This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Emery, C J Articles by Morcos, S K Search for Related Content PubMed PubMed Citation Articles by Emery, C J Articles by Morcos, S K

Effects of radiographic contrast media on pulmonary vascular resistance of normoxic and chronically hypoxic pulmonary hypertensive rats

¹Respiratory Medicine, Sheffield University Medical School, Sheffield S10 2JF and ²Department of Diagnostic Imaging, Northern General Hospital, Sheffield Teaching Hospitals NHS Trust, Sheffield S5 7AU, UK

Correspondence: Dr S K Morcos, X-Ray Department, Northern General Hospital NHS Trust, Sheffield S5 7AU, UK

	Abstract

Top Abstract Introduction Materials and methods Results Discussion References

 
Intravascular radiographic contrast media (RCM) can be associated with significant morbidity in patients with pulmonary hypertension (PH). This study investigated the direct effect of the four main classes of RCM (high osmolar ionic monomer "diatrizoate"; low osmolar ionic dimer "ioxaglate"; low osmolar non-ionic monomer "iopromide"; and iso-osmolar non-ionic dimer "iotrolan") in ex vivo isolated rat lungs perfused with blood at 20 ml min^-1 under basal conditions (air + 5% CO₂ ventilation, pulmonary artery pressure (Ppa) 16–20 mmHg) and when Ppa was raised by hypoxic vasoconstriction in normal rats (2–3% O₂+5% CO₂ ventilation, Ppa increased by 4–14 mmHg). The effects of low osmolar RCM (ioxaglate, iopromide and iotrolan) were also studied in rats with PH induced by chronic hypoxia (3 weeks 10% O₂, Ppa 26–36 mmHg). Increasing volumes (0.05 ml, 0.1 ml, 0.3 ml, and 0.5 ml) of RCM, mannitol (osmolar and pH control) or normal saline (volume control) were added to the 10 ml blood reservoir (n=4–9 per group). In normal rats, RCM caused a dose-dependent slow rise in Ppa. The maximum rise in mean±SEM Ppa at the cumulative dose of 0.95 ml was ioxaglate 13.8±1.6 mmHg>iotrolan 7.3±1.7 mmHg=diatrizoate 9.8±2.2 mmHg>iopromide 3.0±0.8 mmHg (p<0.05). The rise in Ppa induced by ioxaglate and iotrolan was significantly greater than in the mannitol and saline controls (p<0.05). Pre-treatment with endothelin receptor A/B blockade (SB209670) did not abolish the rise in Ppa induced by diatrizoate (0.95 ml) in the normal rat (3.8±1.3 mmHg diatrizoate alone and 3.4±1.1 mmHg in the presence of 40 µM SB209670, n=5 per group). When Ppa was raised by acute hypoxia, ioxaglate and diatrizoate (0.5 ml) caused a fall in Ppa (percentage fall -53±23 and -118±10, respectively, p<0.001) while iotrolan and iopromide caused a small further rise in Ppa, which was significant with iotrolan at a dose of 0.3 ml (percentage rise in pressure 14.2±2.3, p<0.05). In chronic pulmonary hypertensive rats, RCM (0.95 ml) caused an overall slow progressive rise in Ppa (iopromide 6.8±1.7 mmHg< ioxaglate 11.6±2.5 mmHg=iotrolan 12.7±1.1 mmHg). However, ioxaglate initially induced an acute fall of Ppa (maximum fall 4.22±0.9 mmHg, p<0.05) for almost 20 min. In summary, iopromide induced the least change in Ppa of normal and pulmonary hypertensive rats. The pathophysiology of the effects of RCM on the pulmonary circulation remains uncertain.

	Introduction

Top Abstract Introduction Materials and methods Results Discussion References

 
It is well documented that administration of radiographic contrast media (RCM), either high, low or iso-osmolar, into the pulmonary circulation is accompanied by an acute rise in pulmonary artery pressure (Ppa) [1–6]. This has been considered to be contributory to the morbidity and mortality associated with pulmonary angiography, particularly in patients with pulmonary hypertension (PH) [7–13]. The mechanisms responsible for the rise in Ppa remain uncertain and could be secondary to an increase in pulmonary vascular resistance (PVR) owing to pulmonary arterial vasoconstriction either by direct action of RCM on vascular smooth muscle or through modulating the release of endogenous vasoactive mediators such as endothelin (ET), histamine, serotonin or nitric oxide [14–20]. Neural reflexes causing pulmonary vasoconstriction, and increased capillary resistance due to a change in the rheological properties of the red blood cells may also play a role in RCM-induced rise in PVR [20–23]. However, the rise in Ppa induced by RCM is associated with an increase in cardiac output and a fall in PVR [2, 3]. The increase in cardiac output was attributed to reduced peripheral vascular resistance of the systemic circulation owing to RCM-induced vasodilatation [2]. The suggested fall in PVR was calculated from the formula PVR=(pulmonary artery pressure-pulmonary venous pressure)/cardiac output. The fall in PVR could be owing to an increase in the capacity of the pulmonary vascular bed by recruitment of closed vessels and active vasodilatation of pulmonary arteries. However, this observation conflicts with previous findings of the direct effects of RCM on small pulmonary arteries that are characterized by sustained vasoconstriction following an initial transient vasodilatation [14].

