Forearm and finger hemodynamics, blood pressure control, and lipid changes in diabetic hypertensive patients treated with atenolol and prazosin

Forearm Pressure Diabetic Atenolol

and Finger Hemodynamics, Blood Control, and Lipid Changes in Hypertensive Patients Treated with and Prazosin

A Britzf ReporP CHI MINGK\VAN M.D.. AtExmm BRUCE

A. BROCKWAY, M.D. San

M.M. SHEPHERD, M.D., Ph.D.,JOHI~ JOHNSON, Ph.D.;W. FRED TAYLOR, M.S.,

~ntonm

Texas

Hypertension and diabetes mellitus frequently coexist and are independent risk factors for reduced peripheral perfusion. Antihypertensive medications that reduce blood pressure and improve peripheral perfusion would have advantages in diabetic hypertensive patients. In a randomized, two-placebo-period, single-blind, two-way, crossover study, finger and forearm blood flow, lipid levels, and blood pressure control were determined in 19 diabetic hypertensive patients given prazosin and atenolol, with each drug and placebo period lasting four weeks. Both drugs reduced blood pressure (sitting, 157195 to 112181 mm Hg for atenolol and 155195 to 138032 mm Hg for prazosin; standing, 153191 to 111181 mm Hg for atenolol and 151194 to 13381 mm Hg for prazosin). Lipid levels did not change, except that low-density lipoprotein levels decreased from 118 to 127 mgidl with prazosin. Atenolol did not change forearm or finger blood flow or vascular resistance. I’razosin increased blood flow and reduced vascular resistance in both finger and forearm. In conclusion, prazosin demonstrated a potentially more appropriate hemodynamic profile than atenolol in diabetic hypertensive patients in this study.

pitlemiologic surveys indicate that hypertension E occurs frequently in Jlatients \f,ith diabetes mellitus 111 and in subjects \vith a positive family history of diabetes 121. Although essential hypertension accounts for the majority of cases of high blood pressure in the diabetic population, diabetic nephropathy is a significant contributor as well [3]. In addition, diabetes mellitus and systemic hypertension are each risk factors for occlusive peripheral vascular disease [ S’i]. Such patients with hypertension and diabetes are likely to be in the older age g:rouJ) and are therefore more likely to have elevated peripheral resistance (81. Consequently, one concern during treatment for coiicomitant hypertension is to avoid using therapeutic agents that might further reduce peripheral blood flO\\:.

Beta-aclrenoceptor antagonists are commonly used in the treatment of hypertension, but on first administration they can cause reflex peripheral vasoconstriction secondarg to the primary reduction in cardiac output caused by beta,-aclrenoceJ1toi~ blockade. On the other hand, alpha-adrenoceptor blockade should reduce any vasoconstrictor effects of tonic sympathetic activity. Consequently, we hypothesized that equieffective blood pressure-lowering doses of prazosin and atenolol would have differing effects on vascular resistance and blood flow in the forearm and finger of diabetic hypertensive patients. The present study examines the blood flow in two peripheral sites, the forearm (muscle and skin) and finger (skin only), before and during four weeks of treatment with two commonly used antihypertensive agents, prazosin and atenolol. In addition, we \vere interested in examining the effects on blood lipid levels in these diabetic hyperteiisive patients since hyl~er~lil~oJ~roteinemia and a low high-density 1ipoJwotein:total cholesterol ratio are generally recognized as being risk factors for the development of atheromatous disease. The results of this study are presented in greatel* detail elsewhere [ 91. This brief report will present a summary of those findings.

PATIENTS AND METHODS Rventy patients, 11 men and nine women, bvith es-

r

-1 sential hypertension and diabetes mellitus were en-

L

1

Fror the Departments of Medune. Pharmacology and Physiology Uwers~ty of Texas Health Science Center. San Anton~o, Texas *Mater!ai in this a&k was adapted wth perm~won from Clinical Pharmacology and Therapeutics. Requests fat reprints should be addressed to Dr Alexander M M Shepherd. Department of Pharmacology. University of Texas Health Science Center, 7703 Floyd Curl Dwe San Antonio Texas X284-7764

rolled in a randomized, crossover study with foulphases, each lasting at least four Lveeks: Initial Leashout phase, first study drug phase, interim washout phase, and alternate study dmg phase. The average

January 23 1989

The Amerlcan Journal of Medlclne

Volume 86 (suppi 1B)

