the sympathetic nervous system and additional anatomical regions have been considered eligible for these ablations. Nonetheless, the number of ablations required has been controversial and there is no clear consensus on the different protocols used.

The new systems available today offer the possibility to act upon larger territories with every ablation by placing the electrodes around the spiral catheter and other changes in the design.

Other controversial aspects relate to what anatomical sections of the renal arteries should these ablations be performed on or what is the necessary number of ablations to be performed to achieve significant reductions in the renal sympathetic activity. Sympathetic fibers penetrate the kidney through the walls of the main renal arteries, and it is precisely on this location where these ablations were initially performed. The introduction of protocols to perform ablations in the distal branches of the renal arteries seems to achieve a greater sympathetic block and, consequently, higher effectiveness. Currently, studies should include the segments of renal arteries on which the ablations should be performed, and indicate how many of them are required, although there is not such a thing as a standard recommendation yet.

Q.: What would the current indications for this technique be, if any?

A.: The current indications for RD use are also controversial. The first possible indication would be for patients with true RAHT in whom the right pharmacological treatment cannot keep BP under control and below 140/90 mmHg.

Drug-resistance is due to several causes, many of which can be detected and fixed. An easy protocol on this regard should include: a) checking high BP levels using outpatient measures, such as home or 24-hour ambulatory monitoring; b) anamnesis aimed at detecting compliance and adherence to pharmacological treatment; c) evaluating excessive salt and alcohol intake, weight gain in cases of obesity, and use of medication or drugs that may elevate BP levels; d) excluding the presence of sleep apnea; e) assessing the need to rule out secondary arterial hypertension requiring specific treatment; and f) administering a combination of the right drugs at the right doses, especially in the case of diuretic treatment.

When this protocol is implemented, the percentage of patients diagnosed with uncontrollable RAHT does not exceed 10% and it is precisely those patients who may be eligible for RD. However, the various aspects that are still controversial today led the European Society of Cardiology and the European Society of Hypertension to avoid recommending RD outside research protocols in their latest clinical practice guidelines.7

The off-label use of RD for the management of RAHT is still far from being recommended. There is no objective reason to perform RD in situations where the right medication achieves the therapeutic goals suggested by the clinical practice guidelines.

Q.: What studies do we need so that clinical practice guidelines can recommend RD as a therapeutic alternative for the management of AHT?

A.: The development of RD technique still has a long way to go in the near future since we are still unaware of key aspects regarding its effectiveness, safety profile or who are the best candidates for this procedure.

First of all, when performing an ablation it is not possible to know when the denervation is effective enough. Until know we have tried to identify markers capable of determining whether the ablations performed were effective in suppressing adrenergic activity, because the changes in the BP levels that take place during the procedure are not associated with the degree of denervation.Therefore, the procedure is performed in the blind, with no control, and without being able to predict whether it will be successful or not. Also, it is impossible to bring procedural time down to the minimum possible with guaranteed success.

In part connected to this limitation, there is also the number of radiofrequency applications required for this procedure. Since the repetition of these may have long-term repercussions, the ideal number would be the minimum necessary to achieve the denervation we are aiming at, but, as before, it is not possible to ascertain its effectiveness during the procedure.

Similarly, to this day we still do not know the duration of the antihypertensive effect. It is well-known that after cutting the sympathetic nerves, they have the capability of reinnervation as it has been confirmed in post-transplant patients. The duration of the response, if any, is not well-known, and there are no records on its effectiveness beyond 2 or 3 years of follow-up. Although the persistence of the effect can be anticipated, it is difficult to verify because of the very nature of RAHT, as we have already seen and because this persistence is influenced by several factors related to the hypertensive state.

Also, it is important to mention the great variability in patient’s individual response to ablation. The clinical characteristics of patients who may respond better to RD are still to be established; we only have evidence that patients with isolated systolic hypertension or patients with high pulse wave velocity fare worse with RD, maybe because they have an important component of impaired vascular elasticity. Similarly, patients with a good response to anti-aldosterone drugs would also have attenuated responses since the high BP levels seen in these patients are volume-dependant. Apart from these exceptions, there are no validated markers available of a better antihypertensive response to RD.

