Author + information
- Received February 5, 2013
- Revision received April 21, 2013
- Accepted May 14, 2013
- Published online December 24, 2013.
- Rodrigo Estévez-Loureiro, MD, PhD∗,
- Olaf Franzen, MD†∗ (, )
- Reidar Winter, MD, PhD‡,
- Lars Sondergaard, MDSc†,
- Per Jacobsen, MD‡,
- Gary Cheung, MD†,
- Neil Moat, MS∗,
- Nikolaj Ihlemann, PhD†,
- Matteo Ghione, MD∗,
- Susanna Price, MD, PhD∗,
- Alison Duncan, MD∗,
- Tine Streit Rosenberg, RN†,
- Sarah Barker, MSc∗,
- Carlo Di Mario, MD, PhD∗ and
- Magnus Settergren, MD, PhD‡
- ∗National Institute Health Research Cardiovascular Biomedical Research Unit, Royal Brompton Hospital, London, United Kingdom
- †Division of Cardiology, Rigshospitalet, Copenhagen, Denmark
- ‡Unit for Interventional Cardiology, Department of Cardiology, Karolinska University Hospital and Karolinska Institutet, Stockholm, Sweden
- ↵∗Reprint requests and correspondence:
Dr. Olaf Franzen, Department of Cardiology, Rigshospitalet Copenhagen, Blegdamsvej 9, 2100 Copenhagen, Denmark.
Objectives This study aimed to assess the clinical and echocardiographic results of MitraClip implantation in noncentral degenerative mitral regurgitation (dMR) compared with central dMR.
Background It is unknown whether the use of MitraClip therapy in noncentral dMR is as safe and effective as in central dMR.
Methods We analyzed a multicenter registry of 173 patients treated with the MitraClip and compared results of central and noncentral dMR.
Results Seventy-nine patients (age 79.2 ± 8.0 years, 58.2% men) had dMR. Forty-nine patients (62%) had central dMR, with the remainder classified as noncentral dMR (n = 30, 38%). Patients with noncentral dMR had a wider pre-procedural vena contracta (8.5 ± 2.0 mm vs. 6.9 ± 2.2 mm, p = 0.039) and higher systolic pulmonary pressure (57.9 ± 18.0 vs. 47.3 ± 13.0 mm Hg, p = 0.019). Procedural success was the same in both groups (95.5% central vs. 96.7% noncentral, p = 0.866). Post-procedural MR and New York Heart Association (NYHA) functional class at 1 month (MR ≤2, 96.0% vs. 96.6%, p = 0.866, and NYHA functional class ≤II, 81.6% vs. 90.0%, p = 0.335) and 6 months (95.2% central vs. 91.7% noncentral, p = 0.679; and NYHA functional class >II, 21.1% vs. 0%, p = 0.128) did not differ between groups. There were also no differences in serious post-procedural adverse events: partial clip detachment (central n = 1 [2.0%] vs. noncentral n = 1 [3.3%], p = 1.000), death (5.4% central vs. 13.0% noncentral, p = 0.298), or heart failure admission (10.8% central vs. 8.7% noncentral, p = 0.791).
Conclusions In experienced centers, MitraClip treatment can be performed safely and effectively in both central and noncentral dMR.
Severe mitral regurgitation (MR) is associated with a significant increase in morbidity and mortality (1–4). Mitral valve (MV) surgery is the standard of treatment for symptomatic patients with degenerative MR (dMR) (5). However, nearly one-half of patients (6,7) referred for surgery are not operated on, predominantly due to comorbidities, left ventricular (LV) dysfunction, or age (8). In these cases, catheter-based interventions have been proposed as an alternative treatment strategy. The MitraClip system (Abbott Vascular, Menlo Park, California) is such a therapy, which mimics surgical edge-to-edge MV repair (9–11). The EVEREST II (Endovascular Valve Edge-to-Edge Repair II) trial showed that the MitraClip device is less effective than surgery but safer. Only selected patients with central MR were included. Successful use of the device in a more diverse patient population with complex pathologies has been reported recently (12–14). However, little is known regarding applicability of the procedure in noncentral dMR.
The aim of this study was to compare short-term and mid-term safety and efficacy of MitraClip treatment between patients with central versus noncentral dMR.
