EDUCATIONAL REVIEW

Can acute renal failure be prevented?

H.F. GALLEY
Academic Unit of Anaesthesia and Intensive Care, University of Aberdeen, Aberdeen, U.K.

Correction of salt and volume depletion is paramount in the prevention of renal damage. Measures which stimulate intense filtration of glomeruli in acute renal failure, such as the use of atrial natriuretic peptide analogues, theophylline, dopamine, or growth factors should be regarded with caution, since they all increase metabolic workload in the outer medulla and hence aggravate medullary hypoxia. Neither frusemide, dopamine nor dopexamine have been shown to be better than aggressive saline loading in preventing acute renal failure in at risk patients. Until new clinical studies emerge, avoidance of nephrotoxic insults where possible, monitoring of circulating concentrations of potentially nephrotoxic drug levels and volume loading coupled with supportive measures is recommended. When volume depletion persists, usual blood pressure cannot be restored and patients remain oliguric, early referral to the intensive care unit is paramount. The mortality rate in patients with acute renal failure is high; therefore, measures which reduce the incidence and progression of renal dysfunction will be of benefit.

Keywords: renal damage, acute renal failure, hepatic disease, critical care

J.R.Coll.Surg.Edinb., 45, February 2000, 44-50

INTRODUCTION

What is Acute Renal Failure?

Acute renal failure (ARF) is an abrupt deterioration in renal function and includes glomerular nephritis and interstitial- or acute tubular- nephritis. Unlike in experimental models of acute renal failure, in clinical practice it is a complex and poorly understood pathophysiological event, which occurs in response to a series of insults. Renal hypoperfusion and certain nephrotoxins, such as radiocontrast media, cause intra-renal vasoconstriction which may lead to parenchymal ischaemia and the development of acute tubular necrosis. Glomerular filtration in an individual nephron occurs as the result of the difference between the capillary hydrostatic pressure forcing fluid out, on the one hand, and the colloid osmotic pressure and the hydrostatic pressure in Bowman’s capsule (which both oppose filtration) on the other. Glomerular filtration rate falls due to vasoconstriction. Tubular obstruction as a result of parenchymal oedema, desquamated epithelial cells and casts, and precipitation of salts such as oxalates occurs. Backleak of glomerular filtrate through abnormally permeable tubular epithelia can then result. The vulnerability of the tubular epithelium reflects its high metabolic activity and the physiological conditions of borderline hypoxia. Mismatch between energy supply and demand has been implicated in the pathogenesis of experimental acute tubular necrosis and this has formed the basis for addressing the concept of preventing renal failure.

The Scope of the Problem

ARF occurs in about 30% of critically ill patients. Loss of renal function in such patients leads to increased morbidity, increased length of stay and consequently increased costs, with a mortality rate of about 65%. The causes of ARF commonly include post-operative hypovolaemia, congestive cardiac failure, radiocontrast media-induced injury and aminoglycoside therapy. It has been estimated that 20% of episodes of ARF are drug-induced and 55% are potentially avoidable episodes as a result of fluid and/or drug mismanagement.¹ The mortality in patients who develop renal failure through hypoperfusion is much higher than, for example, in patients who receive an overdose of aminoglycosides. Mortality is higher in patients who develop ARF in hospital.² Interestingly, the mortality rate has not decreased over the last ten years. Management of ARF in critically ill patients remains primarily supportive until renal function recovers. Thus, any therapeutic strategy which prevents or reduces either the incidence or morbidity of ARF will be of benefit.

PATHOPHYSIOLOGY

Histological Changes

Microscopic examination of renal biopsy specimens from patients with ARF usually shows histological damage which is not extensive and is localised only to certain areas of the nephron. The histological appearance of a damaged kidney with acute renal failure bears little relevance to the level of renal dysfunction; apparently minimal damage can result in fairly severe functional deficits. Increased susceptibility of particular segments of the proximal tubule to injury probably reflects the relative hypoxia of the inner cortical/outer medullary milieu where these segments lie. The ascending loop of Henlé, which was previously thought to be a major site of injury, is now known not to be severely affected in clinical situations. Injury to the cortical collecting duct following an ischaemic or toxic insult is relatively rare.

