- Email: [email protected]
Effects of prazosin and propranolol on blood lipids and lipoproteins in hypertensive patients
Effects of Prazosin and Propranolol on Blood Lipids and Lipoproteins in Hypertensive Patients
MANUEL EVA
VELASCO,
HURT,
HONOR10
SILVA,
ADALBERTO OTTO EDGARDO GERMAN Caracas,
Prazosin and propranolol were compared in an open, crossover study to determine their effects on plasma lipids and lipoproteins. After a four-week placebo period, 10 hypertensive patients were randomly assigned to prazosin treatment (Group I) and another 10 to propranolol treatment (Group II) for eight weeks. After a second four-week placebo period, treatment in each group was switched to the alternative drug for eight weeks. The mean blood pressure was reduced to normal levels (diastolic blood pressure less than or equal to 90 mm Hg) by both drugs-prazosin (1 to 8 mg per day) and propranolol(40 to 240 mg per day). The results of the study indicate that prazosin decreases serum cholesterol levels. In contrast, propranolol not only increases serum triglyceride levels and very-lowdensity lipoprotein cholesterol, but decreases total high-density Iipoproteln cholesterol, high-density lipoprotein, cholesterol, hlghdensity lipoprotein,, and apoprotein A-l. The data suggest that propranolol may induce significant, potentially atherogenic changes in lipid metabolism. whereas prazosin may represent an advantageous alternative as an antihypertensive agent, especially in subjects with an already atherogenic lipoprotein profile.
M.D.
M.Sc. M.D.
URBINA-QUINTANA,
HERNANDEZ-PIERElTI, FELDSTEIN, CAMEJO,
M.D. M.D.
M.D. Ph.D.
Venezuela
From the Section of Clinical Pharmacology, Division of Cardiology, Vargas Medical School, Laboratorio de Lipoproteinas y Aterogenesis, Centro de Biofisica y Bioquimica, lnstituto Venezolano de Investigaciones Cientfficas, Caracas, Venezuela, and Pfizer International, New York, New York. This work was supported by a grant from Pfizer Intemational. Requests for reprints should be addressed to Dr. Manuel Velasco, Apartado Postal 73333, El Marques, Caracas 1070-A, Venezuela.
February
Although hypertension is a well-established coronary risk factor [l], most controlled, randomized hypertension drug trials with untreated or placebo-treated control groups have failed to show that reducing blood pressure has a definite preventive effect on the incidence of coronary heart disease. In contrast, this effect has been observed in cerebrovascular disease and renal failure [2]. This fact has led to some speculation concerning the possible adverse metabolic effects of commonly used antihypertensive agents on blood lipid levels [3]. Several investigators have demonstrated that propranolol, a known beta-adrenergic blocker, reduces high-density lipoprotein cholesterol and increases total serum triglyceride levels in hypertensive patients [4,5]. Such adverse lipid effects could negate the benefits in blood pressure reduction achieved with propranolol. Furthermore, high-density lipoproteinp cholesterol levels have been inversely correlated with coronary heart disease risk, and therefore this fraction has become an even more important parameter to evaluate than total high-density lipoprotein cholesterol [6]. In an open, crossover trial, we studied the effects of propranolol and prazosin (a selective postsynaptic alpha,-adrenergic blocker) on plasma lipids and lipoproteins in two groups of male patients with mild to moderate essential hypertension. Our results indicate that propranolol may induce significant, potentially atherogenic changes. In contrast, prazosin produced no adverse effects on plasma lipids and lipoproteins.
14, 1966
The American
Journal
of Medicine
Volume
66 (suppl2A)
199
SYMPOSIUM
ON INITIAL ANTIHYPERTENSIVE
THERAPY-VELASCO
ET AL
PATIENTS AND METHODS Twenty male patients with mild to moderate essential hypertension were studied in an outpatient hypertension clinic. The patients, who ranged in age from 39 to 75 years, were given physical examinations at the start of the study. In addition, complete clinical histories were obtained. We defined normal levels of plasma cholesterol as those below 220 mg/dl and normal plasma triglyceride levels as those less than 150 mg/dl. Within the group, type Ila and type Ilb hyperlipidemic patients were identified. Results of other conventional hematologic and urine tests were normal in all’subjects. Experimental Design. The design of the study was an open, crossover, comparative program of prazosin and propranolol. After an initial placebo period of four weeks (Placebo l), 10 patients were randomly assigned to receive prazosin and another 10 to receive propranolol for eight weeks. There were no dropouts. At the end of a second four-week washout placebo period (Placebo 2), treatment in each group of 10 patients was switched to the alternative drug for eight weeks. Dosages ranged from 1 to 8 mg per day for prazosin, and 40 to 240 mg per day for propranolol. As much as possible, patients’ diets, daily activities, weight, and smoking habits were kept constant throughout the study. Clinical Measurements. Blood pressure was measured with a mercury sphygmomanometer, and heart rate was determined by the digital method in supine and standing positions. Clinical evaluations were performed every two weeks. Determination of Blood Lipids and Lipoproteins. At the end of each phase, blood samples were obtained by venipuncture after an overnight fast of 12 to 14 hours. To obtain plasma, blood was collected in tubes containing a mixture of