Haemorheological Effects of Losartan and Enalapril in Patients with Renal Parenchymal Disease and Hypertension

Home , Hemorheology

Journal of Human Hypertension (2000) 14, 305–309  2000 Macmillan Publishers Ltd All rights reserved 0950-9240/00 $15.00 www.nature.com/jhh ORIGINAL ARTICLE Haemorheological effects of losartan and enalapril in patients with renal parenchymal disease and hypertension

BI Shand Department of Nephrology, Christchurch Hospital, Christchurch, New Zealand

The objective of this study was to compare the effects an increase in blood cell deformability. In patients tak- of the angiotensin II (ang II) antagonist, losartan and the ing losartan, the increase in blood cell deformability did angiotensin-converting enzyme inhibitor (ACEI), enalap- not result in a decrease in mean whole blood viscosity ril on haemorheology. Twenty-nine patients with renal due to a concomitant, significant increase in mean parenchymal disease and hypertension were enrolled in plasma viscosity (P Ͻ 0.01). In contrast, the improved the prospective, open, parallel study that involved a 14- cell deformability in patients treated with enalapril day washout period followed by a 120-day treatment resulted in a small and statistically insignificant mean ;0.06 ؍ period. Patients were allocated randomly to receive decrease in mean whole blood viscosity (P ؊0.15 mPa sec). The mechanism of the؍ or enalapril 2.5– change (15 ؍ either losartan 50–100 mg/day (n to achieve blood pressure control increase in blood cell deformability and the rise in (14 ؍ mg/day (n 10 Ͻ140/90 mm Hg. Blood pressure, haemorheology profile plasma viscosity associated with losartan remain and plasma fibrinogen concentration were measured unclear. It is possible but unproven that the improve- after the washout phase and after 2, 10, 60, and 120 days ment in intrinsic blood cell rheology with losartan and of treatment. The data were analysed using ANOVA with enalapril may be the result of changes in cation trans- repeated measures. Twenty-seven patients completed port systems and/or the consequence of the protective the study. Treatment with both losartan and enalapril antioxidant properties of drug metabolites. was associated with a significant decrease (P Ͻ 0.05) in Journal of Human Hypertension (2000) 14, 305–309 relative high shear rate whole blood viscosity, indicating

Keywords: haemorheology; angiotensin II antagonists; angiotensin-converting enzyme inhibitors; erythrocyte deformability; plasma viscosity; renal disease

