Stable-label intravenous glucose tolerance test minimal model.

The minimal model approach to estimating insulin sensitivity (Sl) and glucose effectiveness in promoting its own disposition at basal insulin (SG) is a powerful tool that has been underutilized given its potential applications. In part, this has been due to its inability to separate insulin and glucose effects on peripheral uptake from their effects on hepatic glucose inflow. Prior enhancements, with radiotracer labeling of the dosage, permit this separation but are unsuitable for use in pregnancy and childhood. In this study, we labeled the intravenous glucose tolerance test (IVGTT) dosage with [6,6-2H2]glucose, [2-2H]glucose, or both stable isotopically labeled glucose tracers and modeled glucose kinetics in six postabsorptive, nonobese adults. As previously found with the radiotracer model, the tracer-estimated S*l derived from the stable-label IVGTT was greater than Sl in each case except one, and the tracer-estimated SG* was less than SG in each instance. More importantly, however, the stable-label IVGTT estimated each parameter with an average precision of +/- 5% (range 3-9%) compared to average precisions of +/- 74% (range 7-309%) for SG and +/- 22% (range 3-72%) for Sl. In addition, because of the different metabolic fates of the two deuterated tracers, there were minor differences in basal insulin-derived measures of glucose effectiveness, but these differences were negligible for parameters describing insulin-stimulated processes. In conclusion, the stable-label IVGTT is a simple, highly precise means of assessing insulin sensitivity and glucose effectiveness at basal insulin that can be used to measure these parameters in individuals of all ages, including children and pregnant women.

Correlates of aortic distensibility in chronic aortic regurgitation and relation to progression to surgery.

Aortic distensibility decreases with increasing age. Patients with chronic aortic regurgitation eject a large stroke volume into the proximal aorta. A decrease in distensibility of the aorta may impose a higher afterload on the left ventricule and may contribute to deterioration of left ventricular function over time. Accordingly, aortic distensibility was measured in 33 patients aged 13 to 73 years who had chronic isolated aortic regurgitation with minimal or no symptoms. Ascending aortic diameter was measured 4 cm above the aortic valve by two-dimensional echocardiography and pulse pressure was measured simultaneously by sphygmomanometry. Aortic distensibility was calculated as (Change in aortic diameter between systole and diastole/End-diastolic diameter)/Pulse pressure. Left ventricular systolic wall stress and mass were derived from standard M-mode echocardiographic measurements. Left ventricular volumes and ejection fraction were measured by radionuclide ventriculography. Aortic distensibility decreased logarithmically with increasing age (r = -0.62, p less than 0.001) and also correlated inversely with systolic wall stress, left ventricular mass and end-diastolic volume. Patients who eventually underwent aortic valve replacement for symptoms of left ventricular dysfunction had significantly lower aortic distensibility than did those who did not yet require valve replacement: 0.09 +/- 0.08 vs. 0.22 +/- 0.19 x 1/100 (1/mm Hg) (p less than 0.05). Thus, the reduced aortic distensibility that occurs with increasing age may contribute to the gradual left ventricular dilation and dysfunction seen in patients with chronic aortic regurgitation.

Intertest variability of echocardiographic and chest X-ray measurements: implications for decision making in patients with aortic regurgitation.

Echocardiograms and chest X-ray examinations are commonly employed for serial measurements of left ventricular size and function in patients with chronic aortic insufficiency and often support or even determine therapeutic decisions. This study was undertaken to assess the intertest variability of these measurements made from M-mode echocardiograms and X-ray films performed 3 months apart without intervening clinical or therapeutic changes in 22 patients with significant but asymptomatic aortic insufficiency. End-diastolic and end-systolic dimensions, fractional shortening and cardiothoracic ratios were measured by the same reader, with the initial and 3 month tests being read both independently and together for comparison. The mean values for the initial and 3 month studies were similar, but the intertest variability was substantial, especially when the two tests were read independently. The 95% prediction limits are approximately 50% smaller when the serial studies are read together for comparison. The coefficient of variation for end-diastolic and end-systolic dimensions was 6.1 and 10.1%, respectively, and that for fractional shortening was 17.1%. These findings translate into 95% level prediction limits exceeding +/- 8 mm for left ventricular dimensions and 0.12 for fractional shortening; changes on serial evaluations would have to exceed these values to be considered with a high degree of certainty to represent more than random variability. Although this variability may reflect a number of biologic and technical factors, it emphasizes the need to be cautious in making decisions based solely on changes between two tests, particularly if they are not evaluated together.

