Cardiovascular Effects of Selected Water-miscible Solvents for Pharmaceutical Injections and Embolic Liquids: A Comparative Hemodynamic Study Using a Sheep Model Alexandre Laurent
1,2
*, Florence Mottu 3, René Chapot 1, Olivier Jordan 3,
Daniel A. Rüfenacht 4, Eric Doelker 3, and Jean-Jacques Merland 1,2 1
Laboratory of Neuroradiology and Therapeutic Angiography, Lariboisière Hospital, University of Paris 7, France
2 3
Center for Research in Interventional Radiology, APHP/INRA, Jouy-en-Josas, France School of Pharmaceutical Sciences, Ecole de pharmacie Genève-Lausanne, University of Geneva, Switzerland
4
Division of Diagnostic and Interventional Radiology, University of Geneva, Switzerland
Submitted to: PDA Journal of Pharmaceutical Science and Technology * Corresponding author:
Dr Alexandre Laurent Lariboisière Hospital 2 rue Ambroise Paré, 75010 Paris, France Phone: +33 1 49 95 83 52 Fax: +33 1 49 95 83 56 E-mail:
[email protected]
1
_______________________________________________________________ ABSTRACT:
Generally,
organic
water-miscible
solvents
are
used
intravascularly (both intravenous and intra-arterial) for preparing two types of formulations, namely pharmaceutical injections of poorly soluble drugs and precipitating liquid embolics containing polymeric biomaterials for the minimally invasive treatment of aneurysms, arteriovenous malformations or tumors. Although several of such solvents have been used in both drug delivery and interventional radiology their harmlessness is a concern. In particular there is a lack of comparative investigations of their cardiovascular effects.
Thirteen non-aqueous water-miscible solvents were selected because of their capacity to solubilize drugs or embolic materials, and their described use, at least diluted with water, in pharmaceutical formulations.
Their in vivo hemodynamic toxicity in male adult sheep after infra-renal aorta catheterization has been estimated with respect to the arterial and venous pressures, as well as the heart rate. Saline solution was used as a control. Three different volumes (0.1, 0.5 and 1.0 mL) were infused rapidly to achieve better discrimination between solvents. Increase in arterial pressure and concomitant decrease in venous pressure were observed at differing extent for all organic solvents. Changes in heart rate were negligible.
Based on the intensity of arterial pressure change after 1-mL infusion (the most relevant clinical criteria), a classification of the solvents is proposed: -
solvents devoid of significant cardiovascular toxicity: dimethyl isosorbide, GlycofurolTM 75, polyethylene glycol 200 (PEG 200), diglyme
-
Solvents with moderate cardiovascular toxicity: tetrahydrofurfuryl alcohol, ethanol, acetone, Solketal, glycerol formal, dimethyl sulfoxide (DMSO)
-
Solvents with marked cardiovascular toxicity: propylene glycol, ethyl lactate, (N-methyl-2-pyrrolidone (NMP).
Emphasis is put on the relative character of the proposed ranking and on the lack for certain solvents, at least in the open literature, of data pertaining at 2
other forms of toxic effects (e.g. undesirable pharmacological action, cancerogenicity,
teratogenicity,
mutagenicity,
irritating
and
sensitizing
properties), all factors that have to be considered when selecting a proper solvent.
KEY WORDS: organic solvents, intravascular injection, cardiovascular effects, hemodynamic toxicity, parenteral route, liquid embolics, interventional radiology, sheep model _______________________________________________________________
3
Introduction
When limited solubility of the drug or chemical instability prevents the use of water to prepare injectable solutions, nonaqueous organic solvents can help in developing stable parenteral products [1-10]. Their need has even been reinforced in recent years as the drug discovery process has essentially yielded poorly water-soluble chemicals. The use of water-miscible solvents is in particular one possible way for the development of reference intravenous solutions necessary for determining the bioavailability of drugs administered orally.
Another application of water-miscible solvents is their use as solvent for embolic biomaterials
for
the
endovascular
treatment
of
aneurysms
[11,
12],
arteriovenous malformations [13] or tumors [14] in interventional radiology. The principle of this new generation of embolic liquids relies on the in situ precipitation upon contact with blood of the preformed polymer dissolved in the water-miscible organic solvent (e.g. dimethyl sulfoxide). This approach contrast with that of the in situ polymerization of acrylic monomers or prepolymers.