Introduction decreased RBC aggregation.3,7,8 The mechanism(s) responsible for the improvements in intrinsic RBC The flow properties of blood (haemorheology) are a rheology remain poorly understood but are possibly major determinant of vascular resistance and blood 1 linked to decreased angiotensin II (ang II) pro- pressure. In order for an antihypertensive agent to duction3 or, indirectly, to changes in sympathetic have maximum blood pressure lowering potential it tone and haemodynamics.7 The recent development is important that determinants of haemorheology are of the ang II receptor antagonists, of which losartan not adversely affected. There is now considerable is the prototype, provide a means of studying the evidence that with the exception of the thiazide physiological actions of ang II specifically.9 Two diuretics, all major classes of antihypertensive drugs small studies, including one from our research are associated with beneficial haemorheological 10,11 changes.2,3 These changes include haemodilution group have shown that long-term treatment with with ␣ and ␤ receptor antagonists,4,5 or improve- losartan was associated with a small but clinically ments in red blood cell (RBC) deformability induced significant decrease in whole blood viscosity. by calcium channel antagonists.6 The aim of the present study was to compare the Within the last decade several studies have shown haemorheological effects of losartan and enalapril that angiotensin-converting enzyme inhibitors and to determine whether a decrease in either ang (ACEls) improve haemorheology by a combination II activity or concentration had similar effects on of actions that include haemodilution secondary to determinants of whole blood viscosity. The study vasodilatation, increased RBC deformability and was carried out in patients with renal parenchymal disease and hypertension as this patient group is known to have an increased incidence of haemo- 8,12 Correspondence: Dr Brett Shand, Lipid and Diabetes Research rheological abnormalities. Group, Christchurch Hospital, Private Bag 4710, Christchurch, New Zealand Received 7 September 1999; revised 28 December 1999; accepted 10 February 2000 Haemorheology and inhibition of angiotensin II BI Shand 306 Patients and methods and plasma viscosity measurements, respectively. Plasma fibrinogen concentration was determined by Patients the heat precipitation method.16 Twenty-nine patients (15 males) mean age 45 ± 13 An index of blood cell deformability was obtained years (range 21–75 years) with hypertension, renal by determining the relative whole blood viscosity at parenchymal disease and stable renal function high shear rate. This was calculated as whole blood ± viscosity at 200 per sec standardised to 0.45 haema- (mean plasma creatinine 0.11 0.04 mmol/l; range 17 0.03–0.18 mmol/l) were enrolled in the study. tocrit divided by plasma viscosity. Patients taking diuretics or requiring more than one antihypertensive agent were excluded from the Statistical analysis study. This number of patients provided the study with a power of 95% at the 0.05 significance level Changes in indices over time were determined using assuming a decrease in mean high shear rate whole analysis of variance with repeated measures and blood viscosity of 0.16 mPa/sec with a standard Dunnett multiple comparisons test (GraphPad TM ± deviation of 0.12 mPa/sec. These values were selec- Instat ). Results are given as mean standard devi- ted on the basis of data from an earlier study on the ation. Associations between indices were determ- rheological effects of enalapril8 that was carried in ined by calculating the Pearson correlation coef- patients with renal parenchymal disease. ficient. The effect of the two drugs on plasma viscosity was compared using the Wilcoxin rank sum test. Study design The study was an open, comparative design and was Results approved by the local ethics committee. Following There was no significant difference in age (losartan enrollment, the patient’s antihypertensive therapy 48 ± 13 years; enalapril 43 ± 12 year) or plasma crea- was stopped for 14 days. Blood pressure was meas- tinine concentration (losartan 0.10 ± 0.05 mmol/l; ured midway and at the end of this washout period. enalapril 0.11 ± 0.04 mmol/l) in the two treatment Following the washout period patients were ran- groups. Two patients were withdrawn from the domised to receive either losartan (50–100 mg/day) study, one patient on losartan for inadequate blood or enalapril (2.5–10 mg/day) for 120 days. The dos- pressure control and the other patient on enalapril, age of either drug was titrated during the study to Ͻ who developed cough. The clinical and haemorheo- achieve blood pressure control of 140/90 mm Hg logical data from the remaining 27 patients (14 losar- in the sitting position. Measurement of blood press- tan, 13 enalapril) are summarised in Tables 1 and ure and collection of venous blood samples for hae- 2 respectively. matology, haemorheology (EDTA anticoagulated) Blood pressure was reasonably well controlled and plasma creatinine concentration were carried with either drug. The dosage of the antihypertensive out at day 0 (end of washout period) and after 2, 10, medication was increased during the study in 5 60 and 120 days of treatment. A 24-h urine sample patients on losartan and in 8 patients taking enalap- was also collected at day 0 and 120 to determine ril. Mean plasma creatinine concentration and mean endogenous creatinine clearance. creatinine clearance did not change significantly with either drug. Methods Mean haematocrit and mean whole blood viscosity at both high and low shear rates did not Sitting blood pressure was measured by the same change significantly with either drug. No patient operator in triplicate using a mercury sphygmoman- ometer and the median measurement recorded. Table 1 Time-related changes in clinical parameters in patients Plasma creatinine concentration was measured ± using the picrate acid method (Hitachi treated with losartan or enalapril (mean s.d.) autoanalyser). Blood pressure (mm Hg) Plasma Creatinine Creatinine clearance prior to the study was creatinine clearance 13 2 assessed using the formula of Cockroft and Gault Systolic Diastolic (mmol/l) (ml/sec 1.73 m ) while creatinine clearance during the study was measured by a 24-h urine collection. Haematology Losartan indices were determined using an automatic cell Day 0 153 (18) 100 (13) 0.11 (0.05) 1.88 (0.32) counter (Coulter STKS, Coulter Electronics, Luton, Day 2 140 (21)** 90 (11)** 0.11 (0.05) – UK). Whole blood viscosity at 37°C at the patient’s Day 10 135 (18)** 89 (7)** 0.11 (0.05) – Day 60 137 (16)** 90 (13)** 0.11 (0.06) – haematocrit was measured at shear rates of 200 per Day 120 138 (16)** 88 (8)** 0.11 (0.06) 1.90 (0.32) sec and 20 per sec in a programmable rheometer (Brookfield Engineering, MA, USA) according to re- Enalapril commended standard procedures14 and the data Day 0 141 (14) 96 (13) 0.11 (0.04) 1.82 (0.21) analysed using Rheocalc software. Plasma viscosity Day 2 136 (13) 83 (11)** 0.11 (0.05) – ° Day 10 136 (12) 86 (12)** 0.12 (0.05) – at 37 C was measured using a capillary viscometer Day 60 137 (13) 87 (10)** 0.11 (0.05) – (Coulter Electronics) according to the guidelines of Day 120 134 (10) 87 (10)** 0.11 (0.05) 1.69 (0.20) Harkness.15 The inter-assay coefficients of variation were 1.3% and 1.1% for the whole blood viscosity **P Ͻ 0.01 compared to Day 0.

