© 1997 by European Society of Cardiology
Copyright © 1997, European Society of Cardiology
Cardiac depression after experimental air embolism in pigs: role of addition of a surface-active agent1
aThoraxcenter, Erasmus University Rotterdam, Rotterdam, Netherlands
bThe Thoraxcenter, University Hospital Rotterdam-Dijkzigt, Rotterdam, Netherlands
cDepartment of Anesthesiology, Erasmus University Rotterdam, Rotterdam, Netherlands
dDepartment of Biostatistics, Erasmus University Rotterdam, Rotterdam, Netherlands
* Corresponding author. Laboratory for Haemodynamics, Thoraxcenter Ee 2322, Erasmus University Rotterdam, P.O. Box 1738, 3000 DR Rotterdam, Netherlands. Tel.: +31 (10) 4088044; fax: +31 (10) 4365191; e-mail: slager@tch.fgg.eur.nl
Received 13 August 1996; accepted 28 January 1997
 |
Abstract |
|---|
 Top Abstract 1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
Objective: Air bubbles entering the coronary artery may have harmful effects on cardiac function. From the physical point of view it is the relatively high surface tension of the blood–air interface which causes bubbles to trap in small vessels. The aim of the present study was to reduce depression of myocardial function from air embolism by lowering the surface tension of air bubbles. Methods: The effect of using antifoam as a surface-tension-reducing agent on air bubble entrapment and cardiac function was investigated in 6 anesthetized pigs (27±1 kg) and analyzed using a two-compartment diffusion model. Air bubbles with a diameter of 150 µm were selectively injected into the left anterior descending coronary artery (LADCA) in a carrying fluid in the presence or absence of antifoam. Myocardial systolic segment shortening in the LADCA region (SS-LADCA) was measured by sonomicrometry. Presence of emboli was detected by measuring the amount of reverberation of ultrasound scattered by trapped air bubbles. Results: SS-LADCA transiently decreased after injections of air bubbles in both the absence and presence of antifoam. However, in the presence of antifoam the regional depression recovered to normal sooner, the average depth of the depression was reduced, and bubbles from the embolized area cleared faster. These observations can be explained by a model derived from Laplace's law.
KEYWORDS Coronary air embolism; Surface tension; Antifoam; Surfactant; Pig, anesthetized
|
1 Introduction |
|---|
|
TopAbstract1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
Air embolism in the arterial and pulmonary circulation has been studied for decades because of its potential traumatic threat during various clinical interventions [5, 6, 24]. During catheterization [10, 13]and open-heart surgery [12, 18, 19, 27]air bubbles may inadvertently enter the blood circulation. In the heart catheterization laboratory this generally does not lead to serious complications, but during open-heart surgery it has been suggested that part of the neurological or neuropsychological injury after operation may originate from diffuse brain damage caused by air embolism [17].
In another respect, very small coated gas bubbles have been used as contrast material in ultrasound imaging [7, 23]. Currently the wish exists to use somewhat larger bubbles which are trapped in the microcirculation and persist sufficiently long to enable better visualization, of for example the myocardium, and yet do not lead to irreversible damage.
For obvious reasons more knowledge regarding the mechanisms involved in bubble trapping, bubble persistence, and tissue tolerance to temporary air embolism is highly desired. By applying Laplace's law [4]the physics of air bubble entrapment has been described by 3 parameters: (i) the blood pressure, (ii) the diameter of the vessels in which the air bubble is trapped, and (iii) the surface tension of the air–blood interface. Furthermore, it has been observed that the persistence of embolization depends mainly on the speed of diffusion of air from the emboli into the surrounding blood and tissue [6, 22, 25, 26].
Recently the effects of air bubbles selectively injected into a pig coronary artery on regional myocardial function have been evaluated [25, 26]. In the first study it was shown that smaller bubbles disappeared faster and caused less severe myocardial depression [25]. In the second study diffusion parameters like blood pressure and tissue nitrogen concentration were varied and it was shown that under elevated blood pressure and diminished tissue nitrogen content the deleterious effect of bubbles on myocardial function was greatly reduced [26]. Reduction of the effect of air embolism on myocardial function may not only be accomplished by changes in the parameters affecting diffusion but also by changes in the surface tension of the air–blood interface.
