Effect of Hydralazine on Aortic RuptureInduced by B-Aminopropionitrile in Turkeys

CHARLES F. SIMPSON, D.V.M., PH.D., AND W. JAPE TAYLOR, M.D.

SUMMARY The effects of hydralazine on aortic rupture, hemodynamics and aortic ultrastructure werestudied in turkeys fed B-aminopropionitrile (BAPN). A mortality rate of 24% due to hemopericardium and in-ternal hemorrhage in turkeys fed only BAPN increased to 91% when turkeys were fed both BAPN andhydralazine, despite a significant reduction in blood pressure after both drugs. Death rates among turkeys fedBAPN and hydralazine were lowered by adding either dietary propranolol (53%), which lowered blood pres-sure and dP/dt max, or reserpine (67%), which reduced blood pressure and increased dP/dt max. Striking ul-trastructural alterations of collagenous and elastic fibers of the aortic media, which were additive to the effectsof BAPN alone, were induced by BAPN and hydralazine.

This study demonstrates that a 6-week feeding of high levels of BAPN and hydralazine, which accumulatesin vessel walls, can produce vascular injury and increase mortality from hemorrhage in lathyritic turkeys.

THE LATHYRITIC TURKEY is used as a modelfor evaluating pharmacologic interventions in thetherapy of life-threatening aortic dissections in hu-mans. In both the human disease and the B-amino-propionitrile (BAPN)-induced disease of turkeys,death from aortic rupture into either the pericardialsac or an open space is common. Studies indicate thatreduction of blood pressure or, perhaps more impor-tantly, the rate of pressure increase (dP/dt max)retards or stops the progression of aortic dissection inin vitro models, the turkey and man.'3The interactions between the tensile strength of the

aortic wall and hemodynamic factors are not clearlydelineated in aortic dissections. This study was de-signed to evaluate the effect of hydralazine, which in-terferes with collagen synthesis, accumulates in thearterial wall,4 and alters the elasticity of collagen,5' inthe lathyritic turkey, singly and in combination withother hypotensive agents.

Materials and MethodsOne-day-old male, broad-breasted white turkeys

were obtained from a commercial hatchery. Theywere housed in a specially ventilated turkey house inwhich they were segregated into groups of six andshielded from undue noise or other stress. Control andexperimental groups were in contiguous pens and wereinspected daily to determine mortality and generalhealth.

There were 150 turkeys each in the control groupand in two of the four experimental groups; 90 turkeyswere included in the other 2 experimental groups(table 1). All turkeys were fed turkey mash until theage of 4 weeks, at which time 0.07% BAPN was addedto the experimental diets and continued for theremaining 6 weeks of the study. The turkeys thatreceived BAPN are designated as lathyritic groups.One lathyritic group served as a control; 0.07%

From the College of Veterinary Medicine and the College ofMedicine, University of Florida, Gainesville, Florida.Address for correspondence: C.F. Simpson, D.V.M., Box J-125,

J. Hillis Miller Health Center, College of Veterinary Medicine,University of Florida, Gainesville, Florida 32610.

Received April 8, 1981; revision accepted July 1, 1981.Circulation 65, No. 4, 1982.

hydralazine was fed to three groups either alone or incombination with 0.04% propranolol or 0.0005%reserpine. The average daily consumption of feed was0.14 kg/day, which calculates to daily doses of 95 mgof BAPN, 95 mg of hydralazine, 54 mg of propran-olol, and 0.7 mg of reserpine in the respective groups.

After 2 weeks of the drug regimen, 20% of eachtreatment group was selected at random for hemo-dynamic measurements. A carotid artery was can-nulated to measure heart rate, blood pressure andmaximal rate of arterial pressure increase (dP/dtmax) directly. Heart rate was calculated from thearterial pulse recording, blood pressure was measuredwith a linear-core transducer and dP/dt max with adifferentiator coupler (Narco Biosystems). The ca-rotid artery was ligated after these procedures, and nosubsequent ill effects were noted.Ten percent of the turkeys in each treatment group

were sacrificed for histologic and electron micro-scopic studies of the aorta after 2 weeks of drug ad-ministration. The remaining turkeys continued toreceive the experimental drug regimen for 4 moreweeks. At the age of 10 weeks, 6 weeks after the initia-tion of drug intervention, the remaining turkeys werenecropsied. Mortality was recorded daily, and thecause of death was determined by necropsy. Sectionsof the abdominal aortas of turkeys that died or weresacrificed were prepared for examination by lightmicroscopy. The specimens were fixed in 10% neutralformalin, embedded in paraffin, cut at 5 ,um, andstained with hematoxylin-eosin and also orcein VanGieson stain for elastic fibers. Sections of the aortasfor electron microscopy were fixed in 3% bufferedglutaraldehyde, postfixed in 1%0Y04, and embeddedin Araldite. Thin sections on grids were stained withuranyl acetate and lead citrate before examination byelectron microscopy.

