Haemodynamic effects of frusemide in patients suspected of having acute myocardial infarction

Treatment of Hypertensive Cardiogenic Edema with Intravenous High-Dose Nitroglycerin in a Patient Presenting with Signs of Respiratory Failure: A Case Report and Review of the Literature

BACKGROUND Pulmonary edema is the accumulation of fluid in the lung secondary to increased hydrostatic pressure. Hypertensive cardiogenic pulmonary edema presents with a sudden onset of severe dyspnea, tachycardia, and tachypnea, and can occur when the systolic blood pressure exceeds 160 mmHg in association with acute decompensated congestive cardiac failure (CCF). A case is presented of hypertensive cardiogenic pulmonary edema treated with high-dose nitroglycerin and includes a review of the literature. CASE REPORT A 63-year-old Hispanic male with a medical history of hypertension, coronary artery disease, heart failure with a reduced ejection fraction of 35%, chronic kidney disease (CKD) and diabetes mellitus, presented as an emergency with acute, severe dyspnea. The patient was initially managed with 100% oxygen supplementation and intravenous (IV) high-dose nitroglycerin (30 mcg/min), which was titrated every 3 minutes, increasing by 15 mcg/min until a dose of 120 mcg/min was reached. After 18 minutes of aggressive therapy, the patient's condition improved and he no longer required mechanical ventilation. CONCLUSIONS Hypertensive cardiogenic pulmonary edema is a challenging clinical condition that should be diagnosed and managed as early as possible, and distinguished from respiratory failure due to other causes. Although hypertensive cardiogenic pulmonary edema is usually managed acutely with high-dose diuretics, this case has highlighted the benefit of high-dose IV nitroglycerin, and review of the literature supports this treatment approach.

Usefulness of lung impedance-guided pre-emptive therapy to prevent pulmonary edema during ST-elevation myocardial infarction and to improve long-term outcomes

Patients sustaining an ST-segment elevation myocardial infarction (STEMI) frequently develop pulmonary congestion or pulmonary edema (PED). We previously showed that lung impedance (LI) threshold decrease of 12% to 14% from baseline during admission for STEMI marks the onset of the transition zone from interstitial to alveolar edema and predicts evolution to PED with 98% probability. The aim of this study was to prove that pre-emptive LI-guided treatment may prevent PED and improve clinical outcomes. Five hundred sixty patients with STEMI and no signs of heart failure underwent LI monitoring for 84 ± 36 hours. Maximal LI decrease throughout monitoring did not exceed 12% in 347 patients who did not develop PED (group 1). In 213 patients LI reached the threshold level and, although still asymptomatic (Killip class I), these patients were then randomized to conventional (group 2, n = 142) or LI-guided (group 3, n = 71) pre-emptive therapy. In group 3, treatment was initiated at randomization (LI = -13.8 ± 0.6%). In contrast, conventionally treated patients (group 2) were treated only at onset of dyspnea occurring 4.1 ± 3.1 hours after randomization (LI = -25.8 ± 4.3%, p <0.001). All patients in group 2 but only 8 patients in group 3 (11%) developed Killip class II to IV PED (p <0.001). Unadjusted hospital mortality, length of stay, 1-year readmission rate, 6-year mortality, and new-onset heart failure occurred less in group 3 (p <0.001). Multivariate analysis adjusted for age, left ventricular ejection fraction, risk factors, peak creatine kinase, and admission creatinine and hemoglobin levels showed improved clinical outcome in group 3 (p <0.001). In conclusion, LI-guided pre-emptive therapy in patients with STEMI decreases the incidence of in-hospital PED and results in better short- and long-term outcomes.

Haemodynamic effects of rolofylline in the treatment of patients with heart failure and impaired renal function

AIMS

The direct effects of adenosine A1 receptor antagonists on haemodynamic parameters in patients with acute heart failure (HF) remain largely unknown.

