The simple halogenated alkylamine 2-bromoethylamine has long been known as a nephrotoxic agent. Early work with the amine revealed that it produced renal papillary necrosis in the rat and that these lesions resembled those observed in renal damage caused by nonsteroidal analgesics.1 The fact that the pathologic changes it elicits are remarkably circumscribed to the renal medulla2,3 has more recently led to its use to characterize the putative renal medullary depressor hormone4 as well as to implement a specific model of hypertension.5,6
Another action of bromoethylamine is inhibition of a variety of amine oxidases,7-9 acting in all cases as a suicide inhibitor. Not unexpectedly, it also inhibits by this mechanism another class of amine-oxidizing enzymes collectively known as semicarbazide-sensitive amine oxidase (SSAO; EC 184.108.40.206).10-12
Although a precise physiological or pathologic role of SSAO in mammals has not been established, a number of functions attributed to the enzyme and/or its products have been proposed. Among these are formation of advanced glycation endproducts associated with the vascular complications of diabetes, glucose transport in adipocytes, adipocyte maturation, extracellular matrix deposition in vascular smooth muscle, and stimulation of lymphocyte recirculation and adhesion to endothelial cells.13 The last effect is attributed to the sequence identity of SSAO with vascular adhesion protein-1.14
Because SSAO is particularly abundant in arteries,15 we have proposed a role of the enzyme in the regulation of vascular tone.16 Hydrazine derivatives constitute a group of fairly specific SSAO inhibitors,17 and hydralazine, a member of this group, is also a potent arterial vasodilator. Its hypotensive effect in rats in enhanced by pretrteatment with other hydrazine SSAO inhibitors18 as well as with SSAO substrates.19 These observations, together with the fact that hydralazine is concentrated in vascular smooth muscle after systemic administration,20 led us to postulate that vasodilation by the drug is related to in situ inhibition of vascular SSAO.16
One of the limitations in SSAO research is the lack of sufficiently specific inhibitors that could be used as tools to characterize the functions of the enzyme.17 For example, the antitubercular hydrazine derivative isoniazid at low doses potentiates hydralazine hypotension, presumably because in these conditions it inhibits SSAO.21 However, this drug at higher doses nonspecifically enhances responses to a variety of vasodilators, an effect attributed to decreased cerebral GABA consequent to inhibition of the GABA-synthesizing enzyme glutamic acid decarboxylase.22 Considering its mechanism of SSAO inhibition, bromoethylamine can be expected to exhibit a high degree of specificity for the enzyme because it is SSAO itself that generates the active inhibitor, probably the corresponding aldehyde.8,12 It therefore seemed of interest to determine whether bromoethylamine, like SSAO inhibitors of the hydrazine type, also potentiates hydralazine hypotension.
This possibility was explored in the present work using anesthetized rats, either unpretreated or pretreated with bromoethylamine, challenged with a standard dose of hydralazine and noting any difference in the magnitude of hypotensive responses. Because little is known of the relationship between in vivo enzyme inhibition and dose or time after administration of bromoethylamine, these issues were explored as part of this work. Results with the amine were compared with those with the prototypical nonsuicide SSAO inhibitor semicarbazide. Responses to various combinations of both inhibitors were also analyzed.
Experiments were carried out in adult male Wistar rats weighing between 200 and 300 g, raised in the animal facilities of the School of Medicine, Universidad Nacional Autonoma de Mexico. They were kept in animal rooms maintained at 21-23°C and subjected to 12-hour light/12-hour dark cycles, the light phase occurring between 0700 and 1900 hours. The animals had free access to food pellets (5001 Rodent Laboratory Chow, Agribrands Purina Canada, Woodstock, Canada) and tap water. They were brought daily to the laboratory for the experiments, which were conducted in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, revised 1996). The experimental protocol was approved by the Ethics Committee of the School of Medicine, Universidad Nacional Autonoma de Mexico.