preservants and centrifuged for 10 minutes at 2,500 revolutions per minute at room temperature. The final preservants’ concentrations in plasma were: ethylene diaminetetraacetic acid, 2.5 mmol; thimerosal, 0.05 percent; hydrochloric acidbenzamidine, 60 pmol; and gamma-amino-n-caproic acid, 50 pmol. Plasma samples were stored at 4°C and analyzed within six days. Lipoproteins were fractionated using the exponential potassium bromide gradient of Redgrave et al [7], with some modifications: 3 ml of plasma, prestained with Sudan Black B 0.65 percent (weight per volume) in ethylene glycol, was adjusted to a density of 1.250 g/ml with solid potassium bromide and pipetted into 13.5-ml polyallomer centrifuge tubes. A discontinuous gradient was formed by carefully layering 1 ml of potassium bromide solution of density 1.210 g/ml above the plasma, followed by 3 ml of potassium bromide solution of density 1.063 g/ml, and another 3 ml of density 1.019 g/ml. The final potassium bromide layer was a 1 -ml solution of density 1.006 g/ml. All the solutions contained the same mixture of preservants that were used in preparing the plasma samples. Samples were centrifuged for 24 hours at 20°C in a SW 41 Beckman rotor at 176 x g. Following centrifugation, the lipoprotein bands were clearly visible. The very-low-density lipoproteins, low-density lipoproteins, high-density lipoprotein,, and high-density lipoprotein, were dialyzed over 72 hours with a tris-hydrochloric acid buffer of pH 8.2 containing the
110
February
14,1986
The American
Journal
ol Medicine
mixture of preservants described earlier. The buffer was changed every 24 hours to remove the potassium bromide. After dialysis was completed, all the fractionated lipoproteins were immediately analyzed. An inherent problem of this methodology is that only approximately 75 to 85 percent of cholesterol is extracted during ultracentrifugation. This accounts for the fact that total cholesterol values are not equal to the sum of the values for very-low-density lipoprotein, lowdensity lipoprotein, and high-density lipoprotein. Subsequent statistical calculation of high-density lipoprotein, and highdensity lipoprotein, is, therefore, also not equal to the total high-density lipoprotein value. Quantification of apoprotein A-l in high-density lipoprotein, was made by an immunoelectrophoresis assay [S]. The following parameters were measured after each phase of placebo and drug: total plasma cholesterol, very-low-density lipoprotein cholesterol, low-density lipoprotein cholesterol, highdensity lipoprotein, cholesterol, high-density lipoprotein, cholesterol [9], plasma triglycerides [lo], and plasma unesterified fatty acids [I 11. Statistical Analysis. Averages of clinical and biochemical parameters during each placebo period were compared with those from each drug treatment period by paired Student t tests [12].
RESULTS Effects of Prazosin and Propranolol on Blood Preasure and Heart Rate. Blood pressure and heart rate data for Groups I and II are listed in Tables I and II. There were no significant baseline differences in blood pressure and heart rate between groups. In both treatment groups, significant differences in systolic and diastolic blood pressures (in supine and standing positions) were noted between active treatment and placebo periods, but no significant differences were noted between the drug-treatment periods. Prazosin did not change heart rate significantly, although propranolol did decrease it significantly. Effects of Prazosin and Propranolol on Plasma Lipids and Lipoproteins. In Group I (Table Ill), total plasma cholesterol concentration decreased significantly after prazosin administration, probably as a consequence of the small decrease observed in very-low-density lipoprotein cholesterol and low-density lipoprotein cholesterol. There
was also a nonsignificant
decrease
in total high-
density lipoprotein cholesterol. After treatment for this group was switched to propranoloI, total plasma triglyceride levels increased by 46 percent, and very-low-density lipoprotein triglycerides also increased significantly. Although total plasma cholesterol was unchanged, total high-density lipoprotein cholesterol decreased significantly. In addition, high-density lipoprotein:! cholesterol levels and apoprotein A-l content diminished. The decrease in high-density lipoprotein2 cholesterol was not statistically significant. In Group II (Table IV), the most striking result was a significant 50 percent decrease in high-density lipopro-
Volume
80
(suppl 2A)
SYMPOSIUM
ON INITIAL ANTIHYPERTENSIVE
THERAPY-VELASCO
Effects of Pratosin and Propranolol on Supine Blood Pressure and Heart Rate in Group I
TABLE I
Parameter (mean f SEM)
Placebo 1 (n = 10)
Prazosin (II = 10)
Placebo 2 (n = 9)
Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Heart rate (beatsimin) Mean drug dose (mg/day)
163 -t 5.70 106 + 1.90 70 ‘- 2.53 0
142 2 2.85 90” 1.27 69 2 1.58 3.15 k 0.70
160 2 3.16 106 + 1.58 70 -c 1.90 0
‘p ~0.001 +p ~0.005
compared compared
TABLE II
with value with value
for preceding for preceding
placebo placebo
142 90 54 124
t? P t
4.75 1.90 5.06+ 17.34
period. period.
Placebo 1 (n = 10)
Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Heart rate (beats/min) Mean drug dose (mg/day) compared compared
TABLE Ill
with value with value
for preceding for preceding
placebo placebo
136 89 56 108
2 * ” +-
Placebo 2 (n = 9)
4.26 2.60” 1.54’ 14.68
Prezosin (n = 9)
159 k 3.68 107 2 2.79 76 t 3.77 0
140 91 70 2.33
+ 2.62 2 3.37’ I+_3.32 2 0.78
period. period.
