Microalbuminuria and endothelial dysfunction in essential hypertension

Microalbuminuria and endothelial

dysfunction in essential

hypertension

Summary

Introduction

Microalbuminuria (urinary albumin excretion between 20 and 200 µg/min) and endothelial dysfunction coexist in patients with essential hypertension. To evaluate whether the two phenomena are related and the determinants of that association, we recruited 10 untreated males with essential hypertension and microalbuminuria without diabetes to be compared with an equal number of matched patients with essential hypertension excreting albumin in normal amounts and 10 normal controls. The status of endothelial function was inferred from

A

circulating von Willebrand Factor antigen (vWF), a glycoprotein secreted in greater amounts when the vascular endothelium is damaged. vWF concentrations were higher in hypertensive patients with microalbuminuria than in hypertensive patients without and controls. Individual vWF and urine albuminexcretion values were correlated (r=0·55, p<0·002). Blood pressure correlated with both urinary albumin excretion and vWF. Left ventricular mass index and minimal forearm vascular resistances were comparable in patients with hypertension and higher than in controls; total and low-density lipoprotein cholesterol, triglycerides, lipoprotein-a, Factor VII, and plasminogen activator inhibitor-1 did not differ. Fibrinogen was higher and creatinine clearance lower in microalbuminurics. Albuminuria in essential hypertension may reflect systemic dysfunction of the vascular endothelium, a structure intimately involved in permeability, haemostasis, fibrinolysis, and blood pressure control. This abnormality may have important physiopathological implications and expose these patients to increased cardiovascular risk.

large proportion of non-diabetic patients with hypertension excrete albumin in the microalbuminuric range;’ increased urinary albumin excretion (UAE) is related to systemic disorders of transcapillary escape rate ;2 and epidemiological studies have identified microalbuminuria as a risk factor for illnesses of atherothrombotic origin.3 Endothelial damage may initiate atherosclerosis4 since the endothelium is involved in permeability, fibrinolysis, haemostasis, and blood pressure control. Impaired endothelial function has also been reported in apparently-uncomplicated human essential hypertension both by infusing agonists in regional beds6 and by measuring circulating von Willebrand Factor antigen (vWF)7 a glycoprotein released in greater concentrations when endothelial cells are damaged.8 Thus, endothelial dysfunction and albuminuria coexist in nondiabetic hypertensives and may be linked. Our study was designed to test the hypothesis that endothelial dysfunction, quantified by circulating vWF, is related to microalbuminuria in essential hypertension, and to evaluate the possible determinants of this association.

Subjects and methods Selection criteria were male sex with no malignant or accelerated hypertension, congestive heart failure, obesity (body mass index < 30 kg/m2), diabetes, previous myocardial infarction, renal or connective tissue disease; normal serum creatinine (< 1-4 mg/dL), normal urinary sediment and culture, a good acoustic window for echocardiography, and absence at doppler examination of valvular lesions which might contribute to cardiac hypertrophy. 30 subjects were recruited, divided in three groups 1: 10 hypertensive patients with microalbuminuria who underwent full investigation, for secondary hypertension, including angiography if necessary. Ultrasound showed normal-sized kidneys and no evidence of cortical scarring or obstructive uropathy, and oral glucose (75 grams) tolerance test showed glucose (by glucooxidase) > 200 mg/dL at 60 min. 2: 10 age-matched (± 3 yrs) males with hypertension, with normal UAE, undergoing a similar investigation, either never treated with drugs or on no treatment for at least two weeks, and with blood pressure (BP) consistently > 140/90 mm Hg as outpatients. 3: 10 normal, age-matched male

subjects as controls. Systolic (SBP) and diastolic (DBP) (Korotkoff V) BPs were measured in the morning, supine, by mercury sphygmomanometer as the mean of 10 readings over 30 min. Height and weight were measured after each subject had removed shoes and upper garments. Blood samples were obtained after fasting overnight at 8-00-9-00 am, after 15 min of supine rest. We asked all patients not to smoke that day and we were careful to avoid venous occlusion. To minimise confounding effects of physical activity, subjects Clinica Medica 1 (R Pedrinelli MD, G Dell’omo MD, G Catapano MD) and II (O Giampietro MD, F Carmassi MD, E Matteucci MD, L Talarico MD, M Morale MD, F de Negri MD, V di Bello MD) and Reparto dl Medicina Interna (E Melillo MD),

