Tissue distribution and magnetic resonance spin lattice relaxation effects of gadolinium-DTPA.

Gadolinium-DTPA complex (Gd-DTPA) is a potential clinical magnetic resonance (MR) contrast agent that enhances images primarily by decreasing spin-lattice relaxation time (T1) in tissues in which it localizes. This study was designed to determine the immediate tissue distribution of intravenously administered Gd-DTPA in selected organs of interest as a function of administered dose and tissue Gd-DTPA concentration. An intravenous bolus of Gd-DTPA with a tracer quantity of Gd-153 was administered to three groups of rabbits at the following doses: 0.01 mM/kg (n = 6); 0.05 mM/kg (n = 6); 0.10 mM/kg (n = 6). A control group received sham injections. Five minutes after Gd-DTPA was administered, all animals were killed; samples of serum, lung, heart, kidney, liver, and spleen were analyzed in a 0.25 T MR spectrometer to measure T1, and then in a gamma well counter to determine tissue concentration of Gd-DTPA. Tissue distribution (per cent dose/tissue weight in g) at five minutes after injection was proportionally constant over the range of doses given. Tissue concentration varied linearly with injected dose (r greater than 0.98 for all tissues). Relaxation rate (1/T1) varied linearly with injected dose and with tissue Gd-DTPA concentration (r greater than 0.97 for all tissues). The order of tissue relaxation rate response to a given dose was: kidney greater than serum greater than lung greater than heart greater than liver greater than spleen. We conclude that because of its extracellular distribution and linear relaxation rate versus concentration relationship, Gd-DTPA enhancement in MR images may be a good marker of relative organ perfusion.

Magnetic resonance relaxation times of percutaneously obtained normal and abnormal body fluids.

Seventy-three fluid samples obtained via percutaneous aspiration and drainage were analyzed by proton magnetic resonance spectroscopy. The fluids included abscess contents, bile, ascitic fluid, cyst and pseudocyst fluid, urine, hematomas, pleural fluid, lymphoceles, seromas, cerebrospinal fluid, pancreatic ductal fluid, and necrotic tumor. They were grouped by their clinical etiology and analyzed with respect to their inherent magnetic relaxation properties. In addition, some of the samples were tested for the following chemical parameters which were correlated with T1 and T2 values: total protein content (n = 36), osmolality (n = 24), specific gravity (n = 11), and amylase levels (n = 23). A large overlap was found in the T1 (spin-lattice) and T2 (spin-spin) relaxation times of the fluids; however, the mean T1 values of abscesses and hematomas were significantly lower than those of other fluids. Similar variability was seen in T2 values, though hematomas and abscesses again could be distinguished by shorter relaxation times. The spin-lattice (1/T1) and spin-spin relaxation rates (1/T2) showed a moderate correlation with total protein content, osmolality, and specific gravity. It is concluded that there is some predictability to MR analysis of body fluids, though the overlap in magnetic relaxation times limits specificity.

Nuclear magnetic resonance analysis of acute myocardial infarction in dogs: the effects of transient coronary ischemia of varying duration and reperfusion on spin lattice relaxation times.

The purposes of this study were to evaluate the effects of acute myocardial ischemia with reperfusion on T1 (spin-lattice) nuclear magnetic resonance (NMR) relaxation times in a canine model and correlate these changes with bulk myocardial water content (%H2O). In 15 dogs the left anterior descending coronary artery was occluded for either 40 minutes (n = 5), 1 hour (n = 5), or 2 hours (n = 5). In 15 additional dogs, matched occlusion periods were followed by 3 hours of reperfusion. T1 of tissue from normal and ischemic myocardium was measured in vitro with a 2.5 kg NMR spectrometer. In the reperfusion animals, the 2-hour group showed significant increases in %H2O and T1 when the ischemic segment of myocardium was evaluated (both p less than 0.01). All but one animal in the 1-hour (reperfusion) group showed increases in both %H2O or T1 in the ischemic segment of myocardium when compared to control segments. The mean values from the ischemic myocardium in the 1-hour group were significantly higher than the values from the matched control segment for %H2O (p less than 0.05) and T1 (p less than 0.05). In the group undergoing 40 minutes of ischemia followed by 3 hours of reperfusion, neither %H2O nor T1 changed significantly. In the nonreperfused animals, neither T1 nor %H2O content increased significantly after 40 minutes. Significant increases were seen in the 1-hour (p less than 0.05) and 2-hour groups (without reperfusion) (p less than 0.01). In addition, the 2-hour occlusion followed by reperfusion animals had significantly greater T1 relaxation times and %H2O than control ischemic animals.(ABSTRACT TRUNCATED AT 250 WORDS)

