Comparison of ventricular pressure relaxation assessments in human heart failure
Quantitative influence on load and drug sensitivity analysis
Hideaki Senzaki, MDa,
Barry Fetics, MAa,
Chen-Huan Chen, MDa and
David A. Kass, MDa
a Division of Cardiology, Department of Internal Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland, USA
View larger version (26K):
[in a new window]
|
Figure 1 Comparative sensitivity of pressure-decay assessments (see Methods section for definitions of parameters) to small variations in the input data. Upper panels display the consequence of varying the lower pressure cut-off point from 2 to 10 mm Hg above end-diastolic pressure (EDP). Lower panels display the effects from a pure downward offset displacement of the LV pressure data (LVP) by 0 to –10 mm Hg. Data are displayed as mean percent change (±SEM) relative to a reference state (leftward condition in each plot). * p < 0.05 versus reference, based on RMANOVA (repeated measures analysis of variance).
|
|
View larger version (22K):
[in a new window]
|
Figure 2 Pressure-time (upper panels), dP/dt-time (middle panels) and dP/dt-P plots (lower panels) from representative beats in normal, HCM and DCM hearts. Monoexponential fits are shown by dashed lines and logistic fit by solid lines in each instance. P(t) data were well fit in all instances by both models. However, in DCM patients, the ME model did not predict dP/dt data well, and in all cases the logistic model better predicted this derivative. The lower plots synthesize these two behaviors, with ME fits appearing as a straight line, and logistic fits as a parabola. The latter better approximated the measured data—most notably in DCM ventricles. DCM = dilated cardiomyopathy; HCM = hypertrophic cardiomyopathy; ME = monoexponential.
|
|
View larger version (24K):
[in a new window]
|
Figure 3 Relations between relaxation load sensitivity and contractility. (A) TD-Pes slope versus Ees. The data define a hyperbolic relation, with apparent increasing load dependence of tau at low contractility. (B) The same relation derived using TF. (C) A similar relation employing resting dP/dtmax as the index of contractile function, rather than Ees. This shows that the dependence did not require a specific contractility analysis parameter. (D) The same relation determined using TL. These results are markedly different from those in panels A–C, with virtually no change in the load-sensitivity despite varying Ees. (E) The same analysis employing T1/2, again showing how a simple change in the assessment method of relaxation markedly altered the appearance of enhanced load-dependence in depressed hearts. (F) A similar analysis, in which the SSD for ME versus logistic models is substitute for a tau-Pes slope ordinate. The plot is similar to panels A–C, showing that the major cause for apparent hyperbolic dependence in the latter is due to the enhanced deviation from a ME-relaxation decay with declining Ees. Ees = end-systolic elastance; ME = monoexponential; SSD = sum of the squares.
|
|
|
