(Hypertension. 1998;32:972-975.)
© 1998 American Heart Association, Inc.
Scientific Contributions |
Inverse Relationship Between Fenfluramine-Induced Prolactin Release and Blood Pressure in Humans
From the Center for Clinical Pharmacology (M.F.M.) and the Department of Psychiatry (J.M.P., K.A.M.), University of Pittsburgh School of Medicine, Penn; and the Departments of Neuroscience (A.F.S.) and Psychology (J.D.F., S.B.M.), University of Pittsburgh, Penn.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract—Although substantial evidence from experimental animals suggests that augmentation and reduction in serotonergic neurotransmission both affect arterial blood pressure (BP), it is unknown whether "tonic" central serotonergic activity is related to resting BP variability in humans. We tested this hypothesis in a community sample by evaluating the relationship between resting BP and a neuropharmacologic index of brain serotonergic activity (the fenfluramine challenge test). Subjects were 270 generally healthy men and women aged 25 to 60 years who were not receiving prescribed antihypertensive or psychotropic medications. The sample included 216 non-Hispanic whites and 47 blacks. Resting systolic BP ranged from 85 to 161 mm Hg and diastolic from 58 to 98 mm Hg. Each subject received 0.55 to 0.65 mg/kg D,L-fenfluramine hydrochloride, and the plasma prolactin concentration was measured over 3.5 hours. Analyses revealed a linear, inverse relationship between the maximum fenfluramine-induced prolactin rise and systolic and diastolic BP in whites: r=-0.36 and r=-0.29, respectively (P<0.001 for both). These relationships were not observed in the black participants. In whites, the prolactin response to fenfluramine remained a significant predictor of systolic and diastolic BPs in multivariate models including age, gender, body mass index, physical activity, smoking, and alcohol consumption (P
0.001). When compared with subjects in the
highest quartile of prolactin response, individuals whose prolactin
responses to fenfluramine comprised the lowest quartile were 2.6 times
more likely to have a resting systolic/diastolic BP
of >135/85 mm Hg. These data reveal that in white but not black
adults, fenfluramine-induced prolactin release correlates inversely
with BP and may indicate a role of central serotonergic activity in the
pathogenesis of hypertension.
Key Words: blood pressure • central nervous system • serotonin • fenfluramine • prolactin
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
As part of a clinical investigation of the neurobehavioral correlates of cardiovascular disease risk, we sought to examine the relationship between brain serotonergic function and resting blood pressure (BP). Effects of central serotonergic activity on BP are suggested by a series of studies in experimental animals. Pharmacological stimulation of some brain serotonin (5-hydroxytryptamine [5-HT]) receptors produces robust, if complex, changes in cardiovascular parameters, including both hypotensive and hypertensive responses.1 Conversely, experimental manipulation of BP has also been found to influence serotonergic neural transmission in certain brain loci. However, it has never been demonstrated whether central serotonergic activity, under resting or basal conditions, is related to individual differences in BP.
In the present study, central serotonergic activity was assessed with a neuroendocrine challenge with D,L-fenfluramine hydrochloride. Fenfluramine enhances serotonin neurotransmission by both inducing presynaptic release of serotonin stores and inhibiting synaptic reuptake.2 3 Activation of serotonergic receptors in the hypothalamus in turn promotes the pituitary release of prolactin into the circulation. Therefore, the rise in plasma prolactin concentration after fenfluramine administration reflects "net" serotonergic responsivity, as influenced by variability in presynaptic events (ie, synthesis, storage, release, and reuptake) and activation of hypothalamic 5-HT receptors. In this article we report an inverse correlation between interindividual variability in prolactin response to fenfluramine and resting BP that is independent of traditional risk factors for hypertension.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Subjects
The 270 adult participants described in this report were recruited from the Pittsburgh area through media advertisements and local distribution of brochures and posters. Exclusion criteria for participation included age >60 years; diastolic BP >99 mm Hg; presence of congestive heart failure, angina, stroke, cancer, or hepatic or renal insufficiency; and use of psychotropic medications. Also excluded from the present analyses were persons receiving antihypertensive medications. The protocol was approved by the University of Pittsburgh Institutional Review Board, and subjects gave informed consent.