Nuclear magnetic resonance analysis of acute and chronic myocardial infarction in dogs: alterations in spin-lattice relaxation times.

The purpose of this study was to evaluate the effects of acute and chronic ischemic injury on myocardial spin-lattice relaxation times (T1) in dogs. Ligation of the left anterior descending coronary artery was performed on 23 dogs which were divided into four experimental groups and killed at 3 hours (n = 6), 4 days (n = 6), 21 days (n = 5), and 56 days (n = 6) after coronary occlusion. T1 was measured in vitro with a 2.5 kg nuclear magnetic resonance (NMR) spectrometer using tissue from ischemic and control areas of the myocardium. Both the 3-hour and 4-day groups showed prolongations in T1 (p less than 0.01) for tissue from the ischemic area. In the 21-day group, two infarcts showed an increase in T1, two showed a decrease in T1, and one showed no significant change. The 56-day old infarcts had a lower mean value for T1 than control myocardium (p less than 0.01). The differences in myocardial water content between control and infarcted myocardium were found to parallel changes in T1 values in each experimental group. Pathologic examination of the myocardial scar from 21- and 56-day-old infarcts revealed extensive fibrosis in the infarcts with lower T1 values and tissue water contents than control myocardium. We conclude that myocardial edema in 3-hour and 4-day-old infarcts results in association with prolongations in T1. Twenty-one-day-old infarcts may be edematous (with increased T1) or fibrotic (with T1 values lower than normal myocardium), while 56-day-old infarcts are fibrotic and have shorter T1 values than normal myocardium.(ABSTRACT TRUNCATED AT 250 WORDS)

Sensitivity and specificity of nuclear phase analysis versus ejection fraction in coronary artery disease.

Phase standard deviation (SD) and skew characteristics of the first Fourier harmonic of equilibrium radionuclide volume curves were examined and compared during rest and during supine bicycle exercise with ejection fraction (EF) changes and the development of ischemia in 17 control subjects and in 2 groups of patients (n = 57) with coronary artery disease (CAD). Group I comprised 37 patients with CAD; IA was a subgroup of 20 patients with previous myocardial infarction (MI) and IB a subgroup of 17 patients with CAD without MI (all with coronary stenosis greater than 75% diameter narrowing). Group II comprised 20 patients with CAD who had undergone coronary bypass surgery. In the Group I subjects, phase SD was the most sensitive indicator of CAD at rest (Group I, 56%; Group IA, 70%, and Group IB, 29%), and the EF was the most sensitive indicator at submaximal (Group I, 78%; Group IA, 86%, and Group IB, 64%) and maximal exercise (Group I, 70%; Group IA, 93%, and Group IB, 53%). When phase SD and skewness were combined with EF changes, little increase in sensitivity occurred in Group I (rest 61%, submaximal exercise 88% and maximal exercise 76%). The results from Group II subgroups were qualitatively similar to those observed with Group I subgroups. These data reveal a marginally improved sensitivity for detection of CAD during supine bicycle radionuclide ventriculography when phase measurements were added to changes in global EF values.

Pulmonary blood volume: relationship to changes in left ventricular end-diastolic pressure during atrial pacing.