55

SYMPOSIUM ON CHD AND HYPERTENSION / KWAN ET AL

Clinical and Laboratory Measurements Subjects were seen at least weekly during washout and titration and every two weeks during maintenance. Body weight and blood pressure (three sitting and three standing) were recorded at each clinic visit. Adverse effects were reported at each visit. Blood lipid levels, diabetic control, and blood flow measurements were assessed by the methods described previously [9]. Forearm and finger blood flows were measured at the end of each phase, three times each minute, for a total of four minutes for forearm and five minutes for finger blood flow. For a particular session, the forearm blood flow was taken as the average of the last five measurements of the four-minute period. Finger blood flow was taken as the average of the last seven measurements of the five-minute period. More measurements of finger blood flow were included in the average because of the greater spontaneous variation in the level of blood flow. Finger blood flow was always measurecl first and from the same limb as forearm blood flow. Arterial systolic and diastolic blood pressures were measured noninvasively once per minute, in the right arm with a programmable electrosphygmomanometer (Narco Biosystems PE-300; Houston, Texas).

-18 Mean Change (mmHg) ~,2

1

-6 0 Sitting SSP

Sitting DBP

Standing SEP

Standlng DSP

_ _

_-

Figure 1. Kesponse ot slttlng and standlng blood pressure to therapy. *p
24 Mean forearm blwd flow (mllmlnllw mL) 1.6

Data Analysis Regional resistance was estimated as the ratio of mean arterial pressure to blood flow in that vascular bed, in this case either forearm or finger. Mean arterial pressure was taken at the time when the blood flow was recorded and calculated as diastolic plus one third of the pulse pressure. Group data are expressed as mean 5 SEM. Variation between treatments was analyzed by randomized-block analysis of variance. Duncan’s test was also applied to define the area of difference, whether or not the analysis of variance determined that there was a difference in the sample [lo]. Significance was defined as p ~0.05.

Figure 2. Effects of therapy on forearm blood flow. *p
“r

T

Figure 3. Effects of therapy on forearm vascular reslstance. *p 4025.

age of the group was 57 ? 2 (SEM), with an average of 42 to 68 years. Average weight was 175 + 6 pounds (range, 139 to 210 pounds). Qualifying blood pressure was a sitting diastolic blood pressure (Korotkoff phase 5) between 90 and 115 mm Hg, measured after four weeks of placebo therapy. Written informed consent was obtained from all patients. Following the initial washout period, patients were given either prazosin (titrated at 1, 2, 5, and 10 mg twice daily; maximal dose, 20 mg per day) or atenolol (initial dose, 50 mg per day; maximal dose, 100 mg per day) until the diastolic blood pressure decreased to less than 90 mm Hg, or there was a decrease in diastolic blood pressure of 10 mm Hg (whichever was lower), the maximal dose was given, or unacceptable side effects were reported. The patient then remained receiving the appropriate maintenance dose for a minimal period of four weeks. Following the initial study drug phase, the study drug was tapered off and patients had a drug-free period of four weeks. During the alternate study phase, patients received the alternate study drug for a minimum of four weeks of maintenance, following the titration phase as described earlier. 56

January 23, 1989

The American Journal of Medune

RESULTS Of the 20 subjects enrolled, 19 were evaluated. One subject could not meet scheduling requirements for measurement of blood flows and was not included in data evaluation of any parameter. The mean daily maintenance dose of atenolol was 78.9 5 5.7 mg, and the mean prazosin dose was 6.4 + 1.5 mg. No patients withdrew from the study because of lack of efficacy or adverse drug reactions. Blood pressure responses in the sitting and standing positions are shown in Figure 1. In the sitting position, atenolol reduced blood pressure by 15111 mm Hg, and prazosin reduced it by 17113 mm Hg. In the standing position, atenolol reduced blood pressure by lo/l0 mm Hg, and prazosin effected a 21113 mm Hg reduction. All reductions from their respective baseline values were statistically significant except for the standing systolic blood pressure reduction attributable to atenolol. However, this treatment value was significantly less than the prazosin control value. The effects of therapy on forearm and finger blood flows and vascular resistances are presented in Figures 2 to 5. Prazosin significantly increased forearm blood flow (p ~0.001) despite the reduction in blood pressure. This was attributable to an average reduction in vascular resistance of 28 percent in this vascular bed (p ~0.025). In the finger, prazosin significantly

Volume 86 (suppl 16)

