Increased oxidative stress and cardiomyocyte myofibrillar degeneration in patients with chronic isolated mitral regurgitation and ejection fraction >60%

Abstract

Objectives: This study assessed myocardial damage in patients with chronic isolated mitral regurgitation (MR) and left ventricular ejection fraction (LVEF) >60%.

Background: Typically, MR patients have decreased LVEF after mitral valve (MV) repair despite normal pre-operative LVEF.

Methods: Twenty-seven patients with isolated MR had left ventricular (LV) biopsies taken at time of MV repair. Magnetic resonance imaging with tissue tagging was performed in 40 normal subjects and in MR patients before and 6 months after MV repair.

Results: LVEF (66 +/- 5% to 54 +/- 9%, p < 0.0001) and LV end-diastolic volume index (108 +/- 28 ml/m(2) to 78 +/- 24 ml/m(2), p < 0.0001) decreased, whereas left ventricular end-systolic (LVES) volume index was 60% above normal pre- and post-MV repair (p < 0.05). The LV circumferential and longitudinal strain rates decreased below normal following MV repair (6.38 +/- 1.38 vs. 5.11 +/- 1.28, p = 0.0009, and 7.51 +/- 2.58 vs. 5.31 +/- 1.61, percentage of R to R interval, p < 0.0001), as LVES stress/LVES volume index ratio was depressed at baseline and following MV repair versus normal subjects (0.25 +/- 0.10 and 0.28 +/- 0.05 vs. 0.33 +/- 0.12, p < 0.01). LV biopsies demonstrated cardiomyocyte myofibrillar degeneration versus normal subjects (p = 0.035). Immunostaining and immunoblotting demonstrated increased xanthine oxidase in MR versus normal subjects (p < 0.05). Lipofuscin deposition was increased in cardiomyocytes of MR versus normal subjects (0.62 +/- 0.20 vs. 0.33 +/- 0.11, percentage of area: p < 0.01).

Conclusions: Decreased LV strain rates and LVES wall stress/LVES volume index following MV repair indicate contractile dysfunction, despite pre-surgical LVEF >60%. Increased oxidative stress could cause myofibrillar degeneration and lipofuscin accumulation resulting in LV contractile dysfunction in MR.

Figure 1. LV systolic strain rates and remodeling pre- and six months post-MV repair

LV circumferential and longitudinal systolic strain rates (A) are significantly decreased post-MV repair vs. pre-surgery and vs. normal controls. 3-D LV end-systolic (B) radius/wall thickness ratios did not differ between normal and post-surgery groups. RR denotes R to R interval. Graphs displaying LV mass/volume (C) and LV end-diastolic radius/wall thickness demonstrate reversal of eccentric remodeling following MV repair. * = p <0.05 vs. control group. † = p <0.05 vs. pre-MV repair.

Figure 2. Indices of LV afterload and function pre- and six months post-MV repair

Graphs displaying mean arterial pressures (A), LV end-systolic wall stress (B), LV end systolic volume index (C), and LV end systolic stress/end-systolic volume index (D) in normals and pre- and post-surgical MR patients.

Figure 3. Myofibrillar loss and oxidative stress in patients with isolated MR

Myocardial biopsy findings in controls (n=10) (a,d,g) and MR patients (n=27) demonstrating myofibrillar degeneration (b,c), increased xanthine oxidase (e,f), and increased lipofuscin (h,i). Nitrotyrosine staining in the LV of an MR patient demonstrating increased staining in areas of lipofuscin accumulation (j) and a corresponding image (k) with immunoabsorbed antibody with no uptake of antibody and only autofluorescence of lipofuscin. Scale bar = 20 μm. * = p <0.05 vs. normal control group. Scale bar = 20 μm.

Figure 4. Lipofuscin in electron micrographs of patients with isolated MR

Transmission electron microscopy of endomyocardial biopsy samples demonstrating marked lipofuscin deposition (arrows) in the hearts of MR patients.

Figure 5. Protein quantification of xanthine oxidase in LV of patients with isolated MR

Western blot analysis for xanthine oxidoreductase (A) depicts a band at 145kD which represents both XDH and transiently modified XO and bands at 125kD and 85kD which represent permanently modified XO. Densitometry depicting average band intensity normalized to respective calsequestrin loading control (B). *=p<0.05 vs. normal control group.