Little data exist about the relationship between changes in cardiac end-diastolic pressure and changes in pulmonary blood volume. To assess this relationship, we studied 11 patients with coronary heart disease during atrial pacing in an attempt to produce multiple pressure-volume points. During catheterization, we obtained Millar pressure recordings of end-diastolic pressure along with equilibrium radionuclide angiograms. Cardiac output, ejection fraction, and pulmonary blood volume were obtained by means of recently validated radionuclide techniques. During pacing, substantial changes in pulmonary blood volume occurred only with marked increase in end-diastolic pressure volume (greater than or equal to 15 mm Hg) and rarely exceeded 15% of control pulmonary blood volume. Cardiac output did not change, while ejection fraction declined during pacing. There was a fair correlation between the absolute change in pulmonary activity (or pulmonary blood volume) or the percentage of change in pulmonary activity over the control value with end-diastolic pressure when all the data points were evaluated (n = 74, r greater than 0.70). However, the scatter in the data precluded making accurate estimates of pressure changes from changes in radionuclide volume changes. We conclude that large changes in cardiac filling pressure must occur during atrial pacing, where cardiac output does not change, before visible pulmonary blood volume changes occur. This may limit the extrapolation of presumed pressure changes from known pulmonary blood volume when changes are small.

Quantitation of size of myocardial infarctions by computerized transmission tomography. Comparison with hot-spot and cold-spot radionuclide scans.

The current study evaluated the ability to quantitate the volume of myocardial infarctions when they are outlined by intravenously administered contrast media in the myocardial perfusion phase and in the phase of delayed contrast enhancement of the infarct. Quantitation by contrast media was assessed from computerized transmission tomography (CTT) scans of the ex situ heart and compared with quantitation by technetium-99m (99mTc)pyrophosphate (99mTc PYP) and thallium-201 (201Tl) scans of the same ex situ hearts. True volume was defined by histochemical morphometry. CTT during the contrast perfusion phase uniformly underestimated infarct size but had a good correlation with true volume (r = 0.87). CTT during enhancement phase correlated closely with true volume (r = 0.98) and most precisely measured true size (y = 1.06 X 0.23). The 99mTc PYP scan overestimated infarct volume (predictive overestimation of 6 to 199%) but had a good correlation with true volume (r = 0.87). 201Tl underestimated infarct volume but correlated well with true volume (r = 0.94). Thus, quantitation of infarct volume from CTT scans performed during either the perfusion or infarct enhancement phase after intravenous contrast media provides a good estimate of true infarct volume. Delineation of the infarct by contrast media in the ex situ heart is more precise during the phase of delayed enhancement of the infarct.

Radionuclide analysis of ejection time, peak ejection rate, and time to peak ejection rate: response to supine bicycle exercise in normal subjects and in patients with coronary heart disease.

Using equilibrium radionuclide angiography, we evaluated the ejection time (ET), peak ejection rate (PER), and time to peak ejection rate (TTp) at test and during supine bicycle exercise in 39 subjects, divided into three groups: group 1 = 13 normal subjects; group 2 = 10 patients with a previous infarction (MI); and group 3 = 16 patients with coronary disease without a previous MI. Normal subjects had greater ejection fractions and PERs than the other two groups at rest or peak exercise (p less than 0.05). PER was no more useful than ejection fraction in identifying cardiac dysfunction at either rest or exercise. The time of its occurrence varied with the group studied, and was slightly but significantly later in systole in groups 2 and 3 when compared to normals (p less than 0.05), though substantial overlap between groups occurred. During exercise, absolute ET shortened in all groups, but actually increased as a function of the R-R interval. The time to peak ejection rate (normalized for the R-R interval) was greater in the noninfarct group (group 3) patients (p less than 0.05) when compared to the group 1 or group 2 individuals at peak exercise. In conclusion, equilibrium radionuclide angiography is a useful technique for the quantification and characterization of events during systole, and is capable of providing information on the timing of events during ejection. Tardokinesis, or the delay of ventricular ejection, is not seen in the response of global indices of left ventricular function to exercise stress. While global early systolic indexes may not detect regional dyssynchrony, their timing during stress may occasionally aid in discerning the presence of cardiac dysfunction.

