A characterization of the acute cardiopulmonary toxicity of fenfluramine in the rat

Pharmacological Research, Vol. 22, No . 3, 1990

37 1

A CHARACTERIZATION OF THE ACUTE CARDIOPULMONARY TOXICITY OF FENFLURAMINE IN THE RAT RONALD N. HUNSINGER and DAVID WRIGHT Department of Biology and the School of Pharmacy, Samford University, Birmingham, AL 35229, USA Received in final form 20 October 1989

SUMMARY

Fenfluramine (FN) is a potent serotonin-releasing drug used primarily as an anorectic agent . The symptomatology of its acute lethality has been well documented in animal models such as the rat . A very prominent feature of this lethality profile is hypoxia, as demonstrated by the onset of severe cyanosis just prior to death . It is not clear in the literature whether this hypoxia is the result of a direct pulmonary effect or is secondary to cardiac injury . To further characterize this aspect of FN's toxicity, respiratory and electrocardiographic measurements were taken in anaesthetized rats subjected to high doses of FN (129 . 6 mg/kg, i.p.) . Death occurred in these animals within 15 min of drug administration, apparently as the result of abrupt respiratory cessation, followed by cardiac ischaemia . No significant gross or histopathological lesions were evident in these animals . In other trials, prior treatment with diethylcarbamazine (DEC) was found to potentiate the lethality of FN, while cyproheptadine (CHP) pretreatment attenuated FN's toxic effects . Necropsies, conducted 24 h after FN administration, revealed widespread alveolar and pulmonary interstitial haemorrhage in the CHP-pretreated animals . The data suggest that high doses of FN directly result in pulmonary hypertension, which secondarily induces ischaemic cardiac injury . KEY WORDS : fenfluramine, cardiopulmonary, rat, cyproheptadine, diethylcarbamazine .

INTRODUCTION

Fenfluramine (FN) is a fluorinated beta-phenethylamine which is best known for its anorectic properties [1, 2] . The literature is replete with studies describing serotonin release as FN's major mechanism of action (see review by Garattini [3]) . In recent years, it has been suggested that FN may have therapeutic value in addition to its anorectic use, in that it appears to improve many aspects of the autistic syndrome in children [4, 5] . Considerable attention has been given to the potential toxicity of the drug. For example, Harvey & McMasters [6] reported that a single injection of FN, at 1043-6618/90/030371-08/S03 .00/0

© 1990 The Italian Pharmacological Society



3 72

Pharmacological Research, Vol. 22, No . 3, 1990

dosages ranging from 12 . 5 to 100 umol/kg, resulted in detectable signs of neurotoxicity in the major serotonergic tracts of the brain stem . Additionally, general toxicity studies in laboratory animals have profiled symptoms resulting from the acute administration of high dosages of FN [7] . These include tremors, tonic-clonic convulsions, limb rigidity, opisthotonos, mydriasis, lacrimation, chromodacryorrhoea, and respiratory distress . It is also common to see what is called the 'serotonin syndrome' in rats given high doses of FN [8] . This behavioural paradigm consists of forepaw padding, side-to-side head weaving or head tremor, hindlimb abduction, straub tail, salivation, and hyperpyrexia [9, 10] . The issue of respiratory toxicity has special significance to the clinical use of FN in humans, since several reports suggest an association between the drug and pulmonary hypertension [11, 121 . Also, Hunsinger et al. [13] observed severe respiratory distress and increased heart-to-body weight ratios in rats given high doses of FN . Despite the central role that respiratory distress may play in FN's toxic action, very little is known concerning the mechanism by which it occurs . Is it secondary to some cardiac effect, or does it involve a more direct action of the drug on the respiratory system? The purpose of this study was to further characterize the acute toxicity of FN in terms of basic cardiopulmonary parameters . Additionally, the potential antidotal effectiveness of diethylcarbamazine (DEC) and cyproheptadine (CHP) in acute epsiodes of FN toxicity was assessed . The rationale for the choice of CHP as a putative antidote was based upon its ability to block 5-HT receptors (see Discussion) and its attenuation of several FN effects, such as the blockade of foot shock-induced analgesia [14], release of ACTH and cortisol [15], and anorexia (see review by Garattini [3]) . DEC was chosen because of reports indicating that it can reduce hypoxia-induced pulmonary hypertension (see Discussion) .