A key element of which there is still no evidence is whether RD is better than pharmacological treatment in reducing the incidence of cardiovascular and renal events. Several studies have analyzed the reduction of organ damage, especially in left ventricular hypertrophy, and the improvement of vascular elasticity following RD, but there is no evidence that the impact RD has on morbimortality is higher compared to the one achieved with drug-induced control of the BP levels.

Last but not least, it is important to know what the possible long-term consequences of ablation may be on the renal arteries and, consequently, on renal function. The data available today tells us that in the short term there is no high risk of damage to the renal arteries, but there is no long-term data on the possible side effects.

All these questions will need to be answered in the future if renal denervation hopes to have a prominent place in the management of RAHT.8

CONFLICTS OF INTEREST

None declared.

REFERENCES

QUESTION: There is no doubt that the most significant advances made back in 2018 in interventional cardiology were the longawaited results from clinical trials with MitraClip (Abbott Laboratories, Abbott Park, Illinois, USA): the MITRA-FR presented at the congress organized by the European Society of Cardiology and the COAPT presented at the Transcatheter Cardiovascular Therapeutics (TCT) congress. Both trials have been discussed extensively. Could you please tell us what the basic differences are in the results obtained by these two trials?

ANSWER: The clinical studies MITRA-FR1 and COAPT2 share the honor of being the very first two studies in history with a rigorous design that randomized patients with functional mitral regurgitation (FMR) to receive optimal medical therapy or optimal medical therapy plus the FMR correction device MitraClip. There were opposing results. In the MITRA-FR the device did not produce any benefits over the composite event of death or rehospitalizations after one year of follow-up. In the COAPT, the Mitra- Clip group showed a significant reduction of hospitalizations at 2 years, and a reduced composite endpoint of death/rehospitalizations at 2 years. It seems reasonable to try to analyze the differences between both studies in an attempto to understand these opposing findings. These are the most significant differences I have seen:

• Magnitude of the FMR. This is probably one of the most important aspects that may explain the differences seen. According to the clinical guidelines established by the ESC, the MITRA-FR included patients with severe FMR defined as an effective regurgitant orifice area (EROA) ≥ 20 mm² or regurgitation volumes > 30 mL. This brought these patients’ average EROA to 31 ± 10 mm². On the contrary, according to the American clinical guidelines,3 the COAPT considered significant FMR EROAs ≥ 30 mm² or regurgitation volumes > 45 mL (average EROA 41 ± 15 mm²). Also, in the MITRA-FR over half of the patients (52%) showed EROAs between 20 and 30 and in the COAPT, 87% of the patients showed EROA values > 30 mm². In sum, the COAPT included patients with more severe FMR which would, logically, have a bigger impact on the patients’ events when corrected.
• Ventricular volumes. The patients from the MITRA-FR showed higher average ventricular volumes compared to those from patients from the COAPT (end-diastolic volume index: 135 ± 35 mL/m² versus 101 ± 34 mL/m²). This means that the disease was far more advanced in the French study compared to the American study. This can also be one of the keys that may explain the differences seen. Treating some FMR in patients with very dilated ventricular volumes may have no effect at all on major cardiovascular events.
• Optimal medical therapy. This is another essential aspect if we want to understand the differences seen between both trials. In the MITRA-FR, the medical treatment that patients were receiving before randomization was the one the treating physician considered optimal. Also, this treatment could be modified during follow-up without the study committee knowing about it. Thus, even though the drug doses administered to manage heart failure may have been the correct ones at the beginning of the study, they may have been modified later on. And it is well-known that optimizing drug doses has a major impact on patients’ events and functional class. This effect can be seen in the functional class improvement experienced by the group that received medical treatment in the MITRA-FR. However, in the COAPT an «eligibility committee» monitored that every patient would receive the right drugs and the maximum tolerated doses before randomization. That’s why there were not too many dose modifications in the COAPT follow-up compared to the beginning of the trial.
• MitraClip performance. There are substantial differences when it comes to the level of success and performance achieved by this device. In the MITRA-FR, 9% of the patients never received the device. The number of complications was higher in the MITRA-FR (14.5% versus 8.5%) and the percentage of success at one year was lower in the French trial (17 versus 5% with FMR > 2+).
• Selection of candidates. The COAPT trial paid special attention to the selection of candidates. Patients whose baseline characteristics offered poor prognosis in the short-term and in whom the intervention wouldn’t probably lead to significant clinical improvements were excluded. Irreversible pulmonary hypertension, moderate or severe right ventricular dysfunction, stage D heart failure, hemodynamic instability, and inotropic therapy were cause for exclusion from the trial. These patients were not in the exclusion list presented by the MITRA-FR. Also, all patients from the MITRA-FR should have been hospitalized, at least, one time before joining the trial, but the COAPT never considered this as a prerequisite which favors the selection of a less evolved population. The possible inclusion of these cases added to the inclusion of patients with ventricles in a far more advanced stage of the disease and not too much mitral regurgitation may have been decisive and explain the results obtained.
• Follow-up time. In the COAPT, survival curves start to separate clearly after one year of follow-up. The MITRA-FR has one year of follow-up only. We may see a different progression of these patients in time, in this study, as follow-up goes on.
• Methodological aspects. The minimum primary endpoint of the MITRA-FR was a composite of rehospitalizations or death at one year. In order to analyze this variable, the Kaplan-Meier survival analysis was used which, even though is correct from the methodological point of view, it presents us with one problem: rehospitalization is an event that may occur in time but this analysis does not take it into consideration. This is especially important here since one of the problems of this type of patients is the number of rehospitalizations. Being hospitalized once or five times a year is certainly not the same. This is something that the MITRA-FR did not pay attention to. On the other hand, the COAPT was actually designed to analyzed recurring events. As a matter of fact, the COAPT primary endpoint was the number of rehospitalizations at 2 years (not only if the patient was admitted or not). The COAPT included a composite of death/rehospitalizations as a secondary endpoint, but the way it analyzed this varies from the MITRA-FR. The events death and hospitalization are not exclusionary, but one is more relevant than the other. In the MITRA-FR, both events were considered the same and the only thing that would cancel the survival analysis was suffering from one or the other, whichever would come first. However, the COAPT had a more appropriate way to analyze this type of intercurrent events: the WIN ratio (the win/loss ratio in the treatment group). The WIN ratio analyzes the most important clinical event (death) giving it relevance even though it may have occurred after hospitalization. This way of analyzing the composite event shows differences that the traditional Kaplan-Meier model does not detect or does not detect so sharply.

Finally, we have to say that the COAPT was a more rigorous study than the MITRA-FR. In the French study, 43 patients were excluded from the protocol in the MitraClip group due to several reasons which amounts to 28% of the sample in the intervention group. Also, the numerous losses of secondary variables such as quality of life, analytical values, functional class and, most surprisingly, control echocardiograms, leaves these events un-scrutinized in the study since they may be prone to bias. In this sense, the percentage of significant FMR relapses at one year is an approximate estimate and, therefore, inaccurate.

Q.: The virtues of the COAPT trial have been praised and the limitations from the MITRA-FR brought to everyone’s attention but, in your opinion, which would be the most positive aspects of the MITRA-FR and the most negative aspects of the COAPT?

A.: The positive aspects of the MITRA-FR are that it is the very first study to conduct a rigorous analysis on this issue by teaching us that choosing the optimal therapy for the management of heart failure and optimizing the maximum doses significantly improves patients and reduces their FMR (let’s not forget that we’re dealing with a dynamic process that can change in time). Also, maybe its most positive aspect is that it tells us what patients shouldn’t probably be eligible for this therapy: patients in advanced stages with not too much mitral regurgitation and without an optimized medical treatment. This subgroup of patients may have to be overlooked.

The negative aspects of the COAPT are that it was a lab experiment: all the variables were perfectly controlled, all treatments to their maximum doses, and the MitraClip device with an almost absolute success at 2 years. It will probably be difficult to replicate all this in the real world, but there is no doubt that it is the perfect example that FMR kills and that correcting it may lead to a substantial reduction of cardiovascular events. The COAPT trial sets the ideal we should aspire to in real life.

Q.: How do both studies complement each other to define what the ideal candidate for this technique really looks like?