Analysis of ventricular emptying and filling indexes during acute increases in arterial pressure.

Using equilibrium radionuclide angiography, an evaluation was made of the response of left ventricular ejection and filling rates at rest and during acute increases in afterload in 8 normal volunteer subjects and 10 patients with previous transmural myocardial infarctions. Using the postatropine point for comparison, normal patients increased ejection time and decreased peak ejection rate (-3.90 +/- 0.49 vol/s to -3.41 +/- 0.95 vol/s) and peak filling rate (3.94 +/- 0.88 vol/s to 3.51 +/- 0.38 vol/s). Infarct patients had similar responses, although all indexes were lower than the corresponding values in the normal subjects. At rest, the ratio of peak filling to emptying rate was similar in the normal subjects and the infarct patients (1.01 +/- 0.24 versus 0.99 +/- 0.25, respectively) and maintained that relationship after atropine (0.91 +/- 0.11 versus 0.81 +/- 0.21) and at the peak increase in arterial pressure (1.07 +/- 0.21 versus 1.02 +/- 0.32). The ratio of time to peak filling/time to peak emptying behaved in similar fashion regardless of the differences in the absolute values. In this study, left ventricular filling and emptying behaved in a similar fashion in response to the alteration in arterial pressure in normal subjects and in patients with previous myocardial infarctions.

Quantitative phase analysis of exercise radionuclide left ventriculography in normals and patients with coronary artery disease.

Using multigated radionuclide ventriculography (RNV), the left ventricular (LV) phase histogram (PH) was used to quantitate LV synchrony at rest and at submaximal and maximal supine exercise in 10 normal and 10 coronary (CAD) patients. Each LV PH was obtained by masking the phase image with an LV region, generated semiautomatically and modified if necessary to minimize ambiguity of LV delineation. The shape and spread of each histogram was characterized by its standard deviation (SD) and skewness (SK). Normals had a tight resting PH with slight negative skewness (SD = 9.06 +/- 0.22 (mean +/- SEM; SK = -0.53 +/- 0.05). CAD patients had a wider resting PH with a larger SD and a significantly more positive skewness reflecting greater contractile asynchrony (SD = 16.53 +/- 2.26; SK = 0.35 +/- 0.22). With exercise the PH distribution for the normals remained tight (SD = 8.1 +/- 0.24) while SK changed little (SK = -0.29 +/- 0.04). In contradistinction, with exercise CAD patients markedly increased the spread of their LV PH (SD = 27.42 +/- 3.77) and increased the number of late pixels (SK = 0.75 +/- 0.15). It is concluded that quantitative PH analysis is a useful adjunct for the analysis of regional asynergy during resting or exercise radionuclide left ventriculography.

Computerized tomographic (CT) appearance of the myocardium after reversible and irreversible ischemic injury.

Computerized tomographic abnormalities have been observed within hours after coronary occlusion. It has not been determined whether these CT changes are always indicative of irreversible myocardial damage or whether they can be produced by reversible myocardial injury. Accordingly, in the current study, the circumflex coronary arteries of 16 dogs were occluded for 20 minutes (reversible injury) or 40 minutes (irreversible injury) and then reperfused. The animals were sacrificed 24 hours later. Ten minutes prior to sacrifice, eight of the dogs received contrast material intravenously. The electron microscopic changes in the 20-minute occlusion group consisted of reversible damage, while those in the 40-minute group were irreversible. Two of the animals in the 20-minute group showed slightly decreased X-ray density of the infarct in the absence of contrast media, but none showed contrast enhancement of the infarct in the presence of contrast media. All the animals in the 40-minute group showed decreased X-ray density, and all animals that received contrast media had contrast enhancement of the area of ischemic damage. The results of the current study suggest that the decreased X-ray density of the area of ischemic damage on CT scans is due to cellular edema and may be observed in reversible myocardial injury, while delayed contrast enhancement is only associated with, and is indicative of, irreversible myocardial injury.

