ANAESTHESIA FOR THE HIGH RISK PATIENT - PART 8 pps

do not require long term renal support whereas in ‘medical’ ARF many patientsprogress to chronic renal failure CRF.approxi-• Following cardiac arrest, 30% will develop ARF.. The incidenc

do not require long term renal support whereas in ‘medical’ ARF many patientsprogress to chronic renal failure (CRF).

approxi-• Following cardiac arrest, 30% will develop ARF The incidence riseswith increasing duration of resuscitation and increasing doses of adrenaline administered

• The incidence of ARF in general ICUs varies between 10% and 25%depending on the local patient populations The majority of these arepatients with surgical pathology ‘Medical’ ARF is much less common.Associated risk factors include sepsis, hypotension and iatrogenic factorssuch as the administration of X-ray contrast and aminoglycosides Inone study hypotension was a factor in 85% of patients developing ARF

on the ICU and was the sole factor in 33%.3

The majority of cases of ARF now occur in ICU leading some to question therole of nephrologists in the management of ARF in the ICU However, the role ofnephrologists in diagnosis and therapy of ‘medical’ ARF remains unquestioned.The only systematic review into preoperative risk factors bemoaned the lack ofconsistency in definitions and statistical analysis and could make few firm conclu-sions.4Repeatedly studies have identified preexisting renal dysfunction, poor LVfunction and advanced age as the main predictors of postoperative ARF in generalsurgical populations

The incidence of ARF is much higher in certain surgical settings and these stitute the high risk groups for development of ARF in the perioperative period:

con-Cardiac surgery

Fifteen per cent of cardiac surgery patients may experience elevations of creatinine

in the postoperative period There seems to be a linear relationship betweenbypass time and ARF

With a more stringent definition of ARF a recent study of over 42 500 patientsfound an overall incidence of ARF of 1.1%.5 It is important to note that ARF

following cardiac surgery is a major determinant of mortality – 63.7% mortality

in those patients with ARF compared to only 4.3% in whom renal functionremained normal The conclusion of this study is that ARF is independently asso-ciated with early mortality after cardiac surgery even after adjustments for comor-bidity and postoperative complications

Vascular surgery

Several studies have identified that vascular surgery and especially aortic surgery

is a major risk factor for perioperative renal dysfunction Risk factors within thisgroup are consistent between the various studies and include:

• advanced age,

• elevated creatinine preoperatively,

• large volume of transfused blood (itself equating with length of surgeryand difficulty),

• duration of aortic cross clamping,

• requirement for postoperative ventilation and/or inotropes

It would seem that the approximate incidence of renal impairment after aortic gery is in the order of 25% with a high mortality

sur-Surgery in liver disease

ARF occurs in approximately 10% of patients operated on with obstructive jaundice

Almost 70% of patients with severe liver failure, e.g undergoing liver ation develop ARF

transplant-OT H E R R I S K F A C TO R S

• Elderly patients also constitute a high risk group owing to their reduced

cardiorespiratory and renal reserve although figures on incidence areless certain

• Rhabdomyolysis: Myoglobin from muscle breakdown precipitates out of

solution and blocks the renal tubules

• NSAIDs: Block the vasodilator effect of prostaglanding A recent paper

looking at the epidemiology of ARF found that NSAIDs were used in18% compared with 11% of control patients, i.e the link is probablyimportant.6

• Abdominal compartment syndrome: The presence of increased

intraabdomi-nal pressure from ileus, haematoma, intraabdomiintraabdomi-nal packs literally

squeezes the kidney This causes a reduction in RBF, GFR, direct pression of the renal parenchyma and increased release of ADH and aldos-terone from stimulation of abdominal wall stretch receptors In generalintraabdominal pressures of 15–20 mmHg are associated with oliguriawhile pressures greater than 30 mmHg may be associated with anuria.

com-• X-ray contrast medium: The risk of renal dysfunction is worse if the

patient is diabetic or has baseline renal dysfunction

• Antibiotics especially aminoglycosides: It seems that the rate of renal cortical

uptake of gentamicin is saturatable, i.e lesser side effects including renaldysfunction if the total daily dose is given once rather than in divideddoses

• Excess use of diuretics: May contribute to hypovolaemia.