A.: The interesting thing about these studies is that they should be implemented together to set the foundations of how we should choose the candidates. It seems obvious that we have to forget about patients in advanced stages (greater ventricular dilation), non-severe FMR and without optimal treatments. If we want to have more positive results, our candidates need to be in the early stages of the disease, have a significant degree of FMR (that really contributes to the clinical situation), be perfectly treated, and have good results with the device (which means that the anatomical selection and the experience of the interventional team need to be high). However, we always have to be cautious with assumptions like these. These data show the «average» patient, but a COAPT trial subanalysis reveals that as long as the FMR is very relevant (EROA > 30 mm2) there will always be a benefit, regardless of the degree of ventricular dilatation. That’s why every case should be treated individually, and in my opinion, we should be very serious about the amount of FMR. On top of reducing major events, MitraClip also improves quality of life and functional class. These are goals that we should take into account when selecting patients because this may be the only therapy that will alleviate very advanced symptoms.

Q.: In the last TCT we also saw the results from a smaller clinical trial, the REDUCE-FMR, that evaluated a system of percutaneous annuloplasty. Given the results of MitraClip, which could be the role of percutaneous annuloplasty systems in patients with heart failure and severe FMR?

A.: In the COAPT study, the results from the clip were excellent because the anatomical selection of the candidates was excellent as well. However, in the real world we won’t find cases like this all the time. In my opinion, in patients with great annular dilatation, so big that will prevent leaflet coaptation, the clip may have suboptimal results. It is in these cases where the role played by percutaneous annuloplasty systems may be essential - alone or in combination with the clip. According to the COAPT, the idea is that we should leave as little FMR as possible and that repair should be long-lasting, because this is what will eventually lead to less cardiovascular events. Percutaneous annuloplasty systems will complement our therapeutic arsenal to achieve this goal. However, to be taken into consideration, they need to show the same safety and efficacy profile as MitraClip.

Conflicts of interests

R. Estévez-Loureiro is proctor for MitraClip and declares to have received a research grant from Abbott Vascular.

Bibliografía

QUESTION: There is no doubt that the most significant advances made back in 2018 in interventional cardiology were the long-awaited results from clinical trials with MitraClip (Abbott Laboratories, Abbott Park, Illinois, USA): the MITRA-FR (Multicentre Study of Percutaneous Mitral Valve Repair MitraClip Device in Patients With Severe Secondary Mitral Regurgitation) presented at the congress organized by the European Society of Cardiology, and the COAPT (Cardiovascular Outcomes Assessment of the Mitra- Clip Percutaneous Therapy for Heart Failure Patients With Functional Mitral Regurgitation) presented at the Transcatheter Cardiovascular Therapeutics (TCT) congress. Both trials have been discussed extensively. Could you please tell us what the basic differences are in the results obtained by these 2 trials?

RESPUESTA: That’s right, both trials have put the spotlight on mitral regurgitation (MR) as the therapeutic target in patients with heart failure and reduced ejection fraction (HF-REF). The COAPT trial randomized 614 patients with HF-REF (left ventricular ejection fraction between 20% and 50%) and moderate-to-severe HF treated with the optimal medical therapy in order to follow 2 therapeutic strategies (1:1): a) optimal medical therapy, or b) optimal medical therapy plus percutaneous implant of MitraClip.1.After 2 years there was a 47% reduction in the risk of the primary endpoint of efficacy (hospitalizations due to heart failure) in the intervention group. Similarly, there was a significant drop in allcause mortality (hazard ratio, 0.62; 95% confidence interval, 0.46-0.82; P < .001) and other secondary endpoints1. En cuanto al objetivo primario de seguridad, se confirmó un número relativamente bajo de complicaciones relacionadas con la técnica a los 12 meses de la intervención. El número necesario de pacientes a tratar fue de 3,1 para evitar una hospitalización y de 5,7 para evitar una muerte. Estos datos de eficacia clínica son muy superiores los observados con la inmensa mayoría de los fármacos de contrastada evidencia en la ICFER, y comparables a los de otras medidas como la administración de antibióticos en enfermedades infecciosas o el implante percutáneo de válvula aórtica en la estenosis aórtica grave. When it comes to the primary endpoint of safety, a relatively low number of complications due to the technique 12 months after the intervention was reported. The number of patients needed to treat to avoid hospitalization was 3.1 and 5.7 to avoid mortality. These data of clinical efficacy are far more superior than the ones seen in most drugs with confirmed effectiveness in the management of HF-REF and similar to those from other measures such as the administration of antibiotics for the management of infectious diseases or the percutaneous implant of aortic valve prostheses for the management of severe aortic stenosis.

