Surgical mitral valve (MV) repair to create a double orifice valve was first performed by Alfieri in 1991 (1- 3). A recently introduced MV repair method uses a clip (MitraClip, Abbott Vascular, Menlo Park, California) in a transcatheter closed-chest intracardiac procedure that simulates the Alfieri technique (4). Data on 107 patients of nonrandomized patients enrolled in EVEREST I and in the roll-in phase of EVEREST II demonstrated acute reduction to ≤2+ mitral regurgitation (MR) at discharge in approximately 75% of patients (5- 6).

One of the concerns with surgical correction of MR has been the potential increase in left ventricular (LV) afterload owing to elimination of the low-impedance regurgitant flow into the left atrium (LA). An increase in afterload associated with correction of MR could impair LV systolic performance (7- 9) and may result in an acute post-operative low cardiac output (CO) state (9- 13). The significance of this phenomenon in patients with surgically corrected MR by valve replacement or repair is controversial because the consequences of MR elimination are confounded by numerous factors owing to the nature of open heart surgery and the effects of cardiopulmonary bypass and cardioplegic arrest.

Because the MitraClip procedure does not require cardiopulmonary bypass, it provides a unique opportunity to assess the acute hemodynamic changes solely attributable to reduction of MR. The objective of this analysis was to characterize the impact of reduction in MR with the MitraClip therapy on intraprocedural CO and hemodynamics.

We studied 107 patients, 55 treated in the EVEREST (A Study of the Evalve Cardiovascular Valve Repair System Endovascular Valve Edge-to-Edge Repair Study) phase I feasibility trial (14), and 52 roll-in patients treated in the EVEREST II pivotal trial, who represent the pre-randomization experience.

Patient characteristics, echocardiographic protocol, and procedural technique have recently been described (6,14- 15). Symptomatic patients with moderate to severe (3+) or severe (4+) MR or asymptomatic patients with the same MR severity and abnormal LV systolic function (left ventricular ejection fraction [LVEF] <60% or LV end-systolic dimension >40 mm) underwent MitraClip treatment in accordance with the ACC/AHA guideline indications for MR surgery (16). As shown in (Table 1), MR was graded according to the criteria of the American Society of Echocardiography (17). Three of the 6 criteria listed in (Table 1), at least one of which was quantitative, had to be present for classification of MR as severe or moderate to severe.

The procedure was performed under general anesthesia using transesophageal echocardiography guidance and fluoroscopy. Vascular access was established from the femoral vein. The MitraClip device was advanced following an echocardiography-guided transseptal approach to the LA and across the MV to the LV. The arms of the clip were oriented perpendicular to the line of leaflet coaptation, the leaflets were grasped, and the clip was closed while reduction of MR severity was assessed by Doppler echocardiography. A second MitraClip device was placed if further reduction of MR was needed. Transthoracic echocardiographic (TTE) images were obtained using a detailed echocardiographic protocol after face to face training of sonographers in participating centers. All quantitative measurements including regurgitant volume, regurgitant fraction (RF), left ventricular end-diastolic volume (LVEDV), left ventricular end-systolic volume (LVESV), and forward stroke volume (FSV) as well as MR and TR grades from +1 to +4 were made at the core laboratory (University of California at San Francisco) (15).

All hemodynamic measurements were taken while the patients were under general anesthesia. Patients were studied according to the protocol before and within 15 min following attempted clip deployment regardless of procedural success.

Transducers were balanced by determining zero level at the mid-axillary line. LV peak end-systolic and peak end-diastolic pressures (LVEDP) were reported. Arterial pressures were measured from the femoral artery. Left atrial pressure was obtained from the transseptal catheter positioned in the LA or by measurement of pulmonary capillary wedge pressure (PCWP) using a balloon-tipped Swan-Ganz catheter. Wedge position was verified by fluoroscopy and phasic changes in pressure waveforms. The values of PCWP were assessed at end expiration with the balloon-tipped catheter at steady state with the patient in a supine position. Measurements were based on an average of 5 cardiac cycles. Cardiac output was determined by the Fick or thermodilution method depending on the preference of the site (18). A low CO state was defined as systolic blood pressure <90 mm Hg and cardiac index (CI) <1.4 l/min/m² after the procedure (12).

