Login to view PhD Thesis

Enter your username and password here in order to log in on the website:

Forgot your password?

Reverse remodeling after aortic valve replacement in severe valvular aortic stenosis – effect of blockade of the angiotensin-II receptor.

Jordi Dahl



This ph.d thesis comprises three original papers based on a single randomized clinical trial conducted at the Department of Cardiology, Odense University Hospital, from the 1st of February 2006 to the 1st of December 2009. Patients with aortic valve stenosis have despite successful valve surgery increased mortality and morbidity compared to the general population. This increased risk most probably is associated with the presence of myocardial remodeling including LV hypertrophy and LV fibrosis. It is well established that the remodeling process is not fully reversible; LV hypertrophy is thus often present even after several years, leaving a permanent substrate for increased mortality and morbidity. In hypertension it is well documented that antihypertensive treatment per se and particularly RAAS inhibiting medication can reverse LV hypertrophy and reduce mortality. It has though never been shown if this also applies to patients with aortic valve stenosis. The scope of this study was; to assess if the level of preoperative remodeling has importance for the postoperative clinical outcome; to examine if RAAS blockade with candesartan was able to augment reverse remodeling after surgery; to evaluate if augmented reverse remodeling improves diastolic function and prevents development of atrial fibrillation.


The study was carried out as a single centre, consecutive, investigator initiated study using a prospective randomized blinded endpoint design. A total of 119 patients with symptomatic severe aortic valve stenosis and preserved LV systolic function referred for aortic valve replacement were included. Patients were randomized to either conventional management or candesartan treatment, candesartan treatment was unblinded, but all analyses were later analyzed blinded to treatment allocation. Patients underwent a clinical evaluation; echocardiography; 6 min. walk test and assessment of NT-proBNP the day prior to surgery and again 3, 6 and 12 month after surgery. In addition a 24-hour Holther ECG analysis was performed 3 month postoperatively and was repeated for 48-hours 12 month after surgery. A complete examination of medical records was performed June 2009 and all hospitalizations and deaths were recorded.


Preoperative severe left atrial dilation (LAVi≥40ml/m2) was present in 64 % of patients and was associated with left ventricular (LV) hypertrophy and increased filling pressure. Moreover preoperative LAVi was associated with persistent abnormalities in LV filling pressure and LV mass index at one year after surgery. Event free survival in patients with LAVi≥40ml/m2 at one year was 66% compared with 87% in patients with LAVi<40ml/m2, p=0.002. Also patients with preoperative increased E/e’-ratio and LV hypertrophy were at increased risk. In Cox regression analysis after correcting for standard risk factors LAVi, E/e’ and LV mass index were found to be predictors of the composite endpoint. In a forward conditional multivariable model, LAVi ≥40 ml/m2 (HR 4.2 [1.6-10.7], p=0.003) remained an independent predictor whereas E/e’ was borderline significant (p=0.06) No baseline differences were seen between patients randomized to respectively candesartan and conventional treatment. During follow-up no differences in systolic, diastolic and pulse pressure were seen between groups. Baseline LV mass index was 134±41 g/m2 with no difference between groups. Mean decrease in LV mass index in the control group was 12±28 g/m2 compared with 30±40 g/m2 in the candesartan group during follow up, p=0.015 ptrend =0.001.The effect of candesartan was independent of afterload measures as systolic blood pressure and aortic valve area. After 12 months LV mass index was significantly lower in the candesartan group (103±29 vs. 119±31 g/m2, p=0.01). Despite this difference, no changes in diastolic function or filling pressures were seen between groups. Improvement in LV longitudinal systolic function (s’) was however seen between groups. Patients treated with candesartan had a larger LA size reduction and lower incidence of new-onset atrial fibrillation (7 % vs. 24 %, p=0.02) compared to the control group. Other predictors for the development of atrial fibrillation were LAVi at 12 month and the presence of LV hypertrofi at 12 month.