Proximal tubular cells exposed to either an ischaemic or toxic injury respond in one of three ways. If the injury is sufficiently severe then the result is lethal and the cells will die, either by necrosis or apoptosis (programmed cell death). Necrosis is a chaotic and unregulated process commonly resulting from more severe insults. In these circumstances, rapid loss of sources of intracellular energy and destruction of the cell membrane results in swelling and ultimately rupture of cells with release of cytosolic contents into the interstitial space. The lytic enzymes within the cells can then induce a local inflammatory response. In contrast, apoptosis is an energy-requiring process with no ensuing inflammatory response. If the initial insult is less severe, cells are likely to die predominantly from apoptosis rather than necrosis, particularly if sufficient energy stores are available. Cells which have become apoptotic are effectively removed by phagocytes. As a consequence, it is uncommon to observe apoptotic cells, such that the loss of cells may be greater than is apparent from the amount of necrosis visible in a biopsy specimen. This is well demonstrated by the nephrotoxic drug cisplatin, which in experimental models causes tubular cell necrosis at high concentrations but apoptosis at lower concentrations.³

Haemodynamic Changes

At moderate blood pressures (mean 80-160mmHg) the renal vascular resistance alters with blood pressure so that renal blood flow remains fairly constant. This is called autoregulation (Figure 1). It is important to note that in hypertensive patients the range of autoregulation is shifted. In ARF marked haemodynamic changes occur, such that renal blood flow can be reduced by up to 50%. It should be remembered, however, that renal blood flow varies in different regions of the kidney.

Figure 1: The effect of systolic blood pressure on renal blood flows, demonstrating autoregulation in both normotensive and hypertensive patients

Blood flow in the cortex is much higher than in the medulla. The outer medulla is relatively hypoxic, even in healthy subjects, such that normal PO₂ in the outer medulla is between 5-10mmHg, compared with about 50mmHg in the cortex. Renal blood flow is decreased both through capillary congestion with leucocytes, platelets and red blood cells, and also by intrarenal vasoconstriction, due to imbalance between vascular constricting and dilating agents, predominantly endothelin and nitric oxide.

The autoregulatory responses to reduced renal blood flow are likely to be impaired during prolonged haemorrhagic shock, sepsis and low cardiac output states such as myocardial pump failure and hypovolaemia. Animal studies have shown that the renal vasculature is abnormally unresponsive to angiotensin II or noradrenaline at week one post-experimental insult. In dogs, renal arterial occlusion-induced ARF showed profound autoregulatory impairment and loss of response to changes in blood flow.⁴

Inadequacy of perfusion inevitably results in hypoxia within the outer medulla making the kidney particularly vulnerable to further insults. Providing that cardiac function is adequate, sodium loading will maintain cardiac output and perfusion pressure, thus reducing sodium reabsorption and medullary oxygen demand. In theory, renal function may be less vulnerable if oxygen consumption is reduced within the medulla of the kidney by the use of a diuretic such as frusemide.

PATIENTS AT RISK

Which Patients Are Likely to Get Renal Failure?

Commonly, patients at risk are hypotensive, with severe vascular disease, cardiac failure, sepsis and/or are receiving nephrotoxic drugs. There are two main categories in which pre-disposition to renal damage is increased; firstly, patients with impaired renal perfusion pressure due to sodium depletion, diuretic therapy, low cardiac output or any other condition that tends to promote increased sodium reabsorption and secondly, patients with already impaired renal function, vascular disease, severe infection, diabetes or liver disease (Table 1). 

Table 1: Which patients are likely to get acute renal failure?

Although age may be important in pre-disposing to renal toxicity, age per se may be less important than other co-morbidities inherent in older patients. The assumption that renal function becomes impaired as age increases in the otherwise healthy elderly may not be correct. Patients with pre-existing poor renal function are more likely to develop ARF since the solute load to individual nephrons is higher due to nephron loss. As nephron work increases, oxygen consumption increases and there is often damage to juxtamedullary glomeruli such that functional reserve is severely depleted. Such patients, therefore, are less able to compensate for drug induced effects. Impaired drug elimination may also result in toxic drug levels with consequent positive feedback on the nephrotoxic process.