• ACE inhibitors: ACE inhibitors induce ARF in renal artery stenosis.

Function often improves when the ACE inhibitors are stopped Therise in creatinine is worse if the patient is also taking diuretics

• Aprotinin: There were reports of renal problems from high doses of the

protease inhibitor, aprotinin, in cardiac surgery patients However, themost recent study found no effect of aprotinin on renal function duringhypothermic cardiopulmonary bypass

• ‘Medical’ risk factors: The incidence of renal dysfunction and its severity

will be increased in the presence of recognised risk factors for renal function such as hypertension and diabetes mellitus The presence ofintrinsic renal disease or chronic renal failure will also obviously increasethe likelihood of perioperative deterioration in renal function

dys-Problem with colloids?

The choice of crystalloids versus colloids for fluid resuscitation has been hotlydebated for over 20 years with most reasonable practitioners accepting a role forboth colloids and crystalloids depending on circumstances Even the colloid prot-agonists accept that there is no firm evidence that use of colloid infusions improvesmeasures of outcome In recent years concerns have been raised that the overuse

of colloid fluids may be associated with a worsening of renal function This hasbeen termed ‘hyperoncotic acute renal failure’ The original reports concernedthe use of dextrans (and at that time it was thought that there was either direct tox-

icity or accumulation in the renal tubules) but may be seen with excess use of all

colloids especially if the patient

• has other renal risk factors,

• is dehydrated, and

• is worse with use of colloids in large volumes or high molecular weights.The theory proposed is that the resultant high plasma colloid osmotic pressurecounteracts opposing hydrostatic filtration pressure in the glomerulus.

Interestingly, small studies suggest that colloids should not be used as plasma volume expanders in brain dead organ donors or kidney transplant recipients Thefunction of the kidney may be at risk

Factors involved in pathogenesis

• In high risk surgical and critically ill patients the main factors involved

in pathogenesis are reductions in cardiac output, RBF and GFR

• Hypovolaemia and haemorrhage are obviously potent mechanisms forinterfering with the functioning of the kidney

• Because cardiac output is such an important factor in ARF, cardiac riskfactors also predict for perioperative renal insufficiency and develop-ment of ARF

• Inflammatory mediators and cytokines, especially in septic patients, alsoreduce RBF and GFR partly through interference with prostaglandinpathways Thus, development of surgical sepsis is an important risk fac-tor for perioperative renal insufficiency

Under normal circumstances a single, short insult does not result in ARF inpatients with previously normal renal function However, repeated and prolongedinsults will produce renal dysfunction and if corrective measures are not taken thepatient is more likely to develop ARF A single insult may be enough to produceARF in a ‘population at risk’

Influence of anaesthetic technique

Choice of anaesthetic technique itself is not thought to be a significant factor.Historically, volatile agents which released inorganic fluoride were a problem butthis is not an issue with modern halogenated agents such as isoflurane and sevoflu-rane Techniques which preserve RBF are preferable Central neural blockade,e.g epidural may have a beneficial effect on RBF due to the reduction in periph-eral resistance provided hypotension and hypovolaemia are avoided Conversely,cardiac output and RBF may fall with general anaesthesia There will be muchinterpatient variation

W H Y T H E K I D N E Y I S AT R I S K ?

Renal problems are common in high risk surgical and critically ill patients

In addition, ARF is an early manifestation of developing multiple organ failure

This is despite the kidney arguably being the best perfused and oxygenated organ

in the body – receiving almost 25% of the total cardiac output This high bloodflow is necessary to provide the high volumes of filtrate required for removal ofwaste products Despite this high blood flow the kidney is usually the first organ

to fail in shock or multiple organ failure Why is this? Why is the kidney so nerable that it has been referred to as the body’s ‘innocent bystander’?

vul-Table 13.1 illustrates the problem and its mechanisms

Thus although the kidney itself has a relatively low O2 consumption to deliveryratio, the renal medulla (the metabolically active part of the kidney) has a very highratio Indeed, the highest ratio in the body Higher than the heart, an organ anaes-thetists are often preoccupied with especially regarding the possible development

of ischaemia Unfortunately, there are no symptoms or immediate signs of renalmedullary ischaemia

