An in vitro comparison of ultrasonic contrast agents in solutions with varying air levels

Abstract

The performance, in particular, the stability of ultrasound (US) contrast agents has yet to be assessed. An in vitro system has been set up to investigate the properties of ultrasonic contrast agents under different suspension conditions. This is designed to contribute to the optimal use of agents in clinical practice. In this study, the contrast agents were introduced into solutions of different oxygen concentration levels, as might be encountered in blood, and their relative performance was assessed in terms of decay in the solution environment. The partial pressures of oxygen in those solutions ranged between 1.5 and 26 kPa. Three IV and one arterial contrast agents were used: Levovist®, DMP115, Quantison™ and Myomap™. Levovist® showed the highest sensitivity to oxygen concentration in the solution, and the other three proved tolerant for the above values of oxygen concentrations.

Introduction

Ultrasound (US) contrast agents have been widely recognised not only to improve the outcome of investigations using various US imaging modalities, but also to create a new perspective in ultrasonic imaging. Quantitation of perfusion is one of the major goals of US contrast imaging, and its requirements have already been highlighted in the literature (Wiencek et al. 1993). One important requirement for US contrast agents is that they must not decay at the time of the examination or, at least, decay in a predictable manner. de Jong et al. (1993) demonstrated that the first commercial ultrasonic contrast agent Albunex® could not sustain pulmonary passage and the echo enhancement of the left ventricle was severely reduced. Wilson et al. (1993) produced similar results. Wiencek et al. (1993) observed, in their laboratory, that a degassed saline solution diminished the acoustic enhancement of Albunex® microbubbles.

The pressure within a free bubble at rest is greater than the pressure of the liquid surrounding the bubble because of the surface tension forces (Leighton 1994). Therefore, the internal pressure of a bubble p_i equals:

$$\text{p}{}_{\mathrm{i}}\text{=p}{}_{\mathrm{l}}\text{+p}{}_{\mathrm{\sigma }}\text{,}$$

where p_l is the pressure in the liquid and p_σ is the surface tension. The internal pressure also equals:

$$\text{p}{}_{\mathrm{i}}\text{=p}{}_{\mathrm{g}}\text{+p}{}_{\mathrm{\nu }}\text{,}$$

where p_g is the pressure of the gas in the bubble and p_ν is the pressure of the liquid vapour. Combining , , we get:

$$\text{p}{}_{\mathrm{g}}\text{=p}{}_{\mathrm{l}}\text{−p}{}_{\mathrm{\nu }}\text{+p}{}_{\mathrm{\sigma }}\text{,}$$

which means that the gas pressure in the bubble is greater than p_l − p_ν. Consequently, even in a saturated solution, the gas pressure inside the bubble is greater than that in the liquid and, therefore, the bubble will tend to dissolve according to Henry’s law. Epstein and Plesset (1950) calculated the dissolution times of air bubbles in water and also their change in size. Their calculations showed that, if surface tension is neglected, then a bubble would not dissolve in a saturated solution, taking the surface tension into account would result in a finite dissolution time for the bubble. According to their calculation, a 10-μm air bubble would dissolve in 1.17 s in a degassed water solution, and in 6.63 s in an air-saturated water solution. Using these calculations, de Jong et al. (1991) extended this theory to show that larger bubbles disappear slower than the smaller ones. Dissolution of a bubble means shrinkage and, therefore, decrease of scattering cross-section (resonance is not accounted for).