In this study the direct effects of RCM on PVR, independent of systemic influences or changes in cardiac output, were investigated utilizing the isolated perfused rat lung preparation with constant flow blood in which changes in Ppa reflect changes in resistance [24]. The effects of RCM, not only in the normal pulmonary vascular bed but also in the abnormal pulmonary hypertensive vascular bed, were studied. To our knowledge, this approach utilizing the rat model of pulmonary hypertension secondary to chronic hypoxia has not been used before in studying the effects of RCM on the pulmonary circulation. In this model the rats develop PH with raised PVR, right ventricular hypertrophy and pulmonary vascular remodelling similar to the changes observed in man [24]. The structural changes, which include an increase in the thickness of the muscle wall of the small pulmonary blood vessels due to newly developed muscle cells, may exaggerate the reactivity of the blood vessels to vasoactive stimuli including RCM [24]. Previous studies relied mainly on experimental models of acute pulmonary hypertension induced by multiple emboli of the pulmonary circulation [9, 10, 25]. The chronic hypoxic model is more relevant to the clinical situation of chronic PH where structural changes in the pulmonary arteries are present. The role of ET in mediating the effects of RCM on PVR has also been investigated.

	Materials and methods

Top Abstract Introduction Materials and methods Results Discussion References

 
The effects of RCM on PVR were tested in isolated perfused lungs (PLs) from normal and chronic hypoxic-induced pulmonary hypertensive male Wistar rats. In normal rats the effects were investigated under normoxic and acute hypoxic ventilation.

Materials
Mannitol (osmolar control) solutions were prepared by dissolving mannitol in de-ionized water to match the RCM osmolality. pH was adjusted to equal that of the RCM by addition of 0.1 M sodium hydroxide. For iotrolan, the mannitol control solution pH was 7.2. It was not possible to supersaturate mannitol to achieve the high osmolarity of diatrizoate (2070 mosmol kgH₂O^-1) and thus a saturated solution of mannitol at room temperature (1000 mosmol kgH₂O^-1) was used as the diatrizoate control. Osmolality was measured by an osmometer (Advanced Instruments Inc., Norwood, MA). The drugs used were diatrizoate (Urografin 370 mgI ml^-1, 2070 mosmol kgH₂O^-1, pH 6.8, viscosity 8.5 cP at 37°C; Schering AG, Berlin, Germany), ioxaglate (Hexabrix 320 mgI ml^-1, 580 mosmol kgH₂O^-1, pH 6.2, viscosity 7.5 cP at 37°C; May & Baker Ltd, Dagenham, UK), iopromide (Ultravist 300 mgI ml^-1, 610 mosmol kgH₂O^-1, pH 7.1, viscosity 4.3 cP at 37°C; Schering AG, Berlin, Germany), iotrolan (Isovist 300 mgI ml^-1, 320 mosmol kgH₂O^-1, pH 6.4–8.0, viscosity 8.1 cP at 37°C; Schering AG, Berlin, Germany), D-mannitol (Sigma Chemical Co., St Louis, MO), heparin (Leo Labs Ltd, Buckinghamshire, UK), pentobarbitone (Sagatal, Rhone Merieux, UK), ET-1 (Sigma, Poole, UK) and SB209670 (kindly donated by Dr Ohlestine; SmithKline Beecham, King of Prussia, PA). RCM were kept in the dark at room temperature in the suppliers' sealed glass bottles. Mannitol solutions were kept at 4°C. ET and SB209670 were kept at -18°C.

Normal rats
8-week-old (250–300 g body weight) male Wistar rats (Sheffield strain) were studied.

Chronically hypoxic pulmonary hypertensive rats
A rat model of chronic hypoxic-induced PH was studied. 5-week-old male Wistar rats were kept in a normobaric hypoxic environmental chamber for 3 weeks at 10% O₂, as previously described by Emery et al [24].