For both end-uses - drug delivery and embolization - , there is a need to find new organic solvents. The reasons are manifold. In the case of pharmaceutical parenterals, a range of vehicles is necessary to meet the solubility requirements preventing drug precipitation upon intravenous or intra-arterial injection [3]. With embolic liquids, the nonaqueous solvents should possess good solubilizing ability for the chosen polymer as well as confer appropriate precipitation behavior to the solution upon contact with blood [15].
However, the main concern in using organic solvents is of course their harmlessness. These vehicles must show hemocompatibility, must be devoid, if possible, of any toxicity, cancerogenicity, teratogenicity and mutagenicity, must be non-irritating and non-sensitizing, and must not exert any pharmacological activity. Dimethyl sulfoxide (DMSO) has generally been used for that purpose but investigations have revealed its local, dose-related toxic effect on vessel 4
and swine brain tissue [16-18], as well as its cardiovascular toxicity [19]. Regarding toxicity of other nonaqueous solvents, the literature survey provides little information concerning the cardiovascular effects observed upon intravascular injection. Moreover, most investigations deal with solvents diluted with water and there is a lack of comparative evaluation of solvents using the same methodologies.
Thus, we report here on the possible hemodynamic changes induced by thirteen water-miscible organic solvents selected owing to their previous use, at least at low concentration, in pharmaceutical injectable formulations, and to their ability to dissolve embolizing materials [20]. Among them, three are welldocumented solvents in terms of toxicological data after intravascular administration, six are used for some intravascular applications but are not yet established, and four have unknown effects upon intravascular injection but are potentially useful.
It is worth noting that cardiovascular effects have been
shown for some solvents of the first two categories. The same thirteen solvents were previously investigated for their hemolytic activity [21] and, more recently, the angiotoxicity (induction of vasospasms) of six of them was evaluated [22]. In present work, the solvents were injected in the infra-renal aorta of male adult sheep, in three different volumes, and the hemodynamic parameters were estimated by recording the arterial and venous pressures, as well as the heart rate [19].
Materials and Methods
Organic solvents
The solvents were selected owing to their rather low viscosity, their ability to dissolve drugs and polymeric materials, as well as their described use, at least at low concentration, in pharmaceutical injectable formulations [15, 20]. Ethyl lactate, dimethyl sulfoxide (DMSO), tetrahydrofurfuryl alcohol, N-methyl-2pyrrolidone (NMP), glycerol formal, ethanol, SolketalTM, acetone, diglyme, 5
propylene glycol and polyethylene glycol 200 (PEG 200) were purchased from Fluka (Buchs, Switzerland). GlycofurolTM 75 was a gift from Hoffmann-La-Roche and dimethyl isosorbide was supplied by ICI Chemicals (Essen, Germany). Saline solution was used as a control.
Injection
Four Prealpes male adult sheep weighing 35-40 kg were used for the study, each solvent being tested in three different animals. Sheep were treated by an intravenous injection of 0.5 ml of sodium thiopental (NesdonalTM, Specia RPR, Paris, France). Then, they were intubed and anaesthetized by inhalation of 1 to 2% halothane under artificial breathing. Two 6F introducers were placed in the iliac position by the Seldinger technique, one in the iliac artery, the other in the iliac vein, for measuring arterial and venous pressure. A continuous monitoring of arterial pressure (AP), venous pressure (VP) and heart rate (HR) was carried out during the experiment.
A 5F catheter (Terumo, Tokyo, Japan), with an empty space of 1.0 mL, was introduced in the infra-renal aorta by the iliac-femoral route. Three doses (0.1, 0.5 and 1.0 mL) of saline and solvent were alternatively injected over a few seconds to examine the dose-related effect. The catheter was rinsed by 5.0 ml saline before every injection. Pressure monitoring was carried out for at least 3 min. A total of 316 injections were done during this study, 160 for the saline and 156 for the solvents. At the end of the experiment, the animals were sacrificed by an intravenous injection of sodium thiopental.