USL 12, University of Pisa, 56100, Pisa, Italy

Correspondence to: Dr Roberto Pedrinelli 14

collected urine from 8 pm to 8 am over three consecutive days.9 Urinary albumin was measured by nephelometry (Istituto Behring SpA, Scoppito, Italy) with a limit of detection of 0-6 mg/dL and an interassay variation of 35%.1° Urinary and serum creatinine was measured by standard colorimetric methods. Within-patient variability of creatinine excretion was 13% (average of 30 variation coefficients of triplicate urinary collections), without significant

Figure 1: von Willebrand Factor Antigen (vWF) In the experimental groups differences among groups. UAE variability was 37 % (average of 30 variation coefficients of triplicate urinary collections), a value reflecting the biological variability of this parameter.9 vWF was measured with an immunoenzymatic method (Boehringer Mannheim Corporation, Milano, Italy). Values of vWF were expressed as percentages of normal pooled plasma, the antigen level of which was defined as 100%. Wall thickness and chamber volumes were measured by monodimensional and bidimensional echocardiograms (Hewlett Packard Sonos 1000, Andover, MA, USA) with 2-5 and 3-5 mHz transducers.9 Forearm blood flow (FBF) was measured by straingauge venous plethysmography (DE Hokanson EC 5R, Issaquah, WA, USA). For evaluation of postischaemic reactive hyperaemia, arterial occlusion of the left forearm was obtained by inflating the plethysmograph cuff to 300 mm Hg for 13 min with dynamic exercise (20-30 hand contractions) during the last minute, a procedure which induces maximal hyperaemia.ll Peak FBF was the highest FBF during 3 min following ischemic release; BP was measured in the contralateral arm. Serum concentrations of total and high-density lipoproteins (HDL) (after precipitation of low density lipoprotein [LDL] and very low density lipoprotein [VLDL] fractions with phosphotungstic acid and magnesium chloride), cholesterol, and triglycerides (average coefficient of variation of control pool: 2%, 5%, and 2% respectively) were measured by enzymatic colorimetry (cholesterol oxidase/peroxidase aminoantipirine and

UAE and urinary creatinine were the average of three collections. Microalbuminuria was defined as >20 ug/min and < 200 IlgJmin;13 creatinine clearance was adjusted for 1 73 m2 body surface area. Left ventricular mass was indexed for height (g/m) to take body weight into account (LVMI). Mean BP (MBP), cardiac output (CO, end-diastolic volume to end-systolic volume x heart rate), total peripheral vascular resistance (TPR, MBP x 80/CO) and forearm vascular resistance (FVR) (MBP/FBF) were derived from standard formulae. Minimal forearm vascular resistance (Rmin) was the ratio of MBP to peak post-ischaemic FBF. A smoker was defined as a person smoking habitually at the time of the study; nonsmoker as never smoked or having not done so for the last six months. Skewed UAE distribution and the relationship between standard deviation of triplicated urine collections and mean values (r=0-76) needed log transformation to stabilise variance and for parametric tests. Descriptive statistics were arithmetic means (standard deviation) or medians with range for skewed data. Statistical analysis was based upon comparison among and between matched groups through one-way ANOVA and calculation of standard correlation coefficients. Multiple regression analysis was done conventionally by testing the statistical significance of each regression coefficient through t-test and 95% confidence limits. Frequency distribution was analysed through X2 statistics. p < 0-05 was chosen as statistically significant.

Results

g/min (range: 4-20) in controls, and and 70 g/min (22-171) in nong/min (7-18) microalbuminuric and microalbuminuric hypertensive raised vWF was in patients respectively. microalbuminurics (p < 0-01) (figure 1) and individual vWF and UAE values were correlated (r=0-55, p<0001, n = 30; figure 2). Significant correlations existed also with MBP (r=0-47, p<001), 4 age (r=0-40, p<002). microalbuminurics were nonsmokers vs 6 smokers; 9 vs 1 in non-microalbuminurics, and 8 vs 2 in controls (p < 0-05). Median UAE was 35-8 (6-1-122) vs 14-3 (4-171) g/min in smokers vs non smokers respectively (p < 0-03). Multiple regression analysis with log UAE as dependent variable and Median UAE