However, this fairy tale is shadowed by the almost simultaneous publication (nearly a month earlier) of the results from the open randomized clinical trial MITRA-FR that evaluated the efficacy and safety of MitraClip device in 307 individuals with HF-REF and moderately severe secondary MR2. In this trial, compared to the standard treatment, MitraClip did not improve the risk of the composite endpoint of all-cause mortality or admission due to heart failure at 12 months2.

A priori, both trials included patients with HF-REF and moderately severe secondary MR. Also, we should not forget that the baseline characteristics of the individuals included in both studies are similar in significant clinical and risk features. However, we should not take into consideration certain echocardiographic differences, results from the intervention, medical treatment, and follow-up time that may be behind these conflicting results:

• Echocardiographic differences: this is one of the key issues when it comes to interpreting both trials. As a matter of fact, the MR effective regurgitant orifice area are was lower in the MITRA-FR compared to the COAPT (31 ± 10 mm² versus 41 ± 15 mm²). Similarly, the left ventricular end-diastolic volumes were higher in the MITRA-FR compared to the COAPT (135 ± 35 mL/m² versus 101 ± 34 mL/m²). This suggests that the patients who may benefit the most from this intervention are those with a higher degree of valvular dysfunction and less left ventricular dilatation; in other words, patients with predominant valvular heart disease over left ventricular disease.
• Differences in the results obtained with the intervention: the COAPT trial says that the frequency of perioperative complications or suboptimal results immediately or 12 months after the intervention was much more inferior compared to the MITRA-FR trial. For instance, in the COAPT trial only 5% of the patients showed grade ≥ III MR after 12 months of followup,2,while this percentage rose to 17% in the MITRA-FR trial2. In this sense, and even though in both studies the number of participating centers was high, it would have been useful to know the efficacy and safety results based on the number of implants managed by each center.
• Differences in the pharmacological approach: at baseline there was a high percentage of patients treated with drugs with confirmed effectiveness in the management of HF-REF in both trials. However, in the COAPT trial, the number of patients treated at baseline with angiotensin -converting enzyme (ACE) inhibitors, angiotensin II-receptor antagonists, or sacubitril/valsartan was slightly superior in the intervention group. It is striking to see that these differences became more significant during follow-up. As a matter of fact, after 12 months, 13.4% more patients from the intervention group were receiving ACE inhibitors, angiotensin II-receptor antagonists, or sacubitril/valsartan. Similarly, after 12 months of follow-up, in the intervention group, the percentage of patients treated with beta-blockers was just 6.6% higher. The fact that it is an open trial may be indicative of performance bias (knowing the group the participants are assigned to in the trial has its echo in the form of systemic differences in care and therapies between both intervention groups) and detection bias (knowing the treatment group may affect important clinical decisions such as whether a patient should be hospitalized or not). Another aspect on medical treatment that we should also mention here is the lack of information on the absolute doses of the drugs used for the management of HF-REF. It is shocking to see that the COAPT trial, that required optimal medical therapy for the inclusion of patients, never mentions the absolute doses of the main groups of drugs used at the beginning or while the trial is being conducted (it only mentions relative changes in doses). Similarly, there is very little information on the intensity of clinical followup conducted or the administration of outpatient intravenous diuretic therapy (many hospital admissions can be avoided with close monitoring and intensification of diuretic therapy). Apart from the intervention per se, all these aspects may have tipped the scales towards one of the two groups of treatment, especially when it comes to the risk of hospitalizations.
• Different assessment times: it is important to emphasize that the results relative to the effectiveness of MITRA-FR were reported 12 months after the intervention compared to the results from the COAPT trial that were reported 24 months post-intervention. If we take a closer look into the results from both trials according to their time frames, we will see that the highest benefit from MitraClip in the COAPT study was observed after 12 months of follow-up. Actually, in the COAPT study, allcause mortality after 1 year did not change in either one of the two therapeutic strategies implemented (hazard ratio, 0.81; 95% confidence interval, 0.57-1.15). We should also mention here that follow-up in the COAPT study was longer in the intervention group, due not only to the higher mortality rate of the control group, but also to the higher number of patients who withdrew from the study medical treatment group (attrition bias).