METHODS AND MATERIALS Male Sprague-Dawley rats, weighing 150-200 g, were obtained from Harlan Industries, Indianapolis, IN, USA, and used in this study . In the first phase of this study, rats were anaesthetized with Nembutal' (50 mg/kg, i .p .). Electrodes were attached to the animals and connected to an impedance pneumograph coupler for respiratory rate measurements . Electrocardiograph tracings (lead II) were recorded simultaneously by bridging the impedance coupler to an ECG transducer . The data were recorded on a Narco-Bio System Mark IV physiograph . Fifteen minutes following loss of righting reflexes, the rats were injected i .p . with either 0 . 5 ml of saline or 129 . 6 mg/kg of fenfluramine hydrochloride . Based upon prior work [13], this high dosage ensures a rapid and uniform response from all treated animals . The symptoms seen at this dosage are identical, in every other respect, to those elicited by lower doses within FN's acute lethality range . High speed physiograph tracings were monitored at various intervals during a 15 min period, which immediately followed drug or saline injections . At the conclusion of the cardiopulmonary monitoring periods, the heart and lungs were quickly removed and weighed . The lungs were then either placed in 10% buffered formalin for subsequent histological examination or dried for 24 h by





Pharmacological Research, Vol. 22, No. 3, 1990

37 3

continual inflation in order to obtain a dry lung weight . Inflation was accomplished by passing a gentle stream of air down the trachea using a Pasteur pipette connected to an air hose . The hearts from all experimental animals were preserved in formalin for later histological staining. In phase II of our studies, other groups of rats (n= 8 per group) were given i .p . injections of either 0 . 5 ml saline, 500 mg/kg diethylcarbamazine hydrochloride (Sigma Chemical Co., St Louis, MO, USA), or 5 mg/kg cyproheptadine hydrochloride (Sigma Chemical Co., St Louis, MO, USA) 30 min prior to injection of various dosages of fenfluramine hydrochloride (62 . 5, 75, 90, 108, or 129 . 6 mg/ kg, i.p .) . Acute symptoms were recorded during a 1 h post-fenfluramine dosing period, and the number of deaths in each group were noted for a 24 h time period . Necropsies were performed on animals surviving at the end of the 24 h period .

RESULTS Phase I

A rapid decline in respiratory rate occurred within 1 min following the injection of 129 . 6 mg/kg of FN (Fig . 1). Thereafter, respiratory rate continued to decrease until complete respiratory arrest occurred at approximately 9 min post-dosing . The respiratory rate of control rats did not change significantly throughout the course of the measurements . Figure 1 also depicts a time-related decrease in heart rate associated with acute episodes of FN lethality. This decline in heart rate began to occur 5 min postdosing, as compared to the more immediate effect of FN on respiratory rate .

80

400 350 E

300

ô

250

o

200

â)

60

o ô

n 40

â N 0

1

T

150

FN

100 50~

a

N C

- 20

ô2

FN 1

I

01

I, i I, * , I I r 1 i i I i i •i i I i i 1 0 3 6 9 12 15 11 0 1 3 6 9 12 15 Post-injection time (min)

Post-injection time (min)

Fig. 1 . Heart rate and respiratory rate in animals given either 0 . 5 ml saline (controls) or 129 . 6 mg/kg fenfluramine (FN). An asterisk represents a significant difference from the corresponding control value (PS 0 . 05 ), as determined by the Student's t-test .