Patients with acute or chronic hepatic disease are likely to metabolise drugs abnormally within the liver. These patients often have altered intrarenal haemodynamics with pronounced salt retention which potentiates any nephrotoxicity.

Diabetic patients are especially at risk of ARF. The vascular endothelium is abnormal and within the kidney, the ability of the endothelium to produce endothelin or nitric oxide in response to stress is severely impaired. Such abnormal vascular responses, therefore, may limit any defence against nephrotoxic insults. In patients with sepsis, intrinsic vascular abnormalities which occur as part of the inflammatory response make these patients susceptible to nephrotoxicity.

Why Do Patients Get Renal Failure?

Several drugs are nephrotoxic (Table 2). Reactions to drugs and other compounds are relatively common and have been described for many substances. They are commonly associated with renal dysfunction although the actual incidence of drug-induced renal failure has not been reported, since incidence is complicated by the complexity of the causes of ARF in seriously ill patients. Nephrotoxicity arises through several mechanisms, including general and local vascular effects, direct effects on renal tubules, tubular obstruction and acute interstitial nephritis. Acute glomerulonephritis can also occur although this is less common.

Table 2: Drugs which can cause renal failure

Diuretics and ß blockers can reduce cardiac output and systemic vasodilators (such as sodium nitroprusside or angiotensin converting enzyme (ACE) inhibitor) may decrease renal perfusion. Drugs which have a direct effect on the renal vasculature may also pre-dispose to renal failure. Angiotensin acts directly within the glomerular circulation and the use of ACE inhibitors, therefore, not only inhibits angiotensin production but also interferes with bradykinin, which has an important role in circulatory control within the glomerulus. Vasoconstriction within the renal circulation and consequent effects on medullary blood flow and oxygen delivery can result from the use of non-steroidal anti-inflammatory drugs (NSAIDs) and cyclosporin A. NSAIDs inhibit prostaglandin synthesis affecting vascular perfusion and oxygen delivery within the outer medulla. These drugs selectively inhibit constitutive cyclo-oxygenase, and their effects are potentiated by hypovolaemia, low cardiac output, sepsis, liver disease and pre-existing renal failure.

Antibiotics, (such as aminoglycosides and vancomycin), anti-fungal agents (amphotericin) and cytotoxic agents, such as cisplatin, are well known for their direct damaging action on proximal tubular cells leading to impaired tubular function. Radiocontrast media also have damaging effects and are discussed later in this review. Heavy metals, such as mercury, organic solvents, such as carbon tetrachloride, plant and animal toxins can also cause tubular damage.

In addition to its diuretic actions, mannitol can cause proximal tubular damage at high dosages, caused by excessive osmotic effects on the proximal tubule. The damaged tubules within the oxygen deficient environment of the outer medulla are unable to contend with large quantities of solute. The tubular cells lose polarity, develop vacuoles and eventually separate from the basement membrane. Marked derangement of electrolytes may also occur due to effects on water reabsorption in the distal tubule. Acute anuria has been reported in critically ill patients who were treated with high dose immunoglobulin for Guillain-Barré syndrome, suggesting that immunoglobulin therapy may result in acute renal dys-function in a similar way to mannitol.

Nephrotoxic damage to the distal tubule results in disturbance of sodium, potassium, hydrogen ion and water balance. NSAIDs, ACE inhibitors and cyclosporin A all alter potassium balance resulting in hyperkalaemia. In addition, NSAIDs and cyclosporin A also inhibit the compensatory mechanisms which protect blood flow to the tubule in the volume depleted kidney.

Chronic administration of lithium may have toxic effects on distal tubular function. As many as 1 in 1 000 of elderly people receive lithium therapy. Initially, it can cause functional and then permanent inability to conserve water, causing nephrogenic diabetes insipidus. Acute lithium intoxication causes a similar tubular effect which may be reversible. High dose cyclophosphamide and amphotericin B both have actions on the distal tubule which may result in hyponatraemia by impairing the ability to excrete water.