Ejection fraction response to supine exercise in asymptomatic aortic regurgitation: relation to simultaneous hemodynamic measurements.

The change in ejection fraction during exercise is frequently employed as a measure of left ventricular functional reserve in patients with aortic regurgitation. However, little information is available about its relation to invasive measurements of cardiac performance. Therefore, simultaneous hemodynamic measurements and supine exercise blood pool scintigraphy were performed in 14 patients with severe, asymptomatic or minimally symptomatic aortic regurgitation associated with cardiomegaly but preserved left ventricular function at rest. Their hemodynamic measurements at rest were normal and their exercise capacity was excellent. When the patients were categorized into those patients whose ejection fraction increased or did not decrease by more than 0.05 (Group 1) and those whose ejection fraction decreased by more than 0.05 (Group 2), important differences were apparent. Echocardiographic, radionuclide and hemodynamic measurements at rest in the two patient groups were similar, but Group 1 exhibited a greater increase in cardiac index during supine exercise (2.8 +/- 0.4 to 10.0 +/- 1.8 versus 2.7 +/- 0.5 to 6.9 +/- 1.0 liters/min per m2; p less than 0.005) and a lesser increase in pulmonary capillary wedge pressure (13 +/- 4 to 19 +/- 7 versus 12 +/- 4 to 31 +/- 8 mm Hg; p less than 0.01). The severity of regurgitation decreased during exercise in all patients, but end-diastolic volume decreased and end-systolic volume decreased or was unchanged in Group 1, whereas end-diastolic volume was unchanged and end-systolic volume increased in Group 2.(ABSTRACT TRUNCATED AT 250 WORDS)

Measurement of transmural distribution of phosphorus metabolites in the pig heart by 31P magnetic resonance spectroscopy.

We used the phase modulated rotating frame imaging technique to measure transmural distribution of phosphorus metabolites in 10 anaesthetised ventilated pigs using a double surface coil placed on the surface of the left ventricle. Anaesthesia was maintained in five animals with halothane, barbiturate and nitrous oxide and in five others with intravenous chloralose. 31Phosphorus spectra were acquired, gated to expiration and systole. From phantom experiments the resolution of the experiment was shown to be approximately 2 mm. The anatomical limits of the myocardium were identified by the appearance of 2,3-diphosphoglycerate peaks from red blood cells. The limits of the epicardium were confirmed by obtaining images after placing a phantom containing fluorophosphate on the surface of the heart. The endocardium was identified by inserting a small balloon catheter through the centre of the coil into the left ventricular cavity, filling it with 0.5 ml of fluorophosphate and pulling it gently against the endocardium. No transmural differences in phosphocreatine to ATP ratio were identified in the normal heart. The animals anaesthetised with chloralose showed a significantly higher phosphocreatine to ATP ratio compared to those anaesthetised with halothane and barbiturate. The chloralose animals tended to have a higher blood pressure and a lower heart rate when compared to the other animals. No transmural differences, however, were identified in either group. When regional ischaemia was produced using a snare to occlude the left coronary artery, phosphocreatine fell and the signal from the inorganic phosphate + 2,3-diphosphoglycerate region increased. The inner wall tended to become more acid compared to the outer wall during ischaemia. These experiments show that the phase modulated rotating frame imaging technique can be used to study the effects of changes in workload, ischaemia, or pharmacological intervention on transmural distribution of metabolites in the heart and thus help elucidate factors responsible for subendocardial vulnerability to stress.

Relation of lactate production to postischaemic reduction in function and myocardial oxygen consumption after partial coronary occlusion in swine.

