lable at ScienceDirect

Current Anaesthesia & Critical Care 21 (2010) 65–68

Contents lists avai

Current Anaesthesia & Critical Care

journal homepage: www.elsevier .com/locate/cacc

FOCUS ON: RENAL

Prevention of acute kidney injury in the intensive care unit

Stuart Murdoch*

St James’s University Hospital, Beckett Street, Critical Care, Leeds, West Yorkshire LS9 7TF, United Kingdom

Keywords:Acute kidney injury (AKI)DiureticsDrug preventionN-acetyl cysteineDopamine

* Tel.: þ44 113 206 5789.E-mail address: stuart.murdoch@leedsth.nhs.uk

0953-7112/$ – see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.cacc.2009.10.011

s u m m a r y

Acute Kidney injury (AKI) is a relatively common condition in the intensive care unit and is associatedwith an increase in mortality. Whilst it can be treated by the use of renal replacement therapies anindependent increase in mortality still exists. It is therefore seems intuitive that the prevention of AKIshould be associated with a reduction in both mortality and morbidity and an improvement in patientcare for this reason there has been many attempts to develop strategies to reduce the incidence of AKI.These strategies involve the use of therapeutic agents to prevent renal failure, the avoidance ofnephrotoxic agents and the maintenance of normal hydration and renal perfusion. More recently therehas been focus on the early recognition of patients at risk of developing AKI and focussing care to avoidexacerbation of risk factors.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Acute Kidney Injury (AKI) is a relatively common condition inthe intensive care unit (ICU) and is associated with an increased riskof mortality despite the ability to support the functions of thekidneys by means of renal replacement therapy (RRT). Studieslooking at mortality in patients in intensive care have demon-strated 60% mortality.1 The recent National Confidential Enquiryinto Patient Outcomes and Deaths (NCEPOD) has highlighted theproblem of AKI in the UK associated with acute hospital admissionsand reported that only 50% of patients received what could beclassed as good care. The report claims that a fifth of patients couldhave been prevented from developing AKI.2 Acute kidney injury ismore likely to occur in certain groups which include the elderly andthose with pre-existing medical conditions such as hypertensionand diabetes. However many of these patients have risks which canbe modified to prevent injury to the kidney.

Within the ICU it has been demonstrated that AKI is an inde-pendent risk factor for mortality, which contrasts with the previ-ously held view that patients die with rather than because of AKI.This view may be due to the relative ease with which RRT can nowbe provided, although the NCEPOD report demonstrated that somepatients with AKI never received RRT. A clear correlation betweenthe degree of AKI and worse patient outcomes (length of stay andmortality) in ICU has been demonstrated (Table 1).3 It is thereforeintuitive that the prevention of AKI should be associated witha reduction in both mortality and morbidity and an improvement inpatient care.

ll rights reserved.

Until recently there was no standard definition of what consti-tuted acute kidney injury (AKI), which has limited the generalapplicability of clinical trials aimed at looking at prevention andtreatment of the disease. Recently two very similar definitions andstaging systems for AKI have been proposed by the Acute DialysisQuality Initiative group (ADQI) and the Acute Kidney InjuryNetwork (AKIN).4 The application of these staging systems shouldprovide a better basis for studies which aim to investigate theprevention of AKI in patients. It is important to recognize that theprevention of any insult to the kidney is important particularly invulnerable patients on the ICU.

Acute kidney injury can be due to pre-renal, intrinsic renal andpost-renal causes and methods to minimise kidney damage can beaimed at these three areas. The cause of kidney injury in criticalcare is often multi-factorial secondary to ischaemia and sepsiscausing hypo-perfusion of the kidneys and acute tubular necrosis(ATN). However patients with sepsis can develop AKI without anyother obvious cause and animal studies of AKI secondary to sepsisalone have suggested that hypo-perfusion of the kidney doesnot always occur.5 It is well recognised that histopathologicalexamination of the kidney in patients with AKI does not alwaysdemonstrate frank proximal tubule cell necrosis.6 It has been sug-gested that sublethal cell injury occurs and contributes to proximaltubule cell dysfunction in the absence of obvious cell necrosis.