Hemodynamic variations
Duration of the saline or solvent injections, as well as duration and intensity of the pressure and heart rate modifications were recorded using the AcknowledgeTM system (Biopac System Inc, Santa Barbara, USA).
6
The injection duration was measured on the recording as the time between the beginning and the end of injection. The duration of the hemodynamic changes was estimated on the arterial pressure curve as the time duration between the beginning of the hemodynamic modification and the return to the base line existing before the injection. The intensity of the hemodynamic modification was generally estimated as the difference, expressed as a percentage, between the mean value before injection and the mean value at the highest measurement of the variation. However, when the confidence interval approach was used for the statistical analysis (see below) the ratio of the parameter at the maximum modification and before injection was considered.
Statistical Analysis
As no assumption could be made about the distribution of the data, correlations between factors were examined using the nonparametric Spearman test. The significance limit p was set at 0.05.
For establishing a final ranking of the solvents according to their cardiovascular toxicity (in fact only the arterial pressure), the classical theory of null hypothesis testing aimed at proving the absence of a purely statistical difference between sets of data was not used in the present work. The significance of the changes in hemodynamic parameters was evaluated using the more relevant approach based on the confidence intervals. The method is advocated when a biologically or clinically important difference is to be detected. It is used for evaluating bioequivalence study data [23]. It is preferred to the classical theory of null hypothesis testing where very close sets of data with small variance could be for instance regarded as statistically nonequivalent although their difference would be unimportant with respect to some biological response [24, 25]. The general procedure of the 90 % confidence interval recommended for evaluating bioequivalence data was thus followed. Here, the ratio of the parameter at the maximum modification and before injection was considered.
7
Results
All the selected solvents, except saline used as a control, induced hemodynamic modifications, expressed as a concomitant decrease in arterial pressure and increase in venous pressure. No significant changes in heart rate were observed. An example of a typical recording of the hemodynamic parameters is given in Figure 1.
Figure 1: Example of the hemodynamic modifications of the arterial (AP) and venous (VP) pressures, as well as heart rate (HR).
Injection duration
Intensity
On average, the time necessary to inject the solvents was twice that of the saline (respectively 6.4, 7.7 or 8.6 s for the solvents and 3.6, 4.3 or 4.4 s for saline, when injecting 0.1, 0.5 or 1.0 mL). Among the organic solvents a threefold difference was observed between the lowest (Solketal, diglyme) and the highest values (PEG 200 and DMSO). Correlation between injection duration and solvent viscosity or volume injected was not systematic, but overall the injection rates are referred to as high and are comparable to those reported in a previous works [19].
8
Modification of the arterial pressure
The percent arterial pressure changes induced by injection of the three nominal volumes are shown in Figure 2.
Figure 2: Percent changes in arterial pressure induced by injection of the solvents at three different volumes (mean ± SD).
0 -10 -20
-30 -40
d0,1ml d0,5ml d1ml
Saline
Solketal
Diglyme
DMI
GFormal
GFurol
THFA
PEG200
PG
LE
NMP
Ethanol
-60
Acétone
-50
DMSO
Modification of the arterial pressure dAP [%]
10
Changes in arterial pressure were insignificant for all organic solvents, with a 1 to 7 ratio decrease in arterial pressure observed between the lowest (PEG 200, GlycofurolTM 75, dimethyl isosorbide) and the highest values (DMSO, ethyl lactate, NMP, propylene glycol). The transient decrease in arterial pressure was actually shown to be negatively correlated with the injection volume when considering all solvents (p < 0.0001, Spearman). Incidentally, a negative correlation was observed between change in arterial pressure and injection duration. Physiologically, these results suggest that the solvents induce an acute cardiac insufficiency with a diminution of the post-charge and an increase of the cardiac pre-charge.
Modification of the venous pressure
9
The concomitant elevation effect of injecting the solvents on the venous pressure (shown on Figure 3) was by far less pronounced than that on the arterial pressure.