A simulation of exchanges of multiple gases in vivo showed the complexity of the problem. Burkard and Van Liew (1994) have shown that gases interact with each other and, after an injection of contrast, the bubble may take in several gases that are present in vivo. Calculations based on this model showed that exchanges of O₂, CO₂ and N₂ between the bubble and the blood might affect the size of the bubble (Van Liew and Burkard 1995a). Kabalnov et al. (1998b) used the same approach to calculate the dissolution times of “sparingly water-soluble gases” used in modern contrast agents (like fluorocarbons), taking into account that a gas may condense into a liquid. The results of the above studies showed an initial swelling of the fluorocarbon bubbles, followed by diffusion and probable condensation. Air bubbles, however, would diffuse from the moment of introduction to the bloodstream. Experimental work on rabbits showed little agreement with the above theoretical predictions (Kabalnov et al. 1998a). The persistence of the bubbles was more prolonged than predicted and did not relate to the gas consistency. Such a finding illustrated the importance of the role of the shell in the stabilisation process. It is definite, however, that nonsoluble gases make better contrast agents Kabalnov et al 1998a, Seidel et al 1998.

The case of stabilised microbubbles using surface films or surfactant monolayers to cancel out the surface tension, responsible for the diffusion of the bubbles, was examined in a model (Van Liew and Burkard 1995b). The permeability was, however, considered high enough to allow the exchange of gases. The complexity of shell modelling lies not only in the accurate prediction of pressures exerted, but also in the general mechanical and chemical properties of the shell material.

Increase in hydrostatic pressure in the liquid would also increase the gas pressure in the bubble according to eqn (3), and this would proportionally reduce its volume according to Boyle’s law. Application of pulsatile pressure has shown that backscatter was partially recoverable, which implies that the bubble size changes followed the pressure changes (Padial et al. 1995). Schneider et al. (1992) applied 150 mmHg pressure to air-filled albumin particles and observed a decay in the backscatter signal over time. de Jong et al. (1993) applied 160 mmHg hydrostatic pressure to Albunex® microbubbles and, apart from compression, they microscopically noticed that the bubbles deform and disappear. This is in agreement with Vuille et al. (1994), who noticed that the application of increasing hydrostatic pressure would proportionally reduce the reflectivity of the bubbles in an irreversible way (i.e., the release of pressure application would not recover the reflectivity of the bubbles). This result suggests that compression is not the sole mechanism for the decay of the agents. Vuille et al. (1994), in attempting to explain this irreversibility, suggested that diffusion of the bubbles accelerated by the application of pressure was more likely to be responsible, which also would be the outcome from eqn (3).

Decompression sickness can occur in deep-sea divers when the ascent is rapid. In that situation, exposure to lower pressure oversaturates nitrogen in the blood, resulting in bubble formation. The opposite happens in a rapid descent (West 1985). A gas-saturated solution can become undersaturated under the application of pressure; the gas concentration gradient is increased across the wall of the bubble and, according to Henry’s law, may cause accelerated diffusion, which would, therefore, decrease the dissolution time. Changes in hydrostatic pressure would, therefore, alter the gas concentration gradient across the bubble wall in two different ways, not only due to changes in the partial pressure of gases inside a bubble, but also to partial pressure changes of the dissolved gases in blood. Padial et al. (1995) applied pulsatile pressure to hand-agitated Angiovist® (Berlex Laboratories, Wayne, NJ), which showed a cyclic variation as well as an overall decrease in backscatter, but the same treatment to Albunex® solutions provided a nonrecoverable decay. This paper concluded that increases in hydrostatic pressure or in pulse rate accelerated the diffusion of all types of bubbles. Shi et al. (1999) have demonstrated an excellent negative correlation between subharmonic backscatter signal amplitude and hydrostatic pressure, displaying a different aspect in the influence of the hydrostatic pressure changes on contrast agents.

Experimental in vitro investigations have been carried out to assess the impact of the hydrostatic pressure on the stability of the agents, but little has been carried out experimentally on the role of the gas content of the suspending solution. The variation of gas content in the blood might affect significantly the stability of the contrast bubbles. In this paper, four agents are tested in terms of relative decay in solutions with different oxygen concentrations. Traditional methods of degassing, such as boiling, were not used and, instead, the introduction of helium gas to the dilutant was employed, which is the routine approach for degassing in chromatography and is undoubtedly reproducible Williams and Miller 1962, Sboros et al 2000. The results give information on the tolerance of the agents in the arterial and venous blood environments.