Blood Pressure and Heart Rate Recordings
Rats were anesthetized with a mixture of chloralose, 50 mg/kg, and urethane, 800 mg/kg, administered IP. Appropriate polyethylene cannulas were inserted in the trachea to facilitate spontaneous breathing, in a femoral artery for blood pressure recording, and in a femoral vein for drug administration. The arterial cannula was filled with heparin, 50 units/mL, to prevent clotting; no anticoagulant was administered systemically. Blood pressure was recorded on a Grass model 79 polygraph with a Statham-Gould P23ID transducer (Grass Instrument Division, Astro-Med, Inc, West Warwick, RI). High frequencies of the transducer signal were electronically filtered to obtain a smooth tracing approximately equivalent to mean blood pressure. Heart rate was simultaneously recorded on another channel of the polygraph with a Grass 7P4 cardiotachograph triggered by the pulse waves from the unfiltered transducer signal. Rectal temperature of the rats was maintained beween 36.8°C and 37°C with a model 621-8620 Barnant temperature controller (Barnant Co, Barrington, IL). After a stabilization period of at least 20 minutes, hydralazine, 0.1 mg/kg, was injected IV as a bolus, and recordings were continued for 60 minutes thereafter; blood pressure and heart rate values were tabulated at 10-minute intervals.
To determine the influence of bromoethylamine on hydralazine hypotension and compare it to that of semicarbazide, optimal pretreatment doses and dosing intervals for both SSAO inhibitors were first established. In 2 series of experiments, groups of 6 rats were pretreated IP with bromoethylamine, 50 mg/kg, or semicarbazide, 10 mg/kg, at intervals of 1, 2, 3, 4, or 5 hours before receiving the test dose of hydralazine and recording of the corresponding hypotensive response. These doses were reported to induce clear inhibition of aortic SSAO in the case of bromoethylamine12 and hydralazine potentiation in the case of semicarbazide.18 Once the optimal times for hydralazine potentiation were determined, 3 hours for bromoethylamine and 1 hour for semicarbazide, decreasing and increasing doses administered at these intervals were explored in 2 other experimental series: 25, 100, and 200 mg/kg for bromoethylamine and 5, 20, and 40 mg/kg for semicarbazide. In these and the following experiments, control groups of 12 unprtetreated rats were run concurrently with each series. With the optimal pretreatment parameters of 50 mg/kg and 3 hours for bromoethylamine and 10 mg/kg and 1 hour for semicarbazide, combinations of the 2 inhibitors were then assessed for hydralazine potentiation. In the semicarbazide plus bromoethylamine combinations, the hydrazine derivative was administered 1 hour before the amine, which in turn was given 3 hours before the hydralazine challenge. In the bromoethylamine plus semicarbazide combination, the inhibitors were given 3 hours and 1 hour, respectively, before hydralazine. Similar experiments were run using isoniazid instead of semicarbazide. The pretreatment schedule for the antitubercular drug was 30 mg/kg 30 minutes before hydralazine.21 In a final series, bromoethylamine, semicarbazide, and isoniazid were administered separately at the above schedules, but verapamil, 0.2 mg/kg IV, was used instead of hydralazine for vasodilator challenge.
The hydrochlorides of semicarbazide, hydralazine, and (±)-verapamil, as well as 2-bromoethylamine hydrobromide and isoniazid free base, were obtained from Sigma-Aldrich Quimica SA de CV (Toluca, Mexico). Doses refer to the salts, where applicable. All drugs were dissolved in isotonic NaCl solution and were administered in a volume of 1 mL/kg.
Data Presentation and Statistical Analysis
Results are presented as means ± SEM of 12 rats for the control groups and of 6 rats for the pretreated groups. Posthydralazine or postverapamil blood pressure changes of pretreated groups at each 10-minute tabulation period were compared with the corresponding controls by one-way analysis of variance followed by Dunnett's post-hoc test. In the experiments with combinations of bromoethylamine, semicarbazide, and isoniazid, groups pretreated with the different combinations were compared with those receiving bromoethylamine or isoniazid alone, as appropriate. When nonhomogeneous variances among the groups were detected, the nonparametric Kruskal-Wallis and Dunn procedures were applied. In all cases, a probability level of less than 0.05 (2-tailed tests) was accepted as significant. Statistical evaluation was carried out with a GraphPad Prism 4.02 package (GraphPad Software Inc, San Diego, CA).
Baseline blood pressure and heart rate values recorded in the different control and pretreated groups immediately before vasodilator injection are shown in Table 1. None of the pretreatments influenced blood pressure, but heart rates of the subgroups receiving semicarbazide, 10 mg/kg at 1, 2, and 3 hours, were significantly increased. In the pooled 5 groups of 12 control animals each, the test dose of hydralazine lowered blood pressure by approximately 10 mm Hg, a statistically significant effect that tended to subside toward the end of the 60-minute period of observation (Fig. 1). This response was accompanied by a parallel nonsignificant fall in heart rate of less than 10 beats/min and of greater variability than the blood pressure effect.