Figure 3: Percent changes in venous pressure induced by injection of the solvents at three different volumes (mean ± SD).
d0,1ml 30
d0,5ml d1ml
20 10 0 -10
Saline
Solketal
Diglyme
DMI
GFormal
GFurol
THFA
PEG200
PG
LE
NMP
Ethanol
-30
Acétone
-20
DMSO
Modification of the venous pressure, dVP [%]
40
As a direct result of the increase in venous pressure, a positive correlation was noticed between pressure change and the infusion volume for all solvents (p < 0.0001, Spearman).
The intensities of arterial and venous pressure modifications were negatively correlated for all products and injection volumes taken together (p < 0.0001, Spearman). Modification of the heart rate
Heart rate was only slightly modified by injecting organic solvents. Interestingly, opposite effects were observed. An acceleration of the cardiac rhythm was noticed for the majority of organic solvents, whereas injections of diglyme or dimethyl isosorbide decreased the heart rate (Figure 4). Quite logically, because of the opposite effects on the cardiac no correlation was found when considering all products taken together. 10
Figure 4: Percent changes in cardiac frequency induced by injection of the solvents at three different volumes (mean ± SD).
d0,1ml
30
d0,5ml d1ml
20
10
0
-10
Saline
Solketal
Diglyme
DMI
GFormal
GFurol
THFA
PEG200
PG
LE
NMP
Ethanol
-30
Acétone
-20
DMSO
Modification of the cardiac frequency, dCF [%]
40
Duration of the hemodynamic modifications
The time where hemodynamic parameter changes were recorded lasted around 2 min at maximum (Figure 5). Such a duration was observed upon injecting 1.0 mL of acetone, NMP, ethyl lactate, tetrahydrofurfuryl alcohol, GlycofurolTM 75, diglyme and Solketal.
Duration of the hemodynamic modifications was correlated with the volume injected for all products (p < 0.0001, Spearman). An inverse correlation was noticed between the duration of the hemodynamic effects and changes in arterial pressure (p < 0.0005, Spearman). Correlation between duration of hemodynamic modifications and duration of injection was significant only for the 1 mL infusion volume (p < 0.05, Spearman).
Figure 5: Duration of the hemodynamic modifications (mean ± s.d.).
11
d0,1ml
180
d0,5ml
160
d1ml
140 120 100 80 60 40
Saline
Solketal
Diglyme
DMI
GFormal
GFurol
THFA
PEG200
PG
on
LE
NMP
report
Acétone
0
Ethanol
20 DMSO
Duration of the haeodynamic modification [s]
200
Discussion
Several
studies
hemodynamic
responses
after
parenteral
administration of organic solvents, but most of the time dealing with single solvents. Moreover, such investigations cannot be compared because of differences in model animal, route of administration, dose and methodology. For instance, hemodynamical changes have been reported many times for DMSO. Transient hypotension and sometimes negative as well as positive chronotropic effect were reported in mice [26], rats ([27, 28], rabbits [29] , cats ([27, 30, 31], dogs [27, 31, 32], and sheep [19], swine and monkeys (De La Torre, Kassel et/ou Tsuruda [33 pour nune de ces 3 publis],). Thus, Hameroff et al. [32] showed in mongrel dogs elevated cardiac output accompanied by decrease in mean arterial pressure and increase in heart rate, which lasted 10 mi after iv infusion. Mean arterial pressure was also higher over 5 h when DMSO was given to a canine model of myocardial ischemia compared to saline [34]. In contrast, in few other studies, no or opposite effects were reported. Thus, no significant change in blood pressure or heart rate was noticed by Chalupka et al. [18] upon slow infusion of DMSO (0.5 and 0.8 mL) using a swine rete mirabile embolization model. There was also no acute effect on blood pressure after i.p. administration of DMSO in rats following experimental ischemia [35]. 12
Investigations by [36] Brown showed transient increase of blood pressure in pentobarbitalized cats (cité par David). As for human patients, low dose of DMSO could lower intracranial hypertension without altering the systemic arterial pressure [37].