Models
Rat IPL model (Figure 1)
This model is well established in our laboratories [24]. Each rat was anaesthetized with intraperitoneal pentobarbitone (60 mg Kg^-1). After cannulation of the trachea, a rat was placed over a heated water-bath kept at 40°C. After laporotomy the rat was heparinized (heparin 1000 U kg^-1) and exsanguinated from the inferior vena cava. The thorax was then opened midsternally and fixed ventilation was maintained using a Harvard respiratory pump connected to the cannulated trachea (48 breaths min^-1). The lung was ventilated with air+5% CO₂. The pulmonary artery and left atrium were cannulated and autologous blood from a heated reservoir (8–10 ml, temperature 39°C) was perfused via a variable speed, constant flow roller pump (Watson, Marlow, UK) through the lung at a constant flow of 20 ml min^-1, giving basal Ppa in the range 16–20 mmHg. Blood pH was measured using a Corning (Halstead, UK) blood gas analyser and was adjusted to within the range 7.35–7.45 by adding 0.8% sodium bicarbonate (1 mol l^-1) to the blood reservoir. While normal rats were perfused with autologus blood, chronic hypoxic (CH) rat lungs were perfused with heparinized blood from normal donor Wistar rats (heparin 1000 U kg^-1). Autologous blood of CH rats was not used in order to avoid the haematological changes associated with chronic hypoxia, which include polycythaemia and changes in the circulating vasoactive autacoids. To eliminate these factors that could influence the PVR we used blood from normal rats. This is a well established approach in the experimental use of IPL of CH rats [24]. Ppa was measured from a side tube in the blood circuit close to the pulmonary artery (Figure 1). As flow and left atrial pressure are constant, changes in Ppa reflect changes in PVR [24].

View larger version (21K): [in this window] [in a new window]   Figure 1. Schematic diagram of the isolated blood-perfused rat lung preparation. PA, pulmonary artery; LA, left atrium.

Experimental protocol
Normal rats
The effects of RCM (diatrizoate, high osmolar ionic monomer 370 mgI ml^-1; ioxaglate, low osmolar ionic dimer 320 mgI ml^-1; iopromide, low osmolar non-ionic monomer 300 mgI ml^-1; and iotrolan, iso-osmolar non-ionic dimer, 300 mgI ml^-1), mannitol (osmolar/pH control) or normal saline (volume control) were studied in isolated blood-perfused rat lungs (n=5–9 per group), either under normal ventilation (air+5% CO₂) or when Ppa was raised by hypoxic vasoconstriction (ventilation with hypoxic gas, 2–3% O₂+5% CO₂). The observation period of each experiment was approximately 150 min. From normoxia, 0.05 ml, 0.1 ml, 0.3 ml and 0.5 ml doses of RCM, or equivolume of saline or mannitol were added sequentially to the reservoir. The Ppa was allowed to stabilize between doses.

Acute hypoxia caused Ppa to rise to a stable pressure. The effects of 0.1 ml, 0.3 ml and 0.5 ml doses of RCM or saline on the raised Ppa were assessed sequentially. After each dose, the Ppa was allowed to stabilize before the lungs were ventilated with normoxic gas (air+5% CO₂), causing the Ppa to return to baseline. This was followed by examining the effect of another dose of the test solution during hypoxic ventilation, employing the same protocol.

Each IPL was tested with one radiographic contrast medium, mannitol or saline, either from the normoxic baseline or from acute hypoxia.

Chronic hypoxic rats
The effects of low osmolar RCM (ioxaglate, iopromide and iotrolan) or normal saline (0.05 ml, 0.1 ml, 0.3 ml and 0.5 ml) were studied in IPLs of CH rats (n=3–6 per group) under normal ventilation (air+5% CO₂), utilizing the same protocol as described for normal rats.

The effects of high osmolar diatrizoate on CH rats was not investigated as it is well documented that a further significant increase in Ppa is often observed after the intravascular administration of high osmolar RCM and that non-ionic media are better tolerated in the presence of PH [12, 19–22].

The effect of ET-1 receptor blockade on RCM-induced increase in Ppa during normoxia
In the normal rat IPL, the constrictor response to a bolus dose of ET-1 (100 µg) given into the circuit tubing close to the pulmonary artery during normoxia was abolished by the ET-1 receptor antagonist SB209670 (40 µM). The effect of diatrizoate (0.05 ml, 0.1 ml, 0.3 ml and 0.5 ml) in the IPL with or without pre-treatment with 40 µM SB209670 was studied. We used only the high osmolar diatrizoate in this experiment since this class of RCM is most likely to induce the largest effects on Ppa.

