Search

There is a definite relation of diuretic treatment to impaired glucose tolerance and biochemical diabetes, and a probable relation to insulin resistance. The effect of diuretics on glucose tolerance is dose-related. Spironolactone does not impair glucose tolerance even at high dosage, but apparent differences between other diuretics may well be due to comparison at doses which are not equivalent. Diuretic-induced changes in carbohydrate metabolism are not conclusively related to altered potassium homeostasis, and impaired glucose tolerance occurs when relatively low doses of thiazide are combined with potassium-sparing agents. The effect of diuretics on glucose tolerance is largely and possibly wholly reversible. These disturbances of carbohydrate homeostasis have been detected by detailed biochemical testing, and their clinical importance is uncertain. In established diabetes, diuretics have a rapid and substantial adverse effect on metabolic control. In non-diabetic subjects diuretics may rarely cause or trigger a serious hyperosmolar non-ketotic diabetic syndrome. This apart, it is not known whether the metabolic changes cause clinical diabetes or lead to microvascular complications in the long-term. It is now established that biochemical diabetes, glucose intolerance and insulin resistance probably do not increase the risk of coronary heart disease in treated hypertensive patients. Diuretics should be avoided in patients with diabetes, otherwise they remain an excellent choice for first-line antihypertensive therapy.

This article summarises the new British Hypertension Society guidelines for management of hypertension, which have been published in full.1 Since the previous guidelines 2 3 much new evidence has emerged on optimal blood pressure targets4; management of hypertension in diabetic patients4–7; treatment of isolated systolic hypertension8; comparison of the antihypertensive efficacy and tolerability of different drug classes9–11; the role of non-pharmacological measures for prevention 12 13 and treatment of hypertension14; and additional benefits associated with the use of aspirin and statins. Of concern is that national surveys continue to reveal incomplete detection, treatment, and control of hypertension.15 Furthermore, treated hypertensive patients still die prematurely from cardiovascular disease.16 These guidelines aim to present the best currently available evidence on hypertension management and their implementation. #### Summary points Use non-pharmacological measures in all hypertensive and borderline hypertensive people Initiate antihypertensive drug treatment in people with sustained systolic blood pressure ≥160 mm Hg or sustained diastolic blood pressure ≥100 mm Hg Decide on treatment in people with sustained systolic blood pressure between 140 and 159 mm Hg or sustained diastolic blood pressure between 90 and 99 mm Hg according to the presence or absence of target organ damage, cardiovascular disease, diabetes, or a 10 year coronary heart disease risk ≥15% according to the Joint British Societies coronary heart disease risk assessment programme or risk chart Optimal blood pressure treatment targets are systolic blood pressure

To review the findings of prospective controlled trials of antihypertensive treatment and determine whether the evidence they have provided is embodied satisfactorily in current national and international guidelines for hypertension management.Conventional guidelines all advise prompt treatment of moderate-to-severe hypertension and treatment of even mild hypertension in subjects with cardiovascular disease, target organ damage or diabetes, and in the elderly; and treatment of isolated systolic hypertension in the elderly. All acknowledge that evidence for efficacy and safety of treatment is strongest for thiazide diuretics and beta-blockers.Conventional guidelines all emphasize the importance of long-term blood pressure, measured over some months, for treatment decisions. However the blood pressure for routine treatment varies from 160/100 mmHg (British Hypertension Society) to 140/90 mmHg (Joint National Committee V). This dictates very large differences in the number of patients to be treated to prevent a cardiovascular disease event and in the proportion of the population to be treated, yet the reasons for these differences are not explicit. None of the conventional guidelines is entirely satisfactory. The more conservative British Hypertension Society policy may leave untreated some middle-aged men who ought to be treated. The more aggressive Joint National Committee V policy will lead to treatment of some young subjects who have only a remote chance of benefit, at very high cost, and possibly with adverse harm-benefit consequences. RISK-BASED GUIDELINES: Guidelines developed in New Zealand target absolute cardiovascular disease risk in mild hypertension and have the potential to correct this shortcoming of conventional guidelines. However they require further consideration as regards the number needed to treat which is acceptable to well-informed patients, the appropriate estimate of relative cardiovascular disease risk reduction by treatment in mild hypertension, the pattern of treatment which will emerge and their acceptability in ordinary practice.Comparative evaluation will be needed to determine whether the outcome is better with conventional guidelines, which are simple but at the expense of accuracy, or with risk-targeted guidelines, which are more accurate but at the expense of simplicity.

