Increased Plasma Concentrations of Prostacyclin Metabolite 6-Keto-PGFI in Essential Hypertension Influence of Therapy With Labetalol

LOUIS ROY, MD, JAWAHAR MEHTA, MD, and PAULETTE MEHTA, MD, With the technical

assistance of NANCY OSTROWSKI, CHRISTINE HORALEK, and GEORGIA KELLEY

To evaluate the role of the vasoactive prostaglandins prostacyclin and thromboxane A 2 in essential hy- pertension, the stable metabolites 6-keto-PGFla and thromboxane B2, respectively, were measured in plasma before and after therapy in 7 patients. During the placebo phase, plasma 6-keto-PGFla levels were significantly greater than normal. Plasma thromboxane B 2 levels were not statistically dif- ferent from those in normal subjects. After intrave- nous administration of labetalol to the point of blood pressure reduction, neither plasma 6-keto-PGFla

nor thromboxane B 2 values changed. With prolonged oral labetalol therapy and concurrent regulation of blood pressure, a significant decrease in plasma 6-keto-PGFla levels occurred while thromboxane B 2 values remained unaltered. Elevation of plasma 6-keto-PGFla in untreated hypertensive subjects suggests that enhanced vessel wall prostacyclin synthesis may be a protective mechanism to prevent organ damage. As blood pressure is controlled this increase is no longer needed, and prostacyclin generation returns to normal.

Since the discovery of prostacyclin and thromboxane A2,1-3 these very labile prostaglandins3, 4 have been implicated in many cardiovascular diseases. The former, synthesized mostly in vessel walls, has po ten t vasodi- latory and platelet aggregation inhibitory properties.l,~ T he latter, synthesized mainly in platelets, 2,G is a pow- erful vasoconstrictor 7 and potent platelet aggregant. 2 Recent evidence has demonstrated tha t prostaglandins may also be involved in the regulation of blood pressure. The drugs interfering with cyclooxygenase activity have been repor ted to decrease the effectiveness of antihy- pertensive therapy, s 11 Some investigators have de- scribed a decrease in urinary metabolites of prostacyclin in hyper tensive subjects which suggests an imbalance of pros tacycl in- thromboxane A2 equil ibrium in favor of the latter. 12 Impaired renal product ion of prosta- glandin E2 in essential hyper tension has also been re- ported. ~3 Vessels from spontaneously hypertensive rats have been shown to generate unusually large amounts

From the Divisions of Cardiology and Hematology/Oncology, Depart- ments of Medicine and Pediatrics, University of Florida, Shands Teaching Hospital and Veterans Administration Medical Center, Gainesville, Florida. This study was supported in part by grants from the American Heart Association, Florida Affiliate, St. Petersburg, Florida; Schering Corporation, Bloomfield, New Jersey; and Glaxo Incorporated, Middlesex, England. Dr. Roy was supported by a grant from the McLaughlin Foundation, Toronto, Ontario, Canada. Manuscript received March 22, 1982; revised manuscript received June 23, 1982, accepted June 25, 1982.

Address for reprints: Jawahar Mehta, MD, Box J-277, J. Hillis Miller Health Center, University of Florida, Gainesville, Florida 32610.

of PGI2.14,15 Whether a similar phenomenon occurs in human subjects is not known.

This s tudy was designed to evaluate the role of th romboxane A2 and prostacyclin in hypertensive subjects before and after ant ihypertensive t rea tment with labetalol, which is an alpha- and beta-adrenore- ceptor blocking drug. 1~,17

Methods Patients and study protocol: We evaluated 7 hypertensive

men, aged 44 to 68 years (mean ± standard error of the mean 54 + 7). Diastolic blood pressure was 95 mm Hg or higher after the patients had been untreated for at least 2 weeks. All sec- ondary causes of hypertension were excluded. Other exclusion criteria were lung disease, symptomatic coronary heart dis- ease, malignant hypertension, and cerebrovascular or renal impairment. No concomitant antihypertensive, platelet- active, or prostaglandin-active drugs were permitted during the study. All patients gave a written informed consent ap- proved by Institutional Review Committees.

