Effects of surgical ventricular reconstruction on diastolic function at midterm follow-up
Received 12 June 2009; received in revised form 15 September 2009; accepted 22 October 2009. published online 28 December 2009.
Objective
Limited data are available on the effects of surgical ventricular reconstruction on diastolic function. The aim of the present study was to evaluate changes in diastolic function induced by surgical ventricular reconstruction at 2 time intervals after surgery (discharge and follow-up) and to assess the impact of diastolic changes on clinical outcome.
Methods
A total of 129 patients (65 ± 9 years, 14 women) underwent echocardiographic Doppler evaluation before surgical ventricular reconstruction, at discharge, and at follow-up (median 7 months). Patients with mitral regurgitation were excluded. Diastolic pattern was graded as follows: 0 (normal), 1 (abnormal relaxation), 2 (pseudo normalization), 3 (restrictive, reversible), and 4 (restrictive, irreversible).
Results
At follow-up, 28 (21.7%) of 129 patients showed a restrictive diastolic pattern (grade 3–4; group 1) and 101 did not (diastolic pattern grade 0–2; group 2). Preoperative and postoperative factors strongly associated with late diastolic restriction included sphericity index (higher in group 1), ventricular shape (nonaneurysmal shape more frequent in group 1), internal dimensions (greater in group 1), diastolic pattern (higher in group 1), ejection fraction (lower in group 1); left atrial dimensions (greater in group 1); mitral regurgitation rate (higher in group 1). At multivariate analysis the most powerful predictors of restriction were preoperative pseudonormalization of diastolic pattern (diastolic pattern 2) and septolateral dimensions (short axis). Overall, ejection fraction improved from 33% ± 9% to 40% ± 9% to 40% ± 9%; P = .001; end-diastolic and end-systolic volumes decreased (112 ± 41 to 73 ± 21 to 88 ± 28mL/m2, respectively; P = .001; and 77 ± 38 to 44 ± 17 to 52 ± 24mL/m2, respectively; P = .001); New York Heart Association class improved (2.4 ± 0.8 to 1.6 ± 0.6; P = .001).
Conclusions
Mild preoperative diastolic dysfunction (pseudonormalized pattern) and increased septolateral dimensions are independent predictors of diastolic restriction after surgical ventricular reconstruction.
aDepartment of Cardiac Surgery, IRCCS San Donato Hospital, Milan, Italy
bDepartment of Critical Care Medicine, University of Florence, Florence, Italy
cDepartment of Cardiothoracic-vascular Anaesthesia and ICU, IRCCS San Donato Hospital, Milan, Italy
Address for reprints: Marisa Di Donato, MD, Department of Cardiac Surgery, Policlinico San Donato, IRCCS, Via Morandi 30, 20097 San Donato Milanese, Milano, Italy.
ABSTRACT