BP measurements were obtained at 2 screening appointments, separated by 1 to 3 weeks. Subjects arrived between 8 and 10 AM following an overnight fast. After the subject had rested in the seated position for 20 minutes, a registered nurse obtained a single BP measurement in the right arm using a mercury sphygmomanometer and a regular, large, or extra-large adult cuff, according to the subject's arm circumference. The average of the readings from the 2 screening visits was used as the subject's resting BP. Self-reported measures of alcohol consumption, smoking, and physical activity4 were also obtained.
Fenfluramine Challenge
Participants reported to the laboratory in the morning after a
12-hour fast, and a 20-gauge heparin-locked venous catheter was
inserted. After a 30-minute adaptation period, a heparinized blood
sample was obtained for determination of baseline prolactin
concentration. Subjects then received 30 to 60 mg of
D,L-fenfluramine hydrochloride orally to
achieve a dose in the range of 0.55 to 0.65 mg/kg body wt. Subsequent
blood samples for plasma prolactin were drawn 1, 2, 2.5, 3, and 3.5
hours later. Additional samples were taken at 2.5 and 3.5 hours for
measurement of plasma fenfluramine and norfenfluramine concentrations.
This protocol is shorter than the standard 5-hour fenfluramine test
used in psychiatric research because of practical constraints on
participation and scheduling in our sample of generally healthy
volunteers. In a sample of 42 individuals, we found that peak prolactin
concentrations over 3.5 and 5 hours had a correlation of 0.91,
indicating high concordance between assessments based on the
abbreviated and standard sampling intervals (S.B.M., J.D.F.,
M.F.M., unpublished data, 1997). All blood samples were
centrifuged immediately, separated, and stored at -70°C
until analysis. Methods for determination of plasma prolactin,
fenfluramine, and norfenfluramine (the principal active metabolite)
have been described elsewhere.5
Analysis
Peak prolactin change was calculated as the arithmetic
difference between the highest prolactin value obtained after drug
administration and the prolactin concentration at baseline. The
resulting distribution of prolactin change scores was normalized by
logarithmic transformation and adjusted for covariation with initial
values to yield a baseline-free index of prolactin response to
fenfluramine. Although prolactin levels were not collected after
ingestion of placebo in this study, prior research has shown that this
calculation of peak prolactin response to fenfluramine is correlated
very highly with both placebo-adjusted prolactin responses and the
calculated area under the curve
(r>0.90).6 Associations between
prolactin response to fenfluramine and BP were examined using standard
parametric correlational and multivariate
linear regression analyses. Pearson correlation coefficients
were compared using Fisher's z transformation. BP and
hypertensive status according to quartiles of prolactin response were
compared by 1-way ANOVA (with contrasts among group means performed
using Tukey's honestly significant difference
procedure7) and 
2
analysis.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The 270 subjects included 145 men and 125 women (216 non-Hispanic whites, 47 blacks, 4 Asians, and 3 Hispanics). Sample characteristics included the following: mean age, 45 years (range, 24 to 60 years); mean education, 15.5 years; mean body mass index (BMI), 27 kg/m2 (range, 18 to 41 kg/m2); 62% married; 79% employed full or part-time; 16% smokers; and mean 3.4 alcoholic drinks consumed per week. Among the women, none were pregnant or lactating; 40 (32%) were postmenopausal, and 12 were taking prescribed hormone replacement therapy. Systolic BP ranged from 85 to 161 mm Hg, and diastolic BP varied from 58 to 98 mm Hg. A total of 31 subjects met the criteria for stage I or II hypertension. The mean±SE baseline plasma prolactin concentration was 3.6±0.3 ng/mL, and the mean maximum level after fenfluramine was 7.1±0.2 ng/mL. The mean plasma concentrations of fenfluramine plus norfenfluramine did not correlate significantly with resting BP, indicating that fenfluramine bioavailability did not vary systematically with BP.
Correlations between BP and age, BMI, and maximum prolactin response to
fenfluramine are provided in the Table
. As is
commonly observed, systolic and diastolic BP
increased with increasing age and with increasing BMI. Weak
relationships were observed between BP and self-reported alcohol
consumption (r=0.10 to r=0.11) and between BP and
basal prolactin concentration (r=-0.13 to
r=-0.17). More notable and unique to this investigation,
systolic and diastolic BPs both correlated
negatively with maximum prolactin response to fenfluramine
(P<0.001). Because determinants of BP variability may vary
according to ethnicity, blacks and non-Hispanic whites, the major races
represented in this sample, were also analyzed
independently. The inverse relationship between BP and peak prolactin
response was highly significant in non-Hispanic whites but not in
blacks. This difference in correlations between blacks and whites was
statistically significant (P<0.01 for both systolic
and diastolic BPs).