1. In eight hypertensive patients with ACE inhibitor-induced cough the resolution of the cough was examined in a prospective observational study over 4 weeks duration. Resolution of cough was measured by visual analogue scales and questionnaire at baseline and days 3, 7, 14 and 28. In addition changes in cough sensitivity to inhaled capsaicin, and skin responses to bradykinin and substance-P were measured at the same time points. 2. All patients recorded significant subjective improvement in cough questionnaire scores for severity and night time waking, and by visual analogue scales for severity and frequency of cough (all P < 0.0005 for trend from day 0-28). Significant changes in subjective measures were recorded by 3 to 7 days for most measures, but further reductions were observed up to day 28 (all P < 0.01 day 28 vs day 0). 3. The sensitivity to inhaled capsaicin fell over the 28 days of study after stopping enalapril. The potency of capsaicin relative to day 0 was reduced to 0.25 (95% CI 0.07-0.87) by day 14, and to 0.20 (95% CI 0.06-0.67) by 28 days. 4. After stopping enalapril there was a highly significant reduction in wheal area produced by intradermal bradykinin, with the majority of the effect seen by day 3 (P < 0.0005). The wheal area to intradermal substance-P also declined with time after stopping enalapril, but significant changes were not observed until 14 days (P < 0.01). 5. Multiple regression analysis of the rates of decline for the subjective and objective measures of cough showed significant associations between the response to inhaled capsaicin and the VAS scores for severity of cough (P = 0.005) and frequency of cough (P = 0.011). Capsaicin response was not related significantly to the severity of cough measured by self-administered questionnaire. 6. There was a significant association between bradykinin response and VAS scores for frequency of cough (P < 0.04) and severity of cough (P < 0.05), but not with cough by questionnaire or the capsaicin response. The response to substance-P did not relate significantly to any of the measures of cough. 7. Cough caused by enalapril improved markedly by 14 days but took up to 28 days to resolve. It was associated with increased sensitivity to inhaled capsaicin which decreased over 28 days, and which paralleled changes in subjective cough scores.

One hundred and thirty five non-smoking hypertensive patients with ACE inhibitor cough confirmed by lisinopril rechallenge and placebo dechallenge were recruited into a double-blind random parallel-group comparison of losartan 50 mg, lisinopril 20 mg and hydrochlorothiazide 25 mg each given once daily for a maximum of 8 weeks. The aim of the study was to compare the incidence of cough with the angiotensin II antagonist losartan, the ACE inhibitor lisinopril and the hydrochlorothiazide in hypertensive patients with previous ACE inhibitor cough. Cough detected by self-administered questionnaire was the primary end-point, and cough frequency by visual analogue scale a secondary end-point. The incidence of cough with losartan (29%) was lower than that for lisinopril (72%, P0.01) and similar to that for hydrochlorothiazide (34%). Cough frequency by visual analogue scale was lower for losartan than lisinopril (P0.01) and similar to that for hydrochlorothiazide. The specific selective AT1 angiotensin II receptor antagonist losartan is significantly less likely than lisinopril to cause cough in patients who previously have had ACE inhibitor cough. ACE inhibitor cough is likely to be related to non-specific kininase II inhibition.

Weight reduction, moderate salt restriction and alcohol reduction are effective in lowering blood pressure (BP), and are feasible interventions for long-term management of hypertension. When used in combination these non-pharmacological measures are significantly inferior to drug therapy in anti-hypertensive effect. When they are implemented as a first step in the treatment of mild hypertension, resorting to drug therapy only if non-pharmacological measures fail, anti-hypertensive drugs can be avoided in about 40% of patients. However, BP control is again significantly inferior with this strategy compared with drug therapy without non-pharmacological advice. Those given advice on non-pharmacological measures may therefore have suboptimal protection against cardiovascular complications. This is particularly so when the threshold for drug treatment is set at a DBP ofor = 100 mmHg, as many patients will be left untreated with DBPs between 90 and 99 mm Hg as a result of non-pharmacological measures. Non-pharmacological treatment may thus stand between patients and anti-hypertensive drug therapy, which nowadays is simple, well-tolerated, safe and proven effective in preventing cardiovascular disease. The role of non-pharmacological therapy needs to be reconsidered.