Postischaemic myocardial dysfunction (stunning) induced by partial occlusion of the left anterior descending coronary artery and its relation to lactate production during reperfusion were studied in nine swine. A 40% reduction in regional left ventricular wall thickening, as measured by ultrasonic crystals, was prospectively defined as stunning. A perfusion pressure of 20 mmHg was maintained with a hydraulic occluder for each ischaemic period and was monitored by a distal arterial catheter. To achieve a 40% reduction in function, four animals required three ischaemic periods (mean ischaemic flow reduction 73%), four two (86% flow reduction), and one one (93% flow reduction). At 25 min of reperfusion transmural flow was slightly reduced from 0.67 ml.g-1.min-1 at control to 0.58 ml.g-1.min-1 (p less than 0.05), whereas regional flow endocardial to epicardial flow ratio was unchanged. At 60 min reperfusion, percentage systolic wall thickening was reduced to 25% from a control of 39% (p less than 0.01) and parallel reductions in regional myocardial oxygen consumption from 4.3 ml.min-1 to 2.7 ml.min-1 occurred (p less than 0.01). Lactate extraction was depressed at 15 min reperfusion (-4.0% compared with control +18.0% (p less than 0.05)) but returned to control values by 30 min. It is concluded that postischaemic myocardial dysfunction (stunning) can be induced by partial coronary occlusions and that the extent of dysfunction depends on the degree of flow reduction. The reductions in myocardial oxygen consumption parallel those of wall thickening during reperfusion after stunning. Finally, lactate production occurs during early reperfusion but does not persist with the postischaemic reductions in function and myocardial oxygen consumption.

Effect of coronary sinus occlusion on coronary flow, resistance, and zero flow pressure during maximum vasodilatation in swine.

The effects of coronary sinus occlusion on the relation between coronary artery pressure and flow during maximum vasodilatation were studied in seven swine. The left anterior descending (LAD) coronary artery was instrumented with two catheters, a hydraulic occluder, and a flowprobe. Mean flow was measured at a series of pressures produced by partial LAD occlusion during maximum vasodilatation induced by an intracoronary infusion of adenosine. Observations were made under control conditions and during occlusion of the coronary sinus produced by inflating the balloon on the catheter positioned in the coronary sinus. Systemic haemodynamic variables did not change significantly after the coronary sinus was occluded. The mean right atrial pressure was 4 mmHg. At any given LAD perfusion pressure mean flow during coronary sinus occlusion was always less than during the control state: at LAD pressure 30 mmHg, control flow was 53 ml.min-1 vs occluded flow 24 ml.min-1; at LAD pressure 40 mmHg, control flow 79 ml.min-1 vs occluded flow 49 ml.min-1; and at LAD pressure 50 mmHg, control flow 105 ml.min-1 vs occluded flow 74 ml.min-1; p less than 0.001 for all comparisons. The mean (SD) LAD pressure at which flow stopped (Pzf) when the coronary sinus was unobstructed was 10(2) mmHg. The Pzf during occlusion of the coronary sinus was significantly higher at 20(4) mmHg (p less than 0.001). The slopes of the mean pressure-flow relations were not significantly different during the control state (2.62(0.65) ml.min-1 per mmHg) vs the occluded state (2.47(0.63) ml.min-1 per mmHg), indicating no change in vascular resistance.(ABSTRACT TRUNCATED AT 250 WORDS)

A quantitative study of the anatomy and distribution of coronary arteries in swine in comparison with other animals and man.

The coronary circulation of swine was studied to establish adequate baseline information for using swine in cardiovascular research. Of 65 hearts from domestic and miniature pigs, 45 were injected with a methacrylate plastic and prepared as coronary artery casts whose branches were described and measured, and 20 were injected with different coloured dyes in the right, left anterior descending, and circumflex coronary arteries so that horizontal sections of the heart showed the distribution of each artery and the source of blood supply to particular areas or structures of the heart. Like man, the swine had a left coronary artery that was larger in diameter and longer than the right coronary artery. The right coronary artery was almost always dominant (78%), supplying the posterior septum and atrioventricular node via the posterior descending coronary artery. Eight (17%) of the hearts possessed a balanced blood supply. Two (5%) hearts had a left dominant supply. The intracoronary artery dye injections showed that 72.4% of the right ventricular mass was supplied by the right coronary artery and 27.6% by the left anterior descending coronary artery. In the left ventricle 49% of the mass was supplied by the left anterior descending coronary artery, 25.5% by the right coronary artery, and 25.5% by the circumflex coronary artery. The left anterior descending coronary artery supplied 58% of the interventricular septal mass, while the posterior descending coronary artery supplied 42%. The distribution of the left anterior descending coronary artery branches to the ventricular wall varied inversely in number and size of its diagonal branches (2-9) with the obtuse marginal branches of the circumflex coronary artery which were occasionally more numerous or extended to the apex. The blood supply to the sinoatrial node was always by a branch of the right coronary artery. This analysis shows that not only the coronary anatomy but also the distribution of blood supply to particular areas or structures of the swine heart are very similar to that of humans.