Journal of Human Hypertension Haemorheology and inhibition of angiotensin II BI Shand 307 Table 2 Time-related changes in haemorheological parameters in patients treated with losartan or enalapril (mean ± s.d.)

Haematocrit Whole blood viscosity (mPa.s) Relative viscosity Plasma viscosity Plasma ﬁbrinogen at high shear rate (mPa sec) (g/l) High shear rate Low shear rate

Losartan Day 0 0.40 (0.06) 3.68 (0.58) 5.29 (1.14) 3.31 (0.16) 1.60 (0.07) 3.6 (0.5) Day 2 0.40 (0.06) 3.73 (0.55) 5.43 (1.14) 3.32 (0.18) 1.60 (0.08) 3.7 (0.6) Day 10 0.40 (0.06) 3.71 (0.61) 5.33 (1.20) 3.28 (0.17) 1.62 (0.08) 3.7 (0.5) Day 60 0.40 (0.06) 3.58 (0.57) 5.13 (1.16) 3.20 (0.17) 1.62 (0.08) 3.8 (0.6) Day 120 0.41 (0.05) 3.67 (0.64) 5.47 (1.30) 3.11 (0.23)** 1.66 (0.08)** 3.8 (0.5)

Enalapril Day 0 0.42 (0.04) 3.75 (0.48) 5.48 (1.18) 3.19 (0.23) 1.64 (0.12) 3.9 (0.6) Day 2 0.42 (0.04) 3.76 (0.49) 5.68 (1.02) 3.14 (0.22) 1.64 (0.12) 4.2 (0.6) Day 10 0.43 (0.04) 3.77 (0.51) 5.85 (0.89) 3.13 (0.27) 1.65 (0.11) 3.8 (0.4) Day 60 0.42 (0.04) 3.77 (0.49) 5.61 (1.02) 3.17 (0.21) 1.64 (0.12) 3.7 (0.8) Day 120 0.43 (0.04) 3.60 (0.45) 5.54 (0.84) 3.05 (0.10)* 1.61 (0.08) 3.9 (0.6)

*P Ͻ 0.05; **P Ͻ 0.01 compared to Day 0. had a decrease in haematocrit level greater than 3% showed there was a significant difference (P Ͻ 0.01) or a reduction in high shear rate whole blood vis- in the changes in plasma viscosity associated with cosity greater than 0.06 mPa sec during the study. the two drugs. Both losartan and enalapril were associated with a Mean plasma fibrinogen concentration did not significant, gradual decrease in mean relative high change with either drug. There was a significant cor- shear rate viscosity (Table 2). The improvement in relation between the change in plasma viscosity and this index of blood cell deformability was 6% for plasma fibrinogen concentration from baseline to the losartan and 4% for enalapril. end of the study with both drugs (losartan r2 = 0.49, In patients treated with losartan there was a grad- P Ͻ 0.01; enalapril r2 = 0.60, P Ͻ 0.01) (Figure 2). ual and statistically significant increase in mean Changes in plasma fibrinogen concentration and plasma viscosity. The change in plasma viscosity plasma viscosity between study visits in individual during the study in individual patients is shown in patients showed a weaker but statistically significant Figure 1. An increase in plasma viscosity degree of correlation (losartan r2 = 0.08, P = 0.03; у0.06 mPa/sec was observed in five of the 14 enalapril r2 = 0.09, P = 0.03). patients on losartan. Mean plasma viscosity in patients taking enalapril was unchanged throughout the study with an increase from the baseline level Discussion of у0.06 mPa/sec being observed in only one of 13 This study provides further evidence that inhibition patients. Analysis using the Wilcoxin rank sum test of the renin-angiotensin system may be associated with changes in haemorheology. In our patients, both losartan and enalapril caused a small but statistically significant reduction in relative whole blood viscosity indicating an improvement in the intrinsic rheological properties of blood cells. In patients taking enalapril, this apparent improvement in cell rheology in conjunction with an unchanged plasma viscosity resulted in a statistically insignificant reduction in whole blood viscosity. In contrast, patients taking losartan had an increase in mean plasma viscosity that offset any change in cell rheology with whole blood viscosity remaining unchanged during the study. The cause of the decrease in relative high shear rate whole blood viscosity associated with both losartan and enalapril is poorly understood. The finding that losartan and enalapril had a similar effect on this index suggests that the change was mediated by alterations in ang II activity, either as a decrease in ang II levels (enalapril) or by blockade of the ATI receptors (losartan). The mechanism of how ang II may influence blood cell rheology is also unclear but may possibly be the result of changes in red blood cell (RBC) transmembrane cation flux18,19 induced Figure 1 Change in plasma viscosity in individual patients from by variations in ang II activity or as a non-specific day 0 to day 120 of the study. consequence of blood pressure reduction. It is well