Data were tested by analysis of variance to deter-mine if there was a significant difference between themeasurements of each variable tested, followed byDuncan's multiple-range test to determine whichspecific variables were significantly different.

ResultsMortality was strikingly increased in turkeys fed

hydralazine (table 1) compared with mortality in704

by guest on July 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

HYDRALAZINE AND AORTIC RUPTURE IN TURKEYS/Simpson and Taylor

TABLE 1. Mortality Among Turkeys Fed B-aminopropionitrile and Hydralazine

Mortality

No. of Average day Hemoperi- AorticMedication turkeys* of death cardium rupture Total

Control 150 - 0 0 0

BAPN 150 24 5 (4%) 27 (20%) 32 (24%)BAPN, hydralazine 150 20 48 (36%) 75 (55%) 123 (91%)BAPN, hydralazine, reserpine 90 24 8 (10%) 46 (57%) 54 (67%)

BAPN, hydralazine, propranolol 90 29 3 (4%) 40 (49%) 43 (53%)*Ten percent of the turkeys from each group were sacrificed at 2 weeks of age; the remainder were used to

calculate mortality.Abbreviation: BAPN = B-aminopropionitrile.

TABLE 2. Hemodynamics AmongTurkeys Fed B-aminopropionitrile and Hydralazine

Heart rate Blood pressure (mm Hg) dP/ dt maxMedication (beats/ min) Systolic Diastolic (mm Hg/ sec)

Control 326a 171a 128a 1758a

BAPN 325a 175a 130a 2084a

BAPN, hydralazine 327a 154 Illb 1945a

BAPN, hydralazine, reserpine 286b 137C 82C 2633b

BAPN, hydralazine, propranolol 256b 133c 98C 1277C

Values with different superscripts are significantly different from each other (p< 0.05).

turkeys fed BAPN alone. The 24% death rate fromaortic hemorrhage in turkeys fed BAPN alone wascomparable to that in other studies.7 The addition ofhydralazine produced a 91% mortality rate, which wasonly moderately reduced by reserpine (67%) orpropranolol (53%). This lethality of the BAPN andhydralazine regimen was caused in part by the early

occurrence of hemopericardium, which was signifi-cantly reduced by the concomitant administration ofreserpine or propranolol. Thirty-six percent of deathsamong turkeys fed BAPN and hydralazine resultedfrom hemopericardium, but 4% of the turkeys fedBAPN alone or BAPN, hydralazine and propranololdied of hemopericardium. Ten percent of the turkeysfed BAPN, hydralazine and reserpine died from he-mopericardium. Death from internal hemorrhage wasnot radically influenced by propranolol or reserpine inthe presence of hydralazine and BAPN.

Hydralazine significantly lowered both systolic anddiastolic blood pressures; addition of propranolol orreserpine caused further reductions (table 2). Reser-pine had a more pronounced effect on diastolic pres-sure than did propranolol; however, propranolollowered and reserpine increased dP/dt max. Reser-pine and propranolol, but not hydralazine, loweredheart rate.The ultrastructure of the abdominal aorta of the

turkeys was profoundly altered by the drugs. In theBAPN-fed turkeys, the changes were similar to thosedescribed previously.7 Smooth muscle cells were oftencompressed and adjacent ones were usually separatedby disorganized bundles of collagen fibers, pools ofmucopolysaccharide, and fragmented and swollenelastic fibers with electron-dense surface elevations(fig. 1). These elevations consisted of disorganized,nonfragmented fibrils that overlay central portions ofnormal-appearing elastic fibers of various widths.

FIGURE 1. Aorta of a turkey fed B-aminopropionitrileshowing electron-dense elevations (1-4) on the surface ofelastic fibers. The interior of a fiber (arrow) has weak elec-tron density and is apparently normal. Magnification X5000.

705

by guest on July 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

VOL 65, No 4, APRIL 1982

FIGURE 2. (A) Aorta of turkeyfed B-aminopropionitrile and hydralazine. The elasticfibers (E) are swollen andfragmented.A small interior portion ofa fiber (arrow) is weakly electron dense and apparently unaltered. Magnification X 4000. (B) Promi-nent swellings (arrows) on elastic fibers are granular because offragmentedfibrils. The interior of thefiber (E) is not electrondense, is apparently normal and contains an orderly array offibrils. Magnification X 25,000.

