SUMMARYWe report a functional application of magnetic resonance imaging (MRI) for the quantitative description of left ventricular geometry through systole and diastole in normal anaesthetized Wistar rats that might be applicable for the analysis of chronic changes resulting from pathological conditions. Images of cardiac anatomy were acquired through planes both parallel and perpendicular to the principal cardiac axis at times that were synchronized to the R wave of the electrocardiogram. The images of the transverse sections were assembled into three-dimensional representations of left ventricular geometry at consecutive time points through the cardiac cycle. This confirmed the geometrical coherence of the data sets, that each slice showed circular symmetry, and that the images were correctly aligned with the appropriate anatomical axes. Different models for the three-dimensional geometry of the left ventricle were then tested against the epi-and endocardial surfaces reconstructed from images of the transverse sections of the left ventricle in both systole and diastole using least-squares minimizations in three dimensions. In agreement with previous reports in the human heart, an elliptical figure of revolution offered an optimal fit to the epicardial and endocardial geometry for the rat heart in diastole. This was in preference to models that used spherical, quartic or parabolic geometries. However, in contrast to contraction in the human heart, all these geometrical representations broke down during systolic ejection in the rat heart. We therefore introduced a more general hybrid model which described left ventricular geometry in terms of the variation of the radii r(z), independently determined for each slice, with its position z along the principal cardiac axis. The resulting function r(z) could then be described by a simple ellipsoid of revolution not only during diastole, but also throughout ventricular ejection. The findings also ruled out alternative geometrical representations. It was then possible additionally to reconstruct the luminal and total left ventricular volumes, wall thicknesses and ejection fractions through the cardiac cycle and to confirm that the predicted total ventricular wall volume was conserved throughout the cardiac cycle. Our hybrid model of cardiac geometry may thus be useful for non-invasive serial studies of chronic pathological changes that use the rat as a model experimental system.

No abstract

Patients were eligible if they presented within 24 hours of suspected acute myocardial infarction with no clear indications for, or contraindications to, the study treatments (although planned use of a few days of intravenous or oral nitrates was permitted). About 40% were within 6 hours of pain onset. 75% had ST elevation, 25% were aged 70% 15% had heart failure, and 2% had systolic blood pressure < 100 ramHgPatients were randomly allocated in a 2 x 2 x 2 factorial design between one month of oral captopril (6.25 mg initial dose, 12.5 rag 2 hours later, 25 mg 10-12 hours later and then 50 mg twice daily) versus placebo, one month of oral controlled-release isosorbide mouonitrate (lmdur: 30 mg initial dose, 30 mg 10-12 hours later and then 60 nag each morning) versus placebo, and 24 hours of intravenous magnesium sulphate (8 raraol initial bolus over 15 minutes followed by 72 mmcl)'versus open control. About 75% received fibrinolytic and almost all antiplatelet therapy.The main comparisons are to be of 5-week and longer-term mortality amongst all those allocated each active therapy versus all those allocated the corresponding control. Principal subsidiary comparisons involve subdivisun by planned nitrate at entry and by the other randomly allocated treatments. Mode of death and major morbidity results will also be considered. The decisions by a single cardiologist as to which of 308 consecutive patients to refer for angiography after treadmill testing were compared with their life expectancy gains from bypass surgery predicted by decision analysis. Neither patient age nor gender influenced the decision to perform angiography. The 94 patients sent for angiograms exercised for a significantly shorter time (p < 0.001 ), had more ST deviation (p < 0.001 ), more angina (p < 0.002) and were more likely to have had a prior myocardial infarction (p < 0.001) than the 214 patients not referred. The mean life expectancy gain predicted from bypass surgery was also greater (p < 0.001) in those referred (2.9 + 1.7 QALYs) than in those not referred for angiography (I.0 + 1,7 QALYs). However, 1~.3 patients not referred were predicted to gain up to 5.7 QALYs from bypass surgery. Consequently the overall predicted life expectancy gain from the cardiologist's 388 intuitive decisions was only 0. I + 2.5 QALYs per patient. Had the referral decision been solely directed by decision analysis the overall gain per patient would have been 1.9 4-1.6 QALYs, and 135 extra patients (229 in total) would have been sent for angiography. Use of decision analysis, therefore might help make referral for angiography more efficient and consistent. Persistent chest pain with normal cardiac investigations is not uncommon following treatment of coronary artery disease. Oesophageal problems are often suspected but to date evaluation has proved difficult. Eight patients who had previously undergone successful coronary artery bypass grafting or coronary angioplasty underwent 24hr ambulatory manoraetry, pH and ECG monitoring. Symptoms were correlated with ...

No abstract