was

11

11

glycerol phosphate oxidase/peroxidase aminoantipirine [Menarini, Firenze, Italy]). LDL cholesterol was calculated as total cholesterol minus (HDL-cholesterol minus triglycerides/5). Lipoprotein a (Lpa) concentrations, were assessed by an immunoenzymatic method (Immuno, Pisa, Italy). Venous blood (9 ml) was collected in a syringe containing 3-8% sodium citrate (0-11 mol/L) to a final ratio (v:v) of 1 part sodium citrate to 9 parts blood. Reduced-platelet plasma was obtained by centrifugation at 3000 rpm for 15 min, and stored at 70°C until assayed.12 The increase of absorbance at 405 nm was estimated either kinetically or in end-point tests by factor VIII-deficient plasma and chromogenic substrates Tos-Gly-Pro-Arg-ANBAIPA for coagulation factor VII (FVII) and HD-Nva-CHA-LyspNA for plasminogen activator inhibitor-1 (PAI-1) (Istituto Bhering, Milano, Italy). Fibrinogen was measured by the endpoint method estimating the clotting rate. All measurements, including vWF, were in duplicate; the inter-assay coefficient of variation never exceeded 10% in our laboratory (typically

Figure 2: Scatterplot of urinary albumin excretion Willebrand Factor Antigen (vWF) values

2-6%).

Correlation coefficient was 0 60, p < 0 0001, n = 30.

(UAE) and von

15

CHOtot(mmol/L) CHOLHDL(mmol/L) CHOLLDL(mmol/L) HDL/LDL TG (mmol/L) Lp (a) (µg/mL)†

PAI-1 (mU/mL) FVII (%) FIB

(pmol/L)

Controls

Nonmlcroalbuminurlcs

Microalbuminurics

501 (07) 106(009) 328(069) 034(009) 142(056) 149 (9-38 9) 2(13) 90(16) 9 79 (2 9)

5 71 (0 64) 08(002) 411(064) 021(007) 174(064) 11 (1-89) 214(11) 956(23) 11 7 (1 7)

5 58 (1 09) 0.83 (0 3)t 403(il) 0 24 (0 13)t 157(038) 105 (0-45 5) 17(09) 96(23) 15 1 (6 9)*§

tmedian and range. Means ± standard deviation, *p<0 0 05, tp<0 0 01 vs controls, §p 0 05 non- vs microalbuminurics. Total (CHOLtot), HDL (CHOlHDL) and LDL (CHOLLDL(cholesterol, HDL/ LDL ratio, triglycerides (TG), lipoprotein (a) (Lp[a]), plasminogen activator inhibitor-1 (PAI-1). Factor VII activity (FVII) and fibrinogen (FIB).

Table 3: Laboratory values In and microalbuminurics

Figure 3: von Willebrand antigen (vWF) vs flbrinogen

MBP, vWF, age, and smoking status as independent variables, showed that only MBP and vWF contributed independently to total UAE variability (t=0-45, table 1). LVMI, wall thickness, resting FVR, and Rmin were comparable in the two hypertensive groups (table 2) and greater than in controls. UAE was positively related to MBP (r=0-49, p<0005), but not with LVMI (r=0-22, NS) Rmin (r=0-18). Patients with albuminuria had significantly higher serum creatinine concentrations and lower creatinine clearance (table 1) independently of UAE (r= -0-17). A negative correlation was present between creatinine clearance and vWF (r 0 50, p < 0 - 005, n 30). =

Independent variables

Regression coefficients

-

=

95% confidence limits

< p

MBP (mm Hg) 00210 002-0 044 0 05 0009 0002-0017 002 vWF (%)

Age (yrs) Smoking

-0023 0329

-0057-0011 -0261-090

NS NS

Regression equation: log UAE=0 731+(0 021xMBP)+(0009xvWF)-(0023xage)++ (0 329 x smoking), n=30, r=0 68. NS= not significant. Table 1: Multivariate analysis of mean blood pressure (MBP), von Willebrand Antigen (vWF), age and smoking as determinants of (log) Urinary Albumin Excretion (UAE) Controls

Age (yrs) BMI (kg/m2)

Crserum(µmol/L) Clcr(ml/sec/173m2) SBP (mm Hg) DBP (mm Hg) MBP (mm Hg) LVMI (g/m) IVST (mm) PWT (mm)

CO (L/min) FBF (mL/dL/min)

FBFPeak(mL/dL/min) (units) FVR (units)

TPR

Rmin(units)