Guidelines for the management of hypertension, and those for hyperlipidaemia, advocate measurement of cholesterol in all hypertensive subjects. It is suggested that knowledge of serum cholesterol should influence the choice of anti-hypertensive agent because thiazides and beta-blockers may influence lipids adversely. However, the changes in lipids associated with low-dose thiazides and beta-blockers are small, not sustained and do not appear to affect prognosis adversely. Some believe that knowledge of serum cholesterol may help target the treatment of mild hypertension more accurately by predicting an increased risk of vascular complications of hypertension. However, serum cholesterol does not predict the risk of cardiovascular complications in hypertensive women. It does predict coronary heart disease (but not stroke) in men, but coronary risk can be estimated satisfactorily without knowledge of serum cholesterol. Suggestions that cholesterol should be lowered in hypercholesterolaemic hypertensive patients to reduce the incidence of coronary heart disease are understandable. However, the diet recommended for this has no useful effect on serum cholesterol, and the benefit of lipid-lowering drugs for cholesterol reduction only exceeds the risk in patients at very high risk of coronary mortality, for example those who have had a myocardial infarction. Hypertension alone will not place individuals at high enough risk to warrant drug therapy for cholesterol reduction. We conclude that the current guidelines are incorrect, and that routine measurement of cholesterol in all hypertensive patients is not justified.

To compare the incidence of cough in patients with a history of angiotensin converting enzyme (ACE) inhibitor-related cough who received losartan [a type 1 angiotensin II (Ang II) receptor antagonist], lisinopril (an ACE inhibitor) or hydrochlorothiazide (a diuretic).An international, multicentre, randomized double-blind, parallel-group controlled trial.Outpatient clinics at 20 tertiary care medical centres in 11 countries.One hundred and thirty-five patients with uncomplicated primary hypertension with a history of ACE inhibitor-related cough were randomly assigned to the double-blind treatment phase and completed the study.After confirming that the cough was ACE inhibitor-related by a single-blind rechallenge, followed by a placebo washout period, patients were randomly assigned to receive 50mg losartan, 20mg lisinopril or 25mg hydrochlorothiazide once a day for 8 weeks.Cough incidence, severity and frequency were assessed by a self-administered questionnaire and a visual analogue scale.The percentage of patients who complained of cough was significantly higher with lisinopril than with losartan or hydrochlorothiazide. The mean visual analogue scale scores for patients treated with lisinopril demonstrated that these patients coughed more frequently than those who received losartan or hydrochlorothiazide.The incidence of cough related to the type 1 Ang II receptor antagonist losartan is significantly lower than that observed with lisinopril, and similar to that observed with hydrochlorothiazide in patients with a rechallenged ACE inhibitor cough. Type 1 Ang II receptor antagonists represent a potential new treatment for hypertensive patients in whom ACE inhibitors are indicated, but who develop a cough with these agents.

OBJECTIVE--To audit avoidable deaths from stroke and hypertensive disease. DESIGN--Details of care before death were obtained from general practitioners and other doctors, anonymised, and assessed by two experts against agreed minimum standards of good practice for detecting and managing hypertension. SETTING--Health authority with population of 250,000. SUBJECTS--All patients under 75 years who died of stroke, hypertensive disease, or hypertension related causes during November 1990 to October 1991. MAIN OUTCOME MEASURES--Presence of important avoidable factors and departures from minimum standards of good practice. RESULTS--Adequate information was obtained for 88% (123/139) of eligible cases. Agreement between the assessors was mostly satisfactory. 29% (36/123, 95% confidence interval 21% to 37%) of all cases and 44% (36/81, 34% to 55%) of those with definite hypertension had avoidable factors that may have contributed to death. These were most commonly failures of follow up and continuing smoking. Assessment against standards of minimum good practice showed that care was inadequate but not necessarily deemed to have contributed to death, in a large proportion of patients with definite hypertension. Common shortcomings were inadequate follow up, clinical investigation, and recording of smoking and other relevant risk behaviours. CONCLUSIONS--This method of audit can identify shortcomings in care of patients dying of hypertension related disease.

