Clonidine reduces elevated cerebrospinal fluid catecholamine levels in patients with essential hypertension

Life Sciences, Vol. 35, pp. 1365-1371 Printed in the U.S.A.

Pergamon Press

CLONIDINE REDUCES ELEVATED CEREBROSPINAL FLUID CATECHOLAMINE LEVELS IN PATIENTS WITH ESSENTIAL HYPERTENSION Cubeddu, L.X. , Hoffmann, I.S. +, Davila, J.+, Barbella, Y.R. +, and Ordaz, P.§ Division of Clinical Pharmacology, School of Medicine, University of North Carolina; +Departments of Pharmacology and Cardiology of Central University of Venezuela; §Department of Neurology, Vargas Hospital (Received in final form July 13, 1984) Summary Cerebrospinal fluid (CSF) catecholamines were measured in normotensive patients and in patients with mild to moderate essential hypertension. CSF-norepinephrine (NE) concentrations were 50% lower in the normotensive individuals (127 ± 28 vs. 240 ± 23 pg/ml) (P< 0.01). In hypertensive patients, CSF-NE was inversely related to age (r =-0.68; P<0.01) and directly related to plasma NE (r = 0.61; P<0.05). Clonidine (450 mcg/day for 2 weeks) significantly reduced CSF-NE (-40%) in hypertensive patients. In addition, it decreased blood pressure, plasma and urinary NE. Urinary VMA was not affected by clonidine. No correlation was observed between clonidine effects on BP and on plasma or CSF catecholamines. This study indicates that patients with essential hypertension have elevated levels of CSF-NE which are reduced after treatment with clonidine. The elevation of CSF-NE suggests that central (spinal?) noradrenergic activity may be increased in patients with mild to moderate essential hypertension, and that can be reduced by treatment with clonidine. Central norepinephrine (NE) containing neurons are known to play a role in the regulation of systemic blood pressure as well as in the development and maintenance of experimental hypertension (11,20,21). Chemical destruction of central noradrenergic cells with 6-hydroxydopamine is known to prevent genetic, renal, DOCA-salt and neurogenic hypertension in laboratory animals (3,4,11,20, 21). However, the role that central NE containing neurons play in the pathophysiology of human hypertension and in the response to ant±hypertensive drugs is unknown. This is due in part to the existing difficulties in assessing central noradrenergic activity in humans. Measurements of cerebrospinal fluid (CSF) catecholamines, their metabolites, and enzyme markers have been employed in an attempt to estimate the activity of these neurons (1,5,6,19,23). Clon±dine, a well known centrally acting ant±hypertensive drug, decreases blood pressure by inhibiting sympathetic nerve firing. This drug reduces the turnover of central noradrenergic neurons in experimental animals, probably by activation of alpha 2 adrenoreceptors (7,12,14). If CSF-norepinephrine concentrations (CSF-NE) reflect the activity of central noradrenergic neurons, treatment with clonidine should reduce CSF-NE. Previous human studies demonstrated a reduction in CSF concentrations of 3-methoxy, 4-hydroxy-phenylglycol (MHPG),

Address reprint requests to: L.X. Cubeddu, Div. of Clinical Pharmacology, 104 MacNider Bldg. 202H, Univ. of North Carolina, Chapel Hill, N.C. 27514. This work was supported by the Burroughs Wellcome Clinical Pharmacology Development Fund. 0024-3205/84 $3.00 + .00 Copyright (c) 1984 Pergamon Press Ltd.