Effect of Prazosin and Propranolol Treatment on Plasma Lipids and Lipoproteins in Group I Placebo 1 (n = 10)
Total cholesterol (mg/dl) Total triglycerides (tigidl) Nonesterified fatty acid (mgidl) Very-low-density lipoprotein cholesterol (mgidl) Low-density lipoprotein cholesterol (mg/dl) High-density lipoprotein2 chole$terol (mgidl) High-density lipoprotein, and apoprotein A-l (mg/dl) High-density lipoprotein, cholesterol (mg/dl) Total high-density lipoprotein cholesterol (mg/dl) comparec
TABLE IV
Proprenolol (n = 10)
153 k 3.31 107 t 2.89 71 t 1.84 0
Parameter (mean + SEM)
‘p CO.05
Propranolol (n = 9)
Effects of Prarosin and Propranolol dh Supine Blood Pressure and Heart Rate in Group II
Parameter (mean + SEM)
“p ~0.001 +p ~0.005
ET AL
with value
for preceding
placebo
269 165 12.2 33.0 154 9.9 14.6 24.6 35.5
‘+ 2 ? 2 + -c + 2
Prazosln (n = 10)
13.61 17.09 1.74 4.53 9.48 1.55 2.82 1.96 3.29
239 142 11.5 25.4 142 9.4 10.4 23.5 29.9
Placebo 2 (n = 9)
f 11.38” k 11.71 -r- 0.95’ -t 2.85 t 7.59 2 1.46 2 1.23 k 1.77 t 2.53
228 135 10.2 28.2 145 8.5 10 26.7 35.0
Pmprenotol (n = 9)
k 25 f 19.94 t 1.17 ” 3.58 -t 4.75 of: 1.36 f 1.71 f 1.58 f 2.66
230 198 13.9 29.1 142 6.2 8.8 24.5 30.3
t r 2 r r t + k r
18.67 18.04* 1.20 3.16 4.75 0.89 1.39 1.11 1.90’
period.
Effect of Propranolol and Prazosin Treatment on Plasma Lipids and Lipoproteins in Group II Placebo 1 (n = IO)
Parameter (mean + SEM) Total cholesterol (mg/dl) Total triglycerides (mg/dl) Nonesterified fatty acid (mgldl) Very-low-density lipoprotein cholesterol (mgidl) Low-density lipoprotein cholesterol (mg/dl) High-density lipoprotein, cholesterol (mgidl) High-density lipoprotein2 and apoprotein A-l (mg/dl) High-density lipoproteina cholesterol (mg/dl) Total high-d&&y lipoprotein cholesterol (mg/dl) ‘p ~0.05 compared +p ~0.001 compared
with value for preceding with value for preceding
231 182 14.5 32.7 128 8.3 11.8 24.4 31.4
2 k f ” 2 4 2 f f
Proprenolol (n = 10)
14.24 30.70 2.25 5.70 5.81 0.89 1.65 2.53 2.85
241 212 12.8 41.1 121 4.2 5.2 24.9 28.9
f k k -c -t * * 5 s
13.61 24.68 1.52 8.64 7.28 0.51+ 0.95+ 1.74 1.74
Placebo 2 (n = 7) 242 164 17.0 40.8 135 4.2 8.1 28.4 34.7
+ k t -+ c k f f k
Pramsln (n = 7)
10.57 27.17 3.25 9.70 9.81 1.36 1.32 1.51 2.23
248 150 18.0 33.0 124 6.6 9.2 29.4 35.7
+ 2 5 t k rt 2 2 2
10.76 11.39 2.34 5.16 10.13 0.47 0.92 2.75 2.54
placebo period. placebo period.
February
14, 1986
The
AmericW
Journal
of Medlclne
Volume
80
(suppl2A)
111
SYMPOSIUM
ON INITIAL ANTIHYPERTENSIVE
THERAPY-VELASCO
ET AL
teinp cholesterol and apoprotein A-l levels, in comparison with placebo baseline values, after eight weeks of propranolol administration. However, the contribution of highdensity lipoprotein2 to the total cholesterol concentration is small, which could explain why this significant diminution was not reflected in the total plasma cholesterol level. Total plasma triglycerides ahd very-low-density lipoprotein cholesterol levels increased with propranolol therapy and decreased with prazosin therapy, but these changes were not significant. After prazosin administration, there was a statistically nonsignificant increase in high-density lipoprotein,! cholesterol and apoprotein A-l levels, but these values still remained lower than the baseline Placebo 1 values. COMMENTS
In the current study, we evaluated the effects of prazosin, a postsynaptic alpha,-adrenergic blocker, and propranoloI, a nonselective beta-adrenergic blocker, on lipid and lipoprotein levels in patients with mild to moderate essential hypertension. Clinical and epidemiologic studies have demonstrated that high-density lipoprotein2 cholesterol and apoprotein A-l contents are inversely correlated with coronary heart disease risk [6,13]. High-density lipoproteinp is relatively richer in apoprotein A-l than in other apoproteins and is, therefore, a parameter to be evaluated in addition to high-density lipoprotein2 cholesterol levels. Apoprotein A-l acts as a cholesterol acceptor and activates lecithin-cholesterol acyl transferase, the enzyme responsible for cholesterol esterification in plasma [14]. In agreement with other studies [15], our results indicate that prazosin has no adverse effects on plasma lipid and lipoprotein levels. High-density lipoprotein, cholesterol and apoprotein A-l were not changed significantly by the administration of prazosin. The effects of propranolol on the treatment groups were different. When propranolol was administered first, as in Group II, high-density lipoprotein2 cholesterol and apopro-