To compare low-dose thiazides to beta-blockers, angiotensin converting enzyme (ACE) inhibitors, calcium antagonists and alpha-blockers for simplicity, tolerability, efficacy, safety and cost-effectiveness as first-line treatment for hypertension.Review of short-term comparative studies, and the outcome of long-term trials with vascular complications of hypertension as endpoints. SIMPLICITY: Among the advantages of thiazides are a flat dose-response; no dose titration; effectiveness when used once a day; no first-dose hypotension; and few contra-indications.Thiazides are the best tolerated agents in patients over the age of 60 years and in younger women. They sometimes cause gout and impotence in younger men, in whom beta-blockers are equally acceptable first-line therapy.Thiazide-based regimens have consistently reduced vascular complications of hypertension, the real measure of efficacy. There is little evidence that regimens based on other drugs are effective in this sense.Concerns that thiazide-induced biochemical changes cause coronary events are baseless. An overview of outcome trials shows that thiazide-based treatment reduces coronary events significantly, and the reduction is not significantly different from that predicted by epidemiological data. Thiazide-based therapy has also reduced coronary events significantly and substantially in elderly patients with isolated systolic hypertension.Low-dose thiazide treatment needs minimal monitoring, and has proved most cost-effective in formal analyses.Low-dose thiazide treatment is a clear first-line choice for patients aged over 60 years and younger women, except those with diabetes or gout. In younger men there is little to choose between thiazides and beta-blockers.

Treatment with thiazide diuretics causes impaired glucose tolerance, biochemical diabetes, and insulin resistance. The effect of diuretics on glucose tolerance is clearly dose-related. Spironolactone does not impair glucose tolerance, even at high dosage, but differences among other diuretics could be due to comparisons at doses that are not equal. Diuretic-induced changes in glucose metabolism are not conclusively related to altered potassium homeostasis, and impaired glucose tolerance occurs even when relatively low doses of thiazide are combined with potassium-sparing agents. The effects of diuretics on glucose homeostasis are in large part and probably entirely reversible. These disturbances of glucose metabolism have been detected only by detailed biochemical testing, and their clinical relevance is uncertain. In established diabetes, diuretics have a rapid and substantial adverse effect on metabolic control. In nondiabetic subjects, diuretics rarely cause or trigger a serious hyperosmolar nonketotic diabetic syndrome. Otherwise, it is not known whether the metabolic changes cause clinical diabetes or lead to microvascular complications in the long term. Evidence from large outcome trials suggests that biochemical diabetes, glucose intolerance, and insulin resistance do not increase the risk of coronary heart disease in treated hypertensive patients. Diuretics should be avoided in patients with diabetes unless their use is essential. Otherwise, a low dose of thiazide remains as excellent choice for first-line antihypertensive therapy. Dihydropyridine calcium antagonists, diltiazem, and verapamil appear to have no important effects on glucose homeostasis. There is very limited evidence that selective alpha-antagonists increase insulin sensitivity. The importance of metabolic differences between drug classes will be established only by comparative outcome trials with coronary events as the end point.

The prevalence and severity of cough during long-term enalapril treatment were examined by comparing a cohort of 136 hypertensive patients who started treatment with enalapril with consecutive age and sex-matched patients who commenced nifedipine therapy during the same period. Cough and other symptoms were assessed by a questionnaire designed to avoid bias towards reporting cough. After a mean of 27 months' treatment patients on enalapril had an excess of persistent cough (16 per cent, 95 per cent CI 7-25, p less than 0.01), voice change (14 per cent, 95 per cent CI 2-27, p less than 0.05) and sore throat (10 per cent, 95 per cent CI -0.1 to 20.3 per cent, p less than 0.01) when compared to nifedipine-treated patients. The cough was usually dry, moderate or severe, paroxysmal, and troublesome at night. Cough tended to be more common in women (23 per cent vs. 7.2 per cent), non-smokers, and at higher doses of enalapril, but was not related to age, duration of treatment, or chronic respiratory disease. Dry cough commonly persists as a troublesome side-effect during long-term enalapril treatment, and is often associated with voice change and sore throat.

Drug treatment of hypertension reduced the incidence of stroke by 40% and prevented cardiac and renal failure, but had less effect on coronary artery disease in the major trials. Coronary events were reduced by about 14% (confidence limits of 4-22%), which compares to the reduction of 20-25% anticipated from epidemiological studies. Any shortfall in coronary prevention that could be attributable to adverse effects of antihypertensive drugs would therefore amount to only 6-11% of all coronary events, and could easily arise by chance. At present, there are no grounds for preferring one class of drug over another from the point of view of coronary prevention, or for altering the "target" diastolic pressure during treatment. Strategies other than blood pressure reduction aimed at preventing coronary disease in hypertensive subjects need to be explored. They should, however, be examined in large controlled outcome trials, and should not be introduced to ordinary practice on the basis of the inadequate evidence currently available.