Pretreatment with bromoethylamine, 50 mg/kg, significantly enhanced hydralazine hypotension when the amine was injected 3 or 4 hours previously (Fig. 2). When different doses of bromoethylamine were tested at 3 hours, potentiation was observed only with 50 and 100 mg/kg (Fig. 3). In the case of pretreatment with semicarbazide, 10 mg/kg, increased responses to hydralazine were apparent at 1 through 5 hours of administration (Fig. 4) and with 10, 20, and 40 mg/kg given 1 hour before (Fig. 5). Figure 6 shows responses to hydralazine after bromoethylamine, 50 mg/kg at 3 hours, either alone or preceded or followed by semicarbazide, 10 mg/kg at 1 hour. Semicarbazide completely prevented hydralazine potentiation when administered before bromoethylamine but produced a greater potentiation when given after the amine. The figure also shows that the potentiation elicited by isoniazid, 30 mg/kg at 30 minutes, was unaffected when preceded by semicarbazide but was significantly enhanced when preceded by bromoethylamine. Hypotension after verapamil was not significantly modified by bromoethylamine, semicarbazide, or isoniazid, administered at optimal hydralazine-potentiating doses and pretreatment intervals, although responses after the hydrazine inhibitors tended to be longer-lasting than those in controls (Fig. 7).
None of the pretreatments significantly affected, at any of the 10-minute observation points, the small changes in heart rate accompanying the hypotension elicited by hydralazine or verapamil (not shown). Although the overall rate changes observed in no case exceeded 25 beats/min, some of the groups pretreated with bromoethylamine tended to show increased heart rates in respònse to the vasodilators, whereas those receiving semicarbazide or isoniazid showed decreased rates. In the case of combined pretreatments, bradycardia was the result whenever semicarbazide or isoniazid was administered last.
The present results show that bromoethylamine was able to potentiate the hypotensive response to hydralazine in an apparently selective fashion because this interaction did not occur with verapamil. Some of the experimental findings are compatible with the idea that this potentiation is related to SSAO inhibition by the amine and is thus another example of the action of other inhibitors of the enzyme.21 On 1 hand, the effect was not observed at pretreatment intervals shorter than 3 hours, in accordance with the known mechanism of SSAO inhibition by bromoethylamine requiring prior formation of an active inhibitory product. On the other hand, hydralazine potentiation was prevented by prior administration of the irreversible SSAO inhibitor semicarbazide, which would block generation of this product.
Yu et al,12 using the pretreatment interval of 3 hours, reported aortic SSAO inhibition in mice with doses as low as 1 mg/kg, whereas in the present experiments hydralazine potentiation was observed at 50, but not at 25, mg/kg. It may be that near maximal SSAO inhibition is necessary for potentiation or that the mouse enzyme is more susceptible than that of the rat to inhibition by bromoethylamine. Although there is no information on the latter possibility, the work of Raimondi et al23,24 on white adipose tissue SSAO from the 2 species suggests differences between their enzymes regarding affinity for diverse substrates; such differences could extend to sensitivity toward inhibitors.
Hydralazine potentiation by bromoethylamine showed some differences from that by semicarbazide. As expected of a direct irreversible SSAO inhibitor, the latency of semicarbazide was considerably shorter than that of bromoethylamine, its effect already being present 1 hour after administration. In fact, previous work showed activity at 30 minutes.18 The extensive studies carried out in the 1950s with semicarbazide as a convulsant agent showed that latency to the first seizure was dose-dependent, amounting to more than 1 hour at 120 mg/kg.25 This was attributed to the slow CNS penetration of the drug, a factor that is not an issue in the present study. Duration of hydralazine potentiation was also different with the 2 SSAO inhibitors. The effect of semicarbazide persisted up to 5 hours after administration, whereas that of bromoethylamine was undetectable at this time. The long duration of the semicarbazide effect agrees with its reported persistent convulsant activity of up to 5 hours.25,26 On the other hand, the shorter duration of bromoethylamine action could be a result of its rapid excretion. Holmes et al27 found that after a single dose of the amine in the rat, most was excreted in the urine by 4 hours, both as the original compound and as its spontaneous degradation product aziridine.