Vasopressin-induced coronary constriction at low perfusion pressures.

We studied the effects of intracoronary vasopressin on the relationship between pressure and flow in the coronary circulation of anaesthetised swine. In addition to measurements at control levels, diastolic pressure-flow relationships were constructed from steady-state points below a coronary pressure of 50 mmHg, where endogenous vasodilatation is strongly stimulated. At baseline pressures, flow fell 28% with maximal vasopressin effect. At all levels of diastolic pressure below 50 mmHg vasopressin also decreased flow, eg, at 30 mmHg flow was depressed by 40%. The slope of the steady-state pressure-flow relationship fell from 1.21 to 0.75 ml.min-1.mmHg-1. The diastolic pressure at which coronary flow ceased rose slightly from 13 to 15 mmHg. Intracoronary adenosine completely prevented vasopressin's effect, and the vasodilator response to adenosine was not attenuated by simultaneous administration of vasopressin. The porcine coronary circulation will constrict in response to vasopressin, not only at normal perfusion pressure, but also at low levels when metabolic vasodilatation is intense. Our study has implications about the therapeutic use of vasopressin, and demonstrates interaction of vasoactive stimuli in the coronary circulation.

Role of the pericardium in the regulation of myocardial blood flow and its distribution in the normal and acutely failing left ventricle of the dog.

The effects of the pericardium on the amount and distribution of left ventricular myocardial blood flow were studied. In 10 normal dogs, transfusion of blood from a donor dog resulted in modest increases in coronary flow and ventricular diameter that were greater with an open than a closed pericardium. The ratio of subendocardial to subepicardial flow remained normal with or without the pericardium, at low and high diastolic ventricular pressure, and before and after pharmacological vasodilation with chromonar. In 18 dogs, cardiac failure was induced by constant infusion of the metabolic inhibitor, phenformin. Modest ventricular dilatation occurred if the pericardium was open. A progressive rise in myocardial blood flow developed in those with the pericardium open (1.06 rising to 3.02 ml . g-1 . min-1). A lesser increase (0.62 to 1.75 ml . g-1 . min-1) was seen in dogs with the pericardium closed; they selectively increased subendocardial flow, producing an average subendocardial to subepicardial flow ratio of 2.25. Pharmacological vasodilatation then resulted in uniform transmural flow. The pericardium can influence myocardial flow indirectly by influencing myocardial metabolic demand, when the heart is stressed. It may have a beneficial role in preventing the increased oxygen and coronary flow requirements produced by ventricular dilatation.

Nitric oxide synthesized by gonadotropin-releasing hormone neurons is a mediator of N-methyl-D-aspartate (NMDA)-induced GnRH secretion.

N-Methyl-D-aspartate (NMDA) directly stimulates gonadotropin-releasing hormone (GnRH) neurons to secrete GnRH. It is not known if this stimulatory effect of NMDA is mediated by NO. Northern blot analysis of the immortalized hypothalamic GnRH neuronal cells (GT1-1) mRNA with a neuronal NOS cDNA revealed this clonal cell line expressed neuronal NOS transcripts as a single 10.5-kb band. Immunoblot analysis of GT1-1 proteins with anti-neuronal NOS serum showed that the GT1-1 cells contain neuronal NOS. GT1-1 cells were used to study the effects of NO and NMDA on GnRH release. L-Arginine (10(-2) M), a precursor of NO enhances basal GnRH secretion. Both oxyhemoglobin (Hb)(10(-6)-10(-4) M), a NO scavenger and N omega-nitro-L-arginine (NNA)(10(-3),10(-2) M), a NOS inhibitor and inactivator block basal as well as NMDA-induced GnRH release. Sodium nitroprusside (SNP) (10(-4), 10(-3) M), a NO donor stimulates GnRH release, an effect inhibited by Hb. Incubation of GT1-1 cells in Ca(2+)-free medium abolished the stimulatory effect of NMDA on GnRH release. In contrast, incubation in medium with increasing concentrations of Ca2+ enhances basal GnRH release as well as augments NMDA-mediated GnRH release. The results demonstrate that L-arginine-NO pathway is functional in the GT1-1 cells and an increase in intracellular Ca2+ [Ca2+]i following NMDA receptor activation activates NOS to generate NO. We conclude that endogenous NO mediates, at least in part, basal as well as NMDA-stimulated GnRH release and may play a role as an intercellular messenger in synchronizing pulsatile GnRH release.