Hydralazine and BAPN produced more severealterations in the aortic media than did BAPN alone.Occasionally, a narrow central portion of an elasticfiber had a normal appearance, but in general, almostthe entirety of the elastic fiber was swollen and frag-mented and had irregular, opaque elevations on itssurfaces (fig. 2). Fragmentation and disorganizationof fibrils caused the larger elevations to have agranular appearance (fig. 2). Collagenous fibers in theintercellular spaces were swollen. The cytoplasm ofmost smooth muscle cells appeared to contain moreglycogen than corresponding cells in the aortic mediaof turkeys that received BAPN alone.When reserpine was added to the BAPN and

hydralazine regimen, the elevations on the elasticfibers of the aortic media were not as extensive as withBAPN and hydralazine alone. The central portion ofthese fibers was normal in appearance, but they hadknobby and electron-dense peripheral elevations (fig.3). The elevations were less granular than those of theturkeys fed BAPN and hydralazine. Bundles of col-

FIGURE 3. Aorta of turkey fed B-aminopropionitrile,hydralazine and reserpine. Much ofthe interior ofthe elasticfiber (E) has a normal appearance. Elevations on a fiber areelectron dense and some have a granular appearance(arrow). Bundles of collagenous fibers (C) are dense andmatted. Magnification X 5000.

706 CIRCULATION

by guest on July 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

HYDRALAZINE AND AORTIC RUPTURE IN TURKEYS/Simpson and Taylor

FIGURE 4. Aorta of turkey fed B-aminopropionitrile,hydralazine and propranolol. Elevations on elastic fibers (E)are bridged by collagenous-like fibers (arrows). Collagenousfibers (C) in the intercellular space are swollen. Magnifica-tion X 20,000.

lagenous fibers in the intercellular spaces were denseand matted, and extremely prominent throughout thesections.Some elastic fibers in the aortic media of turkeys

fed BAPN, hydralazine and propranolol were swollenand had nongranular elevations on their surfaces. Thefibrils in such elevations were disorganized but, unlikeelevations on elastic fibers in turkeys fed hydralazine,the fibrils were not fragmented. Unaltered elasticlamellae without elevations also were present in theaortic media. Only in the aortas of turkeys fed BAPN,hydralazine and propranolol was there bridging ofelevations on elastic fibers with collagenous fibers (fig.4). In addition, collagenous fibers in the intercellularspaces were swollen, but bundles of collagen were notdense or matted.

DiscussionHydralazine dramatically increased the mortality

rate from hemorrhage in turkeys also fed BAPN,compared with mortality in turkeys fed only BAPN.Either reserpine or propranolol partially protectedagainst death in this situation, primarily by decreas-ing the incidence of hemopericardium, as the drugs didnot lower the incidence of hemorrhage from rupture ofthe abdominal aorta. Evaluation of potential mecha-nisms of action of hydralazine, reserpine and propran-

olol in the turkeys fed BAPN requires a considerationof both hemodynamic factors and direct chemical ac-tion on the aortic wall.

Previous studies showed that lowering arterialblood pressure and dP/dt max with propranololreduces BAPN-induced mortality in turkeys;8 thisprinciple is now widely used in the treatment of aorticdissections in man.' In the present experiment, thereduction in deaths from hemopericardium, and ac-cordingly in overall mortality, with propranolol andreserpine therapy was probably due to hemodynamicfactors, because the aortic root is particularly vulner-able owing to the slight torsion produced by cardiaccontraction. Heart rate and blood pressure werelowered by both reserpine and propranolol. Propran-olol, but not reserpine, also reduced dP/dt max sig-nificantly, which probably explains its greater effec-tiveness in reducing hemopericardium and overallmortality. The higher mortality rate with hydralazineand BAPN cannot be explained by hemodynamic fac-tors alone, because hydralazine lowered blood pres-sure without altering either dP/dt max or heart rate.Accordingly, hydralazine had an injurious effect onthe aortic wall, which was shown by ultrastructuralstudies to be additive to that of BAPN.There are several mechanisms by which hydrala-

zine can cause increased degeneration of the aorticwall, as compared with BAPN alone. Hydralazine ac-cumulates in the media of the aorta and other bloodvessels in mice.4 By radioautography, radiocarbonfrom hydralazine-l-C,4 has been shown to be con-centrated in the wall of muscular arteries of thekidney, liver, spleen, heart, lung, brain and muscle.4Hydralazine reacts with the aldehyde groups ofcollagen by means of hydralazine groups.9 This mightbe the mechanism by which the elasticity module ofcollagen fibers from the tail tendon of the rat isreduced after incubation in hydralazine.6 Hydralazinealso inhibits both the synthesis and secretion ofcollagen in embryonic chicken cartilage.5