|
Multivariate linear regression analyses were then conducted on the data from the non-Hispanic whites to determine whether the relationship between BP and prolactin response to fenfluramine was independent of recognized predictors of BP. Age, gender, BMI, physical activity, smoking, and alcohol consumption were first entered into regression models for systolic and diastolic BPs. Maximum prolactin response was then entered into the model for systolic BP (ß=-0.266, P<0.001) and into the model for diastolic BP (ß=-0.199, P=0.001). Because women in this study varied with respect to menopausal status, secondary analyses were conducted in women only. When menopausal status was entered into the multivariate regression models, maximum prolactin response was a significant predictor of both systolic and diastolic BPs.
For illustrative purposes, non-Hispanic white subjects were divided
into quartiles of peak prolactin response to fenfluramine. As shown in
the Figure, mean systolic and
diastolic BPs adjusted for age, gender, and BMI varied with
quartile of prolactin response (P<0.05), with individuals
in the lowest quartile having higher BPs than those in the highest
quartile of prolactin response (P<0.05). Finally, the
prevalence of high-normal BP or clinically defined hypertension (BP
135/85 mm Hg) also differed by quartile
[2 (3)=8.5, P<0.05]; subjects'
likelihood of having elevated BP was 2.6 times greater in the lowest,
relative to the highest, prolactin response quartile (bottom panel of
Figure).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
We found that smaller prolactin responses to fenfluramine were associated with elevated resting BPs in data collected for a relatively large community sample not receiving drugs known to affect prolactin release or BP. This relationship was strongest in non-Hispanic whites. Notably, prolactin response to fenfluramine explained as much of the sample variability in resting BP as did several well-recognized determinants of BP (specifically, age and BMI). The association between prolactin response and BP was linear and persisted after adjustment for other predictors of BP (ie, age, gender, BMI, physical activity, and alcohol consumption). Compared with individuals in the highest quartile of prolactin response, subjects in the lowest quartile were 2 to 3 times more likely to have resting BP in the high-normal or hypertensive range. Finally, this relationship was at least weaker in black than in white participants, possibly reflecting differences in the etiology of essential hypertension among racial groups. However, this latter finding was based on a sample of blacks less than one fourth the size of that of whites.
Interpretation of these results depends on the validity of the D,L-fenfluramine challenge as an index of brain serotonergic activity. Numerous studies in animals and humans have shown that plasma prolactin levels increase after administration of serotonin precursors, releasing agents, and direct agonists.3 When administered acutely, fenfluramine serves as a serotonin agonist by stimulating serotonin release from storage granules in presynaptic neurons, by blocking serotonin reuptake, and possibly by activating postsynaptic receptors.2 3 8 In rodents, prolactin responses to fenfluramine are prevented by acute lesions of the raphe nuclei, and in humans and experimental animals, prolactin responses are dose-related and blocked by pretreatment with serotonin antagonists.3 9 10 Thus, the fenfluramine challenge does not distinguish between presynaptic and postsynaptic processes but instead permits assessment of the overall function of neuronal circuits utilizing serotonin. In addition, prolactin responses to the serotonin-selective D-fenfluramine isomer have been found to correlate highly with responses to D,L-fenfluramine.11 The topographical pattern of neuronal activation on positron-emission tomography induced by racemic fenfluramine is also similar to that observed after administration of D-fenfluramine.12 13 Finally, several reports indicate that prolactin response to fenfluramine has acceptable reproducibility14 and correlates significantly with cerebrospinal fluid 5-hydroxy indole–acetic acid.15
Although fenfluramine-evoked prolactin release is mediated by serotonergic systems, it is possible that the response is also influenced by other factors involved in the control of prolactin secretion. For example, blunted fenfluramine-evoked prolactin release might reflect an attenuation of stimulus-secretion coupling in lactotrophs or enhanced dopaminergic inhibition. However, the functional status of lactotrophs and their tonic inhibition by dopamine, as assessed by prolactin secretion after infusion of thyrotropin-releasing hormone, are unrelated to fenfluramine-induced changes in circulating prolactin levels.16 Some previous research has suggested that hypertension is associated with elevated basal prolactin concentration17 and exaggerated prolactin response to the dopaminergic antagonist metaclopramide.18 19 However, prolactin responses in individuals with a modest BP elevation in this study were blunted and therefore are inconsistent with a general increase in basal and stimulus-evoked prolactin release. Moreover, the prolactin responses in this study were adjusted for any covariation with baseline prolactin levels. Our findings therefore point to a unique and most likely serotonergic mechanism for prolactin responses to fenfluramine.