Statistical analysis was performed with SPSS (version 18.0, SPSS Inc., Chicago, Illinois). Continuous variables are presented as means ± SD. Categoric data are presented as absolute numbers or percentages. Hemodynamic data before and following clip deployment were compared using paired t test. Between-group comparison was completed using independent t tests. The Levene's test was performed to assess homogeneity of variance. Chi-square test was used for comparison of categoric data. The p values were not adjusted for the multiple comparisons.

We analyzed hemodynamic data from 107 patients. The patient demographic and baseline clinical data are shown in (Table 2). New York Heart Association (NYHA) functional class III or IV symptoms were present in 46% of patients. Etiology of MR was degenerative in 84 (79%) and functional in 23 (21%) patients.

(Figure 1) demonstrates a flow chart of patients with attempted MitraClip implantation. One or two MitraClips were implanted in 96 (90%) patients.

For the entire study cohort (n = 107), MR grade decreased significantly from 3.3 ± 0.7 to 1.7 ± 0.9 (p < 0.001) and MR was reduced to ≤2+ in 79 (74%) patients (Tables 3, 4).

The CO increased from 5.0 ± 2.0 l/min to 5.76 ± 1.9 l/min (p = 0.003) and an increase in CI increased from 2.7 ± 1.0 l/min to 3.0 ± 1.0 l/min (p = 0.003) (Table 3). Improvement in CO and CI correlated with having a lower baseline CI prior to successful reduction in MR with the MitraClip (r = 0.5, p < 0.001). In patients with APS, mean systemic systolic arterial pressure increased from 109 ± 20 mm Hg to 113 ± 21 mm Hg (p = 0.049); however, diastolic and mean pressures remained unchanged. This hemodynamic response with an increase in mean CO and unchanged mean systemic arterial pressures resulted in a significant reduction in calculated systemic vascular resistance (SVR) from 1,253 ± 529 dyn·s/cm⁵ to 1,058 ± 475 dyn·s/cm⁵ (p < 0.001). The LVEDP decreased from 11.0 ± 8.6 mm Hg to 8.8 ± 6.0 mm Hg (p = 0.016), and RA pressure increased slightly from 8.31 ± 4.7 mm Hg to 9.3 ± 5.6 mm Hg (p = 0.003). There was no change in PASP, mean PCWP, and pulmonary vascular resistance (PVR). In addition, there was a nonsignificant opposite trend from 20.9 ± 12.8 l/min/m² to 18.7 ± 8.5 mm Hg, p = 0.098). None of the study patients developed acute low CO state after transcatheter MV repair.

TTE at discharge demonstrated an increase in FSV (57 ± 17 ml vs. 65 ± 18 ml, p < 0.001) and a reduction in regurgitant volume 52 ± 20 ml vs. 30 ± 20 ml, p < 0.001, and in RF 46 ± 14% vs. 30 ± 15%, p < 0.001. In addition, there was a reduction in LVEF (60 ± 8% vs. 56 ± 9%, p < 0.001) associated with a reduction in LVEDV (172 ± 37 ml vs. 158 ± 38 ml, p < 0.001), while there was no change in LVESV (p = NS).

Ninety six (90%) of the 107 patients who underwent implantation of a MitraClip experienced reduction of MR grade of the same magnitude as entire study cohort, in this group of patients with a MitraClip, MR grade reduced from 3.3 ± 0.7 l/min to 1.6 ± 0.8 l/min, p < 0.001.

There was an increase in CO from 5.0 ± 2.0 l/min to 5.6 ± 1.9 l/min, p = 0.003 and CI from 2.6 ± 1.0 l/min to 3.0 ± 1.0 l/min, p = 0.002, an increase in systemic systolic BP (109 ± 20 mm Hg vs. 114 ± 20 mm Hg, p = 0.034), and an associated reduction in SVR (1,259 ± 531 dyn·s/cm⁵ vs. 1,059 ± 479 dyn·s/cm⁵, p < 0.001). The LV EDP decreased (10.8 ± 8.8 vs. 8.8 ± 6.0 mm Hg, p = 0.026), while RA pressure increased (7.9 ± 4.7 mm Hg vs. 9.1 ± 5.7 mm Hg, p = 0.02). Similarly, there was a trend for reduction in the ventricular “v” wave of PCWP tracing from 20.1 ± 12.2 mm Hg to 17.9 ± 7.6 mm Hg, p = 0.098 and no significant change in the remainder of the hemodynamic variables.