1366

Clonidine and CSF Catecholamines

Vol. 35, No. 13, 1984

a metabolite of NE after clonidine (2,22). In the present study, we investigated whether treatment with clonidine affects the CSF levels of NE, EPI and DA in patients with essential hypertension. Furthermore, it was investigated whether the drug-induced changes in CSF catecholamines correlated to the blood pressure reducing effect of the drug. Eide et al. (8) and Lake et al. (15) had previously reported that CSF-NE was elevated in hypertensive patients. In the present study it was further investigated whether patients with mild to moderate essential hypertension have abnormal levels of NE, EPI and DA in their CSF, and whether the abnormal levels are corrected by treatment with clonidine. Methods CSF catecholamines were measured in hypertensive and normotensive subject~ The hypertensive patients, aged from 26 to 57 years, had a diagnosis of mild to moderate essential hypertension. The patients were admitted solely for the present study to the Division of Cardiology of the Hospital Clinico at the Central University of Venezuela, after giving their informed consent. The study was approved by the National Council for Research in Science and Technology. All subjects had normal serum creatinine, BUN, electrolytes and glucose. Urinalysis, urinary electrolytes, urinary vanillylmandelic acid (VMA), serum and urinary catecholamines were also within normal limits. Physical examination was unremarkable with the exception of the elevated BP, presence of arteriolar narrowing in the retinal vessels (grade I, KW) and in 5 individuals, there was evidence of left ventricular hypertrophy. None of the subjects had a previous history suggestive of coronary artery disease, stroke, renal or hepatic impairment. All patients had a positive family history of hypertension. In all cases, medications were discontinued for at least two weeks prior to the lumbar puncture. CSF and plasma catecholamines were measured in 12 untreated hypertensive patients and in 10 hypertensive subjects fifteen days after initiation of therapy with clonidine (Figure i). Six of these patients had two lumbar punctures, the first while untreated (baseline) and the second during treatment with clonidine (Table If). In order to minimize side effects, the dose of clonidine was increased gradually over one week to'450 mcg/d (150 mcg TID) and maintained for an additional week; subsequently, the drug was tapered over three days. CSF from normotensive patients was obtained from the Department of Neurology of the Vargas Hospital. A total of twelve CSF samples for catecholamine determination were obtained. Six patients had a diagnosis of multiple sclerosis, three had epilepsy, one had optic neuritis, one a Guillain-Barre syndrome and one a Brown-Sequard syndrome. In all cases BP were always below 140/90 mmHg. Two of the epileptic patients were receiving treatment with phenytoin and phenobarbital. The rest of the subjects received occasional therapy with a n a l g e s i c - a n t i i n f l a ~ a t o r y agents. Lumbar punctures were performed after one hour in the supine position. In order to minimize circadian variation of CSF-NE (25), all CSF were obtained between 10am and 12pm. The taps were performed with sterile, disposable equipments, employing a 20 gauge needle, with the patients in right lateral decubitus. The first two mls of CSF were collected for routine cytobiochemical analysis. Subsequently, 8 mls of CSF were collected in a chilled plastic tube containing ascorbic acid 8 mg, and 80 ~i of an antioxidizing solution containing 0.31M EGTA and reduced gluthatione 0.16M. Samples were immediately mixed and frozen at -80°C for later catecholamine determinations. A I0 ml sample was concomitantly obtained by venipuncture from an antecubital vein. The blood was placed on a heparinized tube, containing 10 mg of ascorbic acid and 200 ~I of the antioxidating solution. Plasma was obtained after centrifugation, and then

Vol. 35, No. 13, 1984

Clonidine and CSF Catecholamines

1367

stored at -80°C until assayed. After completion of the lumbar puncture the patients were closely monitored remaining in supine position for three hours. A physical examination was then performed, and the patients discharged if no side effects or abnormalities were encountered. The CSF and plasma levels of NE, EPI and DA were measured employing a radioenzymatic technique (Cat-A-Kit; Upjohn Diagnostic). Briefly, catecholamines were O-methylated in the presence of the enzyme catechol-O-methyl transferase and tritiated-methyl-S-adenosyl methionine. The radioactive products were separated by organic and inorganic solvent extractions and thin-layer chromatography. After oxidation with periodate and organic solvent extraction, the radioactive products were quantified by liquid scintillation spectrometry. Samples were run in duplicates, and corrections were made employing internal standards. The intra-assay variation was of 8% in the 200 pg/ml range. Blanks for norepinephrine, epinephrine and dopamine averaged 150 ± 15, 24 ± 5 and 45 ± 5 cpm, respectively. After blank substraction, i00 picograms of the amines gave 2833 ± 211 cpm, 2242 ± 196 cpm, and 1899 ± 113 cpm for DA, EPI and NE, respectively. Twenty-four hour urine collection into jars containing i0 ml of 0.6 N HCI and ig sodium metabisulfite were performed the day prior to initiation of clon±dine therapy and again on day 14 of clonidine treatment. Urinary catecholamines were concentrated into alumina columns and quantified by the fluorometric assay of Laverty and Taylor (16). Urinary V M A w a s measured by the spectrophotometric assay of Pisano et al. (18), which eliminates dietary interference. Results were analyzed by student or paired "t" test and/or Mann-Whitney U test (10). Results Baseline levels of CSF and plasma catecholamines were determined in 12 untreated essential hypertensive subjects (Table I and II). NE was the major catecholamine present both in plasma and CSF. Very small amounts of EPI and DA were encountered in the lumbar CSF. In many subjects, the latter two catecholamines were not detectable in CSF. Measurements of CSF catecholamines were performed in twelve normotensive individuals (Table I). NE was also the major catecholamine present in the CSF of normotensive patients; however, the CSF-NE concentrations of normotensive subjects were 50% lower than those of hypertensives (P<0.OI). EPI levels were higher in normotensives. In hypertensive subjects, CSF-NE was directly related to plasma NE (r = 0.61; P<0.05 two tailed t test) and inversely related to age (r ~ 0 . 6 8 ; P<0.01 by one tail t test). There was no correlation between age and CSF-NE in normotensives. Systolic and diastolic pressures were not correlated with age, CSF-NE or plasma NE levels in either group. CSF and plasma-NE concentrations were not significantly different in black (CSF: 229 ± 20; plasma: 405 ± 61 pg/ml) and white (CSF: 251 ± 48; plasma: 360 ± 66 pg/ml) subjects.