Digital intravenous ventriculography: analysis and validation of wall thickness changes during ischemia and inotropic stimulation in dogs.

To assess regional ventricular wall thickening, we studied 8 open-chest dogs using digital intravenous ventriculography at a control stage, after subtotal and total coronary occlusion, and then after isoproterenol administration. In 3 dogs, 2 pairs of myocardial thickness crystals were implanted. Correlations of measures of wall thickness and percentage of wall thickening with crystal measures were excellent at the base (r greater than or equal to 0.97) and near the apex (r greater than or equal to 0.97). Four areas of the inferoapical wall muscle were measured: the base, mid-wall, distal wall, and apex. Four chamber dimensions were also examined: the long axis, base decreased with ischemia. At completion occlusion, it was -24.6 +/- 6% at the base, -27.3 +/- 5% at the mid-wall, -27.1 +/- 5% at the distal wall, and -24.7 +/- 5% at the apex. Percent thickening decreased with occlusion, although greater at the base and mid-wall than at the distal wall and apex. With isoproterenol, end-diastolic thickness increased only at the apex, with little change at the base, distal wall, and mid-wall. Percent thickening increased. In general, ischemia produced increases in end-diastolic hemichords with little change in the long axis. Isoproterenol reduced the hemichords, although the long axis did not change. We conclude the digital intravenous ventriculograms can be used to assess changes in wall thickness with high degrees of accuracy. Asymmetric thickening occurred at rest, with ischemia and with inotropic stimulation, being greatest at the apex and least at the base.

Physiologic--angiographic correlates of coronary arterial stenoses in resting and intensely vasodilated states.

The treatment of ischemic heart disease is dependent upon defining the physiologic significance of coronary arterial stenoses demonstrated on coronary arteriograms. Accordingly, physiologic--angiographic correlates were studied in 12 anesthetized dogs prepared with an electromagnetic flowmeter and micrometer-controlled occluder on the circumflex coronary artery, a pair of sonic crystals in the myocardium within the distribution of the circumflex artery, and a catheter in the thoracic aorta. Measurement of arterial diameters on coronary arteriograms was made to define the minimum percent stenosis that caused: (1) decrease in resting coronary blood flow (CBFr); decline in the peak level of coronary blood flow (CBF) produced by intense vasodilatation induced by intracoronary infusion of ATP (CBFATP); and (3) segmental left ventricular (LV) dysfunction. CBFR decreased at an average stenosis of 80.45 +/- 9.13% (SD) (percent reduction in luminal diameter), while CBFATP declined at a stenosis of 31.83 +/- 5.64%. Segmental LV dysfunction was observed at a stenosis of 85.92 +/- 9.83%. In all dogs, the initial stenosis causing decline in CBFATP was a less than or equal to 40% reduction in luminal diameter. The results of this study indicate that coronary arterial stenoses of 40% or less may be hemodynamically significant under situations of augmented CBF. On the other hand, regional contractile function at rest is preserved up to stenosis in excess of 80%.

Quantitation of left ventricular dimensions and function by digital video subtraction angiography.

Digital video subtraction angiography (DVSA) after central intravenous administration of contrast media was used in experimental animals and in patients with suspected coronary artery disease to quantitate left ventricular dimensions and regional and global contractile function. In animals, measurements of left ventricular (LV) volumes, wall thickness, ejection fraction, segmental contraction, and cardiac output correlated closely with sonocardiometry or thermodilution measurements. In patients, volumes and ejection fractions calculated from mask mode digital images correlated closely with direct left ventriculography. Global and segmental contractile function was displayed in patients by ejection shell images, stroke volume images, and time interval difference images. Central cardiovascular function was also quantitated by measurement of pulmonary transit time and calculation of pulmonary blood volume from digital fluoroscopic images. DVSA was shown to be useful and accurate in the quantitation of central cardiovascular physiology.