Surface tension can be lowered by surfactants [3, 8, 11, 20]. In dogs [8], the mixing of surface-active agents with a normally lethal dose of air injected into the left ventricle reduced the initial mortality; antifoam proved to be the most effective [8]. In that study the mechanism of this beneficial effect of surface-active agents was not well understood as not only the surface tension was reduced but also the size of the air bubbles formed in the left ventricle might have been different due to addition of surface-active agents which may affect the distribution of the bubbles in the systemic circulation [20].
The aim of the present study was to evaluate the contribution of lowering surface tension on reduction of the effect of air embolism on myocardial function in pigs. In order to do so, the possible effect of bubble size and distribution in the arterial circulation was reduced by standardizing these parameters by selective injection of uniformly sized air bubbles into a coronary artery [25, 26]. With these precautions we studied the effect of embolization by small air bubbles in the presence or absence of antifoam by measuring simultaneously regional heart function and bubble persistence. Experimental findings were analyzed using a two-compartment diffusion model [25].
|
2 Methods |
|---|
|
TopAbstract1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
2.1 General animal preparation
Experiments were performed in accordance with the guiding principles in the care and use of animals as approved by the Council of the American Physiological Society (DHEW publication No. (NIH) 80-23, 1980) and under the regulations of the Animal Care Committee of the Erasmus University Rotterdam.
After an overnight fast, cross-bred LandracexYorkshire pigs (HVC, Hedel, Netherlands) of either sex (27–29 kg, n=6) were sedated with 700 mg ketamine i.m. (AUV, Cuijk, Netherlands) and anesthetized with intravenous 480 mg sodium pentobarbitone (Apharmo, Arnhem, Netherlands), intubated and connected to a ventilator for intermittent positive pressure ventilation with a mixture of N2 and O2 (3:2 vol/vol). Respiratory rate and tidal volume were adjusted to maintain arterial blood gases within the ranges of: 7.35<pH<7.45; 35 mmHg<pCO2<45 mmHg and 100 mmHg<pO2<150 mmHg. A catheter was positioned in the descending aorta for withdrawal of blood samples and measurement of central aortic blood pressure (AoP). A 7-French (Fr) Sensodyn micrometer-tipped catheter (B. Braun Medical B.V., Uden, Netherlands) inserted via the right carotid artery, was used to measure left ventricular pressure (LVP) and its first derivative (LVdP/dt). A multilumen 7-Fr catheter was placed in the superior caval vein for: (i) injection of 4 mg of muscle relaxant (pancuronium bromide, Organon Teknika B.V., Boxtel, Netherlands) prior to midline sternotomy; (ii) continuous infusion of 5.5–22.2x10–4 mg/s per kg body weight sodium pentobarbitone (Sanofi, Paris, France); (iii) temporary infusion of saline or the gelatin-derived blood plasma substitute, Haemaccel (Behringwerke A.G., Marburg, Germany), to compensate for blood loss. After midline sternotomy, an electromagnetic flow probe (Skalar, Delft, Netherlands) was placed around the ascending aorta. Regional myocardial segment length was measured by sonomicrometry (Triton Technology Inc., San Diego, CA, USA) using two pairs of ultrasonic crystals (Sonotek Corporation, Del Mar, CA, USA). One pair was placed close to the apex in the mesocardial layer of the distribution of the left anterior descending coronary artery (LADCA) which was the area to be embolized by air bubbles. Another pair was placed in the mesocardial layer of the distribution of the left circumflex coronary artery (LCXCA) for control of stability of the preparation. Under fluoroscopic guidance a 3-Fr catheter to be used for the injection of air bubbles was inserted over a guide wire through a guiding catheter in the left carotid artery and positioned in the LADCA with its tip just distal to the first diagonal branch. After removal of the guide wire an intracoronary injection of contrast material, iopamiro 370 (ASTA Medica B.V., Diemen, Netherlands), was used to visualize the perfused region in which the air bubbles were to be injected and to determine whether the LADCA crystals were inside the distribution area of the LADCA. To prevent clotting, heparin (Pharmacy Department of the University Hospital Rotterdam-Dijkzigt, Rotterdam, Netherlands) was given in a bolus of 10 000 IU prior to the coronary catheterization, followed by administration of 5000 IU every 2 h. In order to eliminate the effect of heart rate, all hearts were paced at 100 beats/min using a right atrial lead, after the normal rhythm was slowed down to well below 100 beats/min by administration of the selective negative chronotropic compound, zatebradine [28](a gift from Dr. J.W. Dämmgen, Dr. Karl Thomae GmbH, Biberach a/d Riss, Germany).