The results obtained here with a sheep model confirm the finding of Laurent et al. [19] that a few seconds after intraarterial administration of DMSO acute dose-dependent hemodynamical changes persisting for several minutes are observed, comprising a decrease in arterial pressure and decrease in venous pressure but no significant modification of heart rate. These effects were noted using injection durations of less than 10 seconds as in the present study. A low dose injected over 60 seconds did not induce hemodynamic changes [19]. Later investigations using a swine model have highlighted the dependency of this systemic toxicity on the volume injected and the injection duration [18].
Hypotension is a well known cardiovascular effect caused by ethanol is well documented. Incontrast, reports on other solvents are much less frequent. For example, glycofurol cardiovascular effects somehow differed according to the test animal and volume injected: both increase and decrease in arterial pressure and positive and negative chronotropic effects were observed in rats, cats and dogs [38]. Intravenous administration in the same three animal species of PEG 400 elicited hypotension as well as initial and transient positive chronotropic effect, followed by hypertension and reactional bradycardy [39]. In contrast, PEG 600 did not affect the mean blood pressure of dogs [40]. Intravenous infusion of acetone was reported to produce a fall in blood pressure primarily due to a decrease in the cardiac output [41]. A transient depressive effect on blood pressure was observed in rats after intravenous administration of glycerol formal [42], that was also noticed I the isolated rabbit heart [43, 44]. Hypotention was observed for propylene glycol in dogs, explained by peripheral vasodilation [45]. As for Buden et al. [38], they again found animal- and dosedependent, sometimes biphasic effect on blood pressures well as both negative and positive chronotropism.
13
Voir Glycofurol: Budden
In the present study the thirteen water-miscible organic solvents induced, after infra-renal aorta injection of three different volumes (0.1, 0.5 and 1.0 mL) in male adult sheep transient reduction in arterial pressure, with a concomitant but less pronounced augmented venous pressure. Differing according to the solvent tachycardia or bradycardia was also observed but alterations were anyhow weak. These results suggest that these solvents induced a cardiac insufficiency with a diminution of the post-charge and an increase of the cardiac pre-charge. In order to classify the solvents in terms of cardiovascular effects, the relative changes in arterial pressure from the baseline after injection of 1 mL volume was considered as being most clinically relevant. Significance of the changes was evaluated using the approach based on the confidence interval, advocated when a biological or clinical difference is to be detected and preferred to the classical theory of null hypothesis testing aimed at proving the absence of a purely statistical difference between sets of data. Table 1 lists for the organic solvents and for saline the mean change in arterial pressure (AP), the 95 % confidence interval for the AP ratio before (1) and after injection (2) as well as the minimum significant change in arterial AP. Data are presented by increasing order of the latter parameter. Under the experimental conditions described drop in arterial pressure up to 25 % could be observed using the confidence interval approach.
Table 1: Arterial pressure change after intraarterial infusion of 1 mL solvent and statistical parameters. Solvent
Mean
AP Confidence
change (%)
Minimum
interval for the significant AP AP2/AP1 ratio
change (%) 14
Saline
+0.8
1.005-1.012
+0.5
Dimethyl isosorbide
-4.3
0.884-1.030
0
Glycofurol 75
-5.4
0.876-1.016
0
PEG200
-2.6
0.969-0.979
-2.1
Diglyme
-11.1
0.848-0.930
-7.0
Tetrahydrofurfuryl alcohol
-10.6
0.890-0.898
-10.2
Ethanol
-15.9
0.796-0.887
-11.3
Acetone
-16.9
0.787-0.875
-12.5
Solketal
-15.9
0.809-0.873
-12.7
Glycerol formal
-15.7
0.819-0.867
-13.3
DMSO
-23.5
0.688-0.843
-15.7
Propylene glycol
-27.9
0.651-0.792
-20.8
Ethyl lactate
-25.7
0.739-0.747
-25.3
NMP
-26.8
0.717-0.747
-25.3
Taking a 10 % change in arterial pressure as the maximum tolerable effect [18], a classification is proposed in Table 2 for the thirteen water-miscible organic solvents tested. Note that a purely statistical test has been used by the latter authors and that a 5 % change has been retained in some studies [27, 38,39].