Between August 2009 and November 2012, 173 patients were treated with the MitraClip device at the three centers: Royal Brompton Hospital (United Kingdom), Rigshospitalet (Denmark), and Karolinska University Hospital (Sweden). The current study involved a retrospective analysis according to pre-defined inclusion and exclusion criteria of all patients with dMR. The indication for the MitraClip was agreed upon by a multidisciplinary team comprising cardiologists, cardiac imaging experts, cardiac surgeons, and anesthetists. Inclusion criteria were severe dMR, and a predicted clinical benefit of MR reduction in patients judged to be a high surgical risk. Exclusion criteria were absent secondary chordal support of the prolapsing segment, flail gap >20 mm, active endocarditis, intracardiac thrombus, mitral stenosis, or anatomy unsuitable for septal puncture, device steering, or device placement.
Before intervention in all patients, transthoracic and transesophageal echocardiography (TEE) were performed and New York Heart Association (NYHA) functional class was assessed. During the procedure, the MV was examined with 2-dimensional (2D)- and 3-dimensional (3D)-TEE. Clip entanglement (needed manipulation of the system >120 s) was defined as a complication. Transthoracic echocardiography was repeated and NYHA functional class was reassessed at discharge, 1 month and 6 months. During follow-up, death from any cause and hospitalization due to heart failure were obtained from medical records.
MR severity was assessed by the individual center at baseline and follow-up using a comprehensive analysis of quantitative and semiquantitative echocardiographic parameters as recommend (15,16). Systolic pulmonary artery pressure was measured using the gradient derived from the maximal velocity of tricuspid regurgitation. Measurement of LV volumes and ejection fraction was performed according to Simpson’s biplane method (17). The origin of MR was determined using the color-compare mode in the 2D transesophageal commissural and LV outflow tract view, as well as the short-axis view, at the level of the MV. A 3D reconstruction of the regurgitant jet and MV was used to confirm the findings. Central MR (Figs. 1A and 1B) was considered if the MR originated from the central part of the line of coaptation (A2–P2) and noncentral MR (Figs. 1D and 1E) if it originated from the lateral or medial part of the line of coaptation (A1–P1 and A3–P3). Procedural success was defined as a reduction in the degree of MR to 2+ or less.
The MitraClip device consists of 2 8-mm clip arms and opposing grippers, which can be opened and closed against each other in order to grasp and coapt the leaflets at the origin of the regurgitant jet. The MitraClip procedure has been described elsewhere (9). Briefly, with the patient under general anesthesia, and by using fluoroscopic and 2D- and 3D-TEE guidance (18), the device was advanced via the transseptal route across the MV into the left ventricle. With the 2 arms of the clip extended, the device was retracted to capture, and subsequently closed to coapt, the MV leaflets. Standard mid-esophageal 2D views are not available for noncentral pathology. A 3D “enface” view of the MV was mandatory for alignment of the clip in the lateral or medial aspects of the curved line of coaptation (Fig. 1D). Nonstandard 2D views were used to monitor the closure of the clip arms and to verify good leaflet insertion (Fig. 1E). In order to compensate for the height increment of the system when advanced laterally in the left atrium, 5 mm was deducted from the recommended optimal septal puncture height for central pathology if lateral noncentral pathology was treated and 5 mm was added if medial noncentral pathology was treated. Retraction of the steerable guide in the right atrium was accepted if needed to treat medial pathology. If 2 clips were needed, it was preferred to implant the first clip in the most central aspect of the pathology. Resultant valve morphologies after clip implantation in central and noncentral pathologies are shown in Figures 1C and 1F.
Patients were extubated after the procedure in the catheterization laboratory and transferred to the intensive care unit or recovery room where they were monitored until fit for ward transfer.
Anticoagulation management was based on an individualized protocol. Patients on warfarin continued on an identical regimen after the intervention; in the remainder of patients, acetylsalicylic acid (75 to 150 mg/day) was prescribed for 3 months and clopidogrel (75 mg/day) for 4 weeks. Clinical and echocardiographic follow-up was performed at discharge and 1 and 6 months post-procedure.
Results are presented as mean ± SD for continuous normally distributed variables, as median (interquartile range) for continuous non-normally distributed data, and as percentages for categorical data. Analysis of normality was performed with the Kolmogorov-Smirnov and Shapiro-Wilk tests. Categorical data and proportions were compared using chi-square test or Fisher exact test, as required. Comparisons of continuous variables were analyzed using unpaired Student t test and the Mann-Whitney U test as appropriate. The Kaplan-Meier method was used for survival analysis and curves. Values of p <0.05 were considered significant. All analyses were performed with SPSS 20.0 statistical package for Windows (IBM SPSS, Chicago, Illinois).