2. Prevention of acute kidney injury

The cornerstones of preventing AKI are the maintenance of anadequate circulating volume, an adequate perfusion pressure andthe avoidance of further insults to the kidney. It has long been

Table 1The incidence of acute kidney injury on the intensive care unit and its effect onlength of stay and mortality.

No AKI Risk Injury Failure

Incidence % 65.6 19.1 3.8 12.5Age 60.5 62.1 60.4 61.1ICU mortality 10.7 20.1 25.9 49.6Hospital mortality 16.9 29.9 35.8 57.9ICU length of stay 2 5 8 9

S. Murdoch / Current Anaesthesia & Critical Care 21 (2010) 65–6866

known that if renal blood flow falls below a given level glomerularfiltration rate will fall and this can clearly be linked to the devel-opment of AKI from the RIFLE criteria. This should be achieved byadequate fluid resuscitation and then the use of inotropes orvasopressors.

The surviving sepsis guidelines7 suggest that fluid therapyshould be to a central venous pressure (CVP) greater than or equalto 8 mmHg in non-ventilated patients or 12 mmHg in ventilatedpatients and the administration of fluid boluses until no furtherhemodynamic response is seen. It is beyond the scope of this articleto discuss the end target of fluid resuscitation, but fluid should beadministered to ensure adequate volume expansion of the patientwithout causing harm. There is considerable debate about thebenefits and harm of various fluids, but there is no good evidencecomparing fluid types for the prevention of AKI. Fluid therapy canhowever be pre-emptive, studies have shown that the adminis-tration of fluid to patients prior to exposure to contrast mediaresults in a reduced rise in serum creatinine.8

The ideal blood pressure to aim for in the prevention of AKI isunclear; the surviving sepsis guidelines suggest a mean arterialpressure (MAP) of 65 mmHg. This may seem low for patients whoare normally hypertensive who may expect to need a higher drivepressure, in this situation it may be prudent to aim for a higherpressure and see its effect on kidney function. The evidence tosupport the use of vasopressors to prevent kidney injury is rela-tively sparse and generally consists of single centre studies in smallnumbers of patients. The evidence suggests that norepinephrine issafe to use and is more effective than high dose dopamine inrestoring blood pressure and improving urine output.9,10

The aim of vasopressor therapy is to improve the perfusionpressure to the kidney, in general the resistance to flow from thevenous pressure can be ignored. However in situations of raisedintra-abdominal pressure the resistance to flow can be such thatrenal perfusion is impeded and AKI occurs.11 Early recognition andsubsequent treatment of raised intra-abdominal pressure isessential to restore kidney function.

The release of myoglobin from damaged muscle can result inrhabdomyolysis, which in turn can result in AKI. Rhabdomyolysiswas first described in patients who had suffered crush injuries butis now recognised to have other aetiologies including drugs, exer-cise, infections and ingestion of toxins. The mechanism of AKI inrhabdomyolysis is unclear but may be due to the crystallisation ofmyoglobin crystals in the renal tubule causing obstruction. It hasbeen demonstrated that early fluid therapy in patients who havesuffered a crush injury improves outcome.12 Prevention of AKI inthe high-risk patient consists of volume expansion and the main-tenance of a urine output, alkalinisation of the urine is oftenadvocated as a way to prevent the myoglobin crystallising.

3. Nephrotoxic drugs

Acute kidney injury can be caused by a wide range of drugs, theavoidance of these drugs, where clinically appropriate or thealteration of their dosage can reduce the incidence of AKI. Some

drugs which are nephrotoxic should be avoided in patients with orat risk of AKI. This is most important for non-steroidal anti-inflammatory drugs which in an unwell hypovoleamic patient playa significant role in causing hypo-perfusion of the kidney. Angio-tensin converting enzyme (ACE) inhibitors should also be used withcaution in patients at risk of AKI.