METHODS AND RESULTS

We evaluated the haemodynamic effects of the AA(1)RA rolofylline in 59 HF patients with concomitant renal impairment (estimated creatinine clearance 20-80 mL/min). Placebo or rolofylline 30 mg was administered as a 4 h infusion followed by intravenous (i.v.) loop diuretic administration. Haemodynamic measurements were carried out hourly up to 8 h post-dosing by pulmonary artery catheterization. Urine output, fractional excretion of sodium, potassium, urea, and uric acid, and blood urea nitrogen (BUN) and creatinine levels were also measured. In both groups, the changes from baseline in all haemodynamic indices except mean pulmonary artery pressure (PAP) were not clinically significant. Mean [95% confidence interval (CI)] PAP showed a placebo-adjusted decrease with rolofylline of -1.5 (-4.1, 1.1)mmHg at Hour 4 and -3.5 mmHg (95% CI: -6.2, -0.2) at Hour 8. There was a significant increase with rolofylline in diuresis [placebo-corrected mean (95% CI) change of 68 (20, 116)mL/h at Hour 2-4 and 103 (21, 185)mL/h at Hour 4-8] and in fractional excretion of sodium, potassium, and uric acid. Placebo-corrected changes in plasma levels of creatinine and BUN with rolofylline were non-significant.

CONCLUSION

Single administration of rolofylline in patients with HF and impaired renal function produced a slight decrease in mean PAP and consistently increased diuresis and natriuresis without compromising renal function, both before and after administration of i.v. loop diuretics.

Modern Management of Cardiogenic Pulmonary Edema

Cardiogenic pulmonary edema (CPE) is a life-threatening condition that is frequently encountered in standard emergency medicine practice. Traditionally, diagnosis was based on physical assessment and chest radiography and treatment focused on the use of morphine sulfate and diuretics. Numerous advances in diagnosis and treatment have been made, however. Serum testing for B-type natriuretic peptide (BNP) has improved the accuracy of diagnoses in these patients. Treatment should focus on fluid redistribution with aggressive preload and afterload reduction rather than simply on diuresis. Some specific medications and noninvasive positive pressure ventilation have been shown to be safe and rapidly effective in improving patients' symptoms and improve outcomes.

Diuretic therapy in elderly heart failure patients with and without left ventricular systolic dysfunction

Long term prescription of diuretics for heart failure is very prevalent among elderly patients, although the rationale for such a treatment strategy is often unclear, as diuretics are not indicated if volume overload is absent. The concept of diastolic heart failure in the elderly might particularly change the role of diuretic therapy, since diuretics may have additional adverse effects in these patients. This paper reviews the effects of diuretic therapy in elderly patients with heart failure, emphasising the differences between patients with normal and decreased left ventricular systolic function. Studies on diuretic withdrawal in elderly patients with heart failure are discussed, with emphasis on issues involved in decision making such as diuretic dose reduction and withdrawal in elderly patients and factors that have been established to predict successful withdrawal. Existing guidelines on the prescription of diuretics in elderly patients with heart failure with normal and decreased left ventricular systolic function and in those with diastolic heart failure are also discussed. By reducing intravascular volume, diuretics may further impair ventricular diastolic filling in patients with diastolic heart failure and thus reduce stroke volume. Indeed, preliminary studies demonstrate that diuretics may provoke or aggravate hypotension on standing and after meals in these patients. Therefore, it is suggested that elderly patients with heart failure with intact left ventricular systolic function should not receive long term diuretic therapy, unless proven necessary to treat or prevent congestive heart failure. This implies that physicians should carefully evaluate the opportunities for diuretic dose tapering or withdrawal in all of these patients, and that a cautiously guided intermittent diuretic treatment modality may be critical in the care for older patients with heart failure with intact left ventricular systolic function.

Diuretics in postinfarction heart failure

Severe left ventricular failure, as evidenced by radiographic pulmonary edema or raised left ventricular filling pressure, accompanying acute myocardial infarction, carries a high mortality risk. In this situation, the intravenous loop-diuretic furosemide induces a rapid reduction in the raised left ventricular filling pressure due to an immediate and substantial increase in systemic venous compliance accompanied by increasing diuresis. This diuretic-induced venodilatation is probably due to the release of prostaglandins. The transient systemic arterial constriction and small increase in systemic blood pressure that follows intravenous furosemide probably results from the release of renin and subsequent activation of angiotensin. These diuretic induced hemodynamic changes are accompanied by restoration of the vasodilator reflex, which enables the heart to accommodate an acute volume load. Orally administered loop diuretics achieve slower, but similar, directional hemodynamic changes. There is no information on hemodynamic or neuroendocrine dose-response effects of loop diuretics, and there is no information pertaining to the use of other diuretic groups in this situation. The hemodynamic changes induced by furosemide summate with the changes induced by other anti-heart-failure drugs. In this subset of patients with acute myocardial infarction and severe heart failure, the influence of the diuretics on morbidity incidence and mortality risk remains to be measured.