2.2 Preparation and administration of air bubbles and antifoam
Air bubbles were produced as described previously [25]. Using a bubble generator, air from an adjustable constant-pressure source was injected through a micropipette into a Haemaccel flow between 46.7 and 53.3 µl/s, which is approximately 1/10 of the blood flow in the LADCA [16]. Based on the findings in previous studies [25, 26], air pressure and Haemaccel flow were adjusted to produce bubbles of 150 µm diameter (light microscopy revealed that variation was less than 10 µm). The time required to inject air in a dose of 2 µl/kg body weight as selected in the present study varied from 18 to 23 s.
A 1:250 dilution of silicone antifoam 1510 EU (Dow Corning, Seneffe, Belgium) was prepared in pure water (Milli Q Plus, Millipore, Molsheim, France), and filtered through 10 µm pores. Diluted antifoam in a volume of 1.75 ml was added at an injection rate of 58.3 µl/s to the flow of Haemaccel and air bubbles just distal to the bubble chamber (Fig. 1). Inspection of the bubbles leaving the delivery catheter by light microscopy revealed that bubble size was not altered by antifoam administration. In the absence of antifoam, Haemaccel was added in the same way as antifoam to keep flow conditions constant.
|
2.3 Ultrasound scatter intensity measurement
Ultrasound is reflected by micro-bubbles as small as 2 µm in diameter [7, 23]. In the presence of air bubbles some of the ultrasound energy emitted by the activated sonomicrometry crystal will be scattered by the air bubbles. After detection of the front of the initial pulse, reverberating ultrasound will be received by the receiving crystal. At an ultrasound frequency of 3 MHz, the received ultrasound pulse normally lasts 2–4 µs. The received amount of scattered ultrasound was measured by integration of the rectified signal over a 10 µs period, starting 5 µs after the arrival of the front of the initial pulse. The maximal incremental distance travelled by the integrated ultrasound during this measuring period may add up to
32.5 mm compared to 10 mm for the shortest direct path between the ultrasound crystals.
2.4 Experimental protocols
After completion of the surgical procedures a 30 min stabilization period was allowed. To maintain mean arterial blood pressure (MAP) of approximately 105 mmHg, phenylephrine was infused in a dose of 1.1–6.7x10–4 mg/s (Pharmacy Department of the University Hospital Rotterdam-Dijkzigt, Rotterdam, Netherlands). When MAP was stable, control recordings of AoP, LVP, LVdP/dt, aortic blood flow (AoBF), segment length, and I(echo) were made at 100 samples/s (ACODAS, DATAQ Instruments Inc., Akron, OH, USA).
A sequence of experimental runs was applied according to the scheme in Fig. 2 with a minimal time interval of 15 min between consecutive runs under the additional condition that both regional myocardial contraction and ultrasound reverberation had returned to baseline values. All injection fluids were at room temperature and were selectively injected into the LADCA.
|
During the first part of the sequence, an injection of Haemaccel was first given to eliminate the possibility of unexpected reactions in the individual animal. The two following air bubble injections were given as control and compared for reproducibility.
During the second part of the sequence, Haemaccel and antifoam were first injected to study their effect on the measured variables. A following injection of air bubbles was given; the effect was compared with the injection of air bubbles preceding the antifoam run to exclude the possibility that the antifoam administered into the blood circulation had long-term effects. The next run of air bubbles was carried out with the addition of antifoam and compared with the previous run of air bubbles in the absence of antifoam.
During the third part of the sequence, another injection of air bubbles was given as control for repeatability of the protocol.
For each run recording of data began at 30 s (baseline) before the start of injection (t=0 s), and continued up to 600 s after the start of injection.
2.5 Surface tension measurement
The maximum bubble pressure method [1]was used to measure surface tension in vitro. Air bubbles were produced at the exit of a special glass tube placed in the upright position in a liquid, and the required air pressure to achieve a selected frequency of bubble production was measured.