tein A-l content decreased by 50 percent. In Group I, which received propranolol last, plasma high-density lipoprotein2 levels decreased (but not significantly) and total plasma triglycerides increased by 46 percent. Apparently, the intermediate four-week washout period between the two drugs was insufficient to completely eliminate the effect of the preceding drug. The results suggest that propranolol used as antihypertensive monotherapy may induce significant, potentially atherogenic changes in lipid metabolism. The precise mechanisms by which beta-adrenergic antagonists modify lipid metabolism remain unknown. However, many studies [16,17] agree that the decreased levels of high-density lipoprotein and increased levels of total triglyceride may be due to the inhibition of lipoprotein lipase activity. Increased lipoprotein synthesis would not be expected to occur during propranolol therapy, because the blood levels of free fatty acids, glucose, and insulin, which represent the most important precursors for hepatic lipoprotein production, have been reported to be unaltered or even decreased in most studies [18]. Despite this evidence, the reason for the apparent reduced lipoprotein lipase activity was not observed in this study. The activity of lecithin-cholesterol acyl transferase has been reported to be diminished during propranolol therapy [19], and we belleve that this effect could be a consequence of the low content of apoprotein A-l that we have detected, specifically in high-density lipoprotein2 cholesterol, and that has been detected by Fager et al [4] in total plasma. In conclusion, our results suggest that the use of prazosin as an antihypertensive agent may represent an advantageous alternative to beta blockers, particularly in subjects with already documented atherogenic lipoprotein profiles. ACKNOWLEDGMENT
The secretarial help of Mrs. lngrid Hernandez-Ramirez is greatly appreciated.
REFERENCES 1. 2. 3.
4.
5.
112
Castelli WP: Epidemiology of coronary heart disease: the Framingham Study. Am J Med 1984; 78: 4-12. Zanchetti A: Treatment goals in hypertension. Am J Med 1984; 76: 1-3. Lowenstein J, Neusy A-J: The biochemical effects of antihypertensive agents and the impact on atherosclerosis. J Cardiovasc Pharmacol 1982; 4 (suppl2): S262-S264. Fager G, Berglund G, Bondjers G, et al: Effects of antihypertensive therapy on serum lipoproteins: treatment with metoprolol, propranolol, and hydrochlorothiazide. Artery 1983; 11: 283296. Leren P, Helgeland A, Holme I, et al: Effect of propranolol and prazosin on blood lipids: the Oslo Study. Lancet 1980; II: 4-6.
Februery
14,1986
The Americen
Journal
of Medicine
6.
7.
8.
9.
Volume
Miller NE, Hammett F, Saltissi S, et al: Relation of angiographitally defined coronary artery disease to plasma lipoprotein subfractions and apolipoproteins. Br Med J 1981; 282: 17411744. Redgrave TG, Roberts DCK, West CE: Separation of plasma lipoproteins by density-gradient ultracentrifugation. Anal Biothem 1975; 65: 42-49. Sakai Y, ltajura K, Kanada T, et al: Quantitation of apolipoprotein A-l in pooled human serum by single radial immunodiffusion and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Anal Biochem 1984; 137: 1-7. Bowman RE, Woff RC: A rapid and specific ultramicro method for total serum cholesterol. Clin Chem 1962; 8: 302-309.
80 (suppl2A)
SYMPOSIUM
10.
11.
12. 13.
14.
Biggs HG, Erikson JM, Moorehead WR: A manual calorimetric assay of triglycerides in serum. Clin Chem 1975; 21: 437441. Duncombe WG: The calorimetric micro-determination of nonesterified fatty acids in plasma. Clin %hem Acta 1964; 9: 122125 Snedecor GW, Cochram WH: Statistical methods, sixth ed. Ames, Iowa: The Iowa State University Press, 1984, 59-65. Durrington PN: High-density lipopiotein cholesterol: methods and clinical significance. CRC. Reviews in Clinical Laboratory Sciences 1982; 18: 31-77. Eisenberg S: Lipoproteins and lipoprotein metabolism: A dynamic evaluation of the plasma fat transport system. Klin Wochenschr 1983; 61: 119-132.
Febiuary
14, 1986
15.
16. 17.
18.
19.
ON INITIAL ANTIHYPERTENSIVE
THERAPY-VELASCO
ET AL
Velasco M, Silva H, Morillo J, et al: Effect of prazosin on blood lipids and on thyroid function in hypertensive patients. J Cardiovasc Pharmacol 1982; 4 (suppl 2): S225-S227. Barboriak JJ, Friedberg HD: Propranolol and hypertriglyceridemia. Atherosclerosis 1973; 17: 31-35. Tanaka N, Sakaguchi S, Oshige K, et al: Effect of. chronic administration of propranolol on lipoprotein composition. Metabolism 1976; 25: 1071-1075. Day JL, Simpson N, Metcalfe J, Page RL: Metabolic consequences of atenolol and propranolol in treatment of essential hypertension. Br Med J 1979; 1: 77-80. Goto Y: Effects of alpha- and beta-blocker antihypertensive therapy on blood lipids: a multicenter trial. Am J Med 1984; 76: 72-78.