Hemodynamic changes after valve replacement with Starr-Edwards prostheses.

A brief review of the pathophysiology of aortic and mitral valve disease and the hemodynamic results of valve replacement with caged ball prostheses are described. In most patients intracardiac pressures and restored to normal at rest, although there are small pressure gradients across mechanical valves. Severe pulmonary hypertension, if present, usually will regress. With exercise, abnormalities of left atrial pressure or left ventricular function may be found after valve replacement. The causes of failure to achieve hemodynamic improvement with surgery and the late return of congestive failure are discussed.

Reproducibility of rest and exercise left ventricular ejection fraction and volumes in chronic aortic regurgitation.

To assess the variability of rest and exercise radionuclide ventriculography in patients with aortic regurgitation (AR), 22 patients in stable condition with chronic AR underwent radionuclide ventriculography at rest and during exercise for measurement of left ventricular volumes and ejection fraction (EF) on 2 occasions 3 months apart. For the group, there were no significant differences between the findings of the 2 studies except for a slightly lower exercise end-systolic volume on the second study. The inter-study differences and variabilities (expressed as the standard deviation of differences) for rest and exercise EF and change in EF were 0 +/- 0.04, +0.02 +/- 0.05 and +0.01 +/- 0.05, respectively. Thus, although mean differences were minor, considerable individual variability occurred, the magnitude of which was similar to that reported for other patient populations. In 3 patients who had an initial normal response to exercise (increase in EF greater than or equal to 0.05), the response became abnormal and in 2 who initially had an abnormal response, the response normalized. This variability must be considered in clinical decision-making or in research study design for patients with chronic AR.

Left ventricular response to submaximal and maximal exercise in asymptomatic aortic regurgitation.

In patients with chronic aortic regurgitation the quantitative changes in loading conditions and left ventricular performance from rest to submaximal exercise have not been related to the magnitude of change observed from rest to maximal exercise. Changes in end-diastolic volume index, as a measure of preload, and measures of contractile performance (ejection fraction and the systolic blood pressure/end-systolic volume index ratio) were assessed at rest, submaximal and maximal supine bicycle exercise using radionuclide angiography in 74 patients with chronic moderate to severe aortic regurgitation. With exercise, end-diastolic volume index decreased in a stepwise manner from 166 +/- 47 to 152 +/- 41 to 143 +/- 41 ml/m2 at rest, submaximal and maximal exercise, respectively. For the entire group, these changes were not associated with a significant change in ejection fraction but were associated with stepwise increases in systolic blood pressure to end-systolic volume index ratio. However, when patients were divided into 3 subgroups based on an increase (group I), minimal change (group II) or a decrease (group III) in ejection fraction from rest to maximal exercise, stepwise increases in systolic blood pressure to end-systolic volume index were again observed in groups I and II but not in group III. These changes were significantly greater in group I than in group II at submaximal and maximal exercise levels. Differences in ejection fraction response and end-diastolic and end-systolic volumes with exercise in the 3 groups were evident at the submaximal exercise level.(ABSTRACT TRUNCATED AT 250 WORDS)

Prevalence and etiology of idiopathic dilated cardiomyopathy (summary of a National Heart, Lung, and Blood Institute workshop.

Idiopathic dilated cardiomyopathy (IDC) is the primary indication for cardiac transplantation, with associated costs of approximately $177 million per year. Recognizing the economic implications of IDC, the increasing incidence, and the limited information on pathogenesis and prognosis, the National Heart, Lung, and Blood Institute convened a workshop on the Prevalence and Etiology of Idiopathic Dilated Cardiomyopathy on June 13 to 14, 1991. The difficulties of studying the disease were reviewed, including its relatively low prevalence, its potentially pluricausal nature, and the fact that it is often a diagnosis of exclusion. Still, it presents significant challenges to the cardiovascular scientific community, since the mechanism of myocardial damage and related etiologic and prognostic factors are virtually unknown. The development of more reliable measures of immune-mediated damage and noninvasive measures of impaired cardiac function present new research opportunities in this disorder. Standardized diagnostic criteria for use in observational and interventional trials were developed, and priorities for future research were proposed. Population-based registries and nested case-control studies, where feasible, are appropriate study designs for tracking incidence and prevalence, and for identifying risk factors, respectively. Interventional studies should focus on secondary prevention, through modifying immune-mediated damage in clinically evident dilated cardiomyopathy, and through prevention of sudden death in patients with the disorder. Primary prevention trials must await the identification of modifiable risk factors and of appropriate and effective interventions.