1368

Clonidine and CSF Catecholamines

Vol. 35, No. 13, 1984

TABLE I CSF Catecholamines in Normotensive and Hypertensive Patients SBP

DBP

AGE

B/W

F/M

Normotensive (n = 12)

36±3

5/7

6/6

i16±4

79±4

127±28

11±2

14±4

Hypertensive (n = 12)

41±2

6/6

8/4

163±5

i07±3

240±23

3±2

12±5

Shown are mean SBP = systolic spinal fluid. *Significantly Mann Whitney U

(mmHg)

CSF(pg/ml) NE

EPI

DA

values ± SEM. B = blacks. W = whites. F = females. M = males. blood pressure. DBP = diastolic blood pressure. CSF = cerebroNE = norepinephrine. EPI = epinephrine. DA = dopamine. different from normotensive at P~0.01 by student "t" test and test.

TABLE II Effects of Clonidine on CSF, Plasma and Urinary Catecholamines in Patients With Essential Hypertension NE SBP

DBP

Baseline

162±5

Clonidine

149±5

VMA Urine (mg/d)

EPI Urine (~g/d)

CSF (pg/ml)

Plasma (pg/ml)

Urine (~g/d)

i05±3

270±30

380±40

23±3

5.9±0.6

16±3

98±3

165±13

276±25

20±2

5.9±0.7

13±3

Shown are mean values ± SEM. SBP = systolic blood pressure. DBP = diastolic blood pressure. NE = norepinephrine. VMA = vanillylmandelic acid. EPI = epinephrine. CSF = cerebrospinal fluid. Two measurements were performed on each patient, one prior to treatment and the second fifteen days after initiation of therapy with clonidine. Significantly different at *P<0.05 and **P<0.01 by paired "t" test (two tails) and Mann-Whitney U test (i0).

Hypertensives treated with clonidine for 2 weeks had significantly lower average blood pressures, heart rates, plasma, urinary and CSF-NE levels than untreated hypertensives. VMA was not affected by clonidine treatment (5.8 ± 0.6 vs. 5.8 ± 0.8). Individual values are shown in Figure I. For the whole group, clonidine reduced CSF-NE from 240 ± 23 to 165 ± 16 pg/ml (P
Vol. 35, No. 13, 1984

Clonidine and CSF Catecholamines

1369

plasma catecholamines.

400

800

E300 o

~

O~ O.

6oo

w z

LU z 200 I00

400 :E co <[ .J n 200

0

o

0 NoITx

I

0 No Tx

Clonidlne

i

Clonidine

t~

EiSO 8

6° I v

I00

~r~:ZL40 n~Z ~uD>"i 2c~~0

8s q m

$

,

uJ

No Tx

Clonidine

0 NoTx

Clonidine

FIG. i The effects of clonidine on mean blood pressure and norepinephrine concentrations in the CSF, plasma and urine of hypertensive subjects. The connecting lines indicate six subjects in whom two measurements were obtained prior and during treatment with clonidine. The rest of the subjects had only a single measurement, either under no treatment or during treatment with clonidine. The patient characteristics were: untreated (41 ± 2Kg, 6 blacks and 6 whites, 8 females and 4 males), treated (40 ± 3Kg, 4 blacks and 6 whites, 6 females and 4 males). Individual measurements (n=1~ are shown. Bars depict means ± SEM. MBP: mean blood pressure.