By TTE, clip implantation resulted in increase in FSV from 58 ± 16 ml to 66 ± 18 ml, p < 0.001, a reduction in regurgitant volume from 51 ± 20 ml to 27 ± 17 ml, p < 0.001, and RF from 46 ± 14% to 28 ± 15%, p < 0.001. There was a reduction in LVEF from 60 ± 8% to 56 ± 9%, p < 0.001, associated with decrease in LVEDV from 171 ± 37 ml to 157 ± 37 ml, p < 0.001, while LVESV remained unchanged.

Eleven of 107 patients did not undergo placement of MitraClip. Most of the patients who did not receive a MitraClip did not have repeat hemodynamic measurements, thus, invasive hemodynamic data is only available from a small number of patients (3 patients and CO data from only 1 patient who did not receive a MitraClip). Similarly, echocardiographic stroke volume data was only available in 6 of 11 patients who did not undergo MitraClip procedure. As expected, MR grade did not change significantly in this subgroup of patients without the MitraClip (3.4 ± 0.55 vs. 3.4 ± 0.55, p = 1.0) and similarly, no significant change was noted in the FSV (51.0 ± 9.1 vs. 54.5 ± 9.3, p = 0.14) as well as LVEF (58.5 ± 90 vs. 59.3 ± 5.6, p = 0.74). The percent delta change in FSV did not reach statistical significance between the groups with the MitraClip vs. without MitraClip (18.2 ± 29% vs. 1.5 ± 11%, p = 0.168). No comparison between group changes was made for parameters if data was missing in majority of patients who did not receive MitraClip.

This is the first study to document the acute hemodynamic effects of MitraClip treatment and of mechanical reduction of MR under beating-heart, closed-chest conditions in humans. Several important and unique findings were identified.

First, MitraClip treatment with a reduction of MR severity resulted in an increase in CO and FSV with concomitant reduction in SVR. Second, MitraClip treatment and reduction in MR grade resulted in favorable changes in LV loading conditions manifested by a decrease in LVEDP and LVEDV.

In chronic severe MR, high pre-load and reduced afterload (provided by low-impedance regurgitant flow into the left atrium) result in a high total CO. However, the magnitude of effective forward CO for any given myocardial contractility, LV pre-load, and heart rate has been thought to be primarily determined by the SVR. MitraClip treatment, which reduces the low-impedance mitral regurgitant flow into the left atrium, results in an immediate change in this relationship and leads to an increase in forward flow (CO, SV) and concomitant decrease in SVR, resulting in improvement or near normalization of CO.

Less invasive treatments such as the MitraClip therapy appear to be particularly beneficial in patients with an increased risk for developing a low CO state following MV surgery. Although surgical correction of MR has been used successfully for 4 decades (19- 20), it is commonly associated with a decrease in LV function and not infrequently an acute low CO state ensues (12). An acute low CO state in the immediate postoperative period is associated with an increased risk of death and major complications such as renal failure (10- 13). Causes of diminished post-operative cardiac function following surgical MR correction have been associated with myocardial injury and depression of myocardial contractility owing to the use of cardiopulmonary bypass, cardioplegic arrest (21), and interruption of annular-chordal-papillary muscle continuity in nonvalve-sparing valve replacement (22- 24). Over the last 40 years, several strategies have been implemented to decrease the prevalence of low CO syndrome, including revising the indications for MV surgery, with earlier intervention recommended even in the asymptomatic patient; improving cardioplegia solutions and myocardial protection; wider application of MV repair versus replacement; and subvalvular apparatus preservation whenever repair is not possible. Although the progress in operative care and surgical procedures have resulted in considerable improvement in peri-operative morbidity and mortality, the incidence of an acute CO state in selected subgroups of patients is as high as 12% to 19% (12,25). This risk is highest in elderly patients and in those with a history of congestive heart failure. In this trial of MV repair with the MitraClip system, no patient developed a peri-procedural low CO state. Moreover, 53% of our patients were age 70 years or older and 46% had advanced congestive heart failure symptoms (NYHA functional class III to IV). These findings suggest that closed-chest, off-pump MV repair on a beating heart carries a lower risk of developing a low CO state than previously reported with traditional MV repair and replacement surgery.

It has been argued that an additional mechanism of reduced LV systolic function following surgical MR reduction is the increase in afterload due to the elimination of the low-impedance regurgitant flow into the LA. However, the significance of this phenomenon has been controversial. Earlier publications have promulgated the concept of an acute increase in global LV afterload after MV replacement or repair as a cause of LV impairment (7- 8), whereas some more recent reports have demonstrated that changes in afterload are not uniform and in fact that afterload in some patients may even decrease as a result of surgical correction of MR (26- 27). We propose that the acute hemodynamic consequence of MR reduction is enhancement of FSV and CO and maintenance of mean arterial pressure while the changes seen in SVR are an adjustment to these hemodynamic changes.

