Volume 116 Number 6 ANF and LV size 1237 Table I. Clinical characteristics of normal subjects and patients Normal subjects (n = 49) Aortic Mitral regurgitation regurgitation P patients (n = 33) P patients (n = 15) Age (~1) Sex (% male) Body surface area (m2) 50 + 12 NS 49 rt 15 NS 56 _t 13 63 NS 85 <0.05 47 1.90 + 0.19 NS 1.97 +- 0.22 <0.0005 1.79 * 0.16 NS, Not statistically significant. perimentally-induced hypertrophy and congestive heart failure. Furthermore, high concentrations of atria1 natriuretic factor have been detected in blood from the coronary sinus,8p rl, lg the vessel subserving venous drainage of the left ventricle as well as part of the left atrium. These observations led us to consider whether atrial natriuretic factor production and secretion might also be governed by stimuli primarily affecting ventricular size and load in the absence of important atrial alterations. Thus to assessthe separate influ- ences of left ventricular dilatation and hypertrophy and left atrial dilatation, we compared plasma levels of atrial natriuretic factor in normal adults with those in patients with chronic, compensated aortic regur- gitation or with chronic, compensated mitral regur- gitation, wherein comparable degrees of ventricular dilatation and hypertrophy were associated with strikingly different atrial size and load. METHODS Patient population. Thirty-three patients with chronic aortic regurgitation and 15 patients with chronic mitral re- gurgitation were drawn from an ongoing longitudinal study of the natural history of valvular regurgitation.20l 21 Pa- tients gave informed consent under a protocol approved in 1980 and at regular intervals thereafter by the Committee on Human Rights in Research of Cornell University Med- ical College. All patients were asymptomatic at the time of evaluation (New York Heart Association Functional class 1) and none demonstrated objective evidence of congestive heart failure. Three patients with mitral regurgitation had chronic atria1 fibrillation. Four patients with aortic regur- gitation and seven patients with mitral regurgitation were taking cardiac medications at the time of evaluation, pri- marily antiarrhythmic agents and atrioventricular (AV) nodal blocking agents to control atria1 fibrillation, although three patients were additionally taking diuretics. An addi- tional 49 normal adults were studied as controls. The three groups were similar with regard to age, but patients with mitral regurgitation had significantly lower body surface areas, reflecting the female predominance in this condition (Table I). Echocardiography. All subjects underwent M-mode and two-dimensional echocardiography, from which left ventricular and left atria1 dimensions were measured using standard technique. 22 Left ventricular fractional shorten- ing was calculated as a measure of ventricular performance to determine whether elevated natriuretic factor levels were related to depressed ventricular function. Left ven- tricular mass was calculated according to the Penn convention.23, 24 End-systolic meridional stress was deter- mined with the method of Reichek et al.25 All patients demonstrated left ventricular dilatation.26 The presence of severe (3+ to 4+/4+) valvular regurgitation was confirmed by Doppler echocardiography in all patients2’p 28 Left atrial volumes were determined by means of a biplane ar- ea-length method from orthogonal planes in two-dimen- sional apical two- and four-chamber views.2g Atrial natriuretic factor determination. Venous blood samples were collected on ice in ethylenediaminetetraace- tic acid (EDTA) Vacutainers (Becton, Dickinson & Co., Rutherford, N.J.) after subjects had been supine for at least 15 minutes. Atria1 natriuretic factor was measured, as pre- viously described,5 in Cis Sep-Pak (Waters Associates, Milford, Mass.) extracts of plasma by a radioimmunoassay technique that employed antibody to the 28-residue pep- tide, a-human atrial natriuretic peptide (or atria1 natri- uretic factor 99-126); the same peptide was used as stan- dard and tracer. Bound and free peptide were separated by double antibody precipitation using solid phase goat anti- rabbit IgG (Immunobeads, Bio-Rad Laboratories, Rock- ville Centre, N.Y.). Statistical analysis. Mean values are presented with the standard deviation as the index of dispersion. Intergroup differences were determined by means of analysis of vari- ance followed by the Scheffe test for multiple comparisons. The relationship between continuous variables was exam- ined using linear regression analysis. Independence of cor- relation was tested with multiple linear and stepwise regression analysis. RESULTS Left ventricular size and function. Left ventricular end-diastolic dimension and mass are compared in Fig. 1. As can be seen, compared with normal subjects, patients with chronic aortic and mitral re- gurgitation had increased end-diastolic dimensions (6.4 + 0.7 and 6.7 + 0.7 cm, p < 0.0005, versus 5.0 & 0.5 cm in the normal subjects) and mass (298 f 93 and 293 f 91 gm, versus 147 f 44 gm in the normal subjects, both p < 0.0005). Fractional shortening, an index of left ventricular systolic performance, is depicted in Fig. 2. Although mean values for fractional shortening were within the