To the extent that fenfluramine-induced prolactin release serves as an index of hypothalamic-pituitary serotonergic activity, the present findings extend prior research demonstrating that brain serotonin plays a role in BP regulation in laboratory animals. Serotonergic neurons modulate autonomic nervous activity and vasopressin and renin release,1 and augmentation of serotonin release in obese individuals with fenfluramine may decrease autonomic sympathetic activity.20 However, as described in the classic article by Page,21 the actions of serotonin are "amphibaric." Relatively small doses of serotonin injected into the cerebrospinal fluid of rats increases BP by activating 5-HT2 receptors, whereas stimulation of 5-HT1A receptors, many of which are autoreceptors located presynaptically, decreases BP, particularly in hypertensive animals. Furthermore, in experimental animals elevations or reductions in BP cause alterations in serotonin neurotransmission, likely reflecting compensatory adjustments to raised or lowered BP.22 In conclusion, the inverse association seen here between fenfluramine-induced prolactin release and BP may reflect either a causal relationship involving 5-HT, a permissive effect of diminished serotonergic activity, or, alternatively, blunted prolactin responses may serve as a marker for a separate unidentified disturbance more directly responsible for BP dysregulation.
|
Acknowledgments |
|---|
This research was supported by National Institutes of Health Public Health Service grant HL46328.
|
Footnotes |
|---|
Reprint requests to Dr Muldoon, Old Engineering Hall Room 506, University of Pittsburgh, Pittsburgh, Pa 15260.
Received June 30, 1998; first decision July 7, 1998; accepted July 14, 1998.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. McCall RB. Role of serotonin1A and serotonin2 receptors in the central regulation of the cardiovascular system. Pharmacol Rev. 1994;46:3:231–243.
2. Quattrone A, Tedeschi G, Aguglia U, Scopacasa F, DiRenzo GF, Annunziato L. Prolactin secretion in man, a useful tool to evaluate the activity of drugs on central 5-hydroxytryptaminergic neurons: studies with fenfluramine. Br J Pharmacol. 1983;16:471–475.
3. Yatham LN, Steiner M. Neuroendocrine probes of serotonergic function: a critical review. Life Sci. 1993;53:447–463.[Medline] [Order article via Infotrieve]
4.
Paffenbarger RS, Wing AL, Hyde RT. Physical activity
as an index of heart attack risk in college alumni. Am J
Epidemiol. 1978;108:161–175.
5. Muldoon MF, Manuck SB, Jansma CL, Moore AL, Perel J, Mann JJ. D, L-fenfluramine challenge test: experience in nonpatient sample. Biol Psychiatry. 1996;39:761–768.[Medline] [Order article via Infotrieve]
6. Coccaro EF, Siever LJ, Klar HM, Maurer G, Cochrane K, Cooper TB, Mohs RC, Davis KL. Serotonergic studies in patients with affective and personality disorders. Arch Gen Psychiatry. 1989;46:578–599.
7. Kirk RE. Experimental Design: Procedures for the Behavioral Sciences. Belmont, Calif: Wadsworth Publishing Co; 1968.
8. McCann UD, Hatzidimitriou G, Ricaurte GA. Prolactin response to fenfluramine is independent of serotonin release. Eur J Pharmacol. 1996;312:R1–R2.[Medline] [Order article via Infotrieve]
9. Lewis DA, Sherman BM. Serotonergic regulation of prolactin and growth hormone secretion in man. Acta Endocrinol. 1985;110:152–157.