SYMPOSIUM

Figure 4. Effects of therapy on finger blood flow. *p CO.05

TABLE

ON CHD AND HYPERTENSION I KWAN ET AL

Figure 5. Effects of therapy on finger vascular resistance

I

Group Mean (?SEM) Data for Forearm and Finger Blood Flow and Vascular Resistance during Control and Treatment with Prazosin and Atenolol in 19 Patients

160

T

-t-

Control Blood flow (mlimlnutesidl) Forearm Finger Vascular resistance (mts) Forearm Finger

2.5 i 0.3 21.4 + 5.1

Treatment 3.3 i 0.3* 31 1 t 5.1*

Atenolol Control 2.6 i 0.3 224 i 4.1

120

Treatment

100 LDL levels (wW

2.3 ? 0.2t 21 6 ? 4.3t

8. M) 40 20

51.5 + 4.6 18.4 z 6.3

36.9 + 45* 13 2 ? 5.6*

48.9 i 4.3 184 i 1.8

52.7 i 5.7t 247? 12.71

@j

Aienolol

I

Pramsin

0

Control

T*

140

Prazosin

*p CO.05

0

1

*F1
Figure 6. Effects of therapy on low-density *II 10.05.

llpoproteln (LDL) cholesterol

levels.

increased blood flow compared with placebo and aten0101treatment values (1~~0.05). It was found that the values of finger vascular resistance were not normally distributed but were normalized by using the logarithmic transformation. Analysis of variance of the logtransformed values showed that prazosin treatment was responsible for a significant reduction (1) ~0.01) in finger vascular resistance compared with control and with atenolol treatment. Atenolol caused no statistically significant changes in blood flow or resistance in either forearm or finger. As shown in Table I, the mean numerical values, indicating changes from baseline, for atenolol tended to go in the opposite direction compared with those seen with prazosin. Although the study was of short duration and the number of patients was small, E-hour fasted blood lipid level profiles showed a significant change in lowdensity lipoprotein levels during prazosin treatment (Figure 6). They were reduced from 148 ? 11 to 127 ? 10 mg/dl after four weeks of treatment (p ~0.05). High-density lipoprotein and triglyceride concentrations did not change during prazosin therapy. There were no significant changes in lipid concentrations during atenolol therapy. Diabetic control, measured by twice-a-day urine testing at the end of each period, was not demonstrably altered by either drug. During both atenolol and prazosin therapy, adverse effects were minimal, and none was of sufficient severity to require reduction or cessation of drug dosage. The side effects were consistent with those reported previously in the literature for both drugs. Atenolol was associated with edema in three patients

and generalized pain in two patients. Prazosin caused blurred vision in two, headache in three, nervousness in two, and flatulence in two patients.

COMMENTS Patients with diabetes mellitus have been shown to have reduced capacity for vasodilation to direct and indirect stimuli in both muscle and skin resistance vessels. The basis for this appears to be at least partially autonomic neuropathy but may also be related to a microangiopathic lesion due to thickening and enclothelial proliferation of dermal capillaries of the forearm 111-131. It is also known that blood flow in the limbs is reduced, ancl vascular resistance higher, in patients with diabetes mellitus and atheromatous disease compared with nondiabetic patients as reported by Faris and Lassen 1141. In this group of patients with concomitant diabetes and hypertension, adequate blood pressure control was achieved with both atenolol and prazosin monotherapy. The major difference between the two drugs was in their effects on flow and resistance in the vascular beds studied. In the forearm, prazosin increased blood flow and decreased vascular resistance. Ateno101caused no change in either blood flow or vascular resistance. In the finger, prazosin increased blood flow by 44 percent and reduced vascular resistance by 28 percent. Atenolol caused no significant change in either blood flow or vascular resistance. In this study population, the control blood flow values for forearm and finger tended to be on the low side of what is regarded as normal for nondiabetic and nonhypertensive subjects. Resting forearm flows usually range from 3 to 5 mliminuteidl. Finger flows vary

January 23, 1989

The American Journal of Medune

Volume 86 (suppl 1B)

57

SYMPOSIUM

ON CHD AND HYPERTENSION

/ KWAN ET AL

more widely, from 1 to 100 mliminuteidl. However, the lack of published normative data prevents us from making direct comparisons between this study group and normal subjects. Hypertension and diabetes mellitus are independent and additive risk factors for atherosclerotic vascular disease. Although there is much debate regarding the contribution of blood lipid level changes to atherosclerotic disease, most authors consider that lowdensity lipoprotein is positively correlated with disease [E,lG], high-density lipoprotein is negatively correlated [ 16,171, low-density lipoprotein:highdensity lipoprotein ratio is positively correlated [l&19], and very low-density lipoprotein and triglyceride levels have no independent association with atherosclerotic disease [20]. The present study was not primarily designed to examine blood lipid level changes; however, it was noted that prazosin treatment resultecl in no adverse changes in lipid parameters measured in this study.

