HOW MUCH PROGRESS HAVE WE MADE IN UNDERSTANDING AND TREATING ISCHEMIC MITRAL REGURGITATION?
Ischemic mitral regurgitation (IMR) is the consequence of myocardial ischemia and is associated with increased mortality and morbidity, particularly when the left ventricle (LV) is impaired and in older and/or sicker patients. The degree of chronic mitral regurgitation after a myocardial infarction (MI) is an independent marker of severity of heart disease and a predictor of mortality. This has been well documented by Grigioni and colleagues in their seminal paper  showing only 29% survival at 5 years when effective regurgitant orifice was over 20 mm2.
Basically, there are two important steps to better understand and better treat patients with IMR. The first step is the analysis of the patient history of ischemic heart disease, the degree of LV dysfunction, and the dynamics of mitral regurgitation occurrence and progression. Analyzing this dynamic aspect of mitral regurgitation related to the LV functional and morphologic changes over time is critical and carries a predictive value for long-term benefit of surgery. The second step is the analysis of the severity and mechanism(s) of the mitral regurgitation in light of functional anatomy (Carpentier classification). Those two steps should confirm the indication for surgery, the likelihood of mitral valve repair, and the potential benefit of repair over mitral valve replacement.
From a surgical standpoint, analysis of the mechanism of mitral regurgitation is critical. How tethered is the posterior leaflet (Carpentier type IIIb)? How abnormal is the anterior leaflet motion? Is it false prolapse overshooting the posterior leaflet or just not reaching it? Is the mitral regurgitation jet central or eccentric? Is there apical displacement of both papillary muscles? How dilated is the LV? Is it superior to 65 mm in diastole, or 50 mm in systole? Is the LV dysfunction global or only regional? How scarred is the LV and how reversible is the LV dysfunction? Can further revascularization be beneficial and added to the mitral valve repair to facilitate LV remodeling?
At the present time, we do not really have a common algorithm for surgical management of IMR. The decision is mainly depending on the level of comfort of surgeons in understanding the mechanisms of mitral regurgitation and performing mitral valve repair. If controversy persists in patients with moderate IMR (grade 2+) regarding whether the mitral valve surgery concomitant with combined coronary artery bypass grafting (CABG) is better than CABG alone, I doubt that many cardiac surgeons would now perform CABG alone in patients referred for CABG with 3+ or 4+ mitral regurgitation. I also believe that moderate mitral regurgitation is less and less ignored at the time of concomitant CABG.
Undersizing annuloplasty, a concept introduced by Bolling in 1995, is indeed the most popular technique for mitral valve repair, although mitral valve replacement is frequently perceived and performed as a ‘safer’ alternative in most complex patients. We clearly see increasing repair rates in North America and improved results of combined mitral valve and CABG procedures, which were historically associated with poor outcomes. In the Society of Thoracic Surgeons (STS) database, including basically all US cardiac centers, the mitral valve repair rate has augmented from 51 to 69%. Looking at the added risk of mitral valve surgery when combined with CABG, the STS database shows for combined mitral valve repair and CABG surgery an operative risk of about 5%, which is 3% higher than isolated CABG. The combination of mitral valve replacement and CABG is associated with a two-fold higher risk than combined mitral valve repair and CABG, around 10%.
These numbers raise two questions. First, is the observed 3% difference the result of added cross-clamp time necessary to expose, analyze, and successfully repair the mitral valve? The answer should be no, although surgical expertise definitely impacts early results, and expert mitral valve centers have better performance and lower hospital mortality and morbidity. Second, does a prosthetic valve implant actually double the mortality of the surgery, compared with a mitral valve repair? Most likely, the answer is no, although repairing the valve should be more beneficial to LV function than replacing it, even if the replacement is carried out with subvalvular apparatus preservation.
The answers for both questions more likely reside in patient selection. IMR is commonly found in patients with advanced heart failure, having history of one or several MI, having previous stents or previous CABG, and presenting with morbid conditions such as diabetes, chronic obstructive pulmonary disease, and renal failure. These very sick patients definitely induce a bias in the decision, and one cannot blame a surgeon leaning toward a ‘safe’ replacement, avoiding the risk of an imperfect repair or a repair requiring intraoperative revision and prolonged operative time. It can actually be a sound decision in patients with severe IMR, long history of LV impairment, and previous CABG, for instance.
A recent study  from Mayo Clinic showed 1-year, 5-year, and 10-year survival of 84, 55, and 25%, respectively, with no influence of repair vs. replacement in patients with ejection fraction (EF) equal to or lower than 45%. The authors identified two distinct risk phases based on survival curve kinetics. Previous CABG, emergency, and low EF were the determinants at the early phase, while patient comorbidities were influencing late survival.
Indeed, IMR is actually about integrating the mitral valve problem into the overall patient problem before making a final decision on surgical indication and preferred surgical solution. Chronic IMR is a chronic disease of the LV with subsequent disease of the mitral valve which may decompensate further into left and right heart failure. We must also be looking at pulmonary hypertension, right ventricular distension and function, and subsequent functional tricuspid regurgitation.
Let us take the example of a patient with moderate to severe IMR (2–3+), low EF, and recent progression of symptoms with right heart failure, although the mitral regurgitation was unchanged on serial echocardiograms. In such a case, correcting the mitral regurgitation may have a modest overall impact and no long-term benefit while exposing to an immediate surgical risk. This patient could improve with aggressive medical management, and likely have mitral regurgitation reduction without surgery.