Discussion Measurements of plasma and CSF-NE can give an indication of nerve activity "in vivo". The NE in plasma originates in sympathetic neurons or adrenal medulla and acute changes in sympathetic activity cause corresponding changes in plasma NE levels. By analogy, CSF may serve as a sink for NE released centrally. Because of the effective blood-CSF barrier for NE, there should be very little transfer of NE from plasma to CSF (26). In fact, patients with pheochromocytoma with increased plasma concentrations of NE have normal CSF levels of this catecholamine (8,24). Therefore, measurements of CSF catechola-

1370

Clonidine and CSF Catecholamines

Vol. 35, No. 13, 1984

mines may be helpful in the investigation of the effects of drugs and disease states on the activity of central catecholamine-containing neurons. Clonidine, a centrally acting antihypertensive drug, is known to activate alpha-2 adrenoreceptors and to reduce the turnover rate of central NE-containing neurons. These effects are associated with a reduction in sympathetic nerve firing and a decrease in blood pressure (7,12,14). In the present study, clonidine reduced the elevated CSF-NE levels present in patients with essential hypertension. Clonidine has been previously shown to decrease CSF-MHPG levels, a metabolite of NE (2,22). These results suggest that clonidine reduces central noradrenergic activity in hypertensive patients. Unfortunately, measurements of lumbar CSF-NE cannot predict which brain noradrenergic neuronal groups are affected by drugs or disease states. Previous studies have demonstrated the existence of a gradient for lumbar CSF-NE, when large volumes of CSF (~ 17 ml) were obtained (25). This gradient may be explained by the greater concentrations of NE in the higher centers than in the spinal cord (25). In this investigation, CSF-NE was always measured in the first 10 ml of lumbar CSF, therefore, it is likely that the first few mls of lumbar CSF would reflect spinal rather than cerebral noradrenergic activity. Clonidine has also been shown to reduce noradrenergic activity in the spinal cord (17). It is thus possible that the reduction in CSF-NE produced by clonidine reflects the inhibitory effects of the drug on noradrenergic bulbospinal pathways. These pathways seem to be an important relay between the brainstem vasomotor centers and the intermediolateral columns in the spinal cord, from where the preganglionic sympathetic fibers emerge (9). As previously reported (7), plasma and urinary NE were also decreased during treatment with clonidine, suggesting that the drug reduces central and peripheral noradrenergic activity. Interestingly, urinary VMA was not affected by clonidine. These findings suggest that urinary VMA do not solely reflect the release of NE and EPI during neuronal activity, but may also reflect intracellular metabolism of catecholamines, as well as the spontaneous (non-exocytotic) loss of catecholamines from adrenergic tissues. In fact, in patients with pheochromocytoma, urinary VMA reflect the size and weight of the tumor and not the plasma and/or urinary levels of catecholamines, or the blood pressure levels of the patients (13). Previous studies by Eide et al. (8) and Lake et al. (15) indicated that CSF-NE levels were increased in patients with essential hypertension. Similar findings were demonstrated in the present study. Lake and coworkers employed healthy volunteers as their normotensive controls. Eide et al. (8) and in our study, normotensive controls, were patients who underwent lumbar puncture for evaluation of neurological symptoms. This elevation of CSF-NE suggests that the activity of central (spinal?) noradrenergic neurons may be increased in patients with mild to moderate essential hypertension. This increase in activity seems to be greater in young subjects since a significant negative correlation between CSF-NE and the age of the hypertensive patients was observed in this study. No correlation was demonstrated between both variables in normotensive individuals. Eide et al. (8) also observed that the youngest subjects had the highest CSF-NE levels. No significant correlation between age and CSFNE was demonstrated in Lake et al. (15) study. However, their hypertensive patients were older than those in the other studies (15, present study). These results suggest that elevated CSF-NE seems to occur early in the course of the disease; and favor the idea of the neurogenic origin of essential hypertension. Interestingly, after reduction in CSF-NE by clonidine, the correlation between age and CSF-NE was eliminated (present study). In summary, patients with essential hypertension have elevated levels of CSF-NE which are reduced after treatment with clonidine.