With this method the surface tension is measured at repeatedly new formed bubbles. Each period between the release of successive bubbles is available for surfactants to migrate towards the surface of the next developing bubble. The surface tension was measured at 37°C in Haemaccel, antifoam dilution, arterial blood from pigs, Haemaccel + diluted antifoam in a ratio of 1:1.8, blood + Haemaccel 10:1, and blood + Haemaccel + diluted antifoam 10:1:1.8.
The maximum pressure was determined at a bubble production rate with a period of 10 s because air bubbles in the animal experiment have, after being generated, about 10 s to dwell in the fluid in the delivery catheter before entering the coronary circulation.
We also performed an additional measurement between successive bubbles at intervals of 60 s as in the trapped situation in vivo surfactant may migrate from the fluid to the bubble interface, thus allowing a further reduction of surface tension. The interval of 60 s was chosen because in our experience [25]recovery of function after air embolism starts within 1–2 min giving an indication of the available time for surfactant migration to trapped bubbles.
2.6 Data analysis
Fig. 3 shows a representative example of LVP, integrated reverberating ultrasound (IRU) and segment length in the LADCA region after injection of air bubbles and air bubbles in the presence of antifoam.
|
indicates end-diastole and end-systole in the SL-LADCA signal.The normalized variable I(echo) was calculated from the recorded signal IRU. For each experimental run the baseline value was defined as 0 and the maximum value IRU(max) was set to 1 by expressing the variable I(echo) as:
|
|
From the recordings, segment length was assessed at end-diastole (EDL) and end-systole (ESL). End-diastole was defined as the positive onset of LVdP/dt, and end-systole as the zero crossing of AoBF. From these data the systolic segment shortening (SS) was calculated as:
|
|
The results of the different runs were compared by means of summary statistics [2]. The first summary variable was the area under the curve defined in a previous study [26]as the time integral of the deviation from baseline (TIDB). TIDB was calculated for SS in the LADCA (SS-LADCA) region and for I(echo).
Further summary variables were two time constants characterizing the rate of change of SS-LADCA as a function of time. The values of SS-LADCA were fitted to a two-compartment model as described in a previous study [25]and characteristic time constants 
1 and 2 were obtained for each run, 1 can be related to the period of embolization and the induced functional depression and 2 to a delay in response of the changing function to the state of perfusion.
2.7 Statistics
Results have been presented as arithmetical mean±standard deviation. For each run the changes in variables were analyzed by repeated measures analysis of variance. Subsequently the deviations in time from baseline were tested and corrected for multiple comparisons by Fisher's LSD method. The summary variables 1, 2 and TIDB's were used to compare pairs of runs [2]. To do so, the two values observed for one animal were divided by the value for the first of the two runs. By doing so, the relative difference between two runs was compared instead of the absolute difference. The relative difference between the variables was tested for significance by using Student's paired t-test, corrected for multiple comparisons by Bonferroni's method. The level of statistical significance was set at 0.05.
|
3 Results |
|---|
|
TopAbstract1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
3.1 Stability of the animals
The baseline values did not change statistically significantly during the 7 runs for any of the variables (Table 1), indicating the stability of the animals throughout the experiment.
|
During the 7 runs, there were no statistically significant changes in global hemodynamic variables MAP and peak LVdP/dt's or in the regional variable SS in the distribution of the LCXCA.
3.2 Experimental runs
3.2.1 Haemaccel
Haemaccel injection did not cause statistically significant changes in the variables SS-LADCA and I(echo) (Fig. 4, run 1).
|
3.2.2 First and second injection of air bubbles
Generally, I(echo) increased steadily during injection of air bubbles and reached its peak level just before the end of injection. The signal was still elevated at 6 min (Fig. 4, runs 2 and 3) and reached baseline values within 10 min. In one animal I(echo) was not measured for technical reasons.
After injection of air bubbles myocardial function as expressed by SS-LADCA was significantly decreased between t=30 s and t=180 s for run 2 and between t=30 s and t=240 s for run 3 (Fig. 4). There was no significant difference between the summary variables 1, 2, TIDB/S of SS-LADCA, and TIDB/I of I(echo) (Table 2) for the first two injections of air bubbles which indicates the reproducibility of the intervention.
|
3.2.3 Antifoam
Injection of antifoam caused a short significant elevation in SS-LADCA and I(echo) at t=30 s (Fig. 4, run 4).