After a 2-week placebo phase, subjects were hospitalized and intravenous labetalol (0.5 to 1.0 mg/kg) was administered to achieve at least 10 mm Hg reduction in diastolic blood pressure. The subjects were then discharged on oral labetalol and readmitted for a second evaluation 6 weeks later. During long-term labetalol treatment, they were followed up on a weekly basis and given increasing doses of labetalol until ap- propriate blood pressure control (diastolic pressure _<90 mm Hg) was obtained.

Blood collection: Blood samples were collected in the placebo phase, the intravenous labetalol phase, and at the end of oral labetalol phase. Each time, the first 2 ml of bh)od was

464

February 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 51 465

TABLE I Hemodynamic Effects of Acute Intravenous (IV) and Long-Term Oral Labetalol Therapy in Hypertensive Subjects

Placebo Acute IV Long-Term Phase Labetalol Oral Labetalol

Heart rate (beats/min) 77 4- 3 71 -I- 4 64 4- 4* Systolic blood pressure 162 4- 6 134 -I-6 t 138 4- 6 t

(mm Hg) Diastolic blood pressure 98 4- 4 80 4- 3 t 82 -t- 35

(mm Hg) Cardiac output 8.6 :J: 1.0 8.2 :E 0.9 7.2 4- 0.7

(liters/min) Cardiac index 4.0 -4- 0.4 3.8 -I- 0.3 3.3 4- 0.3

(liters/min/m 2) Total peripheral 1,240 :E 184 901 -I- 184" 1,227 4- 176

resistance (dynes s cm -5)

* p <0.05 versus placebo phase. ~ p <0.01 versus placebo phase. J; p <0.001 versus placebo phase.

discarded, then 4 ml was gently withdrawn in a polystyrene syringe and immediately transferred into a tube containing aspirin and ethylene diaminetetraacetic acid (EDTA) (1 ml) at 4°C. The tube was centrifuged at 150 g for 8 minutes and plasma was transferred for 6-keto-PGF1, and thromboxane B2 determinations.

Thromboxane B2 and 6-keto-PGFh, determination: Plasma thromboxane B2 and 6-keto-PGFl~ were measured by radioimmunoassay as previously described, ls,19 Lyophi- lized standards, 3H-thromboxane B2 and 3H-6-keto-PGFl,, and the antibodies raised in rabbits were obtained from New England Nuclear, Boston, Massachusetts. With use of this method, the cross-reactivity of thromboxane B2 antibody with other prostaglandins was as follows: PGE2 0.2%, PGA2 0.2%, PGF2~ <0.2%, and 6-keto-PGFl~ <0.2%. The cross reactivity of 6-keto-PGF1, antibody with other prostaglandins was as follows: PGF2, 2.7%, PGE2 and TXB2 <0.1%, PGA2 0.1%, and PGA1 <0.3%. The results for both 6-keto-PGFl~ and throm- boxane B2 were expressed in pg/ml.

Normal values: Normal values for plasma 6-keto-PGF~, and thromboxane B2 were determined in venous blood sam- ples withdrawn from 23 healthy subjects (mean age 47 4- 7 years). Blood was withdrawn from the antecubital vein, without tourniquet pressure. The samples were collected and processed in the same manner as described for the patients.

Statistical analysis: Results were expressed as mean 4- standard error of the mean. Student's t test (paired and un- paired data), the Kruskal-Wallis rank sum test, and Wilcoxon signed rank sum tests were used for comparison. Data were examined with regard to relation between changes in prosta- glandin values and the blood pressure (systolic, diastolic, and mean) by regression analysis. Data were also analyzed to identify any relation between blood pressure levels and 6- keto-PGF1, concentrations before and after treatment. A p value <0.05 was considered significant.

Results

L a b e t a l o l and hemodynamic ef fec t s : The hemo- dynamic effects of labetalol are summarized in Table I. Intravenous labetalol (mean dose 0.6 mg/kg) resulted in a mean plasma level of 105 + 21 ng/ml of plasma. Wi th this dose, mean blood pressure was lower in each subject, bu t cardiac ou tpu t and hear t rate were un- changed. The systemic vascular resistance (mean ar- terial pressure × 80/cardiac output) decreased signifi- cantly (p <0.025). With oral labetalol for 6 weeks (1,200 mg/day in 3 subjects and 800 mg/day in 4 subjects), mean plasma levels of 176 + 59 ng/ml were obtained.