585(9) 245(23) 79 6(17 7) 183(048) 1350(8) 810(7) 990(7) 1000(12) 103(08) 104(05) 53(09) 49(16) 494(86) 15300(310) 218(65) 205(04)

Non microalbuminurics

Microalbuminurics

60 1 (9 1) 253(29) 884(88) 17(041) 162(19) 1000(7) 117(74) 120 0 (9)t 12 2 (1 3)t 11 7 (1 2)t 56(16) 3 4 (1 5)t

594(9) 258(36) 102 5 (17 7)&Dag er;¶ 148(04)5* 1640(12) 980(7) 122 1 (11) 120 (7)t 119 (1 2)t 11 5 (0 9)t 5 9 (0 9) 3 6 (1 7)t 424(10) 430(55) 1805(413) 16730(305) 38 0 (10)t 412 (16)t 2 8 (0 5)t29(06)‡

Body mass index (BMI), serum creatinine (Crserum) and creatinine clearance (CrCI), systolic (SBP), diastolic (DBP) and mean (MBP) blood pressure, left ventricular mass index (LVMI), Interventricular (IVST) and posterior (PWT) wall thickness, cardiac output (CO), basal (FBF) and postischaemic (FBF peak) forearm blood flow, total peripheral vascular resistances (TPR) and basal (FVR) and postischaemic (Rmin) forearm vascular resistances in controls, non microalbuminurics and microalbuminurics (n= 10 each). Means ± standard deviation, *p < 0 05, tp < 0 01, tp < 0 001 vs controls, §p< 0 05, p< 0 01 non vs microalbuminurics. Table 2: Characteristics of groups

16

controls, non-microalbuminurics,

Total and LDL cholesterol, triglycerides, Lpa, FVII, and PAI-1 did not differ (table 3). HDL cholesterol and HDL/LDL ratio was higher in controls than hypertensives, and comparable in patients with and without microalbuminuria. Fibrinogen was raised (table 3) in microalbuminurics and correlated with vWF values

(r=0 60, p<00001,

n=30, figure 3).

Discussion Increased vWF in essential hypertensives with microalbuminuria supports the hypothesis that abnormal UAE reflects a systemic dysfunction of vascular endothelium, and intra-individual correlation between the two variables strengthens this possibility. We studied only male patients, and therefore our data do not necessarily apply to females.14 Although age influenced vWF as previously reported,17 our design took this into account by recruiting age-matched subjects. Third, we found no difference in vWF between controls and hypertensive patients without microalbuminuria who might have endothelial dysfunction when tested under different conditions. However, the relation between endothelial dysfunction as measuring by vWF systematically and by infusing agonists into regional beds (as in reference 6) is so far unknown. If endothelial dysfunction accounts for microalbuminuria, several factors may be involved, including blood pressure which explained part of both UAE and vWF variability. Considering that blood pressure recordings were done in daytime and urine was collected mostly at night, when arterial pressure normally reaches its nadir, and that casual blood pressure reflects poorly the 24 hr profile,1S the correlation between BP and UAE probably represents a biologically relevant finding. Thus, blood pressure seems to contribute to microalbuminuria and endothelial damage, an interpretation supported by the trend of vWF to increase in parallel with blood pressure. LVMI, wall thickness, and Rmin were unrelated to UAE in this sample. One might conclude that endothelial dysfunction is influenced by short-term changes in pressure since UAE is sensitive to acute afterload modifications,16 while both ventricular mass and forearm vasodilator reserve may be considered long-term sensors of the prevailing afterload.17 Smoking appeared an important link between microalbuminuria and endothelial dysfunction. Therefore, as in diabetics,18 smoking may be associated with albuminuria, possibly through endothelial damage, as supported by the results of our multiple regression analysis, where smoking status did not appear to contribute

