Increased airway mucosal permeability of smokers. Relationship to airway reactivity

We studied pulmonary epithelial permeability and bronchial reactivity in 10 smoking and 8 nonsmoking adults. Permeability was measured as the disappearance half-life (T 1/2) of aerosolized 99mTc-DTPA from the lungs, and a permeability index (PI) calculated that reflected the appearance of the tracer in the blood. Smokers had increased permeability with a T 1/2 of 44.6 +/- 12.2 min and PI values at 10, 25, and 60 min of 27.3 +/- 13.2, 32.5 +/- 10.2, and 34.3 +/- 9.9, compared with those in nonsmokers with a T 1/2 of 110.0 +/- 62.7 min and PI values of 9.4 +/- 5.7, 14.9 +/- 8.3, and 23.1 +/- 9.0. Bronchial reactivity to histamine was measured with and without prior exposure to aerosolized propranolol (to achieve beta-blockade of airway smooth muscle). Reactivity increased significantly (p less than 0.001) in both groups after beta-blockade, but no difference was found between smokers and nonsmokers. Despite the increased permeability in smokers, there was no evidence of increased reactivity.     

Nailfold capillary permeability in psoriatic arthritis.

This study is the first report of the permeability status of nailfold capillaries in psoriatic arthritis (PA). "Traditional" nailfold capillary microscopy and intravital fluorescence videomicroscopy were carried out at the nailfold of 13 patients with PA. Twenty five healthy subjects served as controls for nailfold capillary microscopy, and 15 out of these for fluorescence videomicroscopy. The following parameters were assessed: capillary length, apex width, maximum loop width, maximum limb width, loop density, visibility of subpapillary venular plexus, loop tortuosity, transcapillary diffusion, and interstitial concentration at different sites and times of Na-fluorescein given in intravenous bolus. Morphometric analysis of capillaroscopic findings showed a significant increase of loop length (mean +/- SD: 290.1 +/- 73.5 microns) when compared to healthy controls (223.3 +/- 51.9 microns) (P < 0.02). Transcapillary passage of Na-fluorescein was homogeneous and symmetric both in PA patients and in controls. Mean transcapillary and interstitial diffusion was not significantly enhanced at the nailfold in PA patients. Our data support the view that PA is not characterized by a specific capillaroscopic pattern and/or significant abnormalities of microvascular dynamics at the nailfold.     

Intravital detection of skin capillary aneurysms by videomicroscopy with indocyanine green in patients with progressive systemic sclerosis and related disorders

Conventional capillaroscopy and infrared fluorescence videomicroscopy with indocyanine green were performed at the nailfold in 12 healthy controls and 38 patients with microangiopathy due to systemic sclerosis or related disorders. Saccular aneurysms featuring head and neck (type 1) and aneurysmatic enlargements (type 2) were defined. Microaneurysms were located at the apex or near the apex of capillary loops and were significantly more common in patients than in controls (p less than 0.02 for type 1 and p less than 0.001 for type 2). Combination of the two lesions was found only in patients and appears to be a valuable new diagnostic sign for the presence of microangiopathy. In comparison with conventional capillaroscopy, about twice as many microaneurysms were detected by videomicroscopy with indocyanine green coupling almost completely to plasma proteins. The new technique allows visualization of capillary aneurysms even when filled only by plasma.     

Dynamic intravital fluorescence microscopy--a novel method for the assessment of microvascular permeability in acute pancreatitis