Journal of Human Hypertension Haemorheology and inhibition of angiotensin II BI Shand 308

Figure 2 Changes in plasma viscosity and plasma ﬁbrinogen concentration from day 0 to day 120 in the two treatment groups (r2 = Pearson’s correlation coefﬁcient). (b) Enalapril, r2 = 0.60; (ć) Losartan, r2 = 0.49.

documented that hypertension is associated with tion, changes that both have the potential to increase adverse changes in RBC cation transport systems20 plasma viscosity.30 In addition, it is unlikely that and that inhibition of ang II activity results in a nor- either losartan or enalapril caused haemoconcen- malisation of these system.21–25 The resulting tration as the vasodilatatory action of both drugs changes in intracellular cation composition may would be expected to result in haemodilution and influence cell deformability by maintaining optimal therefore a reduction in plasma viscosity. intracellular water content and cytoplasmic vis- In conclusion, this study in patients with renal cosity in addition to preventing polymerisation of parenchymal disease and hypertension demon- cytoskeletal and membrane bound proteins (ie, the strated that treatment with both losartan and enalap- Gardos effect).26 ril was associated with an decrease in relative high It is also possible that the decrease in relative shear whole blood viscosity, a change that indicates whole blood viscosity observed in this study was the an increase in blood cell deformability. In patients result of the combined effect of changes in RBC and treated with enalapril, this apparent change in cell white blood cell rheology. There is considerable evi- rheology resulted in a small, but potentially clini- dence that metabolites of ACEls have free radical cally important reduction in whole blood viscosity. scavenging properties27,28 an action that may poten- In contrast, in patients treated with losartan, a con- tially minimise oxidative damage to blood cell mem- current increase in plasma viscosity offset the branes. Whether or not ang II receptor antagonists improvement in blood cell deformability so that have similar antioxidant properties however whole blood viscosity remained unchanged. The remains to be established. mechanisms of these rheological effects remains This study also observed that losartan, but not poorly understood but in the case of the cellular enalapril, was associated with a small but statisti- changes may possibly be the result of changes in cally significant increase in mean plasma viscosity. intracellular cation composition and/or a conse- This is the first report of such an effect of ang II quence of the protective antioxidant properties of antagonism on plasma viscosity. It is possible that drug metabolites. this finding is a consequence of the relatively small number of patients in the study (ie, a type I statisti- Acknowledgements cal error). However, the observation that plasma viscosity increased in 12 of the 14 patients treated with The author is grateful to Merck Sharp and Dohme losartan, with this increase being of physiological (NZ) Ltd for funding this study and to Anne Smith significance in six patients (ie, Ͼ0.06 mPa sec),29 RN for nursing duties. The assistance of Professor indicates that confirmatory data are required from a MG Nicholls and Dr DO McGregor with the prep- larger study. The possible reasons for the different aration of the manuscript is gratefully acknowl- effects of losartan and enalapril on plasma viscosity edged. is not apparent from the data of the present study. An increase in plasma viscosity may occur as a References result of an increase in the concentration of acute phase proteins such as fibrinogen or from a 1 Letcher RL et al. Direct relationship between blood reduction in intravascular fluid volume.15 While pressure and blood viscosity in normal and hyperten- there was a relationship between changes in plasma sive subjects. Am J Med 1981; 70: 1195–1202. viscosity and plasma fibrinogen concentration in 2 Zannad F, Stoltz J-F. Blood rheology in arterial hypertension. J Hypertens 1992; 10 (Suppl 10): S69–S78. individual patients, the degree of correlation was 3 Zannad F et al. Effects of lisinopril and hydrochloro- similar with both drugs. None of the patients treated thiazide on platelet function and blood rheology in with losartan had clinical signs of an inflammatory essential hypertension: a randomly allocated double- condition, as indicated by an abnormal white blood blind study. J Hypertens 1993; 11: 559–564. cell count, nor a significant decline in renal func- 4 Gomi T, Ikeda T, Ikegami F. Beneficial effect of ␣-