10. Goodall EM, Cowen PJ, Franklin M, Silverstone T. Ritanserin attenuates anorectic, endocrine and thermic responses to d-fenfluramine in human volunteers. Psychopharmacology. 1993;112:461–466.[Medline] [Order article via Infotrieve]
11. Coccaro EF, Kavoussi RJ, Cooper TB, Hauger RL. Hormonal responses to d- and d,l-fenfluramine in healthy human subjects. Neuropsychopharmacology. 1996;15:595–607.[Medline] [Order article via Infotrieve]
12. Kapur S, Meyer J, Wilson AA, Houle S, Brown GM. Modulation of cortical neuronal activity by a serotonergic agent: a PET study in humans. Brain Res. 1994;646:292–294.[Medline] [Order article via Infotrieve]
13. Meyer JH, Kapur S, Wilson AA, DaSilva JN, Houle S, Brown GM. Neuromodulation of frontal and temporal cortex by intravenous d-fenfluramine: an (15O)H2O PET study in humans. Neurosci Lett. 1996;207:25–28.[Medline] [Order article via Infotrieve]
14. Stoff DM, Pasatiempo AP, Yeung JH, Bridger WH, Rabinovich H. Test-retest reliability of the prolactin and cortisol responses to d, l-fenfluramine challenge in disruptive behavior disorders. Psychiat Res. 1992;42:65–72.[Medline] [Order article via Infotrieve]
15.
Mann JJ, McBride A, Brown RP, Linnoila M, Leon AC,
DeMeo M, Meiczkowski T, Meyers JE, Stanley M. Relationship between
central and peripheral serotonin indexes in
depressed and suicidal psychiatric inpatients. Arch Gen
Psychiatry. 1992;49:442–446.
16. Coccaro EF, Klar H, Siever LJ. Reduced prolactin response to fenfluramine challenge in personality disorder patients is not due to deficiency of pituitary lactotrophs. Biol Psychiatry. 1994;36:344–346.[Medline] [Order article via Infotrieve]
17. Hauger-Klevene JH. Abnormal prolactin response to thyrotrophic releasing hormone in essential hypertension. Int J Clin Pharmacol Res. 1983;2:111–114.
18. Stern N, Eshkol A, Lunenfeld B, Rosenthal T. Prolactin secretion in essential hypertension. Clin Exp Hypertens A. 1983;5:543–558.[Medline] [Order article via Infotrieve]
19. Os I, Kjeldsen SE, Westheim A, Aakesson I, Norman N, Enger E, Hjermann I, Eide I. Decreased central dopaminergic activity in essential hypertension. J Hypertens. 1987;5:191–197.[Medline] [Order article via Infotrieve]
20. Kolanowski J, Younis LT, Vanbutsele R, Detry J-M. Effect of dexfenfluramine treatment on body weight, blood pressure and noradrenergic activity in obese hypertensive patients. Eur J Clin Pharmacol. 1992;42:599–606.[Medline] [Order article via Infotrieve]
21.
Page IH. Serotonin (5-hydroxytrypamine).
Physiol Rev. 1954;34:563–588.
22. Singewald N, Juen Gou L, Schneider C, Kaehler S, Philippu A. Serotonin outflow in the hypothalamus of conscious rats: origin and possible involvement in cardiovascular control. Eur J Pharmacol. 1995;294:787–793.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  M. F. Muldoon, R. H. Mackey, K. Sutton-Tyrrell, J. D. Flory, B. G. Pollock, and S. B. Manuck Lower Central Serotonergic Responsivity Is Associated With Preclinical Carotid Artery Atherosclerosis Stroke, August 1, 2007; 38(8): 2228 - 2233. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. Muldoon, R. H. Mackey, K. V. Williams, M. T. Korytkowski, J. D. Flory, and S. B. Manuck Low Central Nervous System Serotonergic Responsivity Is Associated with the Metabolic Syndrome and Physical Inactivity J. Clin. Endocrinol. Metab., January 1, 2004; 89(1): 266 - 271. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. D. Stocker, M. F. Muldoon, and A. F. Sved Blunted Fenfluramine-Evoked Prolactin Secretion in Hypertensive Rats Hypertension, October 1, 2003; 42(4): 719 - 724. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. McCaffery, M. Bleil, M. F. Pogue-Geile, R. E. Ferrell, and S. B. Manuck Allelic Variation in the Serotonin Transporter Gene-Linked Polymorphic Region (5-HTTLPR) and Cardiovascular Reactivity in Young Adult Male and Female Twins of European-American Descent Psychosom Med, September 1, 2003; 65(5): 721 - 728. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Matthews, J. D. Flory, M. F. Muldoon, and S. B. Manuck Does Socioeconomic Status Relate to Central Serotonergic Responsivity in Healthy Adults? Psychosom Med, March 1, 2000; 62(2): 231 - 237. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Everson, G. A. Kaplan, D. E. Goldberg, and J. T. Salonen Hypertension Incidence Is Predicted by High Levels of Hopelessness in Finnish Men Hypertension, February 1, 2000; 35(2): 561 - 567. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||