In a cohort of 136 hypertensive patients started consecutively on enalapril the incidence of persistent dry cough by life-table analysis was 14.6% (95% CI 10.2-19.0%). The incidence in women (19.2%; 95% CI 11.3-27.1%) was twice that in men (9.7%; 95% CI 6.6-12.8%). Dry cough was unrelated to age, smoking habit, renal function, or the dose and duration of enalapril treatment. In one half of patients who developed cough enalapril had to be stopped. The incidence of withdrawal due to cough was 6.0% (95% CI 4.5-7.5%), and cough was by far the most common reason for discontinuing enalapril treatment. Reviewing previous studies of enalapril-induced cough, it is evident that postmarketing surveillance studies have grossly underestimated the incidence and importance of this side-effect. Surveys in hospital clinics have slightly underestimated the true incidence through failure to use life-table methods of analysis.

ACE inhibitors have antihypertensive efficacy similar to that of thiazides, beta-blockers and calcium antagonists. They are simple to prescribe and to take, and are relatively free from subjective side-effects apart from persistent dry cough. There are few specific concerns about safety, but experience of long-term use is still relatively limited. They are valuable additional or alternative antihypertensive drugs when standard therapy is contraindicated or fails to control blood pressure.

Summary The association between haemospermia and hypertension was examined in a case-control study comparing 5 hypertensive patients with haemospermia to 20 age-matched hypertensive men. Patients with haemospermia had much higher blood pressures than hypertensive controls (200/131 mmHg vs 147/90 mmHg; P < 0.0005/P < 0.0001), higher left ventricular voltage on ECG (P < 0.02), and higher concentrations of serum creatinine, proteinuria and renovascular disease (all P = 0.06 vs controls). Haemospermia is associated with severe uncontrolled hypertension. It is not, however, associated with hypertension per se, as the prevalence of hypertension in published series of patients with haemospermia is no higher than that expected in the general population. Men presenting with haemospermia should have their blood pressure measured carefully as they may require antihypertensive treatment urgently.

The Dinamap 845 blood pressure recorder has been evaluated over a wide range of blood pressure by comparison with the Hawksley random zero sphygmomanometer in 32 subjects, six of whom had a cardiac arrhythmia. Group mean radings for systolic and phase 5 diastolic pressure were almost identical but Dinamap diastolic values were on average significantly lower (mean difference 3.4 mmHg) than phase 4 diastolic readings obtained with the Hawksley machine. Correlations between readings with the two instruments were high but the slopes and intercepts of the regression for systolic but not diastolic pressure were significantly different from unity and zero, respectively. The Dinamap is easy to use, portable, and capable of rejecting some motion artefact. Its major disadvantage is that the systolic blood pressure measurement is limited to a maximum of 210 mmHg, a point not made clear in the manufacturer's literature. Nevertheless, the Dinamap 845 is acceptable for blood pressure determinations in subjects who are normotensive or who have mild hypertension.

There is a correlation between body weight or obesity and blood pressure in the population, and most hypertensive patients are overweight. Weight reduction of 6 kg or more reduces blood pressure by an average of 15/9 mmHg even if the ideal body weight is not attained, and this is not explained by changes in salt intake or by measurement artefact due to the fat arm. Some form of calorie restriction is the mainstay of managing obesity in hypertensive patients, and those referred to a dietitian lose significantly more weight tham those given a diet sheet by a doctor or given no specific advice. About 50% of patients referred to a dietitian reduce their body weight by 6 kg or more. There is a tendency to regain weight in the long-term, but after four years more than one third of patients remain at least 6 kg below their initial weight. Men seem to lose relatively little weight whether or not they are referred to a dietitian, whereas women referred to a dietitian achieve substantial and sustained weight loss.