A noncomputerized count-based technique for the determination of resting left ventricular ejection fraction.

A noncomputerized, count-based technique for the determination of left ventricular ejection fraction (LVEF), which does not use geometric assumptions of left ventricular shape, was developed. The noncomputerized technique and computerized multigated ventriculography using both fixed and variable region-of-interest (ROI) methods were performed on 16 patients. The LVEFs obtained with the noncomputerized technique correlated well with both the fixed ROI computerized technique (r = .87) and the variable ROI computerized technique (r = .86). It is concluded that when a computer is not available, the noncomputerized technique is a valid alternative for the determination of LVEF in resting patients in stable sinus rhythm.

Regional myocardial hemodynamic and metabolic effects of ionic and nonionic contrast media in normal and ischemic states.

The effects of intracoronary injection of two nonionic contrast media (iohexol and metrizamide) on myocardial contraction and chemical composition of coronary sinus (CS) blood were compared with those caused by the standard ionic contrast material for coronary angiography, sodium meglumine diatrizoate (R76), in 14 anesthetized dogs. The effects of each agent on regional contractility were compared in the normal state and in the presence of a critical coronary artery stenosis. The three contrast media produced equivalent decreases in hematocrit and sodium (both NS), but R76 caused a greater increase in CS osmolality (p less than 0.02). R76 caused a significant decrease in CS potassium and ionized calcium (both p less than 0.001), but neither nonionic contrast medium caused a significant change in either potassium or calcium. In the normal state, R76 caused initial transient (less than 10 seconds) increases in both end-diastolic (p less than 0.006) and end-systolic segment length (p less than 0.02) and a decrease in rate of change of segment length (dL/dt) (p less than 0.002). The nonionic agents caused a mild increase in dL/dt (p less than 0.04) and a decrease in end-systolic segment length (p less than 0.03); both returned to control levels within 1 minute. In the presence of a stenosis, R76 caused a more severe and prolonged increase in end-diastolic and end-systolic segment lengths (p less than 0.03) and a decrease in dL/dt (p less than 0.002), which did not return to control within 2 minutes. The effects of the nonionic agents were similar in both normal and diseased states. We conclude that nonionic contrast media produce fewer alterations than ionic contrast media in coronary sinus blood chemistry and myocardial contractile state. The effect of ionic contrast media on regional contraction is accentuated in the presence of coronary artery stenosis.

Effects of low osmolality contrast materials on coronary hemodynamics, myocardial function, and coronary sinus osmolality in normal and ischemic states.

The effects of intracoronary administration of contrast materials on regional and global left ventricular (LV) function and coronary sinus osmolality were assessed in six anesthetized dogs with segmental myocardial ischemia produced by critical stenosis of the circumflex coronary artery. Effects caused by Renografin (sodium meglumine diatrizoate), two new low osmolality contrast agents (Hexabrix and Hexabrix with added calcium ions), and metrizamide were evaluated. In a nonischemic state, Renografin produced an early (0-10 seconds) decrease in LV contractility followed by a late (10-20 seconds) rebound augmentation in contractility. In the presence of regional ischemia, there are prolongation of the depression of the myocardial contractile state. The monoacid dimer, Hexabrix, demonstrated a similar biphasic response, although the initial depression of myocardial contractility was significantly less than that observed with Renografin. Hexabrix with added calcium ions and metrizamide produced only augmentation in global and regional parameters of LV contractile function. This lack of depressant effects was also observed in the ischemic state. Renografin caused a significantly greater increase in coronary sinus osmolality (Tp) as compared with Hexabrix, Hexabrix-Ca++, and metrizamide. The increases in osmolality in response to the latter three contrast agents were statistically indistinguishable.