374

Pharmacological Research, Vol. 22, No. 3, 1990 (a)

Normal sinus ryhthm

(b)

Negative S-T segment

Increased T-wave

(c)

Junctional rhythm (d)

A-,

1 V%__ ,

2 :1 AV block

Fig . 2 . ECG tracings showing the effects of fenfluramine (129 . 6 mg/kg) upon electrical activity in the rat heart . (a) A normal sinus rhythm is seen before drug injection . (b-d) The various arrhythmias elicited after fenfluramine injection . Besides the notable decrease in heart rate, other cardiac abnormalities can be seen in the selected ECG tracings in Fig . 2. At approximately 6-7 min following FN administration, PR, QRS, and T-wave intervals increased in length . Additionally, greater negative deflections of the S -T segments and increased T -wave amplitudes were noted (Fig . 2b). Thereafter, most of the rats continued into a junctional rhythm (Fig. 2c) until asystole occurred . In one rat, a second-degree heart block (2 :1 AV block) was observed (Fig . 2d). The control rats showed no significant ECG changes throughout the course of the observations . No changes in organ-to-body weight ratios were noted for the heart or lungs of animals succumbing to the acutely lethal dose (129 . 6 mg/kg, i.p .) of FN (Table I). Histological examination of these tissues revealed no observable pathological lesions . Phase II

Table II shows the acute lethality profile for FN in rats pretreated with either saline, DEC, or CHP 30 min before injection of FN . In animals given only saline immediately prior to FN injection, all behavioural aspects of the serotonin syndrome were prominent, as were tonic-clonic convulsions . Acute respiratory failure, as evidenced by severe cyanosis, immediately preceded death . All deaths occurred within 15 min post-FN injection . DEC failed to block the lethal actions of FN and, in fact, actually potentiated FN's lethality (Table II) . It was also noted that DEC mimicked in some animals the respiratory distress syndrome seen with the administration of FN alone . CHIP was effective in decreasing the acute lethal actions of FN (Table II) . However, FN still



Pharmacological Research, Vol. 22, No. 3, 1990

375

Table I Selected weight indices of interest to the acute cardiopulmonary effects of fenfluramine

Body weight

Controls

FenfJuramine-treated

222 . 83 g SE =12-25 g

231 . 00 g sE = 4 . 60 g

1 . 413 g

Lung weight (wet)

sE=0 . 046 g

1 . 390 g sE=0 . 050 g

Lung weight (dry)

0-332g sE=0 . 017 g

0 . 371 g sE=0 . 044 g

Heart weight

0 . 712 g sE=0 . 033 g

0 . 809 g sE=0 .036 g

Heart/body weight ratio

3 . 21 g/kg sE = 0 . 08 g/kg

3 . 50 g/kg SE = 0 . 19 g/kg

Lung/body weight ratio

5 . 92 g/kg sE = 0 . 17 g/kg

SE

6 . 01 g/kg = 0 . 18 g/kg

The values represent the mean and SE of 5-8 rats per group . Table II Lethality profile for fenfluramine in male Sprague-Dawley rats pretreated with either saline (0 . 5 ml), 500 mg/kg DEC (diethylcarbamazine), or 5 mg/kg CHP (cyproheptadine) Fenfluramine dose (mg/kg)

62-5 75-0 90 . 0 108-0 129 . 6

No . deaths in saline pretreated group

No . deaths in DEC pretreated group

No. deaths in CHP pretreated group

4/8 7/8 8/8 8/8 8/8

7/8 8/8 8/8 8/8 8/8

0/8 0/8 3/8 6/8 8/8

Saline, DEC, or CHP were injected 30 min prior to fenfluramine dosing . elicited the serotonin syndrome actions and induced tonic-clonic convulsions in the CHP pretreated animals . Two animals pretreated with CHP survived the acute effects of FN at dosages of 108 mg/kg, but were severely weak and near death within 24 h . Figure 3 shows that in these animals, there were signs of haemorrhage and oedema in the alveoli . Apparently CHP retarded the acute effects of FN, allowing time for histopathological alterations to become manifest .



376

Pharmacological Research, Vol. 22, No. 3, 1990

Fig. 3 . Photomicrograph of a lung segment showing FN-induced areas of alveolar haemorrhage (arrows) . This animal was a CHP-pretreated subject that survived the acute effects of 108 . 6 mg/kg FN injection . (H & E x375) .