An interesting characteristic of hydralazine potentiation by hydrazine SSAO inhibitors is the apparent lack of dose-dependency of this effect. In a previous study with isoniazid,21 in which a wide range of doses of this drug were explored, potentiation appeared to be an all-or-none phenomenon, with the effect at 3 mg/kg being no different than that at 300 mg/kg. This pattern was observed in the present study with semicarbazide, which produced a similar potentiation at 10, 20, and 40 mg/kg. Furthermore, a combination of effective doses of semicarbazide and isoniazid did not lead to a greater potentiation. These results suggest that hydralazine potentiation occurs at doses of the hydrazines producing essentially complete SSAO inhibition so that increasing the dose does not cause greater potentiation. If this is true, it would seem unnecessary to use doses of semicarbazide in the order of 100 mg/kg for in vivo inhibition of SSAO in the rat.28,29
An entirely different dose relationship was observed with bromoethylamine pretreatment. Potentiation occurred at 50 and 100 mg/kg but not at 200 mg/kg. When these results were analyzed by calculating the respective areas under the curve, the response after 200 mg/kg was found to be greater than control (−1254 ± 159 versus −435 ± 153 mm Hg at 10 minutes; P < 0.01). However, this response was also significantly smaller than that after 50 mg/kg (−1942 ± 186 mm Hg at 10 minutes; P < 0.01). Because bromoethylamine is a substrate of SSAO, the enzyme could have been inhibited by an excess of this substrate, a phenomenon found by Kumagai et al8 while studying the inhibition of a SSAO-related fungal amine oxidase by bromoethylamine. In this case, the excess substrate would prevent the enzyme from forming the active inhibitory product.
Pretreatment with bromoethylamine followed by semicarbazide or isoniazid unexpectedly produced a greater degree of hydralazine potentiation than that observed with these drugs alone. This finding could be explained by invoking the existence of 2 isoforms of SSAO, 1 susceptible to inhibition by bromoethylamine and the other to inhibition by the hydrazine derivatives. Although such isoforms have been described in goat, sheep, and horse plasma,30 this does not seem to apply to rat plasma, in which the low concentrations observed precluded adequate characterization by enzyme kinetics. Interaction of the diverse inhibitors with 2 differfent active sites of SSAO is another possibility. In the work on the fungal amine oxidase referred to above, Kumagai et al8 studied the spectral changes undergone by the enzyme during inactivation by bromoethylamine. The authors concluded that the active aldehyde derived from the amine inhibits the enzyme by alkylation of a sulfhydryl group essential for activity. On addition of phenylhydrazine, this inhibitor reacts with the carbonyl group of the enzyme to form the corresponding hydrazone. Absorption peaks were 320 nm for the alkylated adduct and 450 nm for the hydrazone. Indeed, the existence of multiple binding sites in bovine lung SSAO, even for the hydrazine innhibitors, has been postulated.31
The present results show that the suicide SSAO inhibitor bromoethylamine shares with irreversible inhibitors of the enzyme the ability to potentiate hydralazine hypotension but give no indication of the site of this interaction. Although there is no evidence indicating a selective effect of SSAO inhibitors other than hydralazine on the vascular enzyme, it appears that such inhibitors do affect this localization, perhaps as part of their overall influence on tissue and plasma SSAO. For example, in vivo administration of semicarbazide protects rats from vascular injury by the SSAO substrate allylamine, presumably by preventing its conversion by the enzyme to the toxic product acrolein.29 Furthermore, in vivo treatment of mice with bromoethylamine decreases SSAO activity measured directly in homogenates of aorta.12 If, as we have postulated, hydralazine hypotension is related to inhibition of vascular SSAO, it is reasonable to expect a greater hypotensive response on additional inhibition of the enzyme at this site.