Patients with pneumocystis pneumonia as a result of AIDS and other immunosuppressive disorders are increasingly treated with high dosage sulphonamides. Such treatment is associated with increased incidence of crystalluria resulting in tubular obstruction and renal dysfunction. Adequate salt and water loading, so that tubular filtrate flow is preserved, should prevent precipitation of drug and hence renal failure. Other drugs such as the anti-viral agent acyclovir and the protease inhibitor indinavir have similar toxic actions. Treatment of patients with high dose chemotherapy for haematological malignancies can result in rapid cytolytic affects resulting in a hugely increased uric acid load arriving at the kidney. In such patients, acute crystalluria may develop unless adequate urine flow and sodium diuresis is maintained. Ethylene glycol (antifreeze) poisoning can be relatively common in some countries. Metabolism results in a large oxalate load which may crystallise in the tubule; the risk is exacerbated since these patients are often also volume depleted.

Interstitial nephritis can result from antibiotic usage, particularly the b lactams, rifampicin, sulphonamides, vancomycin and ciprofloxacin. In addition, diuretics, including thiazides and frusemide and NSAIDs also can cause acute interstitial nephritis. Less commonly, ranitidine, cimetidine, and phenytoin may also cause similar damage. An acute allergic reaction, with infiltration of immune cells occurs in response to the drugs, causing direct cytotoxicity. In the presence of other factors including volume depletion and hypotension, nephrotoxicity increases. Rarely, acute glomerulonephritis can occur during therapy with penicillamine and NSAIDs. Clearly, with all these drugs, the decision to use a particular agent must take into account the relative risk and benefit for each particular patient.

Risk of Renal Failure after Radiocontrast Administration

Renal damage arising from the nephrotoxic effects of radio-contrast investigations is becoming increasingly common. Radiocontrast media is used in relatively high doses for computerised tomography (CT) scans and some types of vascular surgery. The risk of renal damage is particularly high in patients who already have compromised renal function or those with diabetes mellitus.⁵ In the unfortunate patient with both diabetes and impaired renal function, the incidence of further renal failure following use of radiocontrast agents is over 50%. Both vasoconstriction and direct tubular damage occur. Patients’ cardiovascular state is commonly unstable and they may have existing renal dysfunction, sepsis, diabetes and vascular disease. Preventative measures are limited to saline diuresis prior to investigations involving radiocontrast administration.

Can We Measure Renal Failure?

Measurement of serum creatinine concentrations and creati-nine clearance are useful in assessing glomerular filtration rate and, hence, renal function. However, in situations such as after major surgery, patients may be unstable and renal function is much more difficult to measure accurately. In a recent survey, serial measurement of creatinine clearance was the single most sensitive parameter⁶, but in clinical practice it is serum creatinine concentration which is used to diagnose renal failure, despite the fact that increases in serum creati-nine are seen only when 50% of renal function is lost. Blood urea nitrogen is dependent not only on renal but also non-renal factors. Although urine output alone is not a reliable endpoint, recovering urine output in a previously oliguric patient may be a useful clinical parameter.

TREATMENT

Can Acute Renal Failure be Prevented?

There are several means which have been used to try and prevent renal impairment in critically ill patients. The most important preventative strategy is the identification of patients at risk and elimination of potential contributing factors. Unnecessary exposure to nephrotoxins or agents that may adversely effect medullary oxygen sufficiency, such as NSAIDs or radiocontrast agents, should be avoided where possible. Optimum protection strategies for patients at risk of developing renal failure in whom nephrotoxic exposure is unavoidable remains elusive.

Loop Diuretics

The theoretical advantages of decreased tubular oxygen consumption and clearing of the obstructing tubular debris suggest that diuretics may be beneficial. The loop diuretic frusemide improves medullary oxygenation by inhibition of solute reabsorption and oxygen requirement and improves hypoxic damage in animal models of renal injury. In rats, frusemide abolishes the physiological outer medullary hypoxia, despite a marked fall in regional blood flow. For instance, in rats given radiocontrast agents, medullary PO2 decreases but returns to normal with frusemide treatment. Animal studies certainly suggest that administration of a diuretic (frusemide) and saline may protect against radiocontrast media-induced acute renal failure.⁷ In healthy volunteers, frusemide improves regional oxygenation and reverses medullary hypoxia produced by nephrotoxins.