Edema formation is the first manifestation of acute pancreatitis. Microcirculatory derangements like leukocyte-endothelial cell interaction and perfusion failure result in enhancement of microvascular permeability to large molecules playing a pivotal role in the progression of the acutely altered pancreatic tissue. Due to the lack of suitable methods the crucial mechanisms of enhanced permeability in vivo are not very well investigated. Sprague-Dawley rats were randomly assigned to three groups: (a) sham operated animals with normal pancreas, (b) the pancreatitis group induced by 60 min temporary occlusion of the arterial supply followed by reperfusion and (c) the histamine group in which the pancreas was superfused with 10(-5)M histamine. The pharmacokinetics of tetramethylrhodamine-labelled BSA in the intravital microscopic images of a capillary network of the pancreas were densitometrically quantified over 20 min. From these data the effective microvascular permeability was calculated taking also into account morphology of microvessels, elimination rate of the tracer from the intravascular space and capillary microhematocrit. In addition macromolecular leakage of gold-labelled BSA was investigated by electron microscopy. Microvascular permeability was 0.10 +/- 0.02 x 10(-7) cm/s, 0.49 +/- 0.04 x 10(-7) cm/s and 1.21 +/- 0.29 x 10(-7) cm/s for control, ischemia and histamine group, respectively (P < 0.05 ischemia, histamine vs. control and ischemia vs. histamine). Electron microscopy revealed albumin extravasation in the last two groups. We established a technique allowing to quantify microvascular permeability in pancreatic tissue by dynamic intravital microscopy being independent of the investigator. This technique enabling accurate pathophysiologic characterisation in terms of edema formation can form the basis for evaluating in the future novel treatment strategies directed against acute pancreatitis.     

Behavior and clinical relevance of histamine and leukotrienes C4 and B4 in grass pollen-induced rhinitis

The release kinetics of histamine and leukotrienes C4 (LTC4) and B4 (LTB4) were investigated in nasal secretions of 10 patients with hay fever after antigen challenge. High levels of biologically active histamine were found in nasal washes from asymptomatic allergic and normal subjects. With repeated lavages, the amount of histamine recovered dropped markedly. Grass pollen challenge was followed by a significant (p less than 0.05) dose-dependent and time-limited (5 min) increase in histamine level in 7 of 10 patients; these values, however, were lower than those found in basal conditions. In 8 of 10 patients with hay fever, antigen challenge induced a significant (p less than 0.05) dose-dependent increase in LTC4 level, which persisted for 30 min. The LTC4 generation was well correlated with the appearance of allergic symptoms; LTB4 production was found in 2 patients only. A different pattern of symptoms was observed after in vivo nasal stimulation with histamine and LTC4. Histamine caused sneezing, itching, rhinorrhea, and nasal obstruction; conversely, the main symptom induced by LTC4 was a more pronounced and longer lasting nasal obstruction.     

Transcutaneous oxygen tension and capillary morphologic characteristics and density in patients with chronic venous incompetence

The transparent oxygen electrode, recently developed by Huch and his co-workers, permits monitoring of transcutaneous oxygen tension (tcPO2) at defined sites on the capillaroscopic image obtained by videomicroscopy. This combined system has been applied to study the nutritional skin capillaries of patients with chronic venous incompetence (CVI). The results of 44 studies in 17 patients with CVI demonstrated a direct correlation between tcPO2 and density and morphologic characteristics of the superficial capillaries. The mean tcPO2 was 47.7 +/- 14.4 mm Hg at the site of incompetent perforating veins of the ankle without major trophic changes. There was no statistically significant difference between the mean values obtained in patients and control subjects (56.8 +/- 9.9 mm Hg). Videomicroscopic examination revealed dilated and tortuous capillaries surrounded by halo formations. In areas of hyperpigmentation, induration, and hyperkeratosis, significantly decreased mean tcPO2 (22.5 +/- 7.0 mm Hg; p less than .001) corresponded to reduced capillary density (less than 10 capillaries/mm2). In avascular skin areas (scar tissue, white atrophy) tcPO2 was measured at 0 mm Hg. No capillaries, or a greatly reduced number, were visible at such sites, resulting in a distance between capillary and cathode tip of the oxygen sensor of greater than 100 micron. The combined system of tcPO2 measurement and simultaneous videomicroscopy gives new pathophysiologic information on the development of skin ulcers and may be useful for the objective comparison of different therapeutic modalities at the microcirculatory level.     