The fact that the present study was carried out in anesthetized animals subjected to a single administration of bromoethylamine or semicarbazide poses some limitations as to the implications of the results. Anesthesia undoubtedly influenced responses to hydralazine, as shown by the absence of the reflex tachycardia typical of this vasodilator. Confirmation of the present findings in conscious animals would therefore seem mandatory for their meaningful interpretation. The acute nature of the present experiments is another drawback because steady-state enzymatic inhibition, theoretically achievable by repeated administration of bromoethylamine or semicarbazide, would be a more desirable condition for assessment of responses to hydralazine. The renal and CNS toxicity of these agents could preclude such experiments, but drugs such as isoniazid could be used instead. Finally, our limited knowledge of the physiological role of SSAO in the vasculature, as well as of the relationship between the tissue and plasma forms of the enzyme, precludes direct extrapolation of the present findings to the clinical use of hydralazine as a vasodilator or antihypertensive agent. For example, the drug is widely used to control the increased blood pressure in preeclampsia, a condition in which plasma SSAO is elevated.32 If plasma and vascular levels of the enzyme are positively correlated, and if hydralazine produces vasodilation through SSAO inhibition, the drug would be expected to elicit less hypotension in preeclampsia than in other hypertensive conditions. Early work with hydralazine has shown the reverse to be true,33 a fact compatible with the present results if an inverse relationship between tissue and plasma SSAO is postulated.
This work constitutes an example of the use of bromoethylamine in vivo as a tool in the investigation of the role of SSAO in vascular function. The results suggest a relatively narrow range both of effective doses and of timing after administration that can be used to achieve predictable SSAO inhibition. The known renal toxicity of the amine,1-6 as well as its biochemical effects highlighted by increased urinary excretion of glutaric and adipic acids, interpreted as being the result of disruption of mitochondrial metabolism,27 represent additional limitations to its use.
1. Murray G, Wyllie RG, Hill GS, et al. Experimental papillary necrosis of the kidney. I. Morphologic and functional data. Am J Pathol. 1972;67:285-302.
2. Hedlund LW, Maronpot RR, Johnson GA, et al. Magnetic resonance microscopy of toxic renal injury induced by bromoethylamine in rats. Fundam Appl Toxicol. 1991;16:787-797.
3. Szenasi G, Alcorn D, Anderson WP. Renal and cardiovascular effects of renal medullary damage with bromoethylamine in dogs. Blood Press. 1994;3:127-136.
4. Bergstrom G, Nystrom HC, Jia J, et al. Effects of the ETA/ETB antagonist TAK-044 on blood pressure and renal excretory function after unclipping of conscious one-kidney-one-clip hypertensive rats. J Hypertens. 2001;19:659-665.
5. Haugan K, Shalmi M, Petersen JS, et al. Effects of papillary-medullary lesion on the antihypertensive effect of furosemide and development of salt-sensitive hypertension in Dhal-S rats. J Pharmacol Exp Ther. 1997;280:1415-1422.
6. Siddiq T, Patel VB, Sherwood R, et al. In vivo protein synthetic rates of atrial, ventricular, and pulmonary tissue proteins in aortic constriction, Goldblatt, and bromoethylamine models of hypertension. Exp Mol Pathol. 2001;70:19-30.
7. Neumann R, Hevey R, Abeles RH. The action of plasma amine oxidase on β-haloamines. J Biol Chem. 1975;250:6362-6367.
8. Kumagai H, Uchida H, Yamada H. Reaction of fungal amine oxidase with β-bromoethylamine. J Biol Chem. 1979;254:10913-10919.
9. Medda R, Padiglia A, Pedersen JZ, et al. Inhibition of copper amine oxidase by haloamines: a killer product mechanism. Biochemistry. 1997;36:2595-2602.
10. Kinemuchi H, Jinbo M, Tabata A, et al. 2-Bromoethylamine, a suicide inhibitor of tissue-bound semicarbazide-sensitive amine oxidase. Jpn J Pharmacol. 2000;83:164-166.
11. Kinemuchi H, Kobayashi N, Takahashi K, et al. Inhibition of tissue-bound semicarbazide-sensitive amine oxidase by two haloamines, 2-bromoethylamine and 3-bromopropylamine. Arch Biochem Biophys. 2001;385:154-161.
12. Yu PH, Davis BA, Deng Y. 2-Bromoethylamine as a potent selective suicide inhibitor for semicarbazide-sensitive amine oxidase. Biochem Pharmacol. 2001;61:741-748.
13. O'Sullivan J, Unzeta M, Healy J, et al. Semicarbazide-sensitive amine oxidases: enzymes with quite a lot to do. Neurotoxicology. 2004;25:303-315.
14. Smith DJ, Salmi M, Bono P, et al. Cloning of vascular adhesion protein 1 reveals a novel multifunctional adhesion molecule. J Exp Med. 1998;188:17-27.