Kidneys are also perfusion pressure dependent Critically ill patients especiallywith sepsis lose their ability to autoregulate their RBF and this becomes linearwith blood pressure

Renal defence mechanisms

Tubuloglomerular feedback (TGF) in the juxtaglomerular apparatus in the nephron

is a protective mechanism which attempts to protect the kidney when perfusion islow Efferent glomerular constriction reduces GFR to reduce the oxygen demand

on the medulla This also preserves vascular volume in hypovolaemia The troversial role of increasing RBF versus improving perfusion pressure is discussedbelow

con-W H AT M AY P R E V E N T R E N A L F A I L U R E ?

Good perioperative care and attention to detail will be the best chance of ing deterioration in renal function and development of renal failure

avoid-Table 13.1 – Oxygen balance in different organs (measurements in ml/min/100 g).

Organ O 2 delivery Blood flow O 2 consumption O 2 consumption/

Reproduced with permission from Brezis M, Rosen S, Epstein FH Acute renal failure In Brenner BM, Rector FC Jr (eds),

The Kidney, 4th edn Philadelphia: WB Saunders, 1991: 993–1061.

to apply to other models of renal dysfunction they are of interest Many

of these and other studies have involved crystalloid infusions usually

of saline The role of colloid infusions is questionable as discussed previously

Oxygen transport

In high risk surgical patients the application of an aggressive strategy of ing cardiac output, oxygen delivery and oxygen consumption may improve outcome One of the original randomised controlled trials (RCTs) showed thatthe improvement in survival was associated with a reduction in organ failures, inparticular, renal failure.8 This is not surprising when one considers the reasons discussed above as to why the kidney is at especial risk compared to other organs

maintain-Short cross clamp time in aortic surgery

During aortic cross clamping RBF is reduced even when the clamp is appliedbelow the origin of the renal arteries Thus it is important to minimise the dur-ation of aortic clamping Once the clamp is released the reperfusion of the lowerlimbs will release mediators and oxygen radicals which may impair renal function

Short bypass time in cardiac surgery

Similarly, as short a time as possible spent on cardiopulmonary bypass is important

as RBF is reduced during bypass and red blood cells are damaged Cellular debrisand free haemoglobin are damaging to the kidney

Frusemide

Animal studies clearly indicate that frusemide is protective to the kidney.Patient studies do not Many believe that conversion of oliguric renal failure to

non-oliguric renal failure by use of diuretics is beneficial As summarised above inTable 13.2 the mortality rate for oliguric versus non-oliguric renal failure seemsbetter This does not necessarily mean that one can improve an individual patient’soutcome with diuretics as there is evidence that those patients who respond todiuretics probably have less severe renal damage Among those patients whorequire renal support mortality is not improved by the use of diuretics.

The use of infusions of frusemide rather than intermittent boluses may be able because of decreased requirement for equivalent effect, improved responseand less tolerance with few adverse effects

prefer-Mannitol

The osmotic diuretic mannitol is widely used to promote urine flow Suggestedmechanisms include its osmotic action with greater washout of solutes, plasmavolume expansion and increased RBF Animal studies suggest improved renalfunction but patient studies especially in the high risk surgical patient are less con-vincing Many studies demonstrate increased urine volumes but are less convin-cing in terms of preservation of renal function

Mannitol (in conjunction with bicarbonate) seems beneficial in the special ation of rhabdomyolysis However, IV hydration is superior to hydration pluseither mannitol or frusemide in X-ray contrast induced renal dysfunction.9

situ-Dopexamine

Dopexamine is a relatively new addition to the debate on perioperative renal protection Its actions on dopaminergic receptors undoubtedly increase RBF and creatinine clearance in animal studies and volunteers but its role and efficacy

in prevention of renal failure in critically ill patients is less established Studies

in cardiac surgery and vascular surgery patients indicate some protective

Table 13.2 – Mortality in non-oliguric versus oliguric renal failure.