Q.: The virtues of the COAPT trial have been praised and the limitations from the MITRA-FR brought to everyone’s attention but, in your opinion, which would be the most positive aspects of the MITRA-FR and the most negative aspects of the COAPT?

A.: In the first place, both are open randomized clinical trials not compared with placebo. From a general perspective, we should highlight the limitations and biases that may affect studies like these compared to double-blind clinical trials. Also, given the lack of sham procedures in both trials we cannot estimate the importance of the placebo effect in this context. In our interventional arena, we have the recent examples of the significant and dramatic difference of the results observed for renal denervation between open clinical trials and double-blind studies with a «sham procedure ».

The main negative aspects of the COAPT trial have already been mentioned above, but on this regard, we should also focus on the selection criteria used by this study that do not fully explain in detail what kind of rigorous selection of patients has been followed (out of the 1576 preselected patients, only 614 were randomized). As a matter of fact, the most common explanation to not be eligible after preselection (n = 419) was as vague as «incomplete preselection or others».

In general, I would say that the MITRA-FR has been a sort of comeback to reality after showing us that acting on the mitral valve does not seem to work in all patients with HF-REF and grade III-IV secondary MR. Once again, it has been confirmed that new advances in precision medicine need to be made while we look for the tools that will eventually allow us to better understand the heterogeneity of this complex syndrome and make better treatment selections for every particular case.

Q.: How do both studies complement each other to define what the ideal candidate for this technique really looks like?

A.: As we’ve already said, I think both studies show us that the ideal patient who would benefit the most from this technique is a patient on optimal medical therapy who remains symptomatic and who, in a situation of clinical stability, shows more severe MR (an effective regurgitant orifice area > 30 mm²) and has a not very dilated left ventricle. However, the adequacy of this profile will need to be confirmed in future studies.

Q.: One of the key differences between both trials was the degree of optimization of medical treatment achieved before randomization that influenced the frequence and magnitude of the therapeutic changes conducted during follow-up in both studies. Do you think it is possible to bring the level of adequacy or maximization of the COAPT trial to the routine clinical practice? How would this work from the organizational standpoint?

A.: This is an essential issue. Although in both studies the treatment before randomization could be considered appropriate, the optimization of medical therapy was more liberal in the MITRA-FR. Unfortunately, as we mentioned before, there is a lot of unavailable information on absolute doses, titration over time and intensity of follow-up. Knowing this missing information would shed light on the influence and adequacy of medical therapy in the results obtained by each study. What seems to be clear is that the optimization of medical therapy should be a prior condition before considering MitraClip a therapeutic alternative.

In sum, based on the results from both trials, it seems evident that to obtain satisfactory clinical outcomes a careful selection of patients is required while avoiding generalizing the percutaneous management of secondary MR in most patients with HF-REF. Right now, we run the risk that an inadequate selection of candidates will lead to questionable results. On the one hand, the excessive enthusiasm from some interventional teams that wish to approach MR with percutaneous treatments and, on the other hand, the clinicians’ quest for new therapeutic alternatives in very advanced patients with poor clinical progression may threaten the successful implementation of percutaneous interventional programs for the management of MR in patients with HF-REF.

Luckily, we’ll soon be getting new results that will tip the scales of this war on either side and, eventually, will allow us to make precise selections of the patients who are real candidates to undergo interventional procedures with MitraClip. In the meantime, healthcare providers (imaging technicians, heart failure specialists, and interventionists) should come to terms and set the foundations for the implementation of machines and programs that will help make reasonable and quiet assessments of each particular case. As the old Spanish proverb goes «haste makes waste».

Funding

This study has been funded by CIBERCV 16/11/00420.

Conflicts of interests

J. Núñez declared receiving fees from Abbott for his lectures.

References