Analysis of results
Results are expressed as a mean±SEM. When drugs were given with Ppa raised by acute hypoxia, the results are expressed as a percentage change of the rise in Ppa with hypoxia (HPV), measured from the baseline immediately prior to the hypoxic test. Statistical analysis was performed using ANOVA and paired and unpaired t-test as appropriate. A p-value d0.05 was considered significant.

	Results

Top Abstract Introduction Materials and methods Results Discussion References

 
Normoxic rats
From basal Ppa (16–20 mmHg) under ventilation with air+5% CO₂, all RCM caused a slow sustained rise in Ppa, reaching stability within 20–30 min (Figure 2). This rise was significantly greater (p<0.05) in comparison with saline (volume control), except for iopromide (Figure 3). The order of potency (maximum effect with cumulative dose of 0.95 ml) was ioxaglate>iotrolan=diatrizoate>iopromide. No significant changes in Ppa were observed with saline (volume control) solutions (maximum rise of Ppa was 1.4±0.6 mmHg observed at the cumulative dose of 0.95 ml, n=9).

View larger version (17K): [in this window] [in a new window]   Figure 2. Change (mean±SEM) in pulmonary artery pressure (Ppa mmHg) from baseline in response to cumulative volumes of radiographic contrast media (RCM) in the normal rat isolated perfused lung preparation (normoxic ventilation). ****Iopromide vs other RCM (p<0.05); ***ioxaglate vs iotrolan and diatrizoate and iopromide (p<0.05); **ioxaglate vs iotrolan (p<0.05); *iopromide vs iotrolan (p<0.05).

View larger version (15K): [in this window] [in a new window]   Figure 3. Rise in pulmonary artery pressure (Ppa) (mean±SEM) in response to a cumulative volume of 0.95 ml of radiographic contrast media (RCM), mannitol solutions (osmolar control) or normal saline (volume control) in normal rat isolated perfused lung preparation (normoxic ventilation) from baseline (n=6–9). IOP, iopromide; IOT, iotrolan; IOX, ioxaglate; D, diatrizoate; IOPM, IOTM and IOXM, RCM-matched mannitol (osmolar control) solutions (for diatrizoate the osmolar control solution DM was saturated mannitol); S, normal saline; *Saline vs D, IOX, IOT and SM (p<0.05); **RCM vs matched mannitol (p<0.01). All tested solutions, with the exception of iotrolan mannitol control and normal saline, induced a significant (p<0.05) rise in Ppa compared with the baseline.

Acute hypoxic vasoconstriction in normal rats (Figure 4)
Hypoxic ventilation (2–3% O₂+5% CO₂) caused a stable, reproducible rise in Ppa of 4–14 mmHg, reversible on return to normoxic ventilation. From this pre-constricted state, diatrizoate and ioxaglate (0.5 ml) caused significant (p<0.0001) falls in Ppa (Figures 4a,b). Iotrolan caused a small further rise in Ppa (p<0.05) (Figure 4c). Iopromide caused a biphasic response, an initial transient fall in Ppa followed by a non-significant rise (Figure 4d). Saline volume control solutions did not induce significant changes in Ppa.

View larger version (42K): [in this window] [in a new window]   Figure 4. Effects of (a) diatrizoate, (b) ioxaglate, (c) iotrolan and (d) iopromide on pulmonary artery pressure (Ppa mmHg) when raised by ventilation with hypoxia (2% O2+5% CO2). Results are expressed as percentagechange (mean±SEM) of the rise in Ppa with hypoxia from baseline (HPV). A schematic diagram of the changes in Ppa in response to 0.3 ml of RCM is shown. (N.B. The time-scale of the schematic diagram is the same for each figure.)

Acute response
Iopromide caused an initial fall in Ppa followed by a rapid (within 2 min) dose-dependent rise in pressure (Figure 5a). The initial fall was not obvious with iotrolan, which caused mainly a rapid rise in Ppa (Figure 5b). Ioxaglate caused an immediate dose-dependent fall in Ppa, followed by gradual recovering to baseline over a 20 min period (Figure 5c). The initial response to normal saline was a volume-dependent transient (<1 min duration) fall in Ppa followed by recovery to baseline.