Among the 173 consecutive patients who received MitraClip implantation between August 2009 and November 2012, 79 patients were found to have dMR. Of these, 49 (62%) had central dMR. The remainder were classified as noncentral dMR (n = 30, 38% [23 patients A3/P3 and 7 patients A1/P1]). Baseline characteristics of both groups were similar (Table 1).
Most patients were in NYHA functional class III or IV (noncentral 96.7% vs. central 91.8%, p = 0.644). There was a trend towards a greater severity in the quantitative parameters of MR in the noncentral group that reached significance with regard to the vena contracta. Systolic pulmonary pressure was significantly higher in the noncentral group (Table 2). The proportion of patients undergoing MitraClip implantations in noncentral positions has increased over time (Fig. 2).
Procedural success was obtained in 76 patients (96.2%), with no significant differences between central (n = 47, 95.5%) and noncentral dMR (n = 29, 96.7%, p = 0.866). We found no significant differences in the procedural time, although noncentral patients had slightly longer procedures (central 107.8 ± 61.5 min vs. noncentral 123.6 ± 63.2 min, p = 0.505). There was no significant difference in the number of implanted clips between groups, although numerically more patients with noncentral dMR received more than 1 clip (53.3% vs. 40.8%, p = 0.278). The number of complications (n = 10, 12.6%) was not significantly different between groups (Table 3).
There were no cases of prolonged clip entanglement, clip embolization, or septal complications. Two patients experienced gastrointestinal bleeding requiring transfusions (1 in each group). One patient with central dMR developed cardiac tamponade after the procedure that was successfully drained by pericardiocentesis. One patient needed temporary pacemaker implantation due to transient heart block (central group). Another patient developed pneumonia during admission that was successfully treated with antibiotics. One patient in the central group underwent elective MV surgery <30 days after the procedure due to severe MR. One patient (1.2%) died <30 days after the procedure (noncentral group, 81 years old, ejection fraction 25%, chronic kidney disease, logistic EuroSCORE 52.8%) due to complications of comorbidities.
Post-MitraClip echocardiographic evaluation: effect on MR
We observed a reduction in MR severity in all patients, with most of the patients having MR ≤2. There was no difference in the extent of reduction of MR between the central and noncentral groups (Fig. 3). The reduction in MR severity was sustained in both groups at 6 months (Fig. 3), with nonsignificant differences in the proportion of patients with MR >2 (4.8% central vs. 8.3% noncentral, p = 0.679).
At 1 month, there was overall a positive effect of the treatment with regard to LV dimensions and volumes; the ejection fraction was lower than before intervention. There were no differences between the central and noncentral groups with regard to changes of LV parameters (Table 2).
There was a significant increase of the mean transmitral gradient (p < 0.001) and decrease in the systolic pulmonary pressure (p < 0.001) 1 month after the procedure. There were no differences between the central and noncentral groups with regard to these 2 parameters. At 1 month, 24.8% of patients had a mean post-procedural transmitral gradient ≥5 mm Hg, without significant differences between the central and noncentral groups (26.5% central vs. 20% noncentral, p = 0.510).
At 1 month, NYHA functional class had improved (p < 0.001), with no differences between the central and noncentral MR groups (Fig. 4). There were also no significant differences at 6-month follow-up, although the proportion of patients with NYHA functional class >II was higher in the central group (21.1% vs. 0%, p = 0.128, 32 patients).
Data regarding all-cause mortality (8.3%) and re-admission due to heart failure (10%) were available in 60 patients (37 patients central and 23 patients noncentral, follow-up 15.2 ± 11.0 months). There were no differences between the central and noncentral groups with regard to mortality (n = 2 [5.4%] central vs. n = 3 [13.0%] noncentral, p = 0.298) and hospital re-admission (n = 4 [10.8%] central vs. n = 2 [8.7%] noncentral, p = 0.791). There were no significant differences between the groups with regard to death and readmission (log-rank p = 0.506). Survival curves are shown in Figure 5.
This study analyzed the results of 3 experienced centers using percutaneous edge-to-edge MV repair. Our results show for the first time to our knowledge the feasibility of treating noncentral dMR with the MitraClip. Reduction of MR, LV remodeling, and clinical events was comparable in central and noncentral dMR.