Mechanism of airway narrowing caused by inhaled platelet-activating factor. Role of airway microvascular leakage

In order to study the mechanism of airway narrowing after inhaled platelet-activating factor (PAF) we measured concomitant changes in lung resistance (RL) and in airway microvascular leakage in anesthetized guinea pigs. RL and its recovery after hyperinflation at 5 min were measured until 6 min after PAF aerosol (0.1, 0.3, 1, and 3 mM), and in the case of 3 mM PAF also until 10 min. Microvascular leakage in trachea, main bronchi, and proximal and distal intrapulmonary airways was determined by measurement of extravasated Evans blue dye content. For comparison, the responses to inhaled histamine (3 mM) and 5-hydroxytryptamine (5HT) (3 mM), which act directly on airway smooth muscle, were also examined. Inhaled PAF increased RL dose-dependently, with a maximal response (peak RL) at 4 min after the inhalation, whereas the response to histamine or 5HT was maximal within a few seconds after the inhalation. Peak RL (cm H2O/ml/s) was significantly less after PAF (1.03 +/- 0.09) than after histamine (8.39 +/- 1.07) or 5HT (18.3 +/- 6.48), although there was no significant difference in RL after hyperinflation (recovery RL). No additional increase in RL was seen between 5 and 10 min after exposure. PAF caused a dose-dependent increase in Evans blue dye extravasation; 3 mM PAF induced significantly higher leakage than did histamine or 5HT at all airway levels at 6 min. PAF did not cause any additional extravasation of Evans blue dye at 10 min compared with that at 6 min after exposure.(ABSTRACT TRUNCATED AT 250 WORDS)     

Effects of histamine on coronary hemodynamics in humans: role of H1 and H2 receptors

To evaluate whether histamine exerts a direct effect on coronary hemodynamics in humans, and to investigate the role played by H1 and H2 receptors in this response, intracoronary saline solution or histamine (4 micrograms) was administered in 10 patients with normal coronary arteries during diagnostic cardiac catheterization. Histamine injection was repeated after intravenous cimetidine (400 mg) and diphenhydramine (10 mg). The electrocardiogram, arterial pressure and thermodilution coronary blood flow were continuously monitored during and for 40 seconds after each injection. Immediately after histamine injection there was a significant increase in coronary blood flow (65 +/- 6%) and a decrease in coronary vascular resistance (-40 +/- 3%) (both p less than 0.001), with minor changes in the RR interval and the mean arterial pressure. H2 receptor blockade with cimetidine did not affect these changes, while H1 receptor blockade with diphenhydramine significantly reduced the histamine-induced increase in coronary blood flow and the decrease in coronary vascular resistance (26 +/- 6%, p less than 0.005 and -18 +/- 5%, p less than 0.001, respectively). Twenty to 30 seconds after histamine injection, a significant decrease in mean arterial pressure (-17 +/- 2%, p less than 0.001) and in the RR interval (-4 +/- 1%, p less than 0.01) was observed. These changes persisted after H2 receptor blockade with cimetidine, but were completely abolished after H1 receptor blockade with diphenhydramine. In each case coronary and systemic hemodynamics returned to normal within 40 seconds of the injection.(ABSTRACT TRUNCATED AT 250 WORDS)     