The Amerlcen
Journal
of Medicine
Volume
80 (suppl
2A)
113
MANUEL EVA
VELASCO,
HURT,
HONOR10
SILVA,
ADALBERTO OTTO EDGARDO GERMAN Caracas,
Prazosin and propranolol were compared in an open, crossover study to determine their effects on plasma lipids and lipoproteins. After a four-week placebo period, 10 hypertensive patients were randomly assigned to prazosin treatment (Group I) and another 10 to propranolol treatment (Group II) for eight weeks. After a second four-week placebo period, treatment in each group was switched to the alternative drug for eight weeks. The mean blood pressure was reduced to normal levels (diastolic blood pressure less than or equal to 90 mm Hg) by both drugs-prazosin (1 to 8 mg per day) and propranolol(40 to 240 mg per day). The results of the study indicate that prazosin decreases serum cholesterol levels. In contrast, propranolol not only increases serum triglyceride levels and very-lowdensity lipoprotein cholesterol, but decreases total high-density Iipoproteln cholesterol, high-density lipoprotein, cholesterol, hlghdensity lipoprotein,, and apoprotein A-l. The data suggest that propranolol may induce significant, potentially atherogenic changes in lipid metabolism. whereas prazosin may represent an advantageous alternative as an antihypertensive agent, especially in subjects with an already atherogenic lipoprotein profile.
M.D.
M.Sc. M.D.
URBINA-QUINTANA,
HERNANDEZ-PIERElTI, FELDSTEIN, CAMEJO,
M.D. M.D.
M.D. Ph.D.
Venezuela
From the Section of Clinical Pharmacology, Division of Cardiology, Vargas Medical School, Laboratorio de Lipoproteinas y Aterogenesis, Centro de Biofisica y Bioquimica, lnstituto Venezolano de Investigaciones Cientfficas, Caracas, Venezuela, and Pfizer International, New York, New York. This work was supported by a grant from Pfizer Intemational. Requests for reprints should be addressed to Dr. Manuel Velasco, Apartado Postal 73333, El Marques, Caracas 1070-A, Venezuela.
February
Although hypertension is a well-established coronary risk factor [l], most controlled, randomized hypertension drug trials with untreated or placebo-treated control groups have failed to show that reducing blood pressure has a definite preventive effect on the incidence of coronary heart disease. In contrast, this effect has been observed in cerebrovascular disease and renal failure [2]. This fact has led to some speculation concerning the possible adverse metabolic effects of commonly used antihypertensive agents on blood lipid levels [3]. Several investigators have demonstrated that propranolol, a known beta-adrenergic blocker, reduces high-density lipoprotein cholesterol and increases total serum triglyceride levels in hypertensive patients [4,5]. Such adverse lipid effects could negate the benefits in blood pressure reduction achieved with propranolol. Furthermore, high-density lipoproteinp cholesterol levels have been inversely correlated with coronary heart disease risk, and therefore this fraction has become an even more important parameter to evaluate than total high-density lipoprotein cholesterol [6]. In an open, crossover trial, we studied the effects of propranolol and prazosin (a selective postsynaptic alpha,-adrenergic blocker) on plasma lipids and lipoproteins in two groups of male patients with mild to moderate essential hypertension. Our results indicate that propranolol may induce significant, potentially atherogenic changes. In contrast, prazosin produced no adverse effects on plasma lipids and lipoproteins.
14, 1966
The American
Journal
of Medicine
Volume
66 (suppl2A)
199
SYMPOSIUM
ON INITIAL ANTIHYPERTENSIVE
THERAPY-VELASCO
ET AL
PATIENTS AND METHODS Twenty male patients with mild to moderate essential hypertension were studied in an outpatient hypertension clinic. The patients, who ranged in age from 39 to 75 years, were given physical examinations at the start of the study. In addition, complete clinical histories were obtained. We defined normal levels of plasma cholesterol as those below 220 mg/dl and normal plasma triglyceride levels as those less than 150 mg/dl. Within the group, type Ila and type Ilb hyperlipidemic patients were identified. Results of other conventional hematologic and urine tests were normal in all’subjects. Experimental Design. The design of the study was an open, crossover, comparative program of prazosin and propranolol. After an initial placebo period of four weeks (Placebo l), 10 patients were randomly assigned to receive prazosin and another 10 to receive propranolol for eight weeks. There were no dropouts. At the end of a second four-week washout placebo period (Placebo 2), treatment in each group of 10 patients was switched to the alternative drug for eight weeks. Dosages ranged from 1 to 8 mg per day for prazosin, and 40 to 240 mg per day for propranolol. As much as possible, patients’ diets, daily activities, weight, and smoking habits were kept constant throughout the study. Clinical Measurements. Blood pressure was measured with a mercury sphygmomanometer, and heart rate was determined by the digital method in supine and standing positions. Clinical evaluations were performed every two weeks. Determination of Blood Lipids and Lipoproteins. At the end of each phase, blood samples were obtained by venipuncture after an overnight fast of 12 to 14 hours. To obtain plasma, blood was collected in tubes containing a mixture of preservants and centrifuged for 10 minutes at 2,500 revolutions per minute at