Myocardial metabolic alterations after contrast angiography.

Contrast media used during angiography are known to produce transient alterations in cardiovascular physiology. However, little information is available concerning what alterations, if any, occur in myocardial metabolism after contrast angiography. Sixteen patients with symptoms of ischemic heart disease undergoing elective left ventriculography were studied. Coronary sinus and arterial blood samples were obtained for free fatty acids, glucose and lactate before and after performing left ventriculography with Renografin-76. Coronary blood flow was determined by the thermodilution technique. Five minutes after ventriculography, the arterial level of free fatty acids had decreased by 18.0 +/- 4.9 percent (mean +/- standard deviation) from the baseline (before angiography) samples (probability [p] less than 0.001). Associated with this decrease in arterial free fatty acids was an increase in the myocardial uptake of this substrate. At 5 minutes after left ventriculography, the free fatty acid uptake had increased 48.5 +/- 33.0 percent compared with the baseline value (p less than 0.001). After the injection of contrast medium, there was no significant change in the arterial levels of glucose or lactate. However, significant decreases in the myocardial uptake of glucose and lactate were demonstrated (-72.5 +/- 44.5 percent [p less than 0.001] and -43.2 +/- 22.9 percent [p less than 0.001], respectively) at 5 minutes. The changes in arterial free fatty acids and in the myocardial uptake of the various substrates persisted throughout the sampling period of 20 minutes after ventriculography. These results demonstrate that contrast medium significantly alters myocardial metabolism. These metabolic alterations persist longer than the hemodynamic changes induced by contrast angiography.

Effects of long-term vasodilator therapy on electrocardiographic abnormalities in chronic aortic regurgitation.

Electrocardiographic abnormalities develop in patients with chronic aortic regurgitation (AR). Although vasodilator drugs may reduce left ventricular (LV) volume overload, the effects of such therapy on electrocardiographic abnormalities have not been previously evaluated. Accordingly, electrocardiograms were analyzed before and after double-blind, randomized administration of either hydralazine or placebo in 54 patients with chronic AR. These patients were without limiting symptoms and had preserved ejection fraction on entry in the study. The magnitude of ST-segment depression and Romhilt-Estes point score for LV hypertrophy were assessed. Baseline ST depression and LV hypertrophy scores in the placebo and hydralazine groups were not significantly different. At follow-up, after a mean of 19 +/- 6 months, there was a significant reduction in ST depression in patients taking hydralazine (n = 28) compared with patients given placebo (n = 26): -0.023 +/- 0.044 vs 0.029 +/- 0.055 mV, respectively (p = 0.0001); and in the LV hypertrophy score (-1.1 +/- 2.2 vs 0.9 +/- 2.3 points, respectively; p = 0.002). Hydralazine-treated patients also had significant decreases in LV end-diastolic and end-systolic volume indexes, and a significant increase in ejection fraction. These results suggest that such vasodilator therapy may be beneficial in patients with chronic AR.

Effects of stopping long-term vasodilator therapy in patients with chronic aortic insufficiency.

We studied the effects of stopping long-term vasodilator therapy in 17 patients with chronic stable aortic insufficiency. These patients received hydralazine for 37 +/- 15 months (mean +/- SD) and, as a result, had experienced a significant decrease in left ventricular volumes. All patients were followed clinically and ten of the patients underwent serial radionuclide evaluation at baseline, while receiving drug, and at 20 +/- 7 months after stopping drug therapy. No patient showed evidence of acute clinical deterioration when drug therapy was stopped. The rate of progression to valve replacement due to onset of symptoms or left ventricular dysfunction was not significantly different from that previously reported in a population with similar characteristics. Left ventricular size, however, returned to levels similar to baseline after drug therapy was stopped. We conclude that long-term vasodilator therapy may be discontinued in patients with chronic stable aortic insufficiency without causing clinical deterioration or significant alteration in rate of progression to valve replacement.