variability, once vWF was taken into account. Furthermore, previous reports showed increased vWF in smokers, irrespective of their BP,19 and smoking may damage endothelial function .20 Finally, microalbuminurics had elevated fibrinogen, a factor strongly influenced by smoking,21 the underlying factor which probably explained the correlation of two otherwise unrelated parameters such as fibrinogen and vWF. We found no differences in PAI-1 and FVII activity, which indicates that fibrinolytic and coagulation systems were not greatly altered, although hypercoagulability and abnormal fibrinolysis were found by others in hypertensives compared with normotensive controls.22 This was not the case in our series, possibly because triglyceride levels did not differ markedly.23 Microalbuminurics did not differ as regards total and LDL cholesterol, in contrast with the results obtained in insulin-dependent diabetics with raised UAE.2’ We do not know the reasons, which may be related to the specific disease status of diabetics, nor can we exclude that a real difference might have been obscured by the relatively small size of our sample. Whatever the case, the data confirm that UAE9 in non-diabetic patients is not strongly related to lipid status. Similar considerations apply also to Lpa, a lipoprotein with independent predictive power for myocardial infarction with a strong hereditary component .21 Reduction in HDL cholesterol is a wellrecognised feature of essential hypertension26 and its possible relationship with vWF 14 needs more study. In spite of the requirement for normal serum creatinine, we ended up with an albuminuric group characterised by lower creatinine clearance although still in the normal range. Possibly hypertension lasted longer or was more severe27 in them, but either increased 28 or normal1 creatinine clearance was found in hypertensive patients with microalbuminuria apparently comparable with ours. The level of renal function could not be explained by UAE, nor by age because our experimental groups were matched. Nephrosclerosis might have been more prevalent in our microalbuminuric group, but we did not investigate kidney structure. An inverse correlation also emerged between vWF and creatinine clearance, rather unexpectedly, since vWF is not cleared through the kidneys. The data may suggest that a dysfunctional glomerular endothelium may cause reduced renal function, in line with the recent demonstration of the importance of nitric oxide in longterm regulation of renal haemodynamics and renal

independently

to

UAE

excretory function.29 In conclusion, non-diabetic hypertensive patients characterised by an abnormal UAE showed raised vWF, an indicator of endothelial dysfunction. This finding is relevant to recent evidence showing the essential role of vWF in the development of occlusive thrombosis.30 In addition, patients with cerebrovascular disease had increased vWF,31 and prospective trials have shown its importance as a risk factor for recurrent myocardial infarction and cardiovascular events,32 and one wonders whether this is one of the reasons for the independent nature of microalbuminuria as a cardiovascular risk factor shown in previous large scale trials.3 We thank Bohringer Mannheim (Milano, Italy), Immuno (Pisa, Italy), and Istituto Behring (Milano, Italy) for the gifts of the vWF, Lp(a), PAI-1, and FVII assays. Supported in part by a grant from the Ministero della Universita’ e della Ricerca Scientifica e Tecnologica (MURST, quota 60%, anno finanziario

1991-1992).

References 1

Bigazzi R, Bianchi S, Campese VM, Baldari G. Prevalence of microalbuminuria in a large population of patients with mild to moderate essential hypertension. Nephron 1992; 61: 94-97. 2 Parving HH, Gyntelberg F. Transcapillary escape rate of albumin and plasma volume in essential hypertension. Circ Res 1973; 33: 3 4

5 6

7

643-53. Yudkin JS, Forrest RD, Jackson CA. Microalbuminuria as predictor of vascular disease in non-diabetic subjects. Lancet 1988; ii: 531-34. Ross R. The pathogenesis of atherosclerosis-an update. N Engl J Med 1986; 314: 488-500. Vane JR, Angaard EE, Botting RM. Mechanisms of disease: regulatory functions of the vascular endothelium. N Engl J Med 1990; 323: 27-36. Panza JA, Quyyumi AA, Brush JE, Epstein SE. Abnormal endothelium-dependent vascular relaxation in patients with essential hypertension. N Engl J Med 1990; 323: 22-27. Blann AD, Naqvi T, Waite M, McCollum CN. Von Willebrand Factor and endothelial damage in essential hypertension. J Human Hypert

1993; 7: 107-11. Suffredini JH, Harpel PC, Parrillo JE. Promotion and subsequent inhibition of plasminogen activation after administration of intravenous endotoxin to normal subjects. N Engl J Med 1989; 320: 1165-72. 9 Pedrinelli R, DiBello VA, Catapano G, et al. Microalbuminuria is a marker of left ventricular hypertrophy but not hyperinsulinemia in nondiabetic atherosclerotic patients. Arteriosclerosis 1993; 13: 900-06. 10 Giampietro O, Penno G, Clerico A, et al. Which method for quantifying "Microalbuminuria" in diabetics? Acta Diabetol 1992; 28: 239-45. 11 Pedrinelli R, Spessot M, Salvetti A. Reactive hyperemia during short term blood flow and pressure changes in the hypertensive forearm. J Hypertens 1990; 8: 467-71. 12 Carmassi F, Morale M, Puccetti R, et al. Coagulation and fibrinolytic system impairment in insulin dependent diabetes mellitus. Thromb Res 8