Vol. 35, No. 13, 1984

Clonidine and CSF Catecholamines

1371

Acknowledsments The authors would like to thank Mr. Michael Shakarjian for his help in the determinations of catecholamines, and to Dr. John Rogers and Ms. Deborah Powell for their assistance in the preparation of the manuscript. References 1. 2. 3. 4. 5, 6. 7. 8, 9.

i0. ii. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

G. BAGDY, M. ARATO, K. BARACZKA, and M.I.K. FEKETE, Life Sci. 32:26672676, (1983). L. BERTILSSON, K. HAGLUND, J. OSTMAN, M. RAWLINS, V. REINGBERGER, and F. SJOQVIST, Eur. J. Clin. Pharmacol. 11:125-128, (1977). J.P. CHALMERS and J.L. REID, Circ. Res. 31:789-804, (1972). J.P. CHALMERS, C.T. DOLLERY, P.J. LEWIS, and J.L. REID, J. Physiol. 238: 403-411, (1974). T.N. CHASE, E.K. GORDON, and L.K.Y. NG, J. Neurochem. 21:581-587, (1973). L.X. CUBEDDU, J. DAVILA, D. ZSCHAECK, Y.R. BARBELLA, P. ORDAZ, and J. DOMINGUEZ, Hypertension ~:448-455, (1981). C.T. DOLLERY, D.S. DAVIES, G.H. DRATTAN, M.B. DARGIE, C.R. DEAN, and J.L. REID, Clin. Pharmacol. Ther. 19:11-17, (1976). I. EIDE, R. KOLLOCH, V. DEQUATTRO, L. MIANO, R. DUGGER, and J. VAN DER MEULEN, Hypertension !:255-259, (1979). K. FUXE, T. HOKFELT, P. BOLME, M. GOLDSTEIN, O. JOHANSSON, G. JONSSON, P. LINBRINK, A. LJUNGDAHL, and C.H. SACHS, Central Action of Druss in Blood Pressure Re$ulation, Pittman Medical, London, (1975). A. GOLDSTEIN, Biostatistics; An Introductory Text, MacMillan Publishing Co., New York, (1964). G. HAEUSLER, L. FINCH, and H. THOENEN, Experienta, 28:1200-1203, (1972). L. ISAAC, J. Cardiovasc. Pharmacol. ~:Suppl. I, (1980). K. KOBAYASHI, Y. MIURA, H. TOMIOKA, H. SAKUMA, M. ADACHI, T. SATO, and K. YOSHINAGA, Clin. Chim. Acta 85:154-165, (1978). W. KOBINGER and A. WALLAND, Eur. J. Pharmacol. 2:155-162, (1967). C.R. LAKE, R.J. POLINSKY, H.G. GULLNER, M.H. EBERT, M.G. ZIEGLER, F.C. BARTTER, and I.J. KOPIN, Clin. Sci. 59:229s-233s, (1980). R. LAVERTY and K.M. TAYLOR, Anal. Biochem. 22:269-279, (1978). J. MARWAHA, J. KEHNE, R.L. COMMISSARIS, J. LAKOSKI, W. SHAW, and M. DAVIS, Brain Res. 276:379-382, (1983). J.J. PISANO, J.R. CROUT, and D. ABRAHAM, Clin. Chem. Acta 7:285-291, (1962). R.M. POST, C.R. LAKE, D.C. JIMERSON, W.E. BUNNEY, J.H. W OD, M.G. ZIEGLER, and F.K. GOODWIN, Am. J. Psychiatry 135:907-912, (1978). J.L. REID, J.A. ZIVlN, and l.J. KOPIN, Circ. Res. 37:569-579, (1975). D.J. REIS, N. DOBA, and M.A. NATHAN, Resulation of Blood Pressure by the Central Nervous System, Grune and Stratton, New York, (1976). R.K. SARAN, R.C. SAHUJA, N.N. GUPTA, M. HASAN, K.P. BHARGAVA, K. SHANKER, and K. KISHOR, Sci. 200:317-318, (1978). S. WILK, S. SHOPSIN, and S. GERSHON, Nature 235:440-441, (1972). M.G. ZIEGLER, C.R. LAKE, and I.J. KOPIN, Nature 261:333-335, (1976). M.G. ZIEGLER, C.R. LAKE, J.H. WOOD, and M.H. EBERG, Nature 264:656-658, (1976). M.G. ZIEGLER, C.R. LAKE, J.H. WOOD, B.R. BROOKS, and M.H. EBERT, J. Neurochem. 28:677-679, (1976).