DISCUSSION Our data suggest that the mechanism by which FN exerts its acute lethality involves a serotonin-induced abrupt cessation of respiration, which secondarily predisposes the heart to arrhythmias . The time frame of the respiratory versus cardiac events certainly favours such an hypothesis . Major respiratory distress, as evidenced by the presence of severe cyanosis and significant decreases in respiratory rate, occurred within 1 min following FN injection . Yet, heart rate and ECG patterns remained normal until 5 min post-FN dosing, making a primary heart effect seem unlikely. Also, since no reflexive increase in heart rate was noted during the monitoring period, it is unlikely that FN exerted a negative inotropic effect . The exact mode of action for the respiratory depression is not known . It is possible that a central mechanism is involved, since prior studies have demonstrated that the levels of several brain stem neurotransmitters, including serotonin, decrease following FN administration [16] . Several features of the present study, however, suggest that the action of FN on the lungs might be more direct . First, in the presence of CHP, a serotonergic blocker with strong peripheral actions [17], the acute respiratory depressant effects of FN were attenuated, and the drug's lethality reduced . CHP failed, however, to antagonize FN's other serotonin-dependent symptoms (i.e . the serotonin syndrome and tonic-clonic convulsions) which are thought to be of central origin [18] . This suggests a selective peripheral blockade of FN's effects by the dosages of CHP used in this study. Second, DEC mimicked and potentiated the acute toxic effects of FN on the respiratory system. At first these results were somewhat surprising, especially



Pharmacological Research, Vol. 22, No. 3, 1990

377

since Morganroth et al. [19] indicated that DEC was effective in blocking hypoxiainduced pulmonary hypertension in rats . However, Cesbron et al. [20] and Kowalski et al. [21] indicate that DEC actually induces the release of serotonin from platelets. Thus, although Morganroth and coworkers (vida supra) suggested that DEC blocked pulmonary hypertension via leukotriene synthesis inhibition, it is entirely possible that their results could be explained on the basis of a lowering of platelet serotonin by DEC, which was given prior to exposing the rats to hypoxic conditions . In our study, the dosage of DEC was considerably higher than that used by Morganroth and thus might have elicited pulmonary hypertension by releasing larger amounts of serotonin in a more rapid manner. Nevertheless, because of the similarity in the symptoms of the two drug reactions and because increased peripheral levels of serotonin can constrict the pulmonary vasculature [22], perhaps DEC and FN share a common mechanism of action on the lung . Finally, most suggestive of a peripheral mode of action for FN's respiratory effects are the changes which developed in the lungs of rats in which CHP delayed the toxic actions of high doses of FN . In these animals, alveolar oedema and haemorrhage strongly indicate high pressure development within the pulmonary vasculature. In summary, these results suggest that the primary pathophysiological lesion occurring in episodes of acute FN lethality involves the rapid development of pulmonary hypertension . The serotonergic blocker, CHP, effectively protected animals subjected to various doses of FN, up to 108 mg/kg, i .p . The histological profile of FN's toxicity closely resembles the acute respiratory distress seen in humans following traumatic or septic shock and high altitude sickness [23] . This raises the possibility that acute FN toxicity in rats may have application as an animal model for such disease states . Finally, the results of our study suggest that CHP might be of antidotal benefit in cases of FN overdose and/or disease states involving pulmonary hypertension elicited by serotonin release .

ACKNOWLEDGEMENTS This research was made possible by Samford University Research Grants 17 and 18 . Our gratitude is expressed to Dr W. Mike Howell for his helpful advice in preparing this manuscript. REFERENCES 1 . Le DoVarec JC, Neven C. Pharmacology and biochemistry of fenfluramine . In : Costa E, Garattini S, eds. International symposium on amphetamines and related compounds . New York: Raven Press, 1970 . 2. Rowland NE, Carlton J . Neurobiology of an anorectic drug : fenfluramine . Prog Neurobiol 1986 ; 27 : 13-62. 3 . Garattini S . Recent studies on anorectic agents . Trends Pharmaceut Sci 1980 ; 1 : 354-6 . 4. Ritvo ER, Freeman BJ, Geller E, Yurviler A . Effects of fenfluramine on 14 outpatients with the syndrome of autism . JAm Acad Child Psychiat 1983 ; 22: 549-58 . 5. Campbell M, Deutsch SI, Perry R, Wolsky BB, Palij M . Short-term efficacy and safety of fenfluramine in hospitalized preschool-age autistic children : an open study. Psychopharmacol Bull 1986 ; 1 : 141-7 .