Central hemodynamic effects of diuretic therapy in chronic heart failure

In chronic heart failure diuretic drugs improve central hemodynamic variables and cardiac pumping secondary to altered plasma and extracellular volumes; humoral markers of these changes include increased plasma renin and aldosterone levels. The latter increases are maximal over the first week but decline with chronic therapy. The plasma alpha-ANP levels show a reciprocal effect; these data are compatible with a rapid contraction of the plasma volume which is sustained during chronic therapy. The acute hemodynamic actions of diuretic agents reflect both immediate and direct vascular actions and also effects secondary to diuresis (volume redistribution). At rest substantial reductions in pulmonary "wedge" pressure (-29%), with a consequent fall in cardiac output (-10%), are described. Total systemic vascular resistance initially increases but "reverse autoregulation" over subsequent weeks returns this elevation gradually towards control values. Tolerance to these initial hemodynamic effects does not occur with maintained therapy; moreover, echocardiographic markers of contractility and exercise capacity may increase. The early venodilator effects of diuretic drugs can be attributed to prostaglandin release and the initial pressor actions to activation of the renin angiotensin system; these vascular actions may have limited relevance to long-term beneficial effects on hemodynamics. Direct pulmonary vasodilation and improved pulmonary compliance remain an interesting finding. Although most patients are both symptomatically and hemodynamically improved at rest, the actions during exercise are more varied. Some individuals with severely impaired left ventricular function show little hemodynamic improvement, whereas those with milder dysfunction usually benefit; in the main this is probably related to the latter being on a steeper cardiac function curve.(ABSTRACT TRUNCATED AT 250 WORDS)

Isosorbide-5-mononitrate in the treatment of acute left ventricular failure following acute myocardial infarction

Eleven patients with acute left ventricular failure following acute myocardial infarction (mean pulmonary capillary wedge pressure [PCW] greater than 20 mmHg) were entered into an open, haemodynamic study of oral isosorbide-5-mononitrate (ISMN). Left ventricular failure was resistant to intravenous diuretic therapy. No patient received concurrent cardioactive drugs nor further diuretic therapy during the study period. Haemodynamic data were acquired via a flow-directed thermodilution catheter placed in the pulmonary artery. Baseline data were acquired prior to the intravenous administration of an ISMN challenge. Thereafter, oral ISMN (20 mg, 8 hourly) was administered over 48 h. Following intravenous ISMN challenge, mean PCW fell from 26.2 to 17.5 mmHg. Cardiac index fell from 2.4 to 2.3 l/min/m2 due to a fall in heart rate (103 to 96.8 beats/min) as stroke volume and blood pressure were unchanged. At 8 h following ISMN, two patients were withdrawn due to hypotension (systolic pressure less than 85 mmHg). In the remainder, PCW remained acceptable throughout 48 h (13.1 mmHg at 48 h), cardiac output, systemic blood pressure and heart rate showed no further significant change. These data suggest that ISMN is effective in the treatment of acute left ventricular failure following acute myocardial infarction.

Renal, haemodynamic, and hormonal effects of human alpha atrial natriuretic peptide in healthy volunteers

The effects of atrial natriuretic peptide (ANP) were investigated in six healthy male volunteers taking a constant diet (120 mmol sodium and 60 mmol potassium daily). They were given an intravenous bolus of 100 micrograms human alpha-ANP on one day or placebo on another day 1-3 weeks apart in a double-blind randomised study. After ANP, urinary sodium excretion increased four-fold, and urine volume, calcium, magnesium, and phosphorus excretion doubled within 30 min of the injection. ANP induced an immediate fall in arterial pressure, followed by a longer vasodepressor phase which exceeded the duration of the effect on electrolyte excretion. There were no significant changes in plasma renin activity, aldosterone, antidiuretic hormone, or noradrenaline when compared with placebo.