3.2.4 Air bubbles after antifoam
The responses of regional variables (Fig. 4, run 5) to injection of air bubbles after the antifoam run were not different from previous bubble runs. SS-LADCA statistically significant decreased between t=30 s and t=180 s while I(echo) significantly increased between t=30 s and t=600 s. Time constants 1, 2, and both TIDB's (Table 2) were not significantly different from run 3, indicating that the injection of antifoam in the previous run has no long-lasting effect on the measured variables.
3.2.5 Air bubbles in the presence of antifoam
During the run of air bubbles with antifoam SS-LADCA was significantly decreased between t=30 s and t=60 s while I(echo) significantly increased between t=30 s and t=180 s (Fig. 4, run 6). There was a significant reduction of the time constant 1 by 78% (Table 2) compared to run 5. Furthermore, compared to run 5, TIDB of SS-LADCA and I(echo) were significantly reduced by 76 and 40%, respectively (Table 2). There was no significant difference between the values of 2 (Table 2).
3.2.6 Final injection of air bubbles
Final injection of air bubbles caused a significant reduction of SS-LADCA (Fig. 4, run 7) between t=60 s and t=360 s while I(echo) was significantly increased between t=30 s and t=600 s. The time constants 1, 2, and both TIDB's (Table 2) were not different from run 5.
3.3 Surface tension
The results of the surface tension measurements are listed in Table 3. The surface tension of air bubbles in Haemaccel with antifoam at 10 and 60 s interval was respectively 11 and 16% lower than in pure Haemaccel. The surface tension of bubbles produced in blood mixed with Haemaccel and antifoam was not different from that of bubbles produced in blood.
|
|
4 Discussion |
|---|
|
TopAbstract1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
In the present study it was demonstrated that lowering of surface tension causes a reduction of the depressing effect of air embolism on myocardial function in pigs. Myocardial function as expressed by SS-LADCA transiently decreased after injections of air bubbles in both the absence and presence of antifoam. The depression of myocardial function in the distribution of the LADCA was similar to that reported in our previous studies [25, 26]and cannot be attributed to the Haemaccel injection. In the presence of antifoam myocardial depression was significantly reduced as indicated by the summary variables TIDB/S and TIDB/I. Restoration of SS-LADCA as indicated by the variable
1 was much faster. This effect of antifoam cannot be explained from parameters like bubble size, blood pressure and ventilation gases which are known to influence the course of coronary air embolism [25, 26], as these were kept constant during the present study. The condition of the animals remained stable throughout the experiment as indicated by the unchanged baseline values. The myocardial depression as expressed by SS-LADCA and increase of I(echo) after injections of air bubbles showed good repeatability throughout the experiment. Therefore the observed reduction of myocardial functional depression and faster decrease of I(echo) after injection of air bubbles in the presence of antifoam was induced by the administration of antifoam.
4.1 Measurement of air bubble presence by ultrasound reverberation
The injection of air bubbles caused a clear reaction of I(echo), indicating the presence of air emboli. The ultrasound reverberation was not affected in a systematic way by injection of Haemaccel. Antifoam caused a short-lasting rise of I(echo). However, the maximum amplitude of the IRU was less than 10% of the value observed during injection of air bubbles. This increase may have been caused by scattering properties of the antifoam.
During bubble runs the peak value of I(echo) was reached before the end of injection, indicating some degree of saturation of echo reverberation intensity. From backscattering experiments with ultrasound [23]it is known that with rising concentration of air bubbles the level of the received signal initially increases up to a maximum and subsequently decreases because of increasing attenuation. This probably implies that in the present study the level of the I(echo) signal was not linearly related to the amount of air bubbles and therefore low concentrations of air bubbles would be detected with a relatively high sensitivity [23].
It was observed that I(echo) was elevated over a longer period than SS-LADCA. This indicates that I(echo) is a more sensitive measure for the presence of air bubbles than measuring regional heart function. Apparently SS-LADCA recovers to baseline value by sufficient restoration of blood flow while only the majority of air bubbles have been dissolved. In this respect it has been observed [29]that coronary blood flow can be reduced by approximately 50% before regional cardiac function starts to decrease.