378

PharmacologicalResearch, Vol. 22, No. 3, 1990

6 . Harvey JA, McMasters SE . Fenfluramine : evidence for a neurotoxic action on midbrain and a long-term depletion of serotonin . Pharmacol Commun 1975 ; 1 : 217-28 . 7 . Gilbert DL, Franko BV, Ward JM, Woodard G, Courtney KD. Toxicologic studies of fenfluramine . Toxicol 4ppl Pharmacol 1971 ; 19 :705-11 . 8 . Trulson ME, Jacobs BL . Behavioral evidence for the rapid release of CNS serotonin by PCA and fenfluramine. EurJPharmacol 1976 ; 36 :149-54 . 9 . Hess SM, Doepfner W. Behavioral effects and brain amines content in rats . Arch Int Pharmacodyn Ther 1961 ; 134 : 89-99 . 10 . Grahme-Smith DG . Studies in vivo on the relationship between brain tryptophan, brain 5-HT synthesis and hyperactivity in rats treated with a monoamine oxidase inhibitor and 1-tryptophan . JNeurochem 1971 ; 18 :1053-66 . 11 . Engelhardt A, Kroneberg G, Stoepel K, Statzer H . The effects of acute and chronic administration of sympathomimetic substances on the systemic and pulmonary circulation . Proc Eur Soc Study Drug Toxicol 1970 ; 11 : 110-17 . 12 . Douglas JG, Munro JF, Kitchin AH, Muir AL, Proudfoot AT. Pulmonary hypertension and fenfluramine. Br Med J 1981 ; 283 : 881-3 . 13 . Hunsinger RN, Kibbe AH, Wilson MC . The effects of previous d-amphetamine treatment on the disposition and lethality of fenfluramine in the rat . Toxicol Appl Pharmacol1985 ; 79 : 236-45 . 14 . Hutson PH, Tricklebank MD, Curzon G. Analgesia induced by brief footshock: blockade by fenfluramine and 5-methoxy-N,N-dimethyltryptamine and prevention by blockade by 5-HT antagonists . Brain Res 1983 ; 279 : 105-10 . 15 . Lewis DA, Sherman BM . Serotonergic stimulation of adrenocorticotropin secretion in man. J Clin Endocrinol Metab 1984 ; 58 :458-62 . 16 . Hunsinger RN, Wilson MC . Alteration of the neurochemical effects of fenfluramine by previous exposure to d-amphetamines . Pharmacol Biochem Behav 1983 ; 22 : 127-35 . 17 . Gilman AC, Goodman LS, Gilman A, eds. The pharmacological basis of therapeutics. 6th ed . New York: Macmillan, 1980 :639. 18 . Deakin JEW, Green AR . The effects of putative 5-hydroxytryptamine antagonists on the behaviour produced by administration of tranylcypromine and l-tryptophan or tranylcypromine and l-dopa to rats . BrJPharmacol1978 ; 64 : 201-9 . 19 . Morganroth ML, Stenmark KR, Morris KG, et al. Diethylcarbamazine inhibits acute and chronic hypoxic pulmonary hypertension in awake rats. Am Rev Resp Dis 1985 ; 131 : 488-92 . 20 . Cesbron JY, Capron A, Vorgaftig RB, et al. Platelets mediate the action of diethylcarbamazine on microfilariae . Nature 1987 ; 324 : 533-6 . 21 . Kowalski KA, McConnel LA, Sadoff DA, Weskeid R . Modulation of equine platelet function by diethylcarbamazine (DEC). Am JPathol 1983 ; 113 : 1-7 . 22 . Hechtman HB . Serotonin and the cardiovascular system. In : Vanhouttee PM, ed . Serotonin and the cardiovascular system . New York: Raven Press, 1985 . 23 . Hagland V. Serotonin and shock . In : Vanhoutee PM, ed . Serotonin and the cardiovascular system . New York : Raven Press, 1985 .