Hemodynamic trial of sequential treatment with diuretic, vasodilator, and positive inotropic drugs in left ventricular failure following acute myocardial infarction

The circulatory effects induced by two sequential intravenous treatment programs with a diuretic, arteriolar or venodilator , and a positive inotropic drug were studied in a randomized between-group trial in 20 male patients with radiographic and hemodynamic evidence of left ventricular (LV) failure following acute myocardial infarction (AMI). Furosemide induced a substantial diuresis in both groups of patients, in association with reductions in LV filling pressure (p less than 0.01) and cardiac output (p less than 0.05), without significant change in heart rate or systemic arterial pressure. The addition of isosorbide dinitrate was followed by reductions in the systemic arterial (p less than 0.01) and LV filling pressures (p less than 0.01) without significant change in the heart rate or cardiac output. Hydralazine after furosemide reduced systemic vascular resistance (p less than 0.01), but the fall in mean blood pressure (p less than 0.01) was limited by the increase in cardiac output (p less than 0.01); heart rate was also increased (p less than 0.01) and LV filling pressure fell (p less than 0.05). The final addition of the beta-1 adrenoceptor agonist, prenalterol, increased systemic arterial systolic pressure (p less than 0.05), cardiac output (p less than 0.05), and heart rate (p less than 0.01), and reduced systemic vascular resistance (p less than 0.01) in both groups; these changes were greatest in those pretreated with furosemide and isosorbide dinitrate. In both treatment pathways compared with control the reductions in systemic vascular resistance and left heart filling pressure were accompanied by increases in heart rate and cardiac output without substantial changes in systemic blood pressure. Which of these hemodynamic pathways offers the optimum prognosis awaits further study.

Diuresis with continuous infusion of furosemide after cardiac surgery

We prospectively evaluated the diuretic effect of furosemide administered by bolus injection and by continuous infusion in 18 cardiac surgery patients. Nine patients were randomly assigned to receive 0.3 mg/kg of furosemide as a bolus injection at time 0 and again 6 hours later (nine patients) or 0.05 mg/kg per hour of furosemide as a constant infusion for 12 hours (nine patients). There were no significant differences between groups with respect to age, weight, creatinine clearance, changes in serum sodium and potassium levels, total urinary concentrations of sodium and potassium, or total urine volume for 12 hours. Diuresis during continuous infusion of furosemide was less variable from hour to hour than after bolus injection of furosemide and was sustained throughout the infusion period. Although the continuous infusion of furosemide will not provide the rapid and vigorous diuresis that is necessary in some clinical situations, it may be useful whenever a gentle, sustained diuresis is desired.

Haemodynamic advantages of isosorbide dinitrate over frusemide in acute heart-failure following myocardial infarction

The immediate haemodynamic effects of intravenous frusemide (1 mg/kg) and intravenous isosorbide dinitrate (50-200 micrograms/kg/h) were compared in a prospective, randomised, between-group study in 28 men with radiographic and haemodynamic evidence of left ventricular failure following acute myocardial infarction. The diuresis induced by frusemide reduced the left heart filling pressure and cardiac output and transiently raised systemic blood-pressure. In contrast, isosorbide dinitrate was accompanied by a reduction in systemic blood-pressure and peripheral resistance with the result that the cardiac output was not decreased despite a large fall in the pulmonary vascular and left heart filling pressures. These results indicate that reduction of excessive preload by venodilatation may be haemodynamically superior to that induced by diuresis in terms of both reducing myocardial oxygen consumption and maintaining peripheral perfusion. The influence of these contrasting treatments on the prognosis of these high-risk patients warrants further study.

Correlation between changes in systolic time intervals and left ventricular end-diastolic diameter after preload reduction. Non-invasive monitoring of pharmacological intervention

In order to measure the effect of a decrease in preload on systolic time intervals and left ventricular end-diastolic diameter (LVEDD) measured by echocardiography, eight healthy young subjects were given 40 mg frusemide intravenously. The pre-ejection period index (PEPI) increased and the left ventricular end-diastolic diameter decreased. A correlation between delta PEPI and delta LVEDD was shown. Using changes in systolic time intervals in the evaluation of changes in contractility it is important to correct for changes in preload. For normal subjects it is suggested that the relation between delta PEPI and delta LVEDD as a percentage of the mean values should be used for this correction. A method is suggested for estimating the changes in pre-ejection period index induced by changes in left ventricular end-diastolic diameter.