Compared to our previous studies [25, 26]I(echo) was found to be a highly useful addition to our measurement set-up, giving an indication of the regional presence of air bubbles. To fully exploit this feature, further technical improvements have to be made to study bubble presence in a more quantitative way.
4.2 Possible underlying mechanisms
According to our present knowledge a reduction of bubble surface tension is the most likely reason to explain the beneficial effect on the course of air embolism, as this is the only parameter that is varied during the experiments. Antifoam addition did not prevent the air bubbles from being trapped as indicated by the observed transient decrease in SS-LADCA and the temporary increase in I(echo) during that run. This finding can be explained more quantitatively by modelling of entrapment of air bubbles in the microcirculation. The equilibrium state (Fig. 5) of a so-called slug [8]can be calculated using Laplace's law [4]. From this it follows that for a pressure difference of, for example, 100 mmHg between the proximal (Pp) and distal (Pd) side of the slug and a surface tension of 0.043 N/m, the diameter of the back interface (db) needs to be smaller than 11 µm to allow direct passage of air bubbles through a capillary where the frontal diameter (df) of the slug becomes 6 µm (Fig. 5). However, this situation is not likely to be reached as it is derived from morphometric data of pig coronary arterial trees [14]where the average length of a 10 µm diameter vessel is 0.06 mm while the volume of a 150-µm-diameter air bubble is equivalent to a 10-µm-diameter slug of 22 mm length. Furthermore, in the present experiments MAP and left ventricular pressure did not exceed 106 and 128 mmHg, respectively. For these reasons it is unlikely that bubbles will directly pass the capillaries at a lowered surface tension of 0.043 N/m, and it is expected that trapping will still occur at a level before the capillaries.
|
(N/m).From the relation for the equilibrium state given in Fig. 5 it can be derived that lowering surface tension influences the condition of a trapped air bubble in two ways. First, at a constant pressure difference between Pp and Pd, lowering surface tension allows df and db to be smaller to reach equilibrium. Therefore bubbles trap at a level more close to the capillaries [25]. As a result the area-to-volume ratio of such relatively more elongated bubbles becomes larger, which promotes diffusion and reduces bubble persistence time. Secondly, at lower surface tension, once the front of a slug has reached the capillaries the condition to pass occurs for a larger backward radius, which implies that a larger air volume can pass the capillaries.
Lowering surface tension may also promote the division of air bubbles into smaller ones at vascular branches [25]. Division of air bubbles into smaller ones increases the total area to volume ratio and thus also promotes diffusion.
Therefore all these factors induced by a lower surface tension decrease the period of embolization and thus increase the return to baseline of I(echo) and SS-LADCA in accordance with the present findings.
4.3 Surface tension
Blood surface tension is the primary physical reason why air bubbles withstand the driving blood pressure and become entrapped in small blood vessels [3, 4, 8, 11]. It has been reported that silicone antifoam is an effective surface-tension-reducing agent [8], as well as lung surfactant [3, 11]. In selecting an agent for the present study we performed pilot experiments in pigs by applying the lung surfactant, Alvofact (Boehringer Ingelheim, Dr. Karl Thomae GmbH, Biberach a/d Riss, Germany). Reduction of the effect of air embolism was not consistently demonstrated and it was found that at that time the composition of Alvofact may not have been stable. In a few cases a severe vasoconstriction in the lungs was observed followed by a large drop in left ventricular pressure. For these reasons we selected antifoam as the agent to reduce surface tension in the present study.
The 1:250 antifoam dilution was chosen to achieve a silicone concentration of approximately 50 volume parts per million (ppm) in the coronary blood flow. This is at least 10 times lower than the concentration published previously [8]and somewhat higher than the 20 ppm dose recommended by the manufacturer for other applications of antifoam.
The measurements of surface tension with the maximum bubble pressure method show that the surface tension of Haemaccel mixed with antifoam was clearly reduced, indicating that in animal experiments at the time of bubble entrance into the blood stream the surface tension was lower than that of air bubbles without antifoam. However, possible subsequent changes in surface tension of air bubbles in blood until trapping and during the period of embolization could not be mimicked in the present in vitro measuring set-up.