room temperature. The final preservants’ concentrations in plasma were: ethylene diaminetetraacetic acid, 2.5 mmol; thimerosal, 0.05 percent; hydrochloric acidbenzamidine, 60 pmol; and gamma-amino-n-caproic acid, 50 pmol. Plasma samples were stored at 4°C and analyzed within six days. Lipoproteins were fractionated using the exponential potassium bromide gradient of Redgrave et al [7], with some modifications: 3 ml of plasma, prestained with Sudan Black B 0.65 percent (weight per volume) in ethylene glycol, was adjusted to a density of 1.250 g/ml with solid potassium bromide and pipetted into 13.5-ml polyallomer centrifuge tubes. A discontinuous gradient was formed by carefully layering 1 ml of potassium bromide solution of density 1.210 g/ml above the plasma, followed by 3 ml of potassium bromide solution of density 1.063 g/ml, and another 3 ml of density 1.019 g/ml. The final potassium bromide layer was a 1 -ml solution of density 1.006 g/ml. All the solutions contained the same mixture of preservants that were used in preparing the plasma samples. Samples were centrifuged for 24 hours at 20°C in a SW 41 Beckman rotor at 176 x g. Following centrifugation, the lipoprotein bands were clearly visible. The very-low-density lipoproteins, low-density lipoproteins, high-density lipoprotein,, and high-density lipoprotein, were dialyzed over 72 hours with a tris-hydrochloric acid buffer of pH 8.2 containing the
110
February
14,1986
The American
Journal
ol Medicine
mixture of preservants described earlier. The buffer was changed every 24 hours to remove the potassium bromide. After dialysis was completed, all the fractionated lipoproteins were immediately analyzed. An inherent problem of this methodology is that only approximately 75 to 85 percent of cholesterol is extracted during ultracentrifugation. This accounts for the fact that total cholesterol values are not equal to the sum of the values for very-low-density lipoprotein, lowdensity lipoprotein, and high-density lipoprotein. Subsequent statistical calculation of high-density lipoprotein, and highdensity lipoprotein, is, therefore, also not equal to the total high-density lipoprotein value. Quantification of apoprotein A-l in high-density lipoprotein, was made by an immunoelectrophoresis assay [S]. The following parameters were measured after each phase of placebo and drug: total plasma cholesterol, very-low-density lipoprotein cholesterol, low-density lipoprotein cholesterol, highdensity lipoprotein, cholesterol, high-density lipoprotein, cholesterol [9], plasma triglycerides [lo], and plasma unesterified fatty acids [I 11. Statistical Analysis. Averages of clinical and biochemical parameters during each placebo period were compared with those from each drug treatment period by paired Student t tests [12].
RESULTS Effects of Prazosin and Propranolol on Blood Preasure and Heart Rate. Blood pressure and heart rate data for Groups I and II are listed in Tables I and II. There were no significant baseline differences in blood pressure and heart rate between groups. In both treatment groups, significant differences in systolic and diastolic blood pressures (in supine and standing positions) were noted between active treatment and placebo periods, but no significant differences were noted between the drug-treatment periods. Prazosin did not change heart rate significantly, although propranolol did decrease it significantly. Effects of Prazosin and Propranolol on Plasma Lipids and Lipoproteins. In Group I (Table Ill), total plasma cholesterol concentration decreased significantly after prazosin administration, probably as a consequence of the small decrease observed in very-low-density lipoprotein cholesterol and low-density lipoprotein cholesterol. There
was also a nonsignificant
decrease
in total high-
density lipoprotein cholesterol. After treatment for this group was switched to propranoloI, total plasma triglyceride levels increased by 46 percent, and very-low-density lipoprotein triglycerides also increased significantly. Although total plasma cholesterol was unchanged, total high-density lipoprotein cholesterol decreased significantly. In addition, high-density lipoprotein:! cholesterol levels and apoprotein A-l content diminished. The decrease in high-density lipoprotein2 cholesterol was not statistically significant. In Group II (Table IV), the most striking result was a significant 50 percent decrease in high-density lipopro-
Volume
80
(suppl 2A)
SYMPOSIUM
ON INITIAL ANTIHYPERTENSIVE
THERAPY-VELASCO
Effects of Pratosin and Propranolol on Supine Blood Pressure and Heart Rate in Group I
TABLE I
Parameter (mean f SEM)
Placebo 1 (n = 10)
Prazosin (II = 10)
Placebo 2 (n = 9)
Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Heart rate (beatsimin) Mean drug dose (mg/day)
163 -t 5.70 106 + 1.90 70 ‘- 2.53 0
142 2 2.85 90” 1.27 69 2 1.58 3.15 k 0.70
160 2 3.16 106 + 1.58 70 -c 1.90 0
‘p ~0.001 +p ~0.005
compared compared
TABLE II
with value with value
for preceding for preceding
placebo placebo
142 90 54 124
t? P t
4.75 1.90 5.06+ 17.34
period. period.
Placebo 1 (n = 10)
Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Heart rate (beats/min) Mean drug dose (mg/day) compared compared
TABLE Ill
with value with value
for preceding for preceding
placebo placebo
136 89 56 108
2 * ” +-
Placebo 2 (n = 9)
4.26 2.60” 1.54’ 14.68
Prezosin (n = 9)
159 k 3.68 107 2 2.79 76 t 3.77 0
140 91 70 2.33
+ 2.62 2 3.37’ I+_3.32 2 0.78
period. period.