Effect of afterload reduction on plasma volume during acute heart failure

Previous investigations in our unit indicated that acute cardiogenic pulmonary edema is associated not only with an increase in left ventricular end-diastolic pressure and pulmonary arterial wedge pressure but also with a relative increase in colloid osmotic (oncotic) pressure and peripheral hemoglobin concentration. This combination of changes suggested that acute congestive heart failure with pulmonary edema, unlike chronic congestive heart failure, is associated with a contraction of intravascular blood volume. In this study, plasma volume changes were measured before and during the treatment of acute cardiogenic pulmonary edema in 14 patients with arteriosclerotic heart disease. The plasma volume measurement in all 14 patients before the initiation of treatment was either normal or decreased. After treatment with the alpha adrenergic blocking agent phentolamine, the plasma volume increased rather than decreased when measured 4 and 12 hours after the initiation of treatment. During this time colloid osmotic pressure and peripheral hemoglobin concentration progressively decreased. These findings suggest that acute cardiogenic pulmonary edema is associated with the extravasation of large quantities of plasma water from the intravascular compartment into the interstitial compartment and contraction of the intravascular plasma volume. The treatment of acute cardiogenic pulmonary edema is associated with the return of hypo-oncotic fluid from the interstitial compartment back into the intravascular compartment with expansion of plasma volume and reduction of colloid osmotic pressure and hemoglobin concentration.

Decreased pulmonary transvascular fluid filtration in awake newborn lambs after intravenous furosemide

We studied the effect of furosemide on pulmonary transvascular filtration of fluid and microvascular permeability to plasma proteins by measuring steady-state lung lymph flow and protein flow, pulmonary arterial and left atrial pressures in nine 1-wk-old unanesthetized lambs before and after rapid intravenous infusion of furosemide, 1 mg/kg in 10 experiments and 8 mg/kg in 5 experiments. With rapid diuresis induced by furosemide (an eightfold increase in urine flow), lung vascular pressures decreased, protein concentrations of lymph and plasma increased, and there was a consistent decrease in lymph flow and lymph protein flow, more pronounced after the larger dose. Five additional lambs received 8 mg/kg of furosemide intravenously in the presence of saline-induced pulmonary edema; in these experiments, the decrease in vascular pressures, increase in transvascular protein gradient, and decrease in lymph flow were greater than in lambs without pulmonary edema. These findings suggest that furosemide decreases transvascular filtration of fluid in the lung by diminishing the transvascular hydraulic pressure gradient and increasing the transvascular gradient for protein osmotic pressure. In five acute experiments on anesthetized lambs with kidneys removed, 8 mg/kg of intravenous furosemide decreased lymph flow one-half as much as it did in the presence of kidneys, with no change in lung vascular pressures or protein concentrations. The results of experiments in lambs without kidneys are consistent with a reduction in the vascular surface area for exchange of fluid and protein in the lung.

Hospital treatment of congestive heart failure. Management according to hemodynamic profile

Selection of therapy for subjects with acute congestive dardiac failure usually involves a choice among a diuretic, a vasodilator and an inotropic agent. Three principal questions are involved in the decision: (1) Is cardiac out normal or depressed? (2) Is blood pressure normal or depressed? (3) is regional myocardial ischemia present? Diuretics are safe and easy to administer, but they do not increase cardiac output or relieve hypoperfusion. Inotropic agents increase cardiac output but differ widely in their effects on blood pressure: selection of specific agents is influenced by their blood pressure effect. All inotropic agents, however, potentially aggravate regional myocardial ischemia. In ischemic heart failure, therefore, vasodilators which also increase cardiac output, may be chosen. Vasodilator administration is in turn limited by the decrease in arterial pressure which accompanies increasing infusion rate. When these three questions are considered in combination, an effective therapeutic regimen can be identified. Thus, congestion without hypoperfusion requires a diuretic if blood pressure is normal; and a vasodilator when blood pressure is increased. In the presence of congestion with hypoperfusion, a vasodilator is employed if blood pressure is normal; and a positive inotropic drug when blood pressure is depressed.

Acute effects of furosemide on blood electrolytes and hemodynamics in dogs

The effects of furosemide on the hemodynamics, blood electrolytes, and urinary output in 5 anesthetized dogs were studied. There were no significant changes in blood Na+ or Ca++ levels, but K+ decreased significantly after 15 minutes of furosemide treatment. There were no significant changes in the blood pressure, heart rate, left ventricular systolic pressure, index of left ventricular contractility [(dp/dt)/IIP], or systemic vascular resistance. Left ventricular dp/dt decreased for 30 to 60 minutes. Later the dp/dt and (dp/dt)/IIP of left ventricular pressure exceeded control values, although increases were not significant. Left ventricular work index and stroke volume decreased significantly between 30 and 90 minutes. The cardiac output and cardiac index also decreased. Left ventricular end-diastolic pressure decreased significantly only at 30 minutes. Cardiac function remained unchanged and consistent with the electrolytes changes. Although there was a marked diuresis, which normally must have significantly decreased the effective blood volume and hence the myocardial contractility, the cardiac function remained unchanged. These results suggests that furosemide might have a direct effect on the myocardium. Clinical improvement in patients might be the result of a direct effect on the myocardium aside from its effect due to diuresis.