View larger version (34K): [in this window] [in a new window]   Figure 5. Initial maximum acute changes (mean±SEM) in pulmonary artery pressure from baseline in chronic hypoxic pulmonary hypertensive rats in response to radiographic contrast media. (a)Schematic diagram of the changes in Ppa in response to 0.5 ml of iopromide shows an initial fall followed by a rise in Ppa within 2 min of exposure (*p<0.05 in comparison with baseline). (b) Schematic diagram of the changes in Ppa in response to iotrolan shows an initial fall followed by a rise in Ppa with 0.1 ml and 0.3 ml of iotrolan, but no initial fall and an immediate rise in Ppa with a 0.5 ml dose (*p<0.05 in comparison with baseline). (c) Schematic diagram of the changes in Ppa in response to ioxaglate shows a significant sustained fall in Ppa that lasted for approximately 20 min (**p<0.01 in comparison with baseline). (N.B. The time-scale of the schematic diagram is the same for each figure.)

View larger version (12K): [in this window] [in a new window]   Figure 6. The maximum rise in pulmonary arterial pressure (Ppa) (mean±SEM) from baseline in response to cumulative 0.95 ml of radiographic contrast media and normal saline in chronic hypoxic-induced pulmonary hypertensive rats. IOP, iopromide; IOT, iotrolan; IOX, ioxaglate; S, normal saline. All solutions caused a significant rise in Ppa from baseline (p<0.05). *, Saline vs IOT and IOX (p<0.05).

	Discussion

Top Abstract Introduction Materials and methods Results Discussion References

 
This study demonstrates that RCM can directly affect PVR in rat IPL. These effects are not entirely dependent on osmolality but are also influenced by the physicochemical properties of RCM and the baseline pulmonary vascular tone. In IPL of the normal rat, increasing doses of RCM (iotrolan, iopromide, ioxaglate and diatrizoate) and hypertonic solutions of mannitol caused an overall rise in Ppa, reflecting an increase in PVR. No significant changes were observed in Ppa throughout the observation period in the volume control group that received only normal saline, confirming the absence of volume effect and the stability of the preparation with time. In this preparation, where blood flow and ventilation are controlled, a rise in Ppa usually reflects an increase in PVR whilst a fall reflects a decrease in resistance [24].

The increase in PVR induced by RCM is most likely due to a combination of active vasoconstriction of the pulmonary arteries and possible increase in blood viscosity. The latter could be secondary to cellular effects (increased aggregation of red blood cells with non-ionic media and rigidity with high osmolar solutions) and/or the high viscosity of some of the contrast agents [20, 23, 26, 27]. RCM may also activate adhesion of leucocytes to the endothelium, which causes capillary plugging and stasis of red blood cells in the small vessels, precipitating an increase in vascular resistance [20]. Vasoconstriction is likely to play an important role in the RCM-induced increase in PVR since, in a previous study using the experimental model of the myograph, we have demonstrated that RCM can cause sustained constriction of the small pulmonary arteries [14]. The most potent constrictor was diatrizoate at calculated iodine concentrations below 50 mgI ml^-1, and ioxaglate at calculated iodine concentrations of 200 mgI ml^-1 or more. The least vasoactive radiographic contrast medium was iotrolan at all iodine concentrations [14]. The maximum calculated iodine concentration in the blood perfusate was less than 50 mgI ml^-1 in all experiments in this study (refer to methods). In normoxic rats where basal vascular tone is low, ioxaglate caused the greatest rise in Ppa. However, the vasoconstrictor effect of ioxaglate in the myograph study was smaller in comparison with diatrizoate and iopromide at iodine concentrations below 50 mgI ml^-1. This discrepancy between the myographic observations and the findings of this study is not fully explained. Development of pulmonary oedema may have contributed to the observed increase in Ppa. Accumulation of fluid in the lung can lead to an increase in PVR and a rise in Ppa. It is tempting to postulate that, in addition to vasoconstriction, ioxaglate-induced pulmonary oedema is precipitated by endothelial injury and an increase in the permeability of the microcirculation. Previous studies have shown that ioxaglate is more cytotoxic to the vascular endothelium compared with diatrizoate and non-ionic media [29, 30]. The contribution of pulmonary oedema to the observed increase in PVR was not assessed in this study, since histological examination of the lungs was not performed at the end of each experiment, and this should be the subject of further investigations. The effect of high osmolar diatrizoate was comparable with that of hypertonic saturated mannitol control solutions, suggesting that the effect of this agent is mainly osmolality dependent. High viscosity and rheological effects, which can reduce the erythrocyte velocity in the microcirculation, could be responsible for the increase in the vascular resistance induced by iotrolan in the isolated lung preparation perfused with blood [20, 23, 26]. The haemorheological effects may explain the difference between the findings of this report and that of the myograph study. While iotrolan, had the lowest effect on vascular tone in the myograph [14], in rat IPL it induced a significant increase in PVR that was comparable with the effect of diatrizoate. Iopromide induced the least effects on PVR in this study, compatible with our findings in the myograph study, which demonstrated its low vasoactive properties [14]. Furthermore, iopromide has a low viscosity and a low cytotoxic effect on the endothelium in comparison with ionic RCM [29, 30]. Hence, it is likely that iopromide would not induce an increase in the permeability of the endothelium to cause pulmonary oedema and an increase in the PVR [29, 30].