Transient effects of histamine on the capillary filtration coefficient

The duration of the effect of histamine to increase the capillary filtration coefficient (CFC) was evaluated in isolated, denervated canine forelimb, hindpaw, and gracilis muscle. CFC was estimated at timed intervals during local intraarterial histamine (12 micrograms base.min-1 per 100 ml.min-1 blood flow). Propranolol (3 mg/kg) was administered to inhibit possible catecholamine-mediated inhibition of histamine-induced increases in CFC. The increase in CFC was greatest after 10 min of drug infusion and returned to control values after 25 min of histamine. These data indicate that the effect of histamine to increase CFC is highly transient. The relative contributions of increases in surface area and/or permeability to increases in CFC were assessed by maximally dilating the vasculatures of the three tissues with nitroprusside (increasing surface area to a maximum). Any further increase in CFC produced by combined nitroprusside-histamine infusion would then be due to increased permeability. Histamine, when infused concomitantly with nitroprusside, produced further increases in CFC relative to CFC obtained during infusion of nitroprusside alone. The time course for the transient increase in CFC during combined histamine-nitroprusside infusion was similar to the time course during histamine alone. These data suggest that the transient increase in CFC induced by histamine is primarily mediated by a transient increase in microvascular permeability to fluid in all three tissues. The transient nature of this increase in permeability was probably not related to a beta-antagonistic action of the catecholamines (which may have been increased reflexly) because these tissues were beta-blocked with propranolol. An equation was derived to estimate the ratio of the number of gaps which form between venular endothelial cells to the number of small pores. It was concluded that less than 3% of small pores need increase in radius to form large pores or gaps with radii ranging from 195 to 1000 A to explain the increases in CFC demonstrated in the hindpaw and gracilis muscle and that structures beyond the microvascular endothelium may provide the principle resistance to fluid efflux during histamine.     

Effects of H1-receptor stimulation on coronary arterial diameter and coronary hemodynamics in humans

Effects of H1-receptor stimulation on coronary arterial diameter and coronary hemodynamics were examined in 11 patients with angiographically normal coronary arteries and without variant angina or resting angina. Selective H1-receptor stimulation was achieved by infusing histamine into the left coronary artery at a rate of 2.0 micrograms/min for 5 minutes after pretreatment with cimetidine (25 mg/kg). Plasma histamine concentration in the coronary sinus, coronary sinus blood flow, heart rate, and aortic pressure were measured before, during, and after the histamine infusion. Coronary arterial diameter was measured by cinevideodensitometric analysis of coronary arteriograms performed before and immediately after the histamine infusion. During the histamine infusion, plasma histamine concentration in the coronary sinus increased from 0.33 +/- 0.06 to 5.86 +/- 0.71 ng/ml (p less than 0.01); coronary sinus blood flow increased from 98 +/- 12 to 124 +/- 13 ml/min (p less than 0.01), and coronary vascular resistance decreased from 1,113 +/- 117 to 851 +/- 91 mm Hg.min/l (p less than 0.01). Heart rate and aortic pressure remained unchanged. The mean luminal diameters of the proximal, middle, and distal left anterior descending artery increased by 9.4 +/- 3.6% (p less than 0.05), 19.2 +/- 3.8% (p less than 0.001), and 31.5 +/- 5.6% (p less than 0.001), respectively, after the histamine infusion. The mean luminal diameters of the proximal, middle, and distal left circumflex artery increased by 15.2 +/- 3.6% (p less than 0.01), 17.5 +/- 5.2% (p less than 0.01), and 20.6 +/- 4.3% (p less than 0.001), respectively, after the histamine infusion.(ABSTRACT TRUNCATED AT 250 WORDS)     

The influence of oxidative stress on permeability of capillary vessels in the cheek pouch of hamsters with alloxan-induced diabetes

BACKGROUND: The aim of the study was to assess the influence of oxidative stress on the increase of permeability of capillary vessels in animals with alloxan-induced diabetes. MATERIAL AND METHODS: The studies were performed in microcirculation system of hamster cheek pouch. After the blockade of histamine receptors and administration of diamine oxidase (DAO) and histamine into circulation fluorescein angiography was done. In addition, the influence of superoxide dismutase, aminoguanidine (DAO inhibitor) and trascolan (protease inhibitor) on vascular permeability caused by superoxide radical generation in DAO/histamine system was assessed. RESULTS: The number of extravasal leakages in the group receiving HA and DAO was significantly higher (p < 0.001) than in the groups receiving potential vascular "sealers", e.g. SOD, aminoguanidine or trascolan. In the group receiving aminoguanidine the number of leakages was significantly lower (p < 0.05) compared to the group receiving SOD or trascolan. CONCLUSIONS: The protective effect of aminoguanidine, superoxide dismutase or trascolan decreasing the vascular permeability, suggests that the increased vascular permeability is a result of superoxide radical generation by diamine oxidase.     