3.1. Aminoglycosides

These drugs are excreted by glomerular filtration and it appearsthat peak levels of the drugs correlate with toxicity. Studies havesuggested that aminoglycosides are a common cause of AKI, thistoxicity can be avoided by the choice of another appropriate anti-biotic. Indeed the need for these drugs for certain infections hasbeen questioned, but with emerging antibiotic resistance andconcerns about vulnerability to other infections as a result ofantibiotic treatment aminoglycosides may be used more widelythan previously. It has been suggested that alteration in the dosingregimen of these drugs from divided doses throughout the day toonce daily dosing is as efficacious and possibly less nephrotoxic.13

Once daily regimens tend to rely on nomo-grams which take intoaccount body weight, renal function and existing aminoglycosidelevel.

3.2. Amphotericin B

Conventional treatment with Amphotericin B results in AKI inover a quarter of patients, with the risk increasing with cumulativedose given. Kidney injury results from constriction of the afferentarterioles leading to a drop in renal blood flow and glomerularfiltration rate, which can occur after a single dose of amphotericin.Strategies exist that administer fluid to the patient both prior to andafter administration of amphotericin, which may act to increase GFRbut have not been tested in a randomised controlled trial (RCT).Despite the significant renal toxicity of amphotericin, its effectivenature against fungi have made it, until recently, the main stay oftreatment. This toxicity led to the development of lipid carriers forthe drug which are effective in preventing toxicity as they result inlower renal concentrations and are concentrated in reticuloendo-thelial tissues such as the liver and spleen. It has been demonstratedthat these preparations of amphotericin are less nephrotoxic thanconventional amphotericin.14,15 The introduction of newer anti-fungal agents and reductions in the cost of liposomal amphotericinB, has opened up new strategies to treat fungal infections and at thesame time prevent AKI.

4. Radio-contrast induced nephropathy

This is a common cause of AKI in hospital,16 its mechanism isunclear but appears to due to a combination of direct renal tubularepithelial cell toxicity and renal medullary ischaemia. A transientdecrease in glomerular filtration rate occurs in almost all patientsafter exposure to contrast media. The degree of injury dependsupon the presence of patient risk factors which include; diabetesmellitus, pre-existing renal disease, other drugs being taken,hypovolaemia and factors related solely to the radiographiccontrast. In patients with normal renal function and no other riskfactors the injury is rarely significant. In many patients the degreeof AKI is relatively minor. However the studies that have beenreported are predominantly before the introduction of the RIFLEcriteria. One of the unique factors about radio-contrast inducednephropathy is that the timing of the insult is usually known andpre-emptive action can be taken to reduce the nephrotoxicity. It isalso possible to recruit relatively large numbers of patients tostudies making the results potentially more powerful.

S. Murdoch / Current Anaesthesia & Critical Care 21 (2010) 65–68 67

As with other forms of AKI the maintenance of a well-hydratedpatient reduces the incidence of AKI. Several studies havedemonstrated that the administration of intravenous fluid prior toadministration of the contrast reduces the incidence of AKI. Theadministration of other agents has also been trialled to prevent AKI,these agents include theophylline, N-acetyl cysteine (NAC) andsodium bicarbonate.

