Author + information
- Frank Corrigan1,
- John Chen1,
- John Lisko1,
- Norihiko Kamioka2,
- Shawn Reginauld3,
- Jose Miguel Iturbe2,
- Patricia Keegan4,
- Vinod Thourani5 and
- Vasilis Babaliaros1
- 1Emory University School of Medicine, Atlanta, Georgia, United States
- 2Emory University, Atlanta, Georgia, United States
- 3Georgia Institute of Technology, Atlanta, Georgia, United States
- 4Emory Healthcare, Grayson, Georgia, United States
- 5Emory University Hospital Midtown, Atlanta, Georgia, United States
The effects of hemodynamic changes during percutaneous mitral valve repair (PMVR) with MitraClip (Abbott Vascular) are incompletely characterized. We compared periprocedural invasive and echocardiographic hemodynamic changes with clinical outcomes at 30 days and 1 year. We hypothesized that periprocedural improvement of pulmonary venous (PV) waveforms are predictive of improved outcomes.
We retrospectively reviewed records and periprocedural transesophageal echocardiograms of 121 consecutive patients (age 76±12 years) from 5/2013-1/2017 who underwent PMVR for mitral regurgitation (MR) at our institution. Six patients were excluded as a clip was not placed. The waveform of the “worst” PV (right upper or left upper) was assessed for periprocedural improvement in morphology recorded as systolic reversal, systolic blunting, or normalization. We measured change in MR grade by color Doppler (4+ to 1+ scale) and invasive changes in left atrial pressure (LAP) and V-wave. We analyzed cardiac rehospitalization and all-cause mortality at 30 days and 1 year by bivariate logistic regression and survival by Kaplan-Meier analysis.
During the procedure, LAP and V-wave reduced by mean 4 and 15 mmHg (p <.001). MR grade improved from 3.8 to 1.6 (p <.001). 93 cases (80%) had PV waveforms before and after clip placement sufficient for analysis of which 67 (73%) demonstrated improvement in PV morphology while 25 (27%) did not. 5 patients were deceased after 30 days and 26 after 1 year. By multivariable analysis controlled for preprocedural predictors outcomes (B-natriuretic peptide, serum creatinine, and history of hypertension), improvement in PV morphology predicted cardiac rehospitalization at 30 days and 1 year (OR= .058, p= .005; OR= .178, p= .043) and improved mortality at 1 year (OR= .078, p= .016). Reduction in MR grade predicted 30-day cardiac rehospitalization (OR= .344, p= .014). Reduction in V-wave predicted 1-year cardiac rehospitalization (OR= .949, p= .021). Periprocedural assessment by color Doppler and invasive hemodynamics were not associated with mortality. Kaplan-Meier analysis demonstrated improved survival in those with procedural improvement in PV waveforms (mean survival 70 vs. 35 months, p <.001, Figure 1).
Pulmonary venous waveforms are important hemodynamic markers of procedural success during PMVR and may provide more consistent assessment than color Doppler or invasive hemodynamics. Assessment of periprocedural PV waveforms may predict cardiac rehospitalization and mortality after PMVR. A larger, multicenter cohort will be important for further assessment.
STRUCTURAL: Valvular Disease: Mitral