Table 2: Comparative cardiovascular toxicity of organic water-miscible solvents based on change in arterial pressure
AP reduction
Range (%)
Solvents
Absent
0 – 9.9
Saline, dimethyl isosorbide, GlycofurolTM 75, PEG 200, diglyme
Moderate
10.0 - 19.9
Tetrahydrofurfuryl alcohol, ethanol, acetone, Solketal, glycerol formal, DMSO
Marked
> 20
Propylene glycol, ethyl lactate, NMP
Hemodynamic changes should of course be analysed on a relative prospective and the classification proposed is only valid for comparative purpose, as data would evidently differ if based on the lower injected doses or when using 15
another animal model, a different parenteral route or a different rate of injection. In particular, Laurent et al. [19] and Chaloupka et al. [18] have shown the prominent influence of the latter factor. Here, it is stressed that rapid infusion was used in order to better discriminate the solvents and, according to the present results, dimethyl isosorbide and glycofurol could be good alternatives to DMSO and NMP, two commonly used organic solvents. PEG 200 could also be a suitable alternative but has less solubilizing power towards water-insoluble drugs or polymers.
The present classification based on arterial pressure reduction is claimed to provide a useful tool for solvent selection but of course the exact relevance of the observed cardiovascular effects for humans still needs to be investigated. Proper solvent selection should also take into consideration additional factors of utmost importance, such as possible pharmacological or toxic effects, cancerogenicity, teratogenicity, mutagenicity, non-irritating and non-sensitizing properties.
In
this
respect,
we
published
hemocompatibility [21] and angiotoxicity
comparative
data
on
the
[22] of the solvents studied in the
present work, but it has to be stressed that other relevant toxicity data are missing for some of these solvents, at least in the open literature [20]. Acknowledgments
This work was supported by FNSRS grant # 32-45 823.95.
References
1.
B. Kruss, "Toxikologie and Verträglichkeit von Lôsungsmitteln zur Parenteralen Anwendung". In Flüssige Arzneiformen Schwerlöslicher Arzneistoffe, (Eds:. ? Essig, and ? Stumpf), Wissenschaftliche Verlagsgesellschaft, Stuttgart, 1990, pp. 42-61.
16
2.
W. Füller, "Vehikel und Hilfsmittel zur Herstellung von Injektionslösungen von Steroidhormonen," Pharm. Acta Helv., 47, 449456 (1972).
3.
J. I. Wells, "Pharmaceutical Preformulation" in The Physicochemical Properties of Drug Substances, Wiley, New York, 1988. Eds ????
4.
Y.-C. J. Wang, and R.R. Kowal, "Review of excipients and pH's for parenteral products used in the United States", J. Parent. Drug Assoc., 34, 452-462 (1990).
5.
A. J. Spiegel, and M. M. Noseworthy, “Use of nonaqueous solvents in parenteral products”, J. Pharm. Sci.,52, 917-927.
6.
C. Heintz, and C. Boymond, "Aspects pharmacotoxicologiques et utilisation thérapeutique de quelques solvants injectables non aqueux miscibles à l'eau", STP Pharma, 5. 548-560 (1989).
7.
S. Nema, R. J. Washkuhn, and R. J. Brendel, "Excipients and their use in injectable products", PDA J. Pharm. Sci. Technol., 51, 166-171 (1997).
8.
M. F. Powell, T. Nguyen, and L. Balvian, "Compendium of excipients for parenteral formulations", PDA J. Pharm. Sci. Technol., 52, 238-311 (1998).
9.
B. Kruss, "Die Hilfsstoffauswahl für Arzneiformen unter gesetzlichen und toxikologischen Aspekten", Acta Pharm. Technol., 35, 187-196 (1989).
10. J. Anschel, “Lösungsmittel und Lösungsvermittler in Injektionen”, Pharm. Ind.,27,781-787 (1965).
17
11. S. Mandai, K. Kinugasa, and T. Ohmoto, "Direct thrombosis of aneurysms with cellulose acetate polymer. Part I. results of thrombosis in experimental aneurysms”, J. Neurosurg., 77, 497-500 (1992).
12. K. Kinugasa, S. Mandai, Y. Terai, I. Kamata, K. Sugiu, T. Ohmoto, and A. Nishimoto, "Direct thrombosis of aneurysms with cellulose acetate polymer. Part II. preliminary clinical experience", J. Neurosurg., 77, 501507 (1992).