Normal pulmonary vessels, unlike systemic vessels, have low intrinsic vascular tone allied to a lack of vascular smooth muscle, which allows little active vasodilatation. However, dilatation can be demonstrated if vascular tone is raised [14]. In the experimental model of the rat IPL preparation, pulmonary vascular tone can be raised acutely by active vasoconstriction in response to alveolar hypoxia, while exposure of the rat in vivo to chronic hypoxia will lead to sustained increase in the vascular tone and Ppa. Under these conditions, a fall in Ppa indicative of dilatation can be demonstrated [24]. When basal tone was raised by acute hypoxic vasoconstriction, iotrolan and iopromide caused a further small rise in Ppa whilst diatrizoate and ioxaglate caused an overall fall in Ppa. Previous studies, including our myograph study, have shown that ionic RCM induce more vasodilatation compared with non-ionic agents [14]. This effect became prominent when the vascular tone was raised prior to exposure to RCM [14]. Our results are consistent with these observations, as the ionic diatrizoate and ioxaglate caused a fall in Ppa. It can be argued that ionic media should be used in preference to non-ionic RCM in patients with acute PH, for example in cases with acute pulmonary emboli, since they can induce immediate reduction of Ppa whereas the non-ionic media may cause a further rise in the Ppa. However, clinical experience in patients with PH secondary to thromboembolic disease has shown the absence of major haemodynamic effects with the use of non-ionic RCM in pulmonary angiography [7]. In addition, the general safety profile of the non-ionic agents and their lower systemic effects in comparison with ionic media make them suitable for use in such patients [30].

In isolated lungs from CH rats, where Ppa and resistance is high, there was an overall slow rise in Ppa with all the tested RCM (ioxaglate, iotrolan and iopromide). The rapid initial transient fall in Ppa observed with all tested solutions was most likely a volume effect. This was followed by a rapid rise in Ppa with iopromide and iotrolan, but with ioxaglate there was a sustained fall in Ppa that recovered over 20 min. At the end of the observation period a rise in the pressure was observed with ioxaglate, which was comparable with that of iotrolan but significantly greater than that with iopromide. Again, this could reflect the effects of iotrolan on blood rheology and of ioxaglate on the integrity of the endothelium. The rise in Ppa at the end of the 150 min test period with iopromide was not significantly different to the rise with saline in the control groups. The rise in Ppa observed with normal saline is most likely secondary to pulmonary oedema, indicating some instability of the preparation with time. However, this rise was significantly less in comparison with iotrolan and ioxaglate.

The mechanisms responsible for the dilator and constrictor action of RCM are not fully understood. Direct effect on the blood vessel wall causing cellular fluid/electrophysiological disturbance, leading to hyperpolarization and smooth muscle relaxation, is a possibility [14, 20]. In adddition, modulating the release of endogenous vasoactive substances such as ET, histamine, adenosine, serotonin, prostaglandins, nitric oxide, vasopressin and natriuretic peptides may also mediate the effects of RCM on vascular tone [14, 16–20, 31–35]. ET has been proposed as an important mediator of the vasoconstrictor effect of RCM, particularly in the kidney [32]. However, in this study ETA/B blockade by SB209670 did not prevent the rise in PVR induced by diatrizoate, in spite of using an effective pharmacological dose of the ET antagonist. This is compatible with our previous observations in the myograph, which indicated that ET is an unlikely mediator of RCM-induced pulmonary vasoconstriction [14].

