Predictable changes in left ventricular mass and function during ten years after valve replacement for aortic stenosis.

BACKGROUND AND AIM OF THE STUDY

Left ventricular (LV) hypertrophy is the underlying basis for longevity after aortic valve replacement (AVR) for aortic stenosis (AS). However, a detailed account of changes in LV mass and function in the long term after AVR and identification of the determinants of such changes have not yet been presented.

METHODS

Ninety-one unselected consecutive adult patients with AS underwent AVR and were followed up to 10 years, at which time 41 survivors without new mitral disease underwent repeat measurement of LV mass index (LVMi), ejection fraction (LVEF), fast filling fraction (LVFFF), and end-diastolic volume index (LVEDVi). A subgroup comprising 49 patients was also assessed at eight days, three months, and 1.5 years postoperatively. All measurements were analyzed in a longitudinal regression model for repeated measures.

RESULTS

LVMi fell from 202 +/- 58 g/m2 (n = 91) via 150 +/- 45 g/m2 (n = 39) at 1.5 years to 139 +/- 40 g/m2 (n = 41) at 10 years in all patients, and to 124 +/- 31 g/m2 (n = 29) in non-hypertensive patients. The LVMi falls were paralleled by improvements in LVEF and LVEDVi. LVFFF was not correlated to LVMi before the 10-year study. The longitudinal model indicated progressive reduction of LVMi to 1.5 years, but no change thereafter. The predictor variables were preoperative LVMi and end-systolic dimension index (high values of both related to high postoperative LVMi), hypertension, and male gender. The model for LVEF indicated a rapid increase to three months, followed by a slight decrease to 1.5 years and further to 10 years, predicted by preoperative LVEF and LVFFF. LVFFF fell sharply by three months, had recovered somewhat at 1.5 years and fully at 10 years, positively related to preoperative LVFFF and inversely to end-systolic chamber radius:wall thickness ration and small-sized prosthetic valves. LVEDVi converged from extreme values over time predicted by preoperative LVEF, but rose with hypertension and coronary artery disease. Hemodynamic function of the prosthetic aortic valve at any of the measurement times had no impact.

CONCLUSION

Changes in LV mass and function up to 10 years after AVR for AS were highly predictable. Poorer outcomes were related to preoperative excessive hypertrophy and indices of underlying irreversible myocardial disease and further compromised by hypertension and, to a lesser extent, coronary artery disease. The hemodynamic function of the aortic prosthetic valve did not seem to play a role.