Summary In hypertensive patients single doses of ketanserin 40 mg prolonged the corrected QT interval (QTc) for at least 8 hours, with a maximal increase of 35 ms (P less than 0.001, n = 6) after 2 hours. During chronic dosing (20 and 40 mg b.d.) the QTc was further prolonged, by 46 and 45 ms respectively. QTc prolongation after treatment with a mean dose of 73 mg/day for 7 weeks (n = 26) was significantly related to body weight (r = -0.58, P less than 0.01), and to the dose of ketanserin corrected for body weight (r = 0.63, P less than 0.01), but not to plasma concentrations of ketanserin, ketanserinol, potassium or calcium. High doses of ketanserin (mean dose 167 mg/day, n = 9) increased the QTc by 40 ms (P less than 0.001), with prolongation of up to 80 ms in individual patients. Treatment with ketanserin at doses proposed for clinical use (40-80 mg/day) may carry a risk of ventricular arrhythmias.

Non-gouty arthritis is quite common in hypertensive populations and there have been suggestions that drugs used to treat hypertension may be the cause. This possibility has been investigated in a case-control study. We identified 127 hypertensive patients with arthritis other than gout and matched them for age and sex to 254 hypertensive controls who did not have arthritis. Use of antihypertensive drugs was similar in cases and controls for all patients, and for those with undiagnosed arthritis. There was no significant excess use of any type of antihypertensive drug by cases with arthritis. These findings do not support the hypothesis that drug treatment for hypertension commonly causes arthritis. It is more likely that hypertension and arthritis occur together frequently because non-steroidal anti-inflammatory drugs raise blood pressure, and because both problems are common in middle-age.

Antihypertensive efficacy and tolerability of the dihydropyridine calcium antagonist nisoldipine were examined in a double-blind, placebo-controlled study. Twenty patients with hypertension (greater than 140/95 mmHg) uncontrolled by bendrofluazide plus atenolol were randomly allocated to receive nisoldipine, 10-20 mg daily, or placebo for six weeks. Nisoldipine, compared with placebo, lowered blood pressure (BP) by 31/21 mmHg (P less than 0.02/P less than 0.002) after two weeks. Thereafter its antihypertensive effect declined to 20/9 mmHg (NS/NS) at four weeks and 8/3 mmHg at six weeks (NS/NS). Compared with placebo, the rise in BP with nisoldipine during the last four weeks of the trial was highly significant (P less than 0.005). During this period BP rose by 14/11 mmHg (P less than 0.05/P less than 0.02) in the nisoldipine group so the apparent loss of effect was not attributable to a fall of BP in the placebo group. Compliance, as assessed by tablet counts, was satisfactory and nisoldipine treatment did not cause significant weight gain. The drug was not well tolerated and three patients withdrew from treatment because of side-effects. Tolerance to the antihypertensive effect of nisoldipine appears to develop rapidly and further studies should be performed to confirm this, and evaluate the mechanism.

Summary The antihypertensive effect of the selective serotonin antagonist ketanserin was examined in a double-blind, placebo-controlled, parallel group study in 20 patients with essential hypertension. After 7 weeks treatment with ketanserin (mean dose 71 mg/d) there was a significant fall of both systolic and diastolic blood pressure, as compared to placebo, with a peak effect of 19.1/9.1 mmHg lying (P < 0.01/P < 0.01), and 16.5/11.3 mmHg standing (P < 0.01/P < 0.01); twice daily dosage appeared satisfactory. Subjective side effects were similar in the ketanserin and placebo groups. Ketanserin is an effective antihypertensive drug of moderate potency when given twice daily, with no orthostatic effect.

Hydralazine, labetalol, methyldopa, minoxidil, prazosin and placebo were compared when added to atenolol 100 mg and bendrofluazide 5 mg daily in hypertensive patients inadequately controlled by the beta-blocker/diuretic combination. Atenolol was withdrawn in those allocated to labetalol and minoxidil was given only to men. The order of acceptability was: placebo, hydralazine, prazosin, methyldopa, minoxidil, labetalol. All the active agents were more effective than placebo. Minoxidil was more effective than the other active drugs, which had similar potency to one another. Hydralazine was the most generally suitable third drug, with prazosin a close second. Minoxidil was effective in the milder hypertensives, but in the present regimen caused fluid retention in those with more severe hypertension. Labetalol probably should be introduced at lower dose (150 mg daily) even as replacement for full doses of a previously administered beta-blocker.

A retrospective survey of serum potassium measurements in a blood pressure clinic provided evidence that potassium supplements averaging 24 mmol daily were effective in elevating serum potassium concentration and reducing the incidence of hypokalaemia in diuretic-treated hypertensives. In patients treated with bendrofluazide the serum potassium concentration was related to the dose of diuretic, but not to age, sex, body weight, supine blood pressure or serum creatinine. There was no evidence that drugs other than diuretics or potassium (including beta-adrenergic blockers) had a clinically significant influence on serum potassium. The major part of the variation in serum potassium between patients could not be explained by the variables examined.