13. T. Terada, Y. Nakamura, K. Nakai, M. Tsuura, T. Nishiguchi, S. Hayashi, T. Kido, W. Taki, H. Iwata, and N. Komai, "Embolization of arteriovenous malformations with peripheral aneurysms using ethylene vinyl alcohol copolymer"J. Neurosurg., 75, 655-660 (1991).
14. Y. Matsumaru, A. Hyodo, T. Nose, T. Hirano, and S. Ohashi, "Embolic materials for endovascular treatment of cerebral lesions", J. Biomater. Sci. Polymer Edn., 8, 555-569 (1997).
15. F. Mottu, P. Gailloud, D. Massuelle, D. A. Rüfenacht, and E. Doelker, “In vitro assessment of new embolic liquids prepared from preformed polymers and water-miscible solvents for aneurysm treatment”, Biomaterials, 21,803-811(2000).
16. F. Mottu, D. A. Rüfenacht, A. Laurent, and E. Doelker, "Organic solvents for pharmaceutical parenterals and embolic liquids : A review of toxicity data", PDA J. Pharm. Sci. Technol., 54, 456-469 (2000).
17. J. C. Chaloupka, G. Guglielmi, and F. Vinuela, "Direct thrombosis of aneurysms," J. Neurosurg., 78, 1006-1007 (1993).
18. J. C. Chaloupka, F. Vinuela, H.V. Vinters, and J. Robert, "Technical feasibility and histopathologic studies of ethylene vinyl copolymer
18
(EVAL) using a swine endovascular embolization model," Am. J. Neuroradiol., 15 1107-1115 (1994).
19. J. C. Chaloupka, D.C. Huddle, J. Alderman, S. Fink, R. Hammond, and H.V. Vinters, "A reexamination of the angiotoxicity of superselective injection of DMSO in the swine rete embolization model," Am. J. Neuroradiol., 20, 401-410. (1999).
20. A. Laurent, M. Wassef, H.Q. Zhang, O. Brailliard, E. Houdart, A. L. Bailly, and J. J. Merland, "Solutions gélifiantes pour embolisation: étude animale de la toxicité cardiovasculaire d'un constituant, le diméthylsulfoxyde", J. Neuroradiol., 24, 314 (1997).
21. F. Mottu, M.-J. Stelling, D. A. Rüfenacht, and E. Doelker, "Comparative haemolytic activity of undiluted organic water-miscible solvents for intravenous and intra-arterial injection”, PDA J. Pharm. Sci. Technol., 55, 16-23 (2000).
22. O. Dudeck
“Organic solvents as vehicles for precipitating liquid
embolics: a comparative angiotoxicity study with superselective injections of swine rete mirabile”, to be published.
23. Food and Drug Administration “Bioequivalence guidance”, October 9, 2002.
24. W. W. J. Westlake, “Bioavailability and bioequivalence of pharmaceutical formulations”, in Biopharmaceutical Statistics for Drug Development, Peace, K. (ed), Marcel Dekker, Inc., 329-352 (1988).
25. D. J. Schuirmann, “A Comparison of the two one-sided tests procedure and the power approach for assessing the equivalence of average bioavailability”, J. Pharm. Biopharm., 15, 657-680 (1987).
19
26. K. Kramer, S. A. B. E. van Acker, J. A. Grimbergen, D.-J. van den Berg, w. J. F. van der Vijgh, and A. Bast, “Effect of dimethyl sulfoxide (DMSO) on the electrocardiogram (ECG) in freely moving male balb/c Mice”, Gen. Pharmac., 26, 1403-1407 (1995).
27. R. von Budden, U. G. Kühl, and G. Buschmann, “Ausgewählte Untersuchungen zur pharmakodynamischen Eigenwirkung verschiedener Lösungsvermittler. 1. Mitteilung: Äthyldi!athylenglykok, N,N-Diäthyllactamid, Dimethysulfoxid”, Arzneim.-Forsch./Drug Res., 28, 1571-1579 (1978).