In summary, RCM, especially the ionic dimer ioxaglate, can cause an increase in PVR and Ppa in the normal rat. When the baseline vascular tone of pulmonary arteries was raised acutely by hypoxia, a drop in Ppa reflecting a vasodilatory effect was observed with the ionic RCM ioxaglate and diatrizoate, but a small further rise in Ppa was observed with the non-ionic RCM iotrolan and iopromide. In the chronic model, in which PH is associated with remodelling of the pulmonary vasculature with increased vascular matrix proteins, vascular smooth muscle, and reduction in the vascular bed, a delayed but sustained rise in PVR and Ppa was observed that was pronounced with dimeric RCM (ioxaglate and iotrolan). An initial vasodilatory effect was evident only with ioxaglate, causing a reduction in Ppa that lasted for almost 20 min. Haemorheological effects of iotrolan and endothelial injury with ioxaglate precipitating pulmonary oedema are possible mechanisms for the observed increase in PVR associated with these agents. Iopromide induced the least changes in Ppa and PVR of normal and CH rats, which can be attributed to its low vasoactivity, cytotoxicity and viscosity.

	Acknowledgments

 
We wish to thank the Research Committee of the Northern General Hospital NHS Trust for financial support.

Received for publication March 19, 2001. Revision received June 28, 2001. Accepted for publication July 27, 2001.

	References

Top Abstract Introduction Materials and methods Results Discussion References

 