1992; 67: 643-54. 13 Viberti GC. Mechanisms of diabetic renal and cardiovascular disease. Acta Diabetol 1990; 27: 267-76. 14 Conlan MG, Folsom AR, Finch A, et al. Associations of factor VIII and Von Willebrand Factor with age, race, sex, and risk factors of atherosclerosis. Thromb Haemost 1993; 70: 380-85. 15 Pickering TG. Diumal rythms and other sources of blood pressure variability in normal and hypertensive subjects. Hypertension: pathophysiology, diagnosis and management. Laragh JH, Brenner BM, eds. Raven Press Ltd, New York 1990: 1397-405. 16 Christensen CK. Rapidly reversible albumin and &bgr;2-microglobulin hyperexcretion in recent severe essential hypertension. J Hypertens 1983; 1: 45-51. 17 Dahlof B, Pennert K, Hansson L. Reversal of left ventricular hypertrophy in hypertensive patients. A metaanalysis of 109 treatment studies. Am J Hypertens 1992; 5: 95-110. 18 Corradi L, Zoppi A, Tettamanti F, Malamani GD, Lazzari P, Fogari R. Smoking habit and microalbuminuria in hypertensive patients with type 2 diabetes mellitus. J Hypertens 1993; 11: S190-91. 19 Blann AD. Increased circulating levels of Von Willebrand Factor antigen in smokers may be due to lipid peroxides. Med Sci Res 1991; 19: 535-36. 20 Nitenberg A, Antony I, Foult JM. Acethylcholine-induced coronary vasoconstriction in young, heavy smokers with normal coronary arteriographic findings. Am J Med 1993; 95: 71-77. 21 Kannel WB, Wolf PA, Castelli WP, D’Agostino RB. Fibrinogen and

risk of cardiovascular disease. The Framingham Study. JAMA 1987; 258: 1183-86. 22 Phillips GB, Jing TY, Resnick LM, Barbagallo M, Laragh JH, Sealey JE. Sex hormones and hemostatic risk factors for coronary heart disease in men with hypertension. J Hypertens 1993; 11: 699-702. 23 Mitropoulos KA, Miller GJ, Reeves BEA, Wilkes HC, Cruickshank JK. Factor VII coagulant activity is strongly associated with the plasma concentration of large lipoprotein particles in middleaged men. Atherosclerosis 1989; 76: 203-10. 24 Jensen T, Stender S, Deckert T. Abnormalities in plasma concentrations of lipoproteins and fibrinogen in type 1 (insulindependent) diabetic patients with increased albumin excretion.

Diabetologia 1988; 31: 142-45. 25 Scanu AM, Fless GM. Lipoprotein (a). Heterogeneity and biological relevance. J Clin Invest 1990; 85: 1709-15. 26 Williams RR, Hunt SC, Hopkins PN, et al. Genes, hypertension, and early familial coronary heart disease. Hypertension: pathophysiology, diagnosis and management. Laragh JH, Brenner BM, eds. Raven Press Ltd, New York 1990; 9: 127-36. 27 Lindeman RD, Tobin JD, Shock NW. Association between blood pressure and the rate of decline in renal function with age. Kidney Int 1984; 26: 861-68.

17

28 Cerasola G, Cottone S, D’Ignoto G, et al. Microalbuminuria points out early renal and cardiovascular changes in essential hypertension. Rev Lat Cardiol 1992; 13: 3-7. 29 Javier Salazar F, Pinilla JM, Lopez F, Romero JC, Quesada T. Renal effects of a prolonged synthesis inhibition of endothelium derived nitric oxide. Hypertens 1992; 20: 113-17. 30 Nichols TC, Bellinger DA, Tate DA, et al. Von Willebrand Factor and occlusive arterial thrombosis. A study in normal and Von Willebrand’s

disease pigs with diet induced hypercholesterolemia and atherosclerosis. Arteriosclerosis 1990; 10: 449-61. 31 Mettinger KL. A study of hemostasis in ischemic cerebrovascular disease I. Abnormalities in factor VII and antithrombin. Thromb Res 1982; 26: 183-92. 32 Jansson JH, Nilsson TK, Johnson O. Von Willebrand factor in plasma: a novel risk factor for recurrent myocardial infarction and death. Br Heart J 1991; 66: 351-55.