Despite the observation that the surface tension of bubbles produced in blood mixed with Haemaccel and antifoam was not reduced we do not expect the lowered surface tension of air bubbles delivered in the presence of antifoam to increase between the time when bubbles enter the blood stream and their entrapment a few seconds later because it is likely that surfactant once present at the interface is not inhibited by blood components. It has been observed [15]that the inhibition of surface-tension-lowering activity of surfactants in blood [9, 15, 21]is temporal when measuring surface tension at a single continuously present air bubble [15]with the so-called pulsating bubble method. Further technical improvements are needed in measuring surface tension under such rapidly changing conditions like those in our experiment to further disclose this issue.
For the above reasons we suppose that air bubbles injected in the presence of antifoam have a reduced surface tension and maintain their lowered surface tension after they have entered the blood stream.
4.4 Implications
The results of the present study show that reduction of surface tension of air bubbles in blood can reduce the period of embolization by the air bubbles. In the settings of the present study it was not demonstrated that administration of antifoam into blood before injection of air bubbles is also effective. However, preventive administration of a surfactant may be effective as has been reported in a study on venous air embolism [20]. The option of administering a surfactant after bubbles have been trapped in the circulation is not likely to be very effective. A long period will be required for the surfactant to reach the surface of the emboli through the stagnated blood.
In the field of open-heart surgery it can be considered to use a biocompatible surfactant for reducing the surface tension of remaining air bubbles in the left ventricle [18, 19]to reduce neurological embolization.
Further studies will be needed for selection of suitable surfactants and techniques how to use these surfactants during clinical procedures.
4.5 Conclusion
Silicon antifoam added as surface-tension-lowering agent to the carrying fluid of air bubbles injected into a coronary artery of pigs at normal physiological pressure does not prevent but substantially reduces the period of embolization. Depth and duration of the induced myocardial depression are also reduced. Restoration of myocardial function at rest seems to occur faster than clearance of embolization. These observations can be explained by a model derived from Laplace's law.
Time for primary review 21 days.
|
Acknowledgements |
|---|
The authors thank R.H. van Bremen for assistance during the preparation of the animals. This work was supported by the Netherlands Heart Foundation under grant 91.100.
|
Notes |
|---|
1 Work performed at the Thoraxcenter, Erasmus University, Rotterdam.
2 Present address: Grabowsky Polytechnics BV, Mauritskade 33, 2514 HD Den Haag, Netherlands. Tel.: +31 (70) 3560976; fax: +31 (70) 3607959; e-mail: van.blankenstein@grabowsky.nl
|
References |
|---|
|
TopAbstract1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
- Adamson WA. Physical chemistry of surfaces, 2nd ed. New York: Interscience Publishers, 1967:18–21,141–144.
- Altman DG. Practical statistics for medical research. Padstow: TJ Press, 1991:331,426–433.
- Butler B.D., Hills B.A. Role of lung surfactant in cerebral decompression sickness. Aviat Space Environ (1983) 54:11–15.
- Chang H.K., Weber M.E., Thomson J., Martin R.R. Hydrodynamic features of pulmonary air embolism: a model study. J Appl Physiol (1981) 51:1002–1008.
[Abstract/Free Full Text] - Chang K.S., Yang W.-J. Survey of literature related to the problems of gas embolism in human body. J Biomech (1969) 2:299–312.[CrossRef][ISI][Medline]
- Curtillet E. L'embolie gazeuse artérielle. J Chir (1939) 53:461–482.
- De Jong N., ten Cate F.J., Lancee C.T., Roelandt J.R.T.C., Bom N. Principles and recent developments in ultrasound contrast agents. Ultrasonics (1991) 29:324–330.[CrossRef][ISI][Medline]
- Eiseman B., Baxter B.J., Prachuabmoh K. Surface tension reducing substances in the management of coronary air embolism. Ann Surg (1959) 149:374–380.[ISI][Medline]
- Fuchimukai T., Fujiwara T., Takahashi A., Enhorning G. Artificial pulmonary surfactant inhibited by proteins. J Appl Physiol (1987) 62:429–437.
[Abstract/Free Full Text] - Heupler F.A., Ferrario C.M., Averill D.B., Bott-Silverman C. Initial coronary air embolism in the differential diagnosis of coronary artery spasm. Am J Cardiol (1985) 55:657–661.[CrossRef][ISI][Medline]
- Hills B.A., Barrow R.E. Air embolism: possible role of surfactant on decompression. J Appl Physiol (1982) 52:119–123.