15. Andrés N, Lizcano JM, Rodríguez MJ, et al. Tissue activity and cellular localization of human semicarbazide-sensitive amine oxidase. J Histochem Cytochem. 2001;49:209-217.
16. Vidrio H. Semicarbazide-sensitive amine oxidase: role in the vasculature and vasodilation after in situ inhibition. Autonom Autacoid Pharmacol. 2003;23:275-283.
17. Kinemuchi H, Sugimoto H, Obata T, et al. Selective inhibitors of membrane-bound semicarbazide-sensitive amine oxidase (SSAO) activity in mammalian tissues. Neurotoxicology. 2004;25:325-335.
18. Vidrio H. Potentiation of cardiovascular responses to hydralazine by diverse hydrazine derivatives. J Pharmacol Exp Ther. 1994;271:171-175.
19. Vidrio H, Medina M, González-Romo P, et al. Semicarbazide-sensitive amine oxidase substrates potentiate hydralazine hypotension: possible role of hydrogen peroxide. J Pharmacol Exp Ther. 2003;307:497-504.
20. Baker JRJ, Hedwall PR, Hermsmeyer K. Subcellular distribution of hydralazine in rat single vascular muscle cells. Cell Biol Int Rep. 1992;16:1023-1039.
21. Vidrio H, Medina M, Fernández G, et al. Enhancement of hydralazine hypotension by low doses of isoniazid. Possible role of semicarbazide-sensitive amine oxidase inhibition. Gen Pharmacol. 2002;35:195-204.
22. Sánchez-Salvatori MA, Ríos C, Vidrio H. Interaction between isoniazid and diverse vasodilators. Role of decreased cerebral GABA. Cardiovasc Res. 1998;37:748-755.
23. Raimondi L, Pirisino R, Ignesti G, et al. Semicarbazide-sensitive amine oxidase activity (SSAO) of rat epididymal white adipose tissue. Biochem Pharmacol. 1991;41:467-470.
24. Raimondi L, Pirisino R, Banchelli G, et al. Further studies on semicarbazide-sensitive amine oxidase activities (SSAO) of white adipose tissue. Comp Biochem Physiol B. 1992;102:953-960.
25. Jenney EH, Pfeiffer CC. The convulsant effect of hydrazides and the antidotal effect of anticonvulsants and metabolites. J Pharmacol Exp Ther. 1958;122:110-123.
26. Pfeiffer CC, Jenney EH, Marshall WH. Experimental seizures in man and animals with acute pyridoxine deficiency produced by hydrazides. Electroencephalogr Clin Neurophysiol. 1956;8:307-315.
27. Holmes E, Caddick S, Lindon JC, et al. 1H and 2H NMR spectroscopic studies on the metabolism and biochemical effects of 2-bromoethanamine in the rat. Biochem Pharmacol. 1995;49:1349-1359.
28. Lyles GA, McDougall SA. The enhanced daily excretion of urinary methylamine in rats treated with semicarbazide or hydralazine may be related to the inhibition of semicarbazide-sensitive amine oxidase activities. J Pharm Pharmacol. 1989;41:97-100.
29. Awasthi S, Boor PJ. Semicarbazide protection from in vivo oxidant injury of vascular tissue by allylamine. Toxicol Lett. 1993;66:157-163.
30. Boomsma F, van Dijk J, Bhaggoe UM, et al. Variation in semicarbazide-sensitive amine oxidase activity in plasma and tissues of mammals. Comp Biochem Physiol C. 2000;126:69-78.
31. Lizcano JM, Fernández de Arriba A, Tipton KF, et al. Inhibition of bovine lung semicarbazide-sensitive amine oxidase (SSAO) by some hydrazine derivatives. Biochem Pharmacol. 1996;52:187-195.
32. Boomsma F, Bhaggoe UM, van der Howen AMB, et al. Plasma semicarnazide-sensitive amine oxidase in human (patho)physiology. Biochim Biophys Acta. 2003;164:48-54.
33. Assali NS, Kaplan S, Oighenstein S, et al. Hemodynamic effects of 1-hydrazinophthalazine (Apresoline) in human pregnancy: results of intravenous administration. J Clin Invest. 1953;32:922-930.
© 2005 Lippincott Williams & Wilkins, Inc.