Effect of Prazosin and Propranolol Treatment on Plasma Lipids and Lipoproteins in Group I Placebo 1 (n = 10)
Total cholesterol (mg/dl) Total triglycerides (tigidl) Nonesterified fatty acid (mgidl) Very-low-density lipoprotein cholesterol (mgidl) Low-density lipoprotein cholesterol (mg/dl) High-density lipoprotein2 chole$terol (mgidl) High-density lipoprotein, and apoprotein A-l (mg/dl) High-density lipoprotein, cholesterol (mg/dl) Total high-density lipoprotein cholesterol (mg/dl) comparec
TABLE IV
Proprenolol (n = 10)
153 k 3.31 107 t 2.89 71 t 1.84 0
Parameter (mean + SEM)
‘p CO.05
Propranolol (n = 9)
Effects of Prarosin and Propranolol dh Supine Blood Pressure and Heart Rate in Group II
Parameter (mean + SEM)
“p ~0.001 +p ~0.005
ET AL
with value
for preceding
placebo
269 165 12.2 33.0 154 9.9 14.6 24.6 35.5
‘+ 2 ? 2 + -c + 2
Prazosln (n = 10)
13.61 17.09 1.74 4.53 9.48 1.55 2.82 1.96 3.29
239 142 11.5 25.4 142 9.4 10.4 23.5 29.9
Placebo 2 (n = 9)
f 11.38” k 11.71 -r- 0.95’ -t 2.85 t 7.59 2 1.46 2 1.23 k 1.77 t 2.53
228 135 10.2 28.2 145 8.5 10 26.7 35.0
Pmprenotol (n = 9)
k 25 f 19.94 t 1.17 ” 3.58 -t 4.75 of: 1.36 f 1.71 f 1.58 f 2.66
230 198 13.9 29.1 142 6.2 8.8 24.5 30.3
t r 2 r r t + k r
18.67 18.04* 1.20 3.16 4.75 0.89 1.39 1.11 1.90’
period.
Effect of Propranolol and Prazosin Treatment on Plasma Lipids and Lipoproteins in Group II Placebo 1 (n = IO)
Parameter (mean + SEM) Total cholesterol (mg/dl) Total triglycerides (mg/dl) Nonesterified fatty acid (mgldl) Very-low-density lipoprotein cholesterol (mgidl) Low-density lipoprotein cholesterol (mg/dl) High-density lipoprotein, cholesterol (mgidl) High-density lipoprotein2 and apoprotein A-l (mg/dl) High-density lipoproteina cholesterol (mg/dl) Total high-d&&y lipoprotein cholesterol (mg/dl) ‘p ~0.05 compared +p ~0.001 compared
with value for preceding with value for preceding
231 182 14.5 32.7 128 8.3 11.8 24.4 31.4
2 k f ” 2 4 2 f f
Proprenolol (n = 10)
14.24 30.70 2.25 5.70 5.81 0.89 1.65 2.53 2.85
241 212 12.8 41.1 121 4.2 5.2 24.9 28.9
f k k -c -t * * 5 s
13.61 24.68 1.52 8.64 7.28 0.51+ 0.95+ 1.74 1.74
Placebo 2 (n = 7) 242 164 17.0 40.8 135 4.2 8.1 28.4 34.7
+ k t -+ c k f f k
Pramsln (n = 7)
10.57 27.17 3.25 9.70 9.81 1.36 1.32 1.51 2.23
248 150 18.0 33.0 124 6.6 9.2 29.4 35.7
+ 2 5 t k rt 2 2 2
10.76 11.39 2.34 5.16 10.13 0.47 0.92 2.75 2.54
placebo period. placebo period.
February
14, 1986
The
AmericW
Journal
of Medlclne
Volume
80
(suppl2A)
111
SYMPOSIUM
ON INITIAL ANTIHYPERTENSIVE
THERAPY-VELASCO
ET AL
teinp cholesterol and apoprotein A-l levels, in comparison with placebo baseline values, after eight weeks of propranolol administration. However, the contribution of highdensity lipoprotein2 to the total cholesterol concentration is small, which could explain why this significant diminution was not reflected in the total plasma cholesterol level. Total plasma triglycerides ahd very-low-density lipoprotein cholesterol levels increased with propranolol therapy and decreased with prazosin therapy, but these changes were not significant. After prazosin administration, there was a statistically nonsignificant increase in high-density lipoprotein,! cholesterol and apoprotein A-l levels, but these values still remained lower than the baseline Placebo 1 values. COMMENTS
In the current study, we evaluated the effects of prazosin, a postsynaptic alpha,-adrenergic blocker, and propranoloI, a nonselective beta-adrenergic blocker, on lipid and lipoprotein levels in patients with mild to moderate essential hypertension. Clinical and epidemiologic studies have demonstrated that high-density lipoprotein2 cholesterol and apoprotein A-l contents are inversely correlated with coronary heart disease risk [6,13]. High-density lipoproteinp is relatively richer in apoprotein A-l than in other apoproteins and is, therefore, a parameter to be evaluated in addition to high-density lipoprotein2 cholesterol levels. Apoprotein A-l acts as a cholesterol acceptor and activates lecithin-cholesterol acyl transferase, the enzyme responsible for cholesterol esterification in plasma [14]. In agreement with other studies [15], our results indicate that prazosin has no adverse effects on plasma lipid and lipoprotein levels. High-density lipoprotein, cholesterol and apoprotein A-l were not changed significantly by the administration of prazosin. The effects of propranolol on the treatment groups were different. When propranolol was administered first, as in Group II, high-density lipoprotein2 cholesterol and apopro-