Hydralazine accumulates in the aorta, but ultra-structural studies of this phenomenon have not beenreported. Using electron microscopy, we foundprofound alterations of aortic collagenous and elasticfibers in all the hydralazine-treated groups that werein addition to the effects of BAPN. There were ac-cumulations of dense bundles of extracellular collagenin the aortic media of reserpine-treated turkeys and, toa much lesser extent, in those that received proprano-lol. These accumulations might be attempts at repair.Destruction of elastic fiber was most severe in the aor-tas of turkeys fed hydralazine and BAPN. In ver-tebrate tissues, elastic fibers consist of two morpho-logically distinct components, amorphous elastinand fibrils, with the former being derived in matura-tion from the latter.10 Controlled synthesis and bal-anced interactions between the two components areessential for normal fibrilogenesis. BAPN interfereswith the cross-linking of elastin and with covalent in-corporation of synthesized elastin into fibers." There-fore, the granular elevations on elastic fibers in theseexperiments appear to represent an interference with

707

by guest on July 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

VOL 65, No 4, APRIL 1982

formation of fibrils in the newly formed peripherallayer (elevations) of elastin on an elastic fiber.Destruction of elastic fiber and granularity of theelevations were less marked in turkeys fed propran-olol or reserpine, possibly because of amelioratinghemodynamic stress.The clinical implications of this study are-not clear.

Although the turkeys received quantities of hydral-azine that exceeded the usual human dose, it did notproduce an overwhelming reduction of blood pressureor the extreme toxicity reported in dogs.'2 In combina-tion with other data on the influence of hydralazine oncollagen, our study provides evidence of potentiallyserious adverse effects in addition to the hydralazine-induced lupus syndrome. These undesirable effectsshould be considered, as hydralazine is being morewidely advocated as an afterload-reducing agent inrefractory heart failure and for long-term manage-ment of hypertension.

AcknowledgmentThe technical assistance of J.W. Carlisle and Beth Roche is

acknowledged. Our gratitude to the Florida Agricultural Experi-ment Stations Journal Series No. 3023 for funding reprints.

References1 Wheat MW, Palmer RF: Drug therapy for dissecting

ane-rysins. Prog Cardiovas Res 54: 372, 1968

2. Prokop EK, Palmer RF, Wheat MW: Hydrodynamic forces indissecting aneurysms. In vitro studies in a Tygon model and indog aortas. Circ Res 27: 121, 1970

3. Wheat MW, Palmer RF, Bartley TD, Seelman RC: Treatmentof dissecting aneurysms of the aorta without surgery. J ThoracCardiovasc Surg 50: 364, 1965

4. Moore-Jones D, Perry HM: Radioautographic localization ofhydralazine-1-C,4 in arterial walls. Proc Soc Expl Biol Med122: 576, 1966

5. Rapana RS, Parr RW, Liu TZ, Bhatnagar RS: Biochemicalbasis of skeletal defects induced by hydralazine: inhibition ofcollagen synthesis and secretion in embryonic chicken cartilagein vitro. Teratology 15: 185, 1977

6. Kraemer HP, Nemetschek T, Gross F: Effects of hydralazineon the elasticity of collagen. Experimentia 35: 527, 1979

7. Simpson CF, Pritchard WR, Harms RH, Sautter JH: Electronmicroscopy of the cardiovascular system of the normal and B-aminopropionitrile-fed turkey. Exp Mol Pathol 1: 321, 1962

8. Simpson CF, Boucek RJ, Noble NL: Influence of d-, 1-, and dl-propranolol and practolol on B-aminopropionitrile-inducedaortic ruptures of turkeys. Toxicol Appl Pharmacol 38: 169,1976

9. Paz MA, Seifter S Immunological studies of collagensmodified by reaction with hydralazine. Am J Med Sci 263: 281,1972

10. Karrer HE, Cox J: Electron microscope study of developingchick embryo aorta. J Ultrastruc Res 4: 420, 1960

11. Siegel RC, Pinnell SR, Martin GR: Cross-linking of collagenand elastin. Properties of lysyl-oxidase. Biochemistry 9: 4486,1970

12. Comens P: Experimental hydralazine disease and its similarityto disseminated lupus erythematosus. J Lab Clin Med 47: 444,1956

708 CIRCULATION

by guest on July 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

C F Simpson and W J TaylorEffect of hydralazine on aortic rupture induced by B-aminopropionitrile in turkeys.

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1982 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.65.4.704

1982;65:704-708Circulation. 

http://circ.ahajournals.org/content/65/4/704the World Wide Web at:

The online version of this article, along with updated information and services, is located on

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer information about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. FurtherEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculationpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:

by guest on July 13, 2018http://circ.ahajournals.org/

Dow

nloaded from