Double-blind trial of lignocaine versus sotalol for acute termination of spontaneous sustained ventricular tachycardia

Summary The efficacy of antiarrhythmic drugs for terminating sustained ventricular tachycardia (VT) has been disappointing. Lignocaine is the traditional drug but it is not very effective. Sotalol, one of the most effective drugs in suppressing spontaneous or induced VT, should theoretically be useful in this setting. We have compared lignocaine with sotalol for the acute termination of spontaneous sustained VT not causing cardiac arrest in 33 patients (26 males, 7 females, aged 21-90) whose underlying heart disease was old myocardial infarction (28), acute myocardial infarction (2), dilated cardiomyopathy (1), or idiopathic cardiomyopathy (2). Left-ventricular ejection fraction was 35% (range 18-76%). Patients were randomly allocated in a double-blind fashion to lignocaine 100 mg ( n=17) or sotalol 100 mg ( =16) given intravenously over 5 min. Those with persistent VT 15 min after onset of administration of the first drug were crossed over to the other

drug. Sotalol was significantly more effective than lignocaine whether analysed on an intention-to-treat basis (69% vs 18%; 95% confidence interval for absolute difference of 51% 22-80%, p=0·003) or by analysis limited to the 31 patients with subsequent electrophysiologically proven VT (69% vs 20%). 1 patient in each group required cardioversion after the first drug. Tachycardia persisted in 14 patients in the lignocaine group and 4 in the sotalol group after 15 min. Tachycardia ceased in 7 (50%) patients who crossed over to sotalol, and in 1 patient who crossed over to lignocaine. There was 1 death in each group after the first drug and 1 death after both drugs. We conclude that sotalol was superior to lignocaine for the acute termination of sustained VT. The incidence of adverse effects was similar for the two drugs. Lancet 1994; 344: 18-23

Department of Cardiology, Westmead Hospital, Sydney, NSW 2145, Australia (D SW Ho FRACP, R P Zecchin RN, D A B Richards FRACP, Prof J B Uther FRACP, Prof D L Ross FRACP) Correspondence to: Prof David L Ross 18

Introduction During ventricular tachycardia (VT) some patients may remain conscious long enough for an intervention other than immediate cardioversion. The usual approach is to try an antiarrhythmic drug; if that fails electrical cardioversion or termination by overdrive pacing is used. Lignocaine is the traditional drug for termination of sustained VT despite its low efficacy (18-19% for spontaneous VT and 30% for induced VT 1-3) . Early recurrence is frequent and repeat therapy with lignocaine is rarely effective.’ Sotalol has theoretical advantages over lignocaine. Programmed ventricular stimulation has shown sotalol to be one of the most effective drugs in preventing reinduction of VT4--8 while lignocaine is often ineffective in this setting.9,10 Sotalol prolongs action potentials and the effective refractory period while lignocaine shortens action potential duration.11-13 The effect on action potentials should potentiate the ability of sotalol to terminate tachycardia. On theoretical grounds and in experimental studies, therefore, sotalol seems likely to be more useful than lignocaine.14 Yet sotalol has not been studied for acute termination of spontaneous sustained VT; nor have controlled comparisons between antiarrhythmic drugs been done. We have compared intravenous lignocaine with sotalol in the acute termination of spontaneous sustained VT in patients who were conscious at the time of

presentation. Patients and methods Patients The protocol

was approved by the Western Sydney Area Health Service research and ethics committees for use in the three hospitals in the area. Sustained broad complex tachycardias judged to be VT on the basis of clinical history and a 12-lead electrocardiogram (ECG) in conscious patients in the wards or the emergency rooms were eligible. Patients were excluded if they had previously been entered into the trial (n=4), had received lignocaine (2) or sotalol (1) within the previous 24 hours, had poor haemodynamic status which was judged to require early DC shock (2), had torsade de pointes (none), or if the VT was interrupted by runs of sinus rhythm during the period from first documentation of sustained tachycardia to commencement of trial drugs (1). Patients with asthma, chronic obstructive airway disease, and controlled congestive cardiac failure were not excluded. Lignocaine was not given routinely by our paramedics or nurses for ventricular ectopy or for sustained wide QRS complex tachycardia not causing cardiac arrest, or to prevent ventricular arrhythmias in acute myocardial infarction.