tein A-l content decreased by 50 percent. In Group I, which received propranolol last, plasma high-density lipoprotein2 levels decreased (but not significantly) and total plasma triglycerides increased by 46 percent. Apparently, the intermediate four-week washout period between the two drugs was insufficient to completely eliminate the effect of the preceding drug. The results suggest that propranolol used as antihypertensive monotherapy may induce significant, potentially atherogenic changes in lipid metabolism. The precise mechanisms by which beta-adrenergic antagonists modify lipid metabolism remain unknown. However, many studies [16,17] agree that the decreased levels of high-density lipoprotein and increased levels of total triglyceride may be due to the inhibition of lipoprotein lipase activity. Increased lipoprotein synthesis would not be expected to occur during propranolol therapy, because the blood levels of free fatty acids, glucose, and insulin, which represent the most important precursors for hepatic lipoprotein production, have been reported to be unaltered or even decreased in most studies [18]. Despite this evidence, the reason for the apparent reduced lipoprotein lipase activity was not observed in this study. The activity of lecithin-cholesterol acyl transferase has been reported to be diminished during propranolol therapy [19], and we belleve that this effect could be a consequence of the low content of apoprotein A-l that we have detected, specifically in high-density lipoprotein2 cholesterol, and that has been detected by Fager et al [4] in total plasma. In conclusion, our results suggest that the use of prazosin as an antihypertensive agent may represent an advantageous alternative to beta blockers, particularly in subjects with already documented atherogenic lipoprotein profiles. ACKNOWLEDGMENT
The secretarial help of Mrs. lngrid Hernandez-Ramirez is greatly appreciated.
REFERENCES 1. 2. 3.
4.
5.
112
Castelli WP: Epidemiology of coronary heart disease: the Framingham Study. Am J Med 1984; 78: 4-12. Zanchetti A: Treatment goals in hypertension. Am J Med 1984; 76: 1-3. Lowenstein J, Neusy A-J: The biochemical effects of antihypertensive agents and the impact on atherosclerosis. J Cardiovasc Pharmacol 1982; 4 (suppl2): S262-S264. Fager G, Berglund G, Bondjers G, et al: Effects of antihypertensive therapy on serum lipoproteins: treatment with metoprolol, propranolol, and hydrochlorothiazide. Artery 1983; 11: 283296. Leren P, Helgeland A, Holme I, et al: Effect of propranolol and prazosin on blood lipids: the Oslo Study. Lancet 1980; II: 4-6.
Februery
14,1986
The Americen
Journal
of Medicine
6.
7.
8.
9.
Volume
Miller NE, Hammett F, Saltissi S, et al: Relation of angiographitally defined coronary artery disease to plasma lipoprotein subfractions and apolipoproteins. Br Med J 1981; 282: 17411744. Redgrave TG, Roberts DCK, West CE: Separation of plasma lipoproteins by density-gradient ultracentrifugation. Anal Biothem 1975; 65: 42-49. Sakai Y, ltajura K, Kanada T, et al: Quantitation of apolipoprotein A-l in pooled human serum by single radial immunodiffusion and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Anal Biochem 1984; 137: 1-7. Bowman RE, Woff RC: A rapid and specific ultramicro method for total serum cholesterol. Clin Chem 1962; 8: 302-309.
80 (suppl2A)
SYMPOSIUM
10.
11.
12. 13.
14.
Biggs HG, Erikson JM, Moorehead WR: A manual calorimetric assay of triglycerides in serum. Clin Chem 1975; 21: 437441. Duncombe WG: The calorimetric micro-determination of nonesterified fatty acids in plasma. Clin %hem Acta 1964; 9: 122125 Snedecor GW, Cochram WH: Statistical methods, sixth ed. Ames, Iowa: The Iowa State University Press, 1984, 59-65. Durrington PN: High-density lipopiotein cholesterol: methods and clinical significance. CRC. Reviews in Clinical Laboratory Sciences 1982; 18: 31-77. Eisenberg S: Lipoproteins and lipoprotein metabolism: A dynamic evaluation of the plasma fat transport system. Klin Wochenschr 1983; 61: 119-132.
Febiuary
14, 1986
15.
16. 17.
18.
19.
ON INITIAL ANTIHYPERTENSIVE
THERAPY-VELASCO
ET AL
Velasco M, Silva H, Morillo J, et al: Effect of prazosin on blood lipids and on thyroid function in hypertensive patients. J Cardiovasc Pharmacol 1982; 4 (suppl 2): S225-S227. Barboriak JJ, Friedberg HD: Propranolol and hypertriglyceridemia. Atherosclerosis 1973; 17: 31-35. Tanaka N, Sakaguchi S, Oshige K, et al: Effect of. chronic administration of propranolol on lipoprotein composition. Metabolism 1976; 25: 1071-1075. Day JL, Simpson N, Metcalfe J, Page RL: Metabolic consequences of atenolol and propranolol in treatment of essential hypertension. Br Med J 1979; 1: 77-80. Goto Y: Effects of alpha- and beta-blocker antihypertensive therapy on blood lipids: a multicenter trial. Am J Med 1984; 76: 72-78.
The Amerlcen
Journal
of Medicine
Volume
80 (suppl
2A)
113