Microvascular dynamics at the nailfold in rheumatoid arthritis

Conventional and fluorescence videomicroscopy using Na-fluorescein as tracer was performed in 13 patients with rheumatoid arthritis (RA) and Waaler Rose titers less than 1: 112, 10 patients with titers greater than or equal to 1: 112 and 11 patients with osteoarthritis (OA) (control group). At the nailfold different microvascular diameters, capillary flow pattern, red blood cell velocity and transcapillary diffusion of Na-fluorescein were determined. Mean red blood cell velocity was significantly (p less than 0.02) decreased in patients with RA and high titers (0.23 +/- 0.21 0.21 mm/s) when compared to OA patients (0.59 +/- 0.20 mm/s). Abnormal flow patterns, in 2 cases even a reversal of flow direction, were only observed in RA patients and not in controls. Mean transcapillary and interstitial diffusion of Na-fluorescein was not enhanced at the nailfold in RA-patients. These findings do not exclude increased microvascular permeability in synovial membranes or in RA patients with vasculitis.     

Effects of platelet-activating factor on vascular permeability and granulocyte recruitment in the rat trachea.

Platelet-activating factor (PAF) is a phospholipid mediator known to produce several features of airway inflammation. We examined the effects of intravenous PAF on vascular permeability and granulocyte recruitment in the rat trachea. To assess vascular permeability, anesthetized rats were given injections of Evans blue dye (30 mg/kg, iv) and PAF (1-10 micrograms/kg, iv), and then their tracheas were removed and assayed spectrophotometrically for dye content. We found that a PAF dosage of 6 micrograms/kg increased the tracheal dye content 7-fold compared to controls. The amount of extravasated dye in the tracheas was significantly increased 1 min after PAF injection, was maximal at 5 min, and had returned to control levels by 10 min. To assess granulocyte recruitment, anesthetized rats were given an injection of PAF (6 micrograms/kg, iv), and then their tracheas were removed and stained to reveal myeloperoxidase-containing neutrophils and eosinophils. We found that the number of neutrophils in the tracheal mucosa was increased 7-fold from controls 5 min after PAF injection, but was not significantly increased 6 h later. The number of eosinophils in the tracheal mucosa was not significantly increased from controls at any time after PAF injection. We conclude that intravenous PAF causes a rapid but transient increase in vascular permeability in the rat trachea, and that intravenous PAF also causes a rapid but transient recruitment of neutrophils into the tracheal mucosa without a similar effect on eosinophils.     

The effects of histamine on contraction frequency, sodium influx, and cyclic AMP in cultured rat heart cells

Histamine has been shown to have both positive inotropic and chronotropic effects. To evaluate the chronotropic effects, spontaneously contracting monolayers of cultured rat myocardial cells were treated with histamine, 10(-7) M-10(-4) M. This resulted in a dose-dependent increase in contraction frequency reaching a maximum in 10(-5) M histamine. Contraction frequency (mean +/- SEM) increased from a control of 121 +/- 5 contractions per minute to 153 +/- 4.5, 181 +/- 9, 212 +/- 4, and 216 +/- 1 in 10(-7) M, 10(-6) M, 10(-5) M, and 10(-4) M histamine, respectively (for each n = 10, p less than 0.001). The effect was time-dependent, taking 30 minutes to develop fully. Changes in contraction frequency were accompanied by parallel dose- and time-dependent increases in the verapamil-sensitive sodium influx. Verapamil-sensitive sodium influx (pmol/cm2/sec) increased from a control of 10.45 +/- 1.44 (mean +/- SEM) to 24.34 +/- 2.41 and 32.57 +/- 2.35 at 10- and 30-minute treatment with 10(-6) M histamine (n = 5, p less than 0.001). These data fit the previously described relation between verapamil-sensitive sodium influx and contraction frequency in these cells. Cimetidine (10(-4) M) but not diphenhydramine (10(-4) M) abolished both the contraction frequency and sodium influx response to histamine. Subsequent studies showed a dose- and time-dependent elevation of cyclic adenosine monophosphate (cAMP) with histamine treatment.(ABSTRACT TRUNCATED AT 250 WORDS)     