Number of patients in study* Non-oliguric mortality (%) Oliguric mortality

* ⬎ 450 ml urine/24 h defined as non-oliguric These studies were in general ARF populations.

Reproduced with permission from Thadhani RI, Bonventre JV Acute renal failure In Lee BW, Hsu SI,

Stasior DS (eds), Quick Consult Manual of Evidence Based Medicine Philadelphia: Lippincott-Raven,

1997: 382–413.

effect Dopexamine may have other beneficial effects in the high risk surgicalpatient:

• Reduces the usual increase in gut permeability after cardiac surgery

• May have specific anti-inflammatory actions in surgical patients

Charing Cross protocol

A protocol developed at Charing Cross hospital in the 1990s has been claimed todramatically reduce the need for haemofiltration in ARF in ICU patients Thereare three main aspects of this protocol which, in truth, just restate many of theprinciples already described:

1 Achieving normovolaemia with monitoring guided by pulmonary arterycatheter and echocardiographic measures of preload GTN infusions arealso used to encourage vasodilation and to reduce myocardial ischaemia

2 Achieving normotension (for the patient’s age) with emphasis on the role

• The ‘downside’ of increasing RBF is that this presents more solute forclearance with the requirement for increased Na reabsorption in themedulla and increased medullary oxygen consumption It has even beensuggested that the oliguria resulting from reductions in RBF protectsthe kidney by reducing medullary oxygen consumption!

• Conversely, increasing blood pressure improves renal perfusion sure and may improve GFR.10 Thus increasing blood pressure withnoradrenaline (norepinephrine) may improve creatinine clearance

pres-For example, there are numerous studies of septic shock demonstratingincreased urine flow and creatinine clearance in patients treated withthe addition of noradrenaline to standard therapies The caveats are thatcardiac output and RBF must be maintained (necessitating monitoring

of cardiac output) and hypovolaemia prevented The ‘Charing Cross’protocol involves as one of its key elements the judicious use of noradrenaline

The above may seem confusing but unfortunately many of the controversies andapparent contradictions remain unresolved

W H AT M AY N OT P R E V E N T R E N A L F A I L U R E ?

Diuretics – the dual edged sword

The maintenance of urine flow does seem to exert some preservative effect onrenal function – at least in certain subgroups such as patients at high risk of X-raycontrast medium induced renal failure However, aggressive hydration of thesepatients has been shown to exert a better protective effect than hydration plusdiuretics.9 In general patients do not go into ARF because they are lacking infrusemide! The diuresis produced may be harmful by giving a false sense of secu-rity and by exacerbating any preexisting hypovolaemia

Dopamine – its hard to keep this drug down

Despite more than 20 years of experience and clinical trials the use of ‘renal dosedopamine’ is still dogged with controversy Dopamine is still widely used, at lowrates of infusion, as an agent to protect the kidney and prevent renal failure This

is despite numerous editorials and review articles stating that there is no renal protective effect of dopamine

Dopamine has been claimed to exhibit different effects depending on the infusionrate:

• 0.5–3␮g/kg/min – dopaminergic effect increasing RBF

• ⬎ 3␮g/kg/min – ␤1 effect increasing cardiac output

• ⬎ 7␮g/kg/min – increasing ␣1 effect leading to vasoconstriction

This itself represents a realisation that vasoconstriction occurs at lesser doses thanwas originally believed, i.e older texts refer to vasoconstriction occurring at

⬎ 15␮g/kg/min

The so-called renal dose and effects of increasing dosage described in the textbooks have always been suspected to be less predictable in the real world Sureenough it has now been demonstrated in volunteers that plasma levels ofdopamine vary widely for identical infusion rates in terms of ␮g/kg/min.11

Critically ill patients would be presumed to be even more varied in their drug position and metabolism Thus there is now no real justification to think of onedose of dopamine as being a ‘renal dose’:

dis-• In volunteers dopamine infusions lead to increased RBF, urine tion and creatinine clearance Animal studies suggest a beneficial effect

produc-on renal functiproduc-on – renal protectiproduc-on (the holy grail of nephrology)

• There is still no convincing evidence to support the idea that the use

of dopamine at