Theophylline acts as a vasodilator by being a non-selectiveadenosine antagonist, adenosine being believed to cause vasocon-striction in contrast-induced nephropathy. Meta-analysis oftheophylline to prevent nephropathy demonstrates a reduction inAKI in the treatment group.17 N-acetyl cysteine is an anti-oxidantwhich has been used in a number of different clinical scenarios. Theinitial studies of (NAC) into potential benefits to prevent contrast-induced nephropathy were promising.18 Unfortunately later studieshave not shown such benefit.19 Meta-analysis of data has so far notbeen definitive and there have been calls for larger definitive RCTs.Initial use of sodium bicarbonate rather than 0.9% sodium chloridedemonstrated a reduction in the incidence of nephropathy ina high-risk population.20 However later larger studies have failed tosupport this initial finding.21

One method aimed at reducing AKI is the use of differentcontrast media: non-ionic, low and iso-osmolar contrast agents.However they can still produce AKI and in a similar way toliposomal amphotericin are considerably more costly thanconventional contrast media. The evidence supporting their usewas initially small due to poorly designed under-powered trials.However a large trial22 in 1995 of over 1100 patients demonstrateda significant reduction in AKI in patients given a low-osmolarcontrast agent compared to the higher osmolar contrast agents.This advantage was most pronounced in the group of patient withpre-existing renal impairment and diabetes. A meta-analysissubsequently showed a 50% reduction in contrast-inducednephropathy with low-osmolar contrast agents compared to theolder agents with a higher osmolarity.23 There is as yet littleevidence to recommend the newer iso-osmolar agents compared tothe low-osmolar media. In patients at high risk of contrast-inducedAKI thought should be given to using alternative imaging modali-ties, as so far the risk of AKI after contrast cannot be eliminatedcompletely.

5. Pharmcological treatment to prevent AKI

5.1. Dopamine

Dopamine is a catecholamine that is vasodilatory at low doses, Itis believed to exert its action predominantly on dopaminergicreceptors at low doses 2.5–5 mg/kg/min commonly referred to as‘‘renal dose’’ dopamine. At higher doses it has a beta and alphaeffect. It was a long held view that dopamine was helpful in thetreatment and prevention of AKI. This was based on initial animalstudies which experimentally induced AKI which was treated withdopamine. Glomerular filtration rate, renal perfusion, and natri-uresis were all better in the treatment group.24 After these initialresults there then followed a number of small studies examiningthe effect of dopamine in humans. Dopamine was not consistentlyshown to improve kidney function, though it did improve renalblood flow. These studies included patients undergoing majorsurgery, and patients on the ICU with AKI. In 2000 a large multi-centre RCT reported on the use of low dose dopamine in patientswith clinical evidence of early renal dysfunction (oliguria or a raisedserum creatinine concentration). The treatment and control groupswere well matched however there was no difference in theoutcome between them either in terms of renal function or othervariables such as length of stay or mortality.25 Despite this lack of

supporting evidence dopamine is still used by some specialists fora ‘‘renal protective’’ effect.26

5.2. Fenoldapam

Fenoldapam is a pure dopamine A-1 receptor agonist thatincreases blood flow to the renal cortex and outer medulla and doesnot have the alpha and beta effects of dopamine. A meta-analysis of16 randomised studies showed that fenoldapam significantlyreduced the risk for AKI, the need for RRT, and in-hospital death. Itsadministration was also associated with a shorter intensive careunit stay.27 A large, multicenter, appropriately powered trial willneed to be performed to confirm this result- several of these areplanned.

5.3. Diuretics

If AKI is viewed as a fall in urine output there is logic in usingdrugs such as diuretics to increase urine output. Trials have beenconducted investigating the use of diuretics in patients at risk ofAKI, a systematic analysis of seven trials failed to demonstrate anysignificant advantage from their use in patients receiving radio-contrast media.28 A recent survey of diuretic use in critical caredemonstrated a widespread use with serum creatinine, urineoutput, blood pressure as major determinates of dose. Despite thiswidespread use many respondents doubted the treatment theywere giving had any effect on mortality, a reduced need for, orduration of RRT.29 This scepticism about the use of diuretics wassupported by a recent meta-analysis of their use.30 Five RCTsenrolling 555 patients were eligible and analysed (only twoincluded critically ill patients). There was no statistical differencein mortality or renal recovery with use of loop diuretics comparedwith the control group. However, loop diuretics were associatedwith a shorter duration of RRT. As so often there is a need fora multi-centre RCT to give a definitive answer, diuretics are notbenign and can result in transient tinnitus and deafness whichcan be disturbing for any patient. The use of diuretics aftersurgery has also been investigated with no strong evidence insupport of their use.