Factors predisposing to diuretic-induced acute gout were investigated in a case-control study. Seventy hypertensive patients with acute gout occurring during diuretic treatment were identified and matched for age and sex to 140 hypertensive controls who took diuretics but did not develop gout. Gout was related more strongly to the use of loop diuretics than thiazides, with 30% of cases taking a loop diuretic compared to 14% of controls (P less than 0.01). In a sub-group of men who took a thiazide and no other diuretic, gout was significantly associated with obesity (odds ratio 3.7, 95% confidence interval 1.4, 9.1) and high alcohol intake (odds ratio 3.3, 95% confidence interval 1.1, 9.8). In these patients, 23% of gout was attributable to obesity and 16% to high alcohol consumption. Approximately 40% of acute gout might have been prevented by avoiding thiazides in those 20% of men who weighed more than 90 kg and/or consumed more than 56 units of alcohol per week.

Summary In 93 patients with hypertension uncontrolled by bendrofluazide 5 mg plus atenolol 100 mg daily, the effects of adding nifedipine (up to 60 mg/day, n = 31), prazosin (up to 20 mg/day, n = 31), or hydralazine (up to 200 mg/day, n = 31) were compared in a 6 month open random parallel group study. The three drugs did not differ significantly as regards antihypertensive effect, withdrawal rate, total number of side effects, or effect on serum biochemical variables. The pattern of side-effects differed. Headache, flushing and oedema were common with nifedipine, tiredness and drowsiness with prazosin, and headache with hydralazine. Nifedipine is an acceptable third-line antihypertensive drug which may have some advantage over hydralazine and prazosin.

Patients with mild hypertension who habitually smoked cigarettes and consumed caffeine were examined after they abstained from caffeine and cigarettes overnight. Their mean blood pressure (147/89 mm Hg) was substantially lower than values recorded in the clinic (164/102 mm Hg) and remained so when they continued to abstain (149/94 mm Hg at two hours). Smoking two cigarettes (3.4 mg nicotine) elevated blood pressure by 10/8 mm Hg, but for only 15 minutes. Drinking coffee (200 mg caffeine) elevated blood pressure by up to 10/7 mm Hg between one and two hours. Combined coffee ingestion and cigarette smoking caused a sustained rise in blood pressure from 5 to 120 minutes to levels similar to those measured in the clinic (162/102 mm Hg at two hours). Similar results were obtained in thiazide-treated patients. The interaction of coffee and cigarettes on blood pressure, but not on pulse rate, was significant. The pressor effect of cigarette smoking and caffeine ingestion in combination may be important in the evaluation of patients with mild hypertension.

A possible association between treatment with methyldopa and the occurrence of renal stones was sought by retrospectively comparing 54 hypertensive patients with calculi detected by X-ray examination with a control group matched for age and sex. Treatment with methyldopa was equally frequent in both groups. There was a significant excess of methyldopa use in a subgroup of 16 patients with bilateral calculi, but the drug could be the primary cause of stone formation in, at most, four of these patients. If methyldopa does cause renal stones, it is probably an uncommon event, but precipitation of the drug from the urine may worsen a tendency to stone formation, which will lead to a higher incidence of bilateral calculi.

Summary Two patients developed profound hypotension approximately one hour after taking an initial oral dose of ketanserin 40 mg. The reaction appeared similar to that reported with prazosin, and may have been due to the alpha 1-adrenoceptor antagonist action of ketanserin. Both patients were taking regular beta-blocker therapy, which may exacerbate such a reaction.

Hyperuricaemia was present in 18 out of 73 men with untreated mild hypertension and was related significantly to alcohol intake, serum aspartate transaminase activity, and obesity. In the whole group the mean serum urate concentration correlated highly significantly with alcohol intake and activities of serum aspartate and alanine transferases but not with ponderal index, serum creatinine concentration, age, or blood pressure. Hypertension and hyperuricaemia are related at least in part through their common association with frequent alcohol use. A serum urate concentration exceeding 0.5 mmol/l (8--4 mg/100 ml) in a man with untreated hypertension is highly suggestive of heavy alcohol consumption. There was no evidence that hyperuricaemia had a deleterious effect on renal function.