Effect of histamine on electrical properties of airway epithelium and its antagonism by azelastine]

To elucidate the mechanism of the action of histamine on bioelectric properties of the airway epithelium and the fact that the anti-allergic agent azelastine is possibly antagonistic to it, we studied canine cultured tracheal epithelium under short-circuit conditions in vitro. Addition of histamine to the submucosal but not mucosal bath of Ussing chamber increased short-circuit current (Isc) in a dose-dependent fashion, the maximal increase and the EC50 value being 5.7 +/- 0.9 microA/cm2 (mean +/- SE, p less than 0.001) and 3 x 10(-6) M, respectively. Pretreatment of tissues with pyrilamine abolished the histamine-induced increase in Isc (p less than 0.001), whereas cimetidine had no effect. In addition, the histamine action on Isc was effectively depressed by the Cl channel blocker diphenylamine-2-carboxylate, Cl-free medium, and indomethacin. The release of PGE2 and PGF2 alpha was significantly enhanced by the submucosal application of histamine. Pretreatment of tissues with azelastine dose-dependently attenuated the increase in Isc induced by 10(-4) M histamine. These results indicate that histamine increases the airway epithelial Isc via stimulation of prostaglandin synthesis and subsequent movement of Cl toward the lumen, and that azelastine may be useful in preventing histamine-induced water secretion in the airway.     

An experimental rat model for studying pulmonary microcirculation by in vivo videomicroscopy.

It is unclear what role pulmonary microcirculatory disorders play in the pathogenesis of adult respiratory distress syndrome. The aim of this study was to establish a rat model for the direct visualization of pulmonary microcirculation by in vivo fluorescence videomicroscopy. The pulmonary terminal vascular bed was visualized and the microcirculatory parameters of leukocyte sticking, erythrocyte velocity, capillary permeability, and interalveolar septal diameter were quantified. These parameters were examined simultaneously. The preparation was stable for 120 min. Under hyperthermia, there was increased permeability with a relative fluorescence of 0.39 +/- 0.19 compared to 0.16 +/- 0.13 in the control group, and interalveolar septal diameters were wider (30.7 +/- 2.9 microm) than in control animals (17.3 +/- 3 microm). Under hypothermia and hypovolemia, the erythrocyte velocity was lower (0.351 +/- 0.063 and 0.378 +/- 0.044 mm/s) than in control groups (0.527 +/- 0.07 mm/s). Under hypoventilation, we observed a higher amount of leukocyte sticking (3.1 +/- 1.1 vs 1.8 +/- 0.8 cells/alveolus) and increased permeability (relative fluorescence 1.03 +/- 0.37 vs 0.16 +/- 0.13 in the control group). The model of rat lung exposure for direct examination of microvascular structures in living animals was valuable because it remained stable for 2 h under baseline conditions and demonstrated distinct changes in microcirculatory parameters following specific pathophysiological interventions.     

Influence of left ventricular filling profile on the effect of atrioventricular synchronous pacing.