The osmotic diuretic mannitol has been investigated asa potential therapy to prevent AKI. In the surgical populationa benefit has not been demonstrated for mannitol administration.31

There has been no benefit demonstrated for patients exposed toradio-contrast agents.

6. Relief of ureteric obstruction

Post renal kidney injury can be caused by obstruction of theurinary tract which can be readily identified on ultrasound andtreated, by the insertion of a nephrostomy. Early recognition andtreatment of obstruction can result in rapid improvement in AKIand the potential for complete recovery. The NCEPOD AKI studyrecommended that radiological services should be available 24 ha day in hospitals to provide this important diagnostic and thera-peutic intervention.

7. Conclusion

Acute Kidney Injury is a significant cause of morbidity andmortality in the critically ill patient. The NCEPOD report highlightsthat this is a condition which in a significant number of people islargely avoidable- by adequate volume expansion, stopping neph-rotoxic drugs and early senior medical involvement. As yet there isno treatment which has reliably been shown to prevent or alter thecourse of AKI, once obstruction has been ruled out. Emphasis must

S. Murdoch / Current Anaesthesia & Critical Care 21 (2010) 65–6868

therefore be to improve the early recognition of patients at risk ofAKI, the removal and avoidance of nephrotoxic drugs and ensuringadequate volume expansion and blood pressure.

Conflict of interest statementThe author of the above manuscript has not declared any

conflict of interest which may arise from being named as an authoron the manuscript.

References

1. Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure in critically ill patients:a multi-national multi-centre study. JAMA 2005;294:813–8.

2. Stewart J, Findlay G, Smith N, et al. Adding insult to injury a review of the careof patients who died in hospital with a primary diagnosis of acute kidney injury(acute renal failure). NCEPOD; 2009.

3. Ostermann, et al. Crit Care 2008;12:R144.4. Bellomo R, Kellum JA, Ronoc C. Defining acute renal failure: physiological

principles. Intensive Care Med 2004;30:33–7.5. Langberg C, Wan L, Egi M, et al. Renal blood flow in experimental septic acute

renal failure. Kidney Int 2006;69:1996–2002.6. Langenberg C, Bagshaw SM, May CN, Bellomo R. The histopathology of septic

acute kidney injury: a systematic review. Crit Care 2008;12:R38.7. Dellinger RP, Levy MM, Carlet JM, et al. Surviving sepsis campaign: international

guidelines for management of severe sepsis and septic shock. 2008 [publishedcorrection appears in Crit Care Med 2008; 36:1394–1396]. Crit Care Med 2008;36:296–327.

8. Mueller C, Buerkle G, Buettner HJ. Prevention of contrast media-associatednephropathy: randomised comparison of 2 hydration regimes in 1620 patientsundergoing coronary angioplasty. Arch Intern Med 2002;162:329–36.

9. Martin C, Papazian L, Perrin G, et al. Norepinephrine or dopamine for thetreatment of hyperdynamic septic shock? Chest 1993;103:1826–31.

10. Martin C, Viviand X, Leone M, et al. Effects of norepinephrine on the outcome ofseptic shock. Crit Care Med 2000;28:2758–65.

11. Malbrain ML, Chiumello D, Pelosi P, et al. Incidence and prognosis of intra-abdominal hypertension in a mixed population of critically ill patients:a multiple center epidemiological study. Crit Care Med 2005;33:315–22.

12. Better OS, Stein JH. Early management of shock and prophylaxis of acute renalfailure in traumatic rhabdomyolysis. N Engl J Med Mar 22 1990;322(12):825–9.

13. Barza M, Ioannidis JP, Cappelleri JC. Single or multiple daily doses of amino-glycosides: a meta-analysis. BMJ 1996;312:338–44.