28. M. K. Bazil, C. Krulan, and R. L. Webb, “Telemetric monitoring of cardiovascular parameters in conscious spontaneously hypertensive rats”, J. Cardiovasc. Pharmacol., 22, 897-905 (1993).
29. F. R. Domer, “Cardiovascular and neuromuscular effects of dimethylsulfoxide in anaesthetized rabbits”, J. Pharm. Sci., 66, 269-270 (1977).
30. V. DiStefano, and J. J. Klahn, “Observations on the pharmacology and hemolytic activity of dimethyl sulfoxide”, Toxicol. Appl. Pharmacol., 7, 660-666 (1965).
31. R. K. Varma, R. Kaushal, G. P. Thomas, A. Y. Junnarkar, P. P. Singh, and R. M. Tripathi, “Evaluation of dimethyl sulfoxide as a solvent in pharmacological experiments”, Ind. J. Ex. Biol., 25, 758-760 (1987).
32. C. G. Peterson, and R. D. Robertson, “A pharmacodynamic study of dimethyl sulfoxide”, Ann. N.Y. Acad. Sci., 141, 273-276 (1967).
33. S. R. Hameroff, C. W. Otto, J. Kanel, P. R. Weinstein, and C. D. Blitt, “Acute cardiovascular effects of dimethyl sufoxide”, Crit. Care Med., 9, 855-857 (1981). 20
34. DeLaTorre ou Kassel ou Tsuruda
35. J. M. Levett, L. M. Johns, N. M. Grina, B. F. Mullan, J. F.Kramer, and J. F. Mullan, “Effects of dimethyl sulfoxide on systemic and cerebral hemiodynamic variables in the ischemic canine myocardium”, Crit. Care Med., 15, 656-660 (1987).
36. S. Shimizu, R. P. Simon, and S. H. Graham “Dimethylsulfoxide (DMSO) treatment reduces infarction volume after permanent focal cerebral ischemia in rats”, Neuroscience letters, 239, 125-127 (1997).
37. Brown ref 50 de David ????
38. L. F. Marshall, P. E. Camp, and S. A. Bowers, “Dimethyl sulfoxide for the treatment of intracranial hypertension : A preliminary trial, Neurosurgery, 14, 659-663 (1984).
39. R. von Budden, U. G. Kühl, and G. Buschmann, “Ausgewählte Untersuchungen zur pharmakodynamischen Eigenwirkung verschiedener Lösungsvermittler. 3. Mitteilung: (1,2)-Propandiol, Tetrahydrofurfurylalkohol-poly!athylenglycol!ather (THFP), Polyoxy!athylensorbitanmonooleat (PSM)”, Arzneim.-Forsch./Drug Res., 28, 1586-1593 (1978).
40. R. von Budden, U. G. Kühl, and G. Buschmann, “Ausgewählte Untersuchungen zur pharmakodynamischen Eigenwirkung verschiedener Lösungsvermittler. 2. Mitteilung: Glycerin, NHydroxy!athyllactamid, Poly!athylenglycol 400 ”, Arzneim.-Forsch./Drug Res., 28, 1579-1586 (1978).
21
41. R. D. Heilman, E. W. Bauer, and J. P. DaVanzo, “Effect of Polyethylene glycol on the cardiovascular response of the dog autonomic agents”, Toxicol. Applied Pharmacol., 23, 263-270 (1972).
42. S. L. Cheav, S. Salloum, F. Pieri, and S. Kirkiacharian, " Action du glycérolformal sur le système cardiovasculaire”, Ann. Pharm. Fr., 43, 491-503 (1985).
43. R. Chahine, ans S. L. Cheav, " Effects of glycerolformal on sympathetic neurotransmission in the isolated rabbit heart”, Arzneim.-Forsch./Drug Res., 41, 449-453 (1996).
44. S. L. Cheav, R. Chahine, and M. S. Mroué, " Inotropic and chronotropic effectof glycerol formal on the isolated rabbit heart”, Arzneim.Forsch./Drug Res., 42, 997-1000 (1992).
45. J. Bost, and Y. Ruckebusch, “Toxicité et pharmacologie du propyleneglycol”, Thérapie, XVII, 83-94 (1962).
22