We correlated the percentage of atrial contribution to left ventricular filling (percent AC) assessed by Doppler echocardiography with the hemodynamic benefit from atrioventricular synchronous pacing assessed by direct hemodynamic measurements. Subjects comprised 40 patients who underwent electrophysiologic catheterization because of unexplained syncope or bradycardia (< 40 beats/min). Femoral arterial and pulmonary capillary wedge pressure were recorded by catheterization, and cardiac output was measured by thermodilution during temporary atrioventricular synchronous (DDD, 70 beats/min with 150 ms of atrioventricular delay) and ventricular (VVI, 70 beats/min) pacing. Mitral inflow velocity by pulsed-wave Doppler echocardiography was recorded during DDD pacing and percent AC was obtained by calculating the ratio of mitral inflow velocity area during atrial systole to total mitral inflow velocity area during early diastole and atrial systole. The mean arterial pressure and the cardiac output increased significantly (99 +/- 16 mm Hg vs 90 +/- 15 mm Hg, p < 0.001; 4.6 +/- 1.0 L/min vs 3.9 +/- 0.9 L/min, p < 0.001), and the mean pulmonary capillary wedge pressure decreased (7 +/- 4 mm Hg vs 10 +/- 4 mm Hg, p < 0.001) during DDD compared with VVI pacing. A significant positive correlation was observed between the percent AC and the increase in cardiac output (r = 0.58, n = 40, p < 0.01) or the increase in mean arterial pressure (r = 0.62, n = 38, p < 0.01) during DDD pacing. The percent AC did not significantly correlate with the decrease in pulmonary capillary wedge pressure. In conclusion, patients with larger percent AC may receive major benefit from atrioventricular synchronous pacing.     

Comparison of high and low osmolar roentgen contrast media in quantitative coronary angiography

In 48 patients undergoing diagnostic coronary angiography changes of mean diameters of angiographically "normal" coronary segments after intracoronary injections of diatrizoate 76% or iopromide 370 performed in different intervals, were analyzed with a computer-assisted contour detection system (CAAS). Four study protocols were applied, differing in respect to the type of contrast medium administered and/or to the timing of the reference- and control-angiograms in the course of diagnostic coronary angiography. Coronary angiograms in identical projections were performed before (= reference) and directly after (1. control = C1) diagnostic angiography of the left coronary artery by injection of either diatrizoate 76% (group 1, 10 patients) or iopromide 370 (group II, 11 patients). Additional coronary angiograms were performed 1, 3, 6, and 10 min after C1. During diagnostic angiography in either group about eight dye injections were performed in about seven min. With diatrizoate 76% a significant coronary dilation averaging 18.9 +/- 6.7% (p less than 0.001) was observed at C1, depending on the number of diagnostic dye injections performed per min (mean 1.2 +/- 0.3) and on the interval between the last diagnostic injection and C1 (mean 73 +/- 35 s). Coronary dilation persisted up to the sixth minute (6.2 +/- 4.6%, p less than 0.01). With iopromide 370 a small but significant coronary dilation was observed merely at C1 (5.8 +/- 4.3%, p less than 0.05). In two other studies coronary angiograms were performed in identical projections immediately following complete diagnostic coronary angiography (reference) and in addition after 3, 4, 5, 6, 10, and 20 min (group III, 18 patients) and after 10, 20, and 30 min, respectively (group IV, 9 patients) by administration of diatrizoate 76% as the only contrast medium. Short injection intervals (1 min) resulted in a mild coronary dilation (mean up to 2.4 +/- 4.1% compared to reference; p less than 0.05), longer intervals (3-10 min) resulted in a marked diameter reduction (averaging up to -9.7 +/- 9%; p less than 0.05), probably a consequence of the return of coronary vasomotor tone to baseline levels. These results suggest that in quantitative coronary angiographic studies (e.g., testing coronary vasomotility) non-ionic contrast media should preferably be applied, and adequate injection intervals (greater than 2 min) are mandatory. In intervention- and follow-up studies based on repeated coronary angiograms dye-induced changes of coronary vasomotor tone can be avoided by premedication with vasodilating drugs, e.g. nitrates, and/or calcium antagonists.