In this edition, Dr Klautz and colleagues provided us with their expertise on surgical correction of IMR with low EF. One of the messages delivered was the necessity to integrate more mitral valve and LV imaging in order to offer better solutions to patients with IMR. Indeed, referrals should not be made without a thorough echocardiogram assessment including transesophageal echocardiogram (TEE), three-dimensional TEE, and viability studies. For instance, mitral valve leaflet tenting height superior to 10 mm indicates that mitral valve repair with undersizing annuloplasty may yield suboptimal results with inadequate line of mitral valve leaflet coaptation. Further analysis of mitral valve tenting area, posterior leaflet angle, annular measurements, and detection of a ‘false’ anterior leaflet prolapse (seagull sign) is a part of the decision-making process and a predictor of immediate and late success of repair. It is unlikely that a mitral valve annulus measured less than 36 mm with large tenting area would benefit from restrictive annuloplasty. ‘A good replacement is better than a bad repair, and may be a good replacement is just as good as a good repair.’ We recently published  a propensity-based, case-matched analysis to determine whether patients who undergo mitral valve repair or mitral valve replacement for IMR have similar long-term outcomes. Long-term follow-up showed 23% recurrence of mitral regurgitation, but mostly limited to moderate 2+ mitral regurgitation, with only one severe recurrent mitral regurgitation and no significant difference in 5-year survival between mitral valve repair (79%) and mitral valve replacement (68%). Replacement was therefore more protective from recurrent mitral regurgitation, but a large majority of repair patients had no, mild, or moderate residual mitral regurgitation and did well. In our study, we performed edge-to-edge Alfieri technique when patients were showing false anterior leaflet prolapse. We usually placed two or three valve-to-valve PTFE sutures between P2–P3 and A2–A3 junctions, in order to adjust to the asymmetrical posterior dilatation of mitral valve annulus which is common after posteroinferior MI. This simple adjunct technique possibly had a protective effect for a subset of patients with more severe anatomy, without creating flow obstruction.
In a previous paper published in Current Opinion in Cardiology, it was pointed out that, in absence of level one evidence, mitral valve repair remained the operation of choice, but replacement could be preferred for select patients, particularly for high-risk and sickest patients, and the authors concluded there was a need for randomized prospective trials to compare directly those two techniques. These trials are underway.
What about catheter-based procedures? While surgery aims at eradicating more than reducing mitral regurgitation, catheter-based techniques resulting in 2+ residual mitral regurgitation are considered as ‘success.’ The Everest II trial investigated a catheter-based approach (MitraClip) mimicking the Alfieri technique. It was presented as a clinical success, although 50% of patients had at least moderate residual mitral regurgitation and 20% went to surgery for severe postprocedural mitral regurgitation, with a large number of patients requiring replacement, possibly due to the valve damage created by the clip. On the other hand, it may be a reasonable goal to reduce functional mitral regurgitation and symptoms in patients with advanced LV failure. Percutaneous mitral clipping should then be proposed as a heart failure solution instead of a true valve repair tool, and be randomized against medical treatment alone. Moreover, the Alfieri technique is recommended to surgeons in conjunction with annuloplasty for most cases. Experience has shown that ‘one technique does not fit all.’ Similarly, for surgery, undersizing annuloplasty may not address the diversity of mechanisms responsible for IMR, and may leave the mitral valve tethering and papillary muscle displacement unsolved. In this edition, Dr Klautz and colleagues support a more tailored approach and comment on the role of new surgical approaches beyond reductive annuloplasty. In this regard, various adjunct techniques may add benefit, such as external ventricular constraint devices, techniques addressing papillary muscle positioning, and leaflet approaches such as edge-to-edge or posterior leaflet augmentation and second chord cutting. Such techniques are conceptually looking beyond mitral valve annulus distortion. However, those ‘nonannular techniques’ are not yet well disseminated and they are performed in conjunction with undersizing annuloplasty, which remains the basis of surgical treatment.
Finally, there is an ongoing surgical debate between advocates of flexible rings and those of rigid rings, full vs. incomplete rings, and 2D-shaped vs. 3D-shaped rings, mimicking saddle-shaped annulus. To date, whatever annuloplasty device or technique surgeons have been using, the recurrence of at least moderate mitral regurgitation for mitral valve repair seems much higher in IMR than in organic mitral regurgitation, and several studies have shown that severe residual mitral regurgitation was associated with dismal 2-year survival.
Obviously, to better understand IMR or to better evaluate new technology, randomized studies are needed, as far as they are based on a standard and uniform surgical approach, which is not easy to obtain from investigators. Equally important could be prospective nonrandomized studies evaluating innovative techniques or new devices. Such studies should contain detailed longitudinal information on mitral regurgitation history as related to LV remodeling and preoperative vs. postoperative myocardial viability. A full set of preoperative and postoperative echocardiography data would be needed, regarding mitral valve annulus, leaflet tethering, and LV function and size. Patient follow-up should include serial echocardiograms simultaneously with longitudinal clinical follow-up. Future studies should also focus on clinical relevance of residual moderate mitral regurgitation, should separate IMR patients undergoing mitral valve repair alone from patients having mitral valve repair and revascularization (CABG or angioplasty), and should not be mixing IMR with nonischemic functional mitral regurgitation.
Finally, I would like to give a word of caution. Mitral valve repair for IMR is generally perceived as easier to perform than mitral valve repair for organic disease, although mitral valve repair in the context of IMR is far less successful than in degenerative disease. Adjunct techniques as mentioned above are often beyond usual surgical skills, and should not be tried by less experienced surgeons. Considering the complexity of the decision-making process and the complexity of IMR patients, innovative surgical approaches should be reserved to expert centers and teams having multidisciplinary expertise in valve disease and in heart failure management.
To answer my very initial question, ‘work in progress and a lot more needs to be done.’
Conflicts of interest
There are no conflicts of interest.