1. Almen T, Aspelin P, Levin B. Effect of ionic and non-ionic contrast medium on aortic and pulmonary arterial pressure. An angiocardiographic study in rabbits. Invest Radiol 1975;10:519–25.
2. Peck WW, Slutsky RA, Hackney DB, et al. Effects of contrast media on pulmonary hemodynamics: comparison of ionic and non-ionic agents. Radiology 1983;149:371–4.[Abstract/Free Full Text]
3. Sunnegardh O, Hietala SO, Wirell S, Ekelund L. Systemic, pulmonary and renal haemodynamic effects of intravenously infused iopental. A comparison in the pig of a new low osmolar non-ionic medium with saline and iohexol. Acta Radiol 1990;31:395–9.[Medline]
4. Cohan RH, Dunnick NR. Intravascular contrast media: adverse reactions. AJR 1987;149:665–70.[Free Full Text]
5. Dawson P. Cardiovascular effects of contrast agents. Am J Cardiol 1989;64:2E–9E.[Medline]
6. Sorensen L, Sunnegardh O, Svanegard J, et al. Systemic and pulmonary haemodynamic effects of intravenous infusion of non-ionic isoosmolar dimeric contrast media. An investigation in the pig of two ratio 6 contrast media. Acta Radiol 1994;35:383–90.[Medline]
7. Pitton MB, Duber C, Mayer E, Thelen M. Hemodynamic effects of nonionic contrast bolus injection and oxygen inhalation during pulmonary angiography in patients with chronic major-vessel thromboembolic pulmonary hypertension. Circulation 1996;94:2485–91.[Abstract/Free Full Text]
8. Tajima H, Kumazaki T, Tajima N, Murakami R, Gemma K. Effect of iohexol on pulmonary arterial pressure at pulmonary angiography in patients with pulmonary hypertension. Radiat Med 1994;12:197–9.[Medline]
9. Rees CR, Palmaz JC, Garcia O, Alvarado R, Siegle RL. The hemodynamic effects of the administration of ionic and nonionic contrast materials into the pulmonary arteries of a canine model of acute pulmonary hypertension. Invest Radiol 1988;23:184–9.[Medline]
10. Schrader R, Hellige G, Kaltenbach M, Kober G. The haemodynamic side-effects of ionic and non-ionic contrast media in the presence of pulmonary hypertension: experimental and clinical investigation. Eur Heart J 1987;8:1322–31.
11. Nicod P, Peterson K, Levine M, Dittrich H, Buchbinder M, Chappuis L, et al. Pulmonary angiography in severe chronic pulmonary hypertension. Ann Intern Med 1987;107:565–8.
12. Mills SR, Jackson BF, Older RA, Heaston DK, Moore AV. The incidence, etiologies and avoidance of complications of pulmonary angiography in a large series. Radiology 1980;136:295–9.[Abstract/Free Full Text]
13. Frisinger G, Schaffer J, Criley M, Gartner R, Ross J. Haemodynamic consequences of the injection of radiopaque material. Circulation 1965;31:730–40.[Abstract/Free Full Text]
14. Wang YX, Emery CJ, Laude E, Morcos SK. Effects of radiographic contrast media on the tension of isolated small pulmonary arteries. Br J Radiol 1997;70:1229–38.[Abstract]
15. Szolar DH, Saeed M, Flueckiger F, et al. Effects of iopromide on vasoactive peptides and allergy-mediated substances in healthy volunteers. Invest Radiol 1995;30:144–9.[Medline]
16. Szolar DH, Saeed M, Flueckiger F, et al. Response of vasoactive peptides to a non-ionic contrast media in patients undergoing pulmonary angiography. Invest Radiol 1995;30:511–6.[Medline]
17. Lasser EC, Walter A, Reuter SR, Lang I. Histamine release by contrast media. Radiology 1971;100:683–6.[Medline]
18. Assem ES, Bray K, Dawson P. The release of histamine from human basophils by radiological contrast agents. Br J Radiol 1983;56:647–52.[Abstract]
19. Ring J, Sovak N. Release of serotonin from human platelets in vitro by radiographic contrast media. Invest Radiol 1981;16:245–8.[Medline]
20. Morcos SK, Dawson P, Pearson JD, et al. The haemodynamic effects of iodinated water soluble radiographic contrast media: a review. Eur J Radiol 1998;29:31–46.[Medline]
21. Almen T, Aspelin P, Nilsson P. Aortic and pulmonary arterial pressure after injection of contrast media into the right atrium of the rabbit. Acta Radiol 1980 (Suppl. 362):37–41.
22. Dawson P, Harrison MJ, Weisblatt E. Effects of contrast media on red cell filterability and morphology. Br J Radiol 1983;56:707–10.[Abstract]
23. Liss P, Nygren A, Olsson U, Ulfendahl HR, Erikson U. Effects of contrast media and mannitol on renal medullary blood flow and red cell aggregation in the rat kidney. Kidney Int 1996;49:1268–75.[Medline]
24. Emery CJ, Bee D, Barer GR. Mechanical properties and reactivity of vessels in isolated perfused lungs of chronically hypoxic rats. Clin Sci 1981;61:569–80.[Medline]
25. Schrader R, Wolpers HG, Korb H, Hoeft A, Klepzig H, Kober G, et al. Central venous injection of large amounts of contrast media, advantages of a low osmolar contrast medium in experimentally induced pulmonary hypertension. Z Kardiol 1984;73:434–41.[Medline]
26. Spttzer S, Munster W, Sternitzky R, Bach R, Jung F. Influence of iodixanol-270 and iopentol-150 on the microcirculation in man: influence of viscosity on capillary perfusion. Clin Hemorheol Microcirc 1999;20:49–55.[Medline]
27. Lerche D, Hennicke G. The effects of different ionic and non ionic x-ray contrast media on the morphological and rheological properties of human red blood cells. Clin Hemorheol Microcirc 1992;12:341–55.
28. Beynon HLC, Walport MJ, Dawson P. Vascular endothelial injury by intravascular contrast agents. Invest Radiol 1994;29(Suppl. 2):S195–7.
29. Zhang H, Holt CM, Malik N, Shepherd L, Morcos SK. Effects of radiographic contrast media on proliferation and apoptosis of human vascular endothelial cells. Br J Radiol 2000;73:1034–41.[Abstract]
30. Thomsen HS, Morcos SK. Radiographic contrast media. BJU Int 2000;86(Suppl. 1):1–10.
31. Oldroyd SD, Morcos SK. Endothelin: what does the radiologist need to know? Br J Radiol 2000;73:1246–51.[Abstract]
32. Heyman S, Goldfarb M, Carmeli F, Shina A, Rahmilewitz D, Brezis M. Effect of radiocontrast agents on intrarenal nitric oxide (NO) and NO synthase activity. Exp Nephrol 1998;6:557–62.[Medline]
33. Morcos SK, Oldroyd S, Haylor J. Effect of radiocontrast media on endothelium derived nitric oxide dependent renal vasodilatation. Br J Radiol 1997;70:154–9.[Abstract]
34. Oldroyd SD, Fang L, Haylor JL, Yates MS, El Nahas AM, Morcos SK. Effects of adenosine receptor antagonists on the responses to contrast media in isolated rat kidney. Clin Sci 2000;98:303–11.[Medline]
35. Peachell PT, Morcos SK. Effect of radiographic contrast media on histamine release from human mast cells and basophils. Br J Radiol 1998;71:24–30.[Abstract]

This article has been cited by other articles:

				L. V. Hofmann, D. S. Lee, A. Gupta, A. Arepally, S. Sood, R. Girgis, and J. Eng Safety and Hemodynamic Effects of Pulmonary Angiography in Patients with Pulmonary Hypertension: 10-Year Single-Center Experience Am. J. Roentgenol., September 1, 2004; 183(3): 779 - 785. [Abstract] [Full Text] [PDF]

				S K Morcos Effects of radiographic contrast media on the lung Br. J. Radiol., May 1, 2003; 76(905): 290 - 295. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Emery, C J Articles by Morcos, S K Search for Related Content PubMed PubMed Citation Articles by Emery, C J Articles by Morcos, S K