Acute effects of furosemide and mannitol on central haemodynamics in the early postoperative period

The effects of furosemide and mannitol on central haemodynamics in the early postoperative period were investigated in 16 patients, who had undergone upper abdominal surgery. Measurements were performed prior to, and then 10, 30, 50 and 90 min after postoperative drug administration. Furosemide administration resulted in reductions of cardiac output, mean pulmonary arterial, pulmonary capillary wedge, and mean systemic arterial pressures, while systemic vascular resistance increased. Mannitol administration on the other hand, caused increases of cardiac output, mean pulmonary arterial and pulmonary capillary wedge pressures. Systemic vascular resistance decreased. It is concluded that mannitol should be used as the diuretic of choice in the treatment of postoperative (post-traumatic) oliguria in patients without known cardiovascular disease.

Correlative classification of clinical and hemodynamic function after acute myocardial infarction

To characterize the relation between clinical and hemodynamic state in acute myocardial infarction, 200 patients with acute infarction were evaluated with clinical and hemodynamic criteria. Patients were classified clinically on the basis of peripheral hypoperfusion (hypotension, tachycardia, confusion, cyanosis, oliguria) and pulmonary congestion (rales, abnormal chest roentgenogram). Four clinical subsets were defined that correlated with cardiac index (Cl, liters/min per m2) and pulmonary capillary pressure (PCP, mm Hg): (see article). Parallel hemodynamic subsets were developed independently on the basis of depressed cardiac index (2.2 liters/min per m2 or less) and elevated pulmonary capillary pressure (greater than 18 mm Hg). The rate of accuracy of clinical examination in predicting hemodynamic abnormalities was 83 percent. Mortality rates were similar in the clinical and hemodynamic subset calssifications, averaging 2.2 percent in subset I, 10.1 percent in subset II, 22.4 percent in subset III and 55.5 percent in subset IV. Drug interventions in the course of hospitalization resulted in a 38 percent increase in depressed cardiac index and 34 percent decrease in elevated pulmonary capillary pressure. Resolution of clinical abnormalities paralleled this hemodynamic improvement in 70 percent of patients. These data suggest that clinical performance and both clinical and hemodynamic subsets are directly relevant to establishing prognosis and the selection of therapy in patients with acute myocardial infarction.

Unimpeded plasma renin increase after intravenous furosemide during saline replacement

The effect on plasma renin activity of intravenous furosemide combined with saline replacement of the volume depletion was studied in twelve patients with insignificant heart disease. In ten of the patients the investigation was repeated without saline replacement. It was found that saline infusion, reducing or eliminating hemoconcentration, had no significant influence on the marked plasma renin increase. In eight of the patients the combined furosemide-saline study was performed during right-hear catheterization. Decrease in atrial pressures, known to occur within 15 min after furosemide intravenously, was virtually absent with the saline replacement. It is concluded that plasma volume reduction after intravenous furosemide is responsible for decreased filling pressures of the ventricles but not for plasma renin increase.

Diuretics: mechanism of action and clinical application

Despite the bewildering number of diuretics available to the physician, these drugs can be divided into 4 main groups, characterised by their site of action on sodium reabsorption in the kidney. Drugs acting on the ascending limb of the loop of Henle have a powerful but short acting diuretic effect; they include frusemide, ethacrynic acid and bumetanide. The benzothiadiazines and related compounds have a moderate diuretic action spread over a longer period, whilst the potassium-sparing diuretics, triamterene, amiloride and spironolactone, have only a weak diuretic effect but a marked ability to diminish urinary potassium excretion. The fourth group is made up of miscellaneous substances which function as vasodilator or osmotic agents. The pathogenesis of oedema formation in heart failure is outlined and a logical approach to treatment suggested. Duiretics are being increasingly used in the treatment of non-oedematous states, in particular hypertension, diabetes insipidus and hypercalciuria; their exact role in pregnancy and acute renal failure remains controversial. Side-effects can be related to their effect on electrolyte excretion and include hypokalaemia, hyponatraemia, hyperkalaemia and hyperuricaemia. The incidence of disturbed carbohydrate tolerance in previously normal individuals is low. Other less common side-effects are also discussed.