Despite the observed beneficial effects of loop diuretics in experimental settings their administration to patients have failed to prevent renal failure in clinical practice. Clinical studies reveal that loop diuretic or mannitol treatment is no better than saline loading and even suggest that diuretics may adversely affect renal function when used to prevent radio-contrast nephropathy.⁸ In a randomised controlled trial, treatment of established ARF with diuretics (torosamide or frusemide) and dopamine was not superior to saline loading alone.⁹ These findings suggest that excessive diuresis may result in volume depletion; aggressive cardiovascular support and sodium loading may offer advantages over diuretic treat-ment. Despite adequate rehydration protocols, frusemide induces volume depletion and results are further confused by lack of comparability between studies. These issues include the degree of fluid replacement in diuretic treated patients, the patient groups themselves, the administration of other drugs such as dopamine and mannitol, and the way in which the frusemide is administered. Moreover, radiocontrastinduced nephropathy in high risk patients may be worsened by frusemide. In conclusion, there is no evidence that loop diuretics may ameliorate or prevent acute tubular necrosis and additional experimental and clinical studies are required to address their potential protective properties.

Dopamine and Dopexamine

Dopexamine is a dopaminergic agonist with b- but no aadrenergic actions. Dopamine, in contrast has unpredictable a-adrenergic actions, which are particularly apparent at higher doses. Although the renal actions of both agents are similar the vasodilatory effects of dopexamine are preferable to those of dopamine.

Dopamine inhibits tubular Na/K-ATPase and promotes vasodilatation, thereby increasing renal blood flow, glomerular filtration rate, natriuresis and diuresis. At low doses, actions are predominantly dopaminergic but as doses increase there is activation of b-adrenergic and then a-adrenergic receptors which leads to unpredictable increases in cardiac output, tachyarrhythmias, myocardial ischaemia and systemic vasoconstriction. In addition, hypovolaemia through excess diuresis, hypokalaemia, hypophosphataemia, respiratory depression and hyperprolactinaemia can also occur. Randomised prospective controlled trials have studied the renoprotective role of dopamine following major surgery and radiocontrast studies.

Whereas dopamine increases mean arterial pressure and cardiac index as a result of increased stroke volume, dopexamine has little effect on mean arterial pressure and increases cardiac index by reduction of systemic vascular resistance. These haemodynamic effects cause both natriuretic and diuretic actions. Tachycardia, nausea and vomiting can also occur. Although there are theoretical benefits of dopaminergic inhibition in patients with renal hypoperfusion or those exposed to nephrotoxic agents many clinical studies of dopamine and dopexamine remain inconclusive. This reflects the failure to use precise measures of renal injury and also inadequate haemodynamic monitoring. There are also concerns due to adverse cardiac effects of dopamine in patients with ischaemic heart disease¹⁰ and the suggestion that dopamine may be nephrotoxic in diabetics exposed to radiocontrast media.⁵ A clear renoprotective role for dopexamine in patients undergoing major surgery and those at risk of radiocontrast nephropathy has, as yet, to be shown.

Other Drugs Used to Prevent Acute Renal Failure

A large number of different drugs, including vasoactive agents and free radical scavengers, have been used clinically to prevent renal failure. These drugs address well defined pathophysiological mechanisms, including vasoconstriction, tubular obstruction and transtubular backleak. Vasoactive agents attempt to increase renal blood flow (atrial natriuretic factor); diuretics increase urine flow and hence reduce tubular obstruction. Some compounds act on more than one mechanism at once, such as mannitol, which reduces tubular obstruction by reducing epithelial swelling, and also acting as an antioxidant. Interventions to enhance regional blood supply and decrease local tubular reabsorption offers an approach for the prevention of outer medullary oxygen insufficiency and hypoxic injury. Agents such as theophylline, mannitol or atrial natriuretic peptide analogues result in increased glomerular filtration and tubular oxygen consumption and, therefore, may increase tubular hypoxic damage.