14. Walsh TJ, Yeldandi V, McEvoy M, et al. Safety, tolerance, and pharmacokineticsof a small unilamellar liposomal formulation of amphotericin B (AmBisome) inneutropenic patients. Antimicrobial Agents Chemother 1998;42:2391–8.

15. Mills W, Chopra R, Linch DC, Goldstone AH. Liposomal amphotericin B in thetreatment of fungal infections in neutropenic patients: a single-center experi-ence of 133 episodes in 116 patients. Br J Haematol 1994;86:754–60.

16. Hou SH, Bushinsky DA, Wish JB, Cohen JJ, Harrington JT. Hospital-acquired renalinsufficiency: a prospective study. Am J Med 1983;74:243–8.

17. Ix JH, McCulloch CE, Chertow GM. Theophylline for the prevention of radio-contrast nephropathy: a meta-analysis. Nephrol Dial Transplant 2004;19:2747–53.

18. Fishbane S, Durham JH, Marzo K, Rudnick M. N-acetylcysteine in the preventionof radiocontrast-induced nephropathy. J Am Soc Nephrol 2004;15:251–60.

19. Durham JD, Caputo C, Dokko J, et al. A randomized controlled trial of ace-tylcysteine to prevent contrast nephropathy in cardiac angiography. Kidney Int2002;62:2202–7.

20. Merten GJ, Burgess WP, Gray LV, et al. Prevention of contrast-inducednephropathy with sodium bicarbonate: a randomized controlled trial. JAMA2004;291:2328–34.

21. Brar SS, Shen AY, Jorgensen MB, et al. Sodium bicarbonate vs sodium chloridefor the prevention of contrast medium-induced nephropathy in patientsundergoing coronary angiography. JAMA 2008;300(9):1038–46.

22. Rudnick MR, Goldfarb S, Wexler L, et al. The Iohexol cooperative study.Nephrotoxicity of ionic and nonionic contrast media in 1196 patients:a randomized trial. Kidney Int 1995;47:254–61.

23. Barrett BJ, Carlisle EJ. Meta-analysis of the relative nephrotoxicity of high- andlow-osmolality iodinated contrast media. Radiology 1993;188:171–8.

24. Iaina A, Solomon S, Gavendo S, et al. Reduction in severity of acute renal failure(ARF) in rats by dopamine. Biomedicine 1977;27:137–9.

25. Bellomo R, Chapman M, Finfer S, et al. Low dose dopamine in patients withearly renal dysfunction: a placebo-controlled randomised trial. Lancet 2000;356:2139–43.

26. Brienza N, Malcangi V, Dalfino L, et al. A comparison between fenoldapam andlow-dose dopamine in early renal dysfunction of critically ill patients. Crit CareMed 2006;34(3):707–14.

27. Landoni G, Biondi-Zocca GG, Tumlin JA, et al. Beneficial impact of fenoldapamin critically ill patients with or at risk for acute renal failure: a meta-analysis ofrandomized clinical trials. Am J Kidney Dis 2007;49(1):56–68.

28. Solomn R, Werner C, Mann D, et al. Effects of saline, mannitol and furosemideto prevent acute decrease in renal function induced by radio-contrast agents. NEng J Med 1994;331:1416–20.

29. Ronco C, Bellomo R, Kellum JA, editors. Acute kidney injury. Contrib Nephrol.Diuretics in the management of acute kidney injury: a multinational survey, vol.156. Basel: Karger; 2007. p. 236–49.

30. Bagshaw SM, Delaney A, Haase M, et al. Loop diuretics in the management ofacute renal failure: a systematic review and meta-analysis. Crit Care Resusc2007 Mar;9(1):60–8.

31. Peter H, Poullis M. Does the administration of mannitol prevent renal failure inopen abdominal aortic aneurysm surgery? Interact CardioVasc Thorac Surg2008;7:906–90.