The distinct effect of MitraClip treatment and its MR reduction on LV and LA hemodynamics are also intriguing. LVEDP and LVEDV after clip implantation fall, suggesting that the magnitude of reduction in regurgitant flow was substantial and sufficient to result in LV unloading.

Although reduction of regurgitant flow would be anticipated to have similar effects on LA and LV pressure there was no decrease in mean PCWP, and the PCWP “v” wave showed only a trend for reduction. The stability of the PCWP can be explained by the mild transmitral diastolic gradient associated with the creation of the double orifice valve. We previously reported that, following MitraClip treatment, the MV area was reduced from 6.0 ± 1.3 cm² to 3.6 ± 1.2 cm² and there is a small increase in the MV diastolic gradient from 1.7 ± 0.9 mm Hg to 4.1 ± 2.2 mm Hg by TTE Doppler (1.2 ± 6.8 days following MitraClip insertion) (28).

The LVEF decreased post-MitraClip by approximately 4 percentage points. This observation is not unique for MitraClip treatment as a reduction in LVEF of comparable magnitude is commonly observed following surgical treatment MR and is attributed to reduction of retrograde flow to the left atrium (29). Despite the reduction in total LVEF, the effective LVEF has increased as evidenced by an increase in forward stroke volume. Of note, the LVESV did not change, but LVEDV showed a statistically significant reduction, so that LVEF decreased slightly but not clinically significant. It is likely that the lack of change in the LVESV reflects preserved LV function and that LVEDV reduction is associated with the reduction in MR and consequently less transmitral flow filling the LV, resulting in a smaller LVEDV. The stroke volume goes up because more blood is ejected forward due to reduction in MR. The small rise in right atrial pressure may be related to intravenous hydration during the procedure, which is performed under general anesthesia. However, a contribution from the newly created left to right shunt after the transeptal puncture, which is requisite for this procedure, cannot be excluded.

The MitraClip therapy provides a unique model to study the acute hemodynamic changes secondary to MR reduction in isolation from the other changes that occur with open heart surgery. This allows for the assessment of the pure hemodynamic consequences of reduction of regurgitant flow. Patients who undergo the MitraClip procedure were not subjected to the hemodynamic insults related to thoracotomy, cardioplegic arrest and cardio-pulmonary bypass, well known to cause transient ventricular dysfunction and abnormal septal motion (30- 31). Additionally surgery is often associated with major changes in loading conditions (32- 33), rhythm disturbances (34- 35), the use of vasoactive medications in the post-operative period, and post-operative bleeding.

All hemodynamic and CO data were obtained under general anesthesia before and after MitraClip placement. It is not possible to completely dissociate the effects of anesthesia from the effects of clip implantation on the hemodynamics. The role of intravenous hydration during this procedure on patients who had been fasting overnight and possibly became relatively hypovolemic at the beginning of the procedure is uncertain. It is, however, possible that in hypovolemic patients, intraprocedural intravenous hydration accounted in part for the rise in mean PCWP and systolic PA pressure following successful reduction in MR. Another limitation of this study is that stroke volume data was not available at the time of cardiac catheterization. Heart rate data for the purpose of the original EVEREST study were not required to be collected at the same time as determination of CO. Consequently, stroke volume could not be reliably calculated retrospectively from the nonsimultaneous CO and heart rate data. Therefore, the FSV data were obtained from the screening and pre-discharge TTE studies. However, it is reassuring that the effect of MitraClip treatment on FSV by the echo Doppler parameters and CO by thermo dilution or Ficks method are concordant. Namely, with post-MitraClip implantation, there is an increase in both CO and FSV.

The reduction in the regurgitant volume achieved by MitraClip treatment results in an acute augmentation in forward CO and FSV and maintenance of mean systemic arterial blood pressure. In addition, MitraClip treatment promotes LV unloading by reducing LVEDP and LVEDV. These hemodynamic effects are likely to reduce the risk of acute low CO state immediately after reduction in MR with the MitraClip device. These findings may have especially important implications for the use of the MitraClip therapy in patients with a high risk for a post-operative low CO state.