Within the kidney, atrial natriuretic peptide (ANP) raises glomerular filtration rate, lowers renin production and causes natriuresis and diuresis; water intake and salt appetite is decreased. ANP has no effect prophylactically on radiocontrast-induced nephropathy¹¹ or as a treatment for patients with acute tubular necrosis¹², but it may improve dialysis-free survival in patients with oliguria.¹² However, ANP may be detrimental in those without oliguria, probably through an increased incidence of hypotensive episodes and may not be the way forward in either preventing or treating acute renal failure.

Endothelin is thought to be responsible for the long lasting vasoconstriction of the vas afferens, causing a disproportionate decrease in glomerular filtration rate compared with renal blood flow in acute renal failure. In animal models, endothelin receptor antagonist therapy improved renal function but there have been no clinical studies.13 However, a recent study of an oral endothelin receptor antagonist in patients with essential hypertension suggests that the hypotensive effect of this agent may be a disadvantage in renal disease.14 Calcium channel blockers have been proven to be effective in animal studies, but there are limited data on their use in patients. Studies in malaria-associated nephropathy, renal transplantation, cyclosporin-induced nephropathy and radiocontrast nephropathy have all found no difference in renal function regardless of calcium channel blocker therapy.

Other Considerations

Difficulties in interpretation of studies of renal protection strategies are confounded by the variation in the clinical characteristics of the patients groups. The original cause of the renal failure has profound implications for both the manifestations and course of renal dysfunction. The pathogenesis of renal dysfunction, for example ischaemic or nephrotoxic, is important, since nephrotoxic renal failure seems to confer a better prognosis, although acute renal failure may often be multi-factorial. An example of this is aminoglycoside toxicity, where both nephrotoxic and ischaemic mechanisms are involved. Patients with oliguria frequently have a worse course than those who continue to pass urine. Although renal function has an effect on mortality, co-morbidity is probably more important and those factors which influence course, prognosis and outcome should be considered.

Prevention of Impending Renal Failure in an Oliguric Patient

The timing of any intervention protocol is vital. Prevention or prophylaxis means intervening before the insult which leads to renal dysfunction. Treatment is intervention after renal dys-function occurs. Severe volume depletion results in decreased glomerular filtration, but restoring volume alone cannot restore renal function when ARF is established. Prevention of such volume depletion is crucial. Assessing volume status is not easy, however, since a patient may appear to be fluid over-loaded in the face of an inadequate intravascular volume. Postural hypotension and low pulmonary artery occlusion pressure (PAOP) are indicative of intravascular depletion. Interpretation of restored PAOP, however, may be difficult since this can be due to either adequate blood volume or the inability of the heart to accommodate the volume presented to it. Infusion of crystalloid or colloid can be used to adjust PAOP, but it is essential to monitor cardiac function since primary heart failure may itself precipitate ARF. Clearly, on a surgical ward or high dependency unit, monitoring of cardiac function and PAOP may not be possible.

Although there is little evidence that prophylactic administration of dopamine or diuretics prior to a nephrotoxic insult in at risk patients is of benefit, in patients with evidence of impending renal failure, ie poor urine output, the use of these agents may be helpful. In a patient with poor urine output (characteristically less than 0.5ml/kg/h) after restoration of intravascular volume and cardiac output, it is recommended that the algorithm outlined in Figure 2 is followed. It cannot be over emphasised that restoration of intravascular volume and blood pressure (remembering that in some patients the usual blood pressure may be considerably higher than normal) is the most important first step to limiting impending renal failure. If blood pressure cannot be restored the patient should be referred to the intensive care unit for more aggressive management with vasoconstrictors and inotropes.