Journal of Human Hypertension Haemorheology and inhibition of angiotensin II BI Shand 309 blocker on hemorheology in patients with essential 18 Saitta MN et al. [Na+,K+,Cl−]-cotransport function and hypertension. Am J Hypertens 1997; 10: 886–892. dysfunction in different forms of primary hyperten- 5 Boogaerts MA et al. Effect of beta-blocking drugs on sion. Am J Hypertens 1988; 1: 60S–63S. red cell adhesive and rheological properties. J Lab Clin 19 David-Dufilho M et al. Control of the erythrocyte free Med 1983; 102: 899–908. Ca2+ concentration in essential hypertension. Hyper- 6 Slonim A, Paran E, Cristal N. Effect of isradapine on tension 1992; 19: 167–174. factors affecting blood viscosity. Am J Hypertens 1991; 20 Touyz RM, Milne FJ. Alterations in intracellular cat- 4: 172S–174S. ions and cell membrane ATPase activity in patients 7 Ernst E, Bergmann H. Influence of cilazapril on blood with malignant hypertension. J Hypertens 1995; 13: rheology in healthy subjects. Am J Med 1989; 87 867–874. (Suppl 6B): 70S–71S. 21 Malini PL, Strocchi E, Ambrosioni E, Magnani B. 8 Shand BI, Bailey RR, Lynn KL, Robson RA. Effect of Long-term antihypertensive, metabolic and cellular enalapril on hemorheology in hypertensive patients effects of enalapril. J Hypertens 1984; 2 (Suppl 2): with renal disease. Clin and Exp Hypertens 1995; 17: S101–S105. 689–700. 22 Fujito K et al. Acute effects of angiotensin-converting 9 Goa KL, Wagstaff AJ. Losartan potassium. A review of enzyme inhibitor on erythrocyte sodium ion transport its pharmacology, clinical efficacy and tolerability in in essential hypertension. Life Sci 1992; 51: 2079– the management of hypertension. Drugs 1996; 51: 2087. 820–845. 23 de la Sierra A et al. Effect of long-term antihyperten- 10 Moan A et al. Effects of losartan on insulin sensitivity sive therapy with angiotensin converting enzyme in severe hypertension: connections through sympath- inhibitors on red cell sodium transport. Am J Hyper- etic nervous system activity. J Hum Hypertens 1995; 9 tens 1995; 8: 622–625. (Suppl 5): S45–S50 24 Golik A et al. Erythrocyte Na+,K+, and Ca2+, Mg(2+)- 11 Shand BI, Gilchrist NL, Nicholls MG, Bailey RR. Effect ATPase activities in hypertensives on angiotensin- of losartan on haematology and haemorheology in eld- converting enzyme inhibitors. Clin Biochem 1996; 29: erly patients with essential hypertension: a pilot 249–254. study. J Hum Hypertens 1995; 9: 233–235. 25 Sanchez RA et al. Recovery of erythrocyte Na(+)-K(+]- 12 Shand BI, Bailey RR, Simpson LO. Blood rheology in Cl-cotransport activity by enalapril. Hypertension IgA nephropathy. Clin Nephrol 1988; 29: 288–293. 13 Gault MH, Longerich LL, Harnett JD, Wesolowski C. 1991; 17: 334–339. Predicting glomerular function from adjusted serum 26 Stuart J, Ellory JC. Rheological consequences of creatinine. Nephron 1992; 62: 249–256. erythrocyte dehydration. Br J Haematol 1988; 69:1–4. 14 International Committee for Standardization in Hae- 27 Chopra M et al. Antioxidant effects of angiotensin- matology (Expert panel on blood rheology). Guidelines converting enzyme (ACE) inhibitors: free radical and for measurement of blood viscosity and erythrocyte oxidant scavenging are sulfhydryl dependent, but deformability. Clin Hemorheol 1986; 6: 439–453. lipid peroxidation is inhibited by both sulfhydryl- and 15 Harkness J. The viscosity of human blood plasma; its nonsulfhydryl-containing ACE inhibitors. J Cardiovasc measurement in health and disease. Biorheology 1971; Pharmacol 1992; 19: 330–340. 8: 171–193. 28 Aruoma OL. Characterisation of drugs as antioxidant 16 Millar HR, Simpson JG, Stalker AL. An evaluation of prophylactics. Free Radical Biol Med 1996; 20: 675– the heat precipitation method for plasma fibrinogen 705. estimation. J Clin Path 1971; 24: 827–830. 29 Lowe GDO. Blood rheology in vitro and in vivo. Bail- 17 Matrai A, Whittington RB, Ernst E. A simple method liere’s Clin Haematol 1987; 1: 597–636. of estimating whole blood viscosity at standardized 30 Shand BI, Bailey RR, Simpson LO. Blood rheology in hematocrit. Clin Hemorheol 1987; 7: 261–265. IgA nephropathy. Clin Nephrol 1988; 29: 288–293.

Journal of Human Hypertension