Diastolic and systolic blood pressures showed a mean decrease of 16 and 24 m m Hg, respectively. The hear t rate decreased significantly. Cardiac ou tpu t decreased in 6 pat ients and increased in the remaining pat ient . Systemic vascular resistance increased in 4 and de- creased in 3 patients. Changes in cardiac ou tpu t and vascular resistance were not significant compared with the placebo phase.

Thromboxane B2 concentrations (Fig. 1): Plasma t h r o m b o x a n e B2 concen t ra t ions in hype r t ens ive subjects in the placebo phase showed a wide variat ion

4 4 0 - TxB 2

4 0 0 -

3 6 0 -

3 2 0 -

2 8 0

2 4 0 -

0 200- o

0

160

E 120

8 0

4 0

6 Keto PGFt~

m

P IV ORAL

L _ _ LABETALOLj

HYPERTENSIVE $L~JECTS

P L A S M A T x B 2 AND 6 K E T O - P G F L ~

IN NORMAL AND HYPERTENSIVE SUBJECTS

IV ORAL

LABIA LOL j

HYPERTENSIVE SUBJECTS

FIGURE 1. Plasma TXB2 and 6-keto-PGFl~ concentrations in normal and hypertensive subjects. TXB2 levels did not differ in the 2 populations. Plasma 6-keto-PGFl~ levels were higher in hypertensive subjects before therapy, and decreased significantly with long-term antihypertensive therapy. Data are expressed as mean -I- standard error of the mean. IV = intravenous; P = placebo.

466 PROSTACYCLIN IN HYPERTENSION

(range 82 to 1,220 pg/ml, mean 318 =t= 183 pg/ml). One patient with very high plasma thromboxane B2 con- centrations, 1,220 pg/ml, continued to exhibit these high concentrations after short- and long-term labetalol administration (570 and 840 pg/ml, respectively) and on another occasion after discontinuation of labetalol (1,100 pg/ml). The mean plasma thromboxane B2 con- centrations in hypertensive subjects in the placebo phase were similar to those in the normal volunteers (175 4- 35 pg/ml). After short-term intravenous and long-term oral labetalol administration, the mean plasma thromboxane B2 concentrations decreased to 200 =t= 76 pg/ml and 215 • 132 g/ml, respectively. None of these values were significantly different from each other or from the normal values.

Concentrations of 6-keto-PGFl~ (Fig. 1): In the 23 normal subjects, 6-keto-PGFl~ concentrations were extremely low: they were undetectable in 7 subjects, and the mean value was 64 =L 14 pg/ml. In contrast, 6 hy- pertensive subjects in the placebo phase had detectable and markedly elevated plasma 6-keto-PGFl~ concen- trations (range 190 to 500 pg/ml), and 1 patient had undetectable concentrations throughout the study. The mean value was significantly higher in hypertensive subjects than in the normal volunteers (321 :t: 67 versus 64 =t= 14 pg/ml, p <0.01). With immediate decrease in blood pressure after intravenous labetalol administra- tion, plasma 6-keto-PGFl~ concentrations decreased in only 2 patients compared with the placebo phase, but were still very high (range 0 to 550 pg/ml, mean 344 ± 82 pg/ml; difference not significant compared with the placebo phase). With prolonged oral labetalol therapy and blood pressure reduction, plasma 6-keto-PGFl~ concentrations showed a significant decrease from the placebo phase in all 6 patients (range 0 to 500 pg/ml, mean 179 :l= 74, p <0.05) or from the short-term intra- venous phase (mean 179 =E 74 versus 344 • 82 pg/ml, p <0.05). These plasma 6-keto-PGFl~ concentrations after oral therapy and prolonged blood pressure re- duction did not differ significantly from those in the normal subjects.

Analysis of data showed no significant relation be- tween levels of 6-keto-PGFl~ and blood pressure before and after treatment. Similarly, there was no correlation between changes in plasma concentrations of 6-keto- PGFI~ and reduction in systolic, diastolic, or mean ar- terial pressure.