Figure 2: Algorithm for preventing acute renal failure in an oliguric patient

Once blood pressure is restored, catherisation of the bladder followed by measurement of urine volume after an hour is the first step. If there is no increase in urine output after a further fluid challenge sufficient to increase the central venous pressure (CVP) to 15mmHg, then administration of 2-5µg dopamine may be given, followed by infusion of a low dose of a loop diuretic such as frusemide. This approach may prevent or reverse the onset of ARF by decreasing pre-glomerular vasoconstriction and tubular cellular oxygen requirement. If the patient remains oliguric, another fluid challenge to generate a PAOP of 18mmHg is the next step. Sodium and fluid loading not only tend to increase renal blood flow, as a result of improved cardiac output, but also stimulate atrial natriuretic peptide (ANP) release, thus counteracting the effect of angiotensin on the afferent arteriole. If there is still no urine output, and abdominal ultrasound excludes ureteric obstruction, then dialysis or haemofiltration, based on fluid and potassium overload and urea levels, will be required.

REFERENCES

1. Hou SH, Bushinsky DA, Wish JB, Cohen JJ, Harrington JT. Hospital-acquired renal insufficiency: a prospective study. Am J Med 1983; 74: 243-8
2. Shusterman N, Strom BL, Murray TG, Morrison G, West SL, Maislin G. Risk factors and outcome of hospital-acquired acute renal failure. Clinical epidemiologic study. Am J Med 1987; 83: 65-71
3. Lieberthal W. Biology of acute renal failure; therapeutic implications. Kidney Int 1997; 52:1102-15
4. Schlichtig R, Kramer DJ, Boston JR, Pinsky MR. Renal oxygen consumption during progressive hemorrhage. J Applied Physiol 1991; 70: 1957-62
5. Weisberg LS, Kurnik PB, Kurnik BR. Risk of radiocontrast nephropathy in patients with and without diabetes mellitus. Kidney Int 1994; 45: 259-65
6. Kellen M, Aronson S, Roizen MF, Barnard J, Thisted RA. Predictive and diagnostic tests of renal failure: a review. Anesth Analg 1994; 78: 134-42
7. Heyman SN, Brezis M, Greenfield Z, Rosen S. Protective role of furosemide and saline in radiocontrast- induced renal failure in the rat. Am J Kidney Dis 1989; 14: 377-85
8. Heyman SN, Rosen S, Epstein FH, Spokes K, Brezis M. Loop diuretics reduce hypoxic damage to proximal tubules of the isolated perfused kidney. Kidney Int 1994; 45: 981-5
9. Solomon R, Werner C, Mann D, D’Elia J, Silva P. Effects of saline, mannitol and furosemide on acute decreases in renal function induced by radiocontrast agents. N Engl J Med 1994; 331:1416-20
10. Baldwin L, Henderson A, Hickman P. Effect of postoperative low-dose dopamine on renal function after elective major vascular surgery. Ann Intern Med 1994; 120: 744-7
11. Kurnik BR, Allgren RL, Genter FC, Solomon RJ, Bates ER, Weisberg LS. Prospective study of atrial natriuretic peptide for the prevention of radiocontrast-induced nephropathy. Am J Kidney Dis 1998; 31: 674-80
12. Allgren RL, Marbury TC, Rahman SN, Weisberg LS, Fenves AZ, Lafayette RA, Sweet RM, Genter FC, Kurnik BR, Conger JD, Sayegh MH. Anaritide in acute tubular necrosis. N Engl J Med 1997; 336: 828-34
13. Kohan DE. Endothelin in the normal and diseased kidney. Am J Kidney Dis 1997; 29: 2-26
14. Krum H, Viskoper RJ, Lacourciere Y, Budde M, Charlon V. The effect of an endothelin-receptor antagonist, bosentan, on blood pressure in patients with essential hypertension. Bosentan Hypertension Investigators. N Engl J Med 1998; 338: 784-90

Copyright date: 20th December 1999

Correspondence: Dr Helen F Galley, Academic Unit of Anaesthesia & Intensive Care, Institute of Medical Sciences, University of Aberdeen, Aberdeen AB25 2ZD, UK. Email: h.f.galley@abdn.ac.uk

©2000 The Royal College of Surgeons of Edinburgh, J.R.Coll.Surg.Edinb.,45; 1: 44-50