SUMMARY Both drugs adequate!y decreased blood pressure to a similar degree and with similar lack of symptomatic side effects. They had different effects on forearm and finger vascular beds in this group of patients. Atenolol caused no significant changes in any parameter, although the trend was to increase resistance and decrease blood flow. Prazosin significantly increased flow and decreased resistance in both beds. There were no adverse effects on any of the measured lipid parameters following prazosin treatment. In this study, the hemodynamic and metabolic profile of prazosin appears to be suitable for diabetic hypertensive patients.

58

January 23, 1989

The American Journal of Medlclne

REFERENCES 1. Chwlleb AR. Diabetes and hypertenwe vascular disease Mechanisms and treatment Am J Card10 1973, 32 592-606 2. Berntorp K. Llndgarde F Far& aggregation of type 2 diabetes mellltus as an etlological factor in hyperterwon Diabetes Res Clan Pratt 1986: 1 307-313 3. The Working Group on Hypertension in Diabetes: Statement on hypertension in dlabetes mellltus Final report. Arch Intern Med 1987, 147. 830-842 4. Crlqul MH Epidemiology of atherosclerow an updated overview. Am J Cardlol 1986. 57 18C-23C. 5. KIngsbury KJ: The relation between glucose tolerance and atherosclerotic vascular disease Lancet 1966. II, 1374-1379 6. KIngsbury KJ. Glucose tolerance, age and atherosclerosis Postgrad Med J 1968. (suppl 5). 944-954 7. Kingsbury KJ. Concept of human atherosclerows. Nature 1969, 224 146-149 8. Lund-Johansen P. Hemodynamlc alterations in hypertension-spontaneous changes and effects of drug therapy A review Acta Med Stand [Suppl] 1977 603: l-14 9. Kwan CM, Shepherd AMM, Johnson J, Taylor WF. Forearm and finger hemodynamics. blood pressure control, and lipid changes in patlents wth diabetic hypertension treated with atenolol and prazosw Clan Pharmacol Ther 1988: 44 202-210. 10. Chew v Comparisons among treatment means in an analysis of variance Department of Agriculture No. ARS/H/6. Washington. Department of Agriculture, 197i 11. Handelsman MB, Momone TG. Ghltman B Skin vascular alterations in diabetes mel lltus Arch Intern Med 1962. 110. 70-77 12. Banson BB. Lacy PE: Dlabetlc microanglopathy in human toes wth emphasis on the ultrastructural change in dermal capillaries Am J Pathol 1964, 45. 41-58. 13. Greeson TP, Freedman RI. Levan NE, Wong WH: Cutaneous vascular responses in dlabetlcs Mlcrovasc Res 1975, 10. 8-16. 14. FarIs IB, Lassen NA, Increased vascular resistance in vasodllated skin an lndlcator of dlabetlc mlcroanglopathy? Cardiovasc Res 1982; 16 607-609. 15. Pllger E, Prlstautz H, Pfelffer KP. Kostner G Risk factors for peripheral atherosclerosis. Retrospecttve evaluation by stepwtse dlscrlmlnant analysis. Arterlosclerosls 1983, 3. 5763. 16. Gordon T, Caste111WP, Hjortland MC, Kannel WB. Dawber TR: High derwty llpoproteln as a protective factor against coronary heart disease. the FramIngham Study. Am J Med 1977 62 707-713 17. Miller NE. Thelle DS, Forde OH, Mojos OD The Tromso Heart-Study High density l~poprotw and coronary heart disease Lancet 1977 I: 965-968. 18. Mertz DP: “Atherosklerose-Index” (LDL/HDL) Rlslkolndlkator beI Storungen des fettstoffwechsels. Med Kl~n 1980, 75 159-161 19. Stossel J-P Wle entsteht Arterlosklerose’ Med Kiln 1980, 75 348-357 20. Hulley SB, Rosenman RH, Bawol RD. Brand RJ Epldemlology as a guide to cllnlcal declslons The assoclatlon between trlglycerlde and coronary heart disease N Engl J Med 1980, 302. 1383-1389

Volume 86 (suppl 1B)