There has been an increasing interest in extending the use of the MitraClip system to patients that do not meet the inclusion criteria of the EVEREST trials (9–11). Good results in patients not amenable for surgery (19) have been reported, with a procedural success of 92% and 83% of the patients having MR ≤2+ at a median follow-up of 1 year. The mortality at 1 year was 25%. In patients with end-stage heart failure, similar results have been observed (13), with a procedural success of 94% and 87% of the patients having MR ≤2+ at 6 months. The mortality at 6 months was 16%. The MitraClip has been shown to be an acceptable alternative in nonsurgical patients (20). However, these studies have demonstrated the benefit of percutaneous repair in a certain subset of clinical profiles, but none of them have focused on noncentral MV pathology. In a recent article by George et al. (21), commenting on the results from the EVEREST II trial, it was stressed that the MitraClip would only be suitable for a selected subgroup of patients with central MR. In opposition to this, it has been suggested that in a real-world population with wider valve suitability criteria, results of the MitraClip therapy can be comparable to the EVEREST experience (12).
Medically treated patients with severe dMR have a 6% annual mortality rate (22). Surgical noncentral repair is considered more challenging and has been shown to be associated with increased risk for reoperation and MV replacement (23,24). Consequently, MitraClip therapy for MR jets originating from the medial or lateral aspect of the line of coaptation may also be more challenging and less safe. There is a chordae-free zone in the central part of the anterior leaflet. The structure of the chordae tendineae in the commissural areas is more complex (Fig. 6) and the anatomical orientation in 2D echocardiography is more difficult in the noncentral area. This may impose a higher risk for clip entanglement.
Our results show that the MitraClip can be safely implanted in a noncentral position of the MV, with a procedural success rate similar to central positioning. Noncentral implantation was associated with a sustained MR reduction, LV remodeling, and clinical improvement during follow-up, similar to previous reports of high-risk patients (12,19,20). There was no difference in the rates of mortality or readmission for heart failure between the groups with a risk profile comparable to other series (19).
These findings challenge the concept that noncentral positioning of a MitraClip is associated with worse clinical outcome because of interference with the subvalvular apparatus (25,26). There was no prolonged clip entanglement in the noncentral group. It suggests that the indication for MitraClip implantation may be expanded to noncentral dMR. However, certain practical issues should be considered with respect to noncentral dMR in this study. First, all patients underwent 3D-TEE study before the procedure because accurate valve analysis is key for successful mitral repair (27,28). In noncentral pathology, intraprocedural 3D-TEE was felt to be essential for correct clip orientation. Procedural changes must be considered in regard to septal puncture, guide catheter positioning, and grasping. Additionally, it needs to be stressed that the rate of noncentral MitraClip placement is increasing over time. This probably reflects a “learning curve” effect and a tendency to accept more challenging valve anatomy with growing experience.
The study is small in size, and a considerable number of patients were not available for follow-up at 6 months. Our results are the compilation of data from highly experienced centers and operators. MR grading is self-reported without use of a core laboratory.
MitraClip treatment of noncentral dMR is safe and associated with similar short-term and mid-term outcomes as the treatment of central MR. Our findings suggest that the valve suitability criteria for MitraClip therapy can be expanded to patients with noncentral origin of MR.
The authors thank Michael Schlueter, PhD, for his valuable comments and corrections.
Mr. Moat is a consultant and proctor for Medtronic. Dr. Price has an educational contract with Medtronic and is on the advisory board of Abbott. Dr. Di Mario has received speaker’s fees for Abbott Vascular. Dr. Settergren is a proctor for and on the advisory board of Abbott Vascular. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Estévez-Loureiro and Franzen contributed equally to this work.
- Abbreviations and Acronyms
- degenerative mitral regurgitation
- left ventricular
- mitral regurgitation
- mitral valve
- New York Heart Association
- transesophageal echocardiography
- Received February 5, 2013.
- Revision received April 21, 2013.
- Accepted May 14, 2013.
- 2013 American College of Cardiology Foundation
- Feldman T.,
- Wasserman H.S.,
- Herrmann H.C.,
- et al.
- Feldman T.,
- Kar S.,
- Rinaldi M.,
- et al.
- Auricchio A.,
- Schillinger W.,
- Meyer S.,
- et al.
- Lang R.M.,
- Bierig M.,
- Devereux R.B.,
- et al.
- Rudolph V.,
- Knap M.,
- Franzen O.,
- et al.
- Whitlow P.L.,
- Feldman T.,
- Pedersen W.R.,
- et al.
- George J.C.,
- Varghese V.,
- Dangas G.,
- Feldman T.E.
- Barabas M.,
- Cormier B.,
- Iung B.,
- et al.
- Van Mieghem N.M.,
- Piazza N.,
- Anderson R.H.,
- et al.