[Abstract/Free Full Text] - Justice C., Leach J., Sterling Edwards W. The harmful effects and treatment of coronary air embolism during open-heart surgery. Ann Thorac Surg (1972) 14:47–52.[Medline]
- Kahn J.K., Hartzler G.O. The spectrum of symptomatic coronary air embolism during balloon angioplasty: causes, consequences, and management. Am Heart J (1990) 199:1374–1377.
- Kassab G.S., Rider C.A., Tang N.J., Fung Y.C. Morphometry of pig coronary arterial trees. Am J Physiol (1993) 265:H350–H365.[ISI][Medline]
- Keough K.M.W., Parson C.S., Tweeddale M.G. Interaction between plasma proteins and pulmonary surfactant: pulsating bubble studies. Can J Physiol Pharmacol (1989) 67:663–668.[ISI][Medline]
- McFalls E.O., Duncker D.J., Krams R., Ward H., Gornick C., Verdouw P.D. Endothelial dependent vasodilatation following brief ischaemia and reperfusion in anesthetized swine. Cardiovasc Res (1991) 25:659–665.[ISI][Medline]
- Moody D.M., Bell M.A., Challa V.R., Johnston W.E., Prough D.S. Brain microemboli during cardiac surgery of aortography. Ann Neurol (1990) 28:477–486.[CrossRef][ISI][Medline]
- Okamura Y., Takeuchi Y., Gomi A., Nagashima M., Mori H., Hattori J. Clinical evaluation of preoperative myocardial infarction as a complication of valve replacement. Kyobu Geka (1989) 42:1012–1015.[Medline]
- Orihashi K., Matsuura Y., Hamanaka Y., et al. Retained intracardiac air in open heart operations examined by transesophageal echocardiography. Ann Thorac Surg (1993) 55:1467–1471.[Abstract]
- Perry J.C., Munson E.S., Malagodi M.H., Shah D.O. Venous air embolism prophylaxis with a surface-active agent. Anesth Anal (1975) 54:792–799.
[Abstract/Free Full Text] - Seeger W., Grube C., Günther A., Schmidt R. Surfactant inhibition by plasma proteins: differential sensitivity of various surfactants. Eur Respir J (1993) 6:971–977.[Abstract]
- Stegmann Th., Daniel W., Bellmann L., Trenkler G., Oelert H., Borst H.G. Experimental coronary embolism. Assessment of time course of myocardial ischemia and the protective effect of cardiopulmonary bypass. Thorac Cardiovasc Surg (1980) 28:141–149.[ISI][Medline]
- Sverre H., Myhrum M., Hoff L. Modelling of the ultrasound return from Albunex microspheres. Ultrasonics (1994) 32:123–130.[CrossRef][ISI]
- Van Allen C.M., Hrdina L.S., Clark J. Air embolism from the pulmonary vein: a clinical and experimental study. Arch Surg (1929) 19:567–599.[ISI]
- Van Blankenstein J.H., Slager C.J., Schuurbiers J.C.H., Strikwerda S., Verdouw P.D. Heart function after injection of small air bubbles in a coronary artery of pigs. J Appl Physiol (1993) 75:1201–1207.
[Abstract/Free Full Text] - Van Blankenstein J.H., Slager C.J., Soei L.K., Boersma H., Verdouw P.D. Effect of arterial blood pressure and composition of ventilation gasses on cardiac depression induced by air embolization of the left coronary artery. J Appl Physiol (1994) 77:1896–1902.
[Abstract/Free Full Text] - Van der Linden J., Casimir-Ahn H. When do cerebral emboli appear during open heart operations? A transcranial Doppler study. Ann Thorac Surg (1991) 51:237–241.[Abstract]
- Van Woerkens L.J., van der Giessen W.J., Verdouw P.D. The selective bradycardic effects of UL-FS 49 do not adversely effect left ventricular function in conscious pigs with chronic coronary artery occlusion. Cardiovasc Drug Ther (1991) 6:59–65.[CrossRef][ISI]
- Waytt H.L., Forreste J.S., Goldner J.V., Logan S.E., Parmley W.W., Swan H.J.C. Effect of graded reductions in regional coronary perfusion on regional and total cardiac function. Am J Cardiol (1975) 36:185–192.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||