Discussion Increase in prostacyclin in hypertension: This

study shows that enhanced prostacyclin but normal thromboxane A2 synthesis is present in patients with essential hypertension. With acute reduction of blood pressure, no significant changes in plasma concentra- tions occur. However, after prolonged treatment as the blood pressure is controlled for several weeks, prosta- cyclin metabolite in plasma decreases. Because 6- keto-PGFl~ is a major metabolite of prostacyclin, in- creased levels in the untreated hypertensive subjects could imply either enhanced degradation of prostacyclin with subsequent increased synthesis, or a primary in- crease in prostacyclin synthesis with concomitant in-

creases in its metabolite concentration in plasma. The latter hypothesis is supported by animal experiments that demonstrate enhanced basal prostacyclin synthesis by vessels from spontaneously hypertensive rats.~4,1~ The fact that plasma 6-keto-PGFl~ concentrations re- turn to normal after prolonged antihypertensive ther- apy leads us to believe that enhanced prostacyclin synthesis by vessel walls in hypertensive patients may be a protective mechanism against organ damage. As blood pressure is adequately controlled, this compen- satory mechanism is no longer necessary, and the prostacyclin synthesis by vessel walls may gradually return toward the normal state. In agreement with this "protective mechanism" concept is the observation that in stroke-prone, spontaneously hypertensive rats with an increase in vascular prostacyclin synthesis, prosta- cyclin-synthesizing capability is lost just before the evolution of stroke. 14

Mechanism of increased prostacyclin synthesis: The mechanism of enhanced prostacyclin synthesis is still unclear, It is known that prostacyclin synthesis increases markedly with trauma 2° and increased in- travascular stress. 14,~ Stress states such as exercise, 21 smoking, 22 ischemia, and venous stasis 2:~ also cause an increase in prostacyclin synthesis. Whether labetalol itself affects PGI2 generation is not known. However, we observed no change in plasma 6-keto-PGFl~ with intravenous labetalol administration. With prolonged antihypertensive therapy, a marked decrease in blood pressure was accompanied by a decline in 6-keto-PGFl~ levels. It is tempting to postulate that a new state of prostacyclin synthesis develops with prolonged anti- hypertensive therapy. Although plasma concentrations of labetalol with oral therapy were similar to those with short-term intravenous administration, direct effects of labetalol on prostacyclin synthesis and release cannot be excluded. We found no correlation between decline in plasma concentrations of 6-keto-PGFl~ and decrease in blood pressure. Further studies are needed in a large number of patients to determine whether the decrease in plasma 6-keto-PGFl~ concentration is observed with other antihypertensive drug regimens and if plasma 6-keto-PGFl~ concentrations correlate with changes in blood pressure.

Platelets and thromboxane A2 in hypertension: In some previous studies, certain platelet function variables have been reported to be abnormal in hyper- tensive patients. 24,25 These abnormalities may be cor- rected with control of blood pressure. However, plasma thromboxane B2 concentrations in the normal range in treated or untreated hypertensive subjects as seen in this study'do not suggest a primary role for thrombox- ane A2 in the genesis of essential hypertension. In a re- cent study, Frishman et a126 administered labetalol to patients with angina pectoris and hypertension, but failed to observe any significant antiplatelet effects of labetalol. This is in agreement with our observations of unchanged plasma thromboxane B2 concentrations in hypertensive patients.

Implications: We therefore conclude that prosta- cyclin synthesis is increased in hypertensive human subjects; this may be a protective mechanism against

February 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 51 467

organ damage. Increase in prostacyclin synthesis occurs most probably in response to decreased blood flow and increased intravascular stress. As blood pressure is de- creased, prostacyclin synthesis returns to normal levels. However, as in other biologic systems, the adjustment of vascular tone and regulation of prostaglandin syn- thesis by vessel wall does not occur immediately.

Acknowledgment: We thank Pamlette E. Kinsey for sec- retarial assistance.

References

1. Gryglewski RJ, Bunting S, Moncada S, Vane JR. Arterial walls are protected against deposition of platelet thrombl by a substance (PGX) which they make from endoperoxide Prostaglandins 1976;12:685-713.

2. Hamber9 M, Svenson J, Samuelseon B. Thromboxanes. A new group of biologically active compounds derived from prostaglandin endoperoxides. Proc Natl Acad Sol 1975;72:2994-2998.

3. Moncada S, Gryglewski RJ, Bunting S, Vane JR. An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Nature 1976;263:663-665.

4. Pifer DD, Cagen LM, Chesney CM. Stability of prostaglandin 12 in human blood. Prostaglandms 1971 ;21:165-175

5. Moncada S, Vane JR. The discovery of prostacyclin: a fresh insight mto arachidonic acid metabolism. In. Kharasch N, Fried J, eds. Biochemical aspects of prostaglandms and thromboxanes. New York: Academic Press, 1968:155-177

6. Svenson J, Hamberg M, Samuelsson B. On the formation and effects of thromboxane A2 in platelets. Acta Physlol Scand 1976;98:285-295.

7. Needleman P, Minkes M, Raz A. Thromboxanes: selective biosynthesis and distinct biological properties Science 1976;193:163-165.

8. Patak RV, Mookerjee BK, Bentzel CJ, Hysert PE, Babej M, Lee JB. An- tagonism of the effects of furosemlde by indomethacin in normal and hy- pertensive man Prostagland=ns 1975; 10:649-659.

9. Durao V, Prata MM, Goncalves LMP. Modff=cation of anti-hypertensive effect of beta-adrenoreceptor blocking agents by inhibition of endogenous pros- taglandin synthesis Lancet 1977;2:1005-1007.

19. Slack BL, Warner ME, Keiser HR. The effect of prostaglandin synthetase inhibition on the action of hydralazine (abstr). Circulation 1978;58:Suppl

11:11-21 11. Easton PA, Koval A. Hypertension reaction with sulindac. Can Med Assoc

J 1980;122:1273-1274 12. Grove JH, Lebel M, Gbeassor FM. Variations in urinary metabohtes of

prostacyclin and thromboxane A2 in essential hypertension (abstr). Clin Res 1980;28:685 A.

13. Tan SY, Sweel P, Mulrew PJ. impaired renal production of prostaglandin E2" a newly identified lesion in human essential hypertension. Prostaglandins 1979;15:139-149

14. Okuma M, Yameri Y, Ohta K, Ichino H. Production of prostacyclin like substance in stroke-prone and stroke resistant spontaneously hypertensive rats Prostaglandins 1979;17"1-7.

15. Pace-Asciak CR, Carrara MC, Rangaraj G, Nicolaou KC. Enhanced for- mation of PGI2, a potent hypotensive substance, by aortic rings and ho- mogenates of spontaneously hypertensive rats. Prostaglandins 1978; 15: 1005-1012.

16. Collier JG, Dawnay NAH, Nachev CH, Robinson BF. Clinical investigation of an antagonist of c~ and ~-adrenorecepors-AH 5158 A. Br J Pharmacol 1972;44:286-293.

17. Mebta J, Cohn JN. Hemodynamic effects of labetalol, an alpha and beta blocking agent, in hypertensive subjects. Circulation 1977;55:370-375.

18. Mehta J, Mehta P. Effect of propranolol therapy on platelet release and prostaglandin generation in patients with coronary heart disease. Ctrculation 1982; 66:1294-1299.

19. Mehta J, Mehta P. Dipyndamole and aspirin in relation to platelet aggre- gation and vessel wall prostaglandin generation J Cardiovasc Pharmacol 1982;4:688-693.

20. Mehta P, Mehta J. Prostacyolin and thromboxane release following me- chanical trauma to human vessels (abstr). Blood 1981;58 Suppl:236A.

21. Mehta J, Mehta P, Horalek C. The significance cf platelet-vessel wall prostaglandin equilibrium during exercise-induced stress. Am Heart J 1982, in press.

22. Mebta P, Mehta J. Influence of nicotine on platelet and vessel wall pros- taglandms (abstr). Clin Res 1981;29.814 A.

23. Neri Serneri GG, Masotti G, Poggeni L, Galanti G. Release of prostacyclin rote the blood stream and ~ts exhaustion m humans after local blood flow changes (ishcemia and venous stasis). Thromb Res 1980; 17:197-208.

24. Poplawski A, $korulska M, Niewiarowski S. Increased platelet adhe- sweness in hypertensive cardiovascular disease. Atherosclerosis 1981; 1:721-723.

25. Mehta J, Mehta P. Platelet function in hypertens=on and effect of therapy. Am J Cardio11981;47:331-334.

26. Frlshman WH, Strom JA, Kirschner M, Poland M, Klein N, Halprin S, Le- Jemtel TH, Kram M, Sonnenblick EH. Labetalol therapy in patients with systemic hypertension and angina pectoris: effects of combined alpha and beta-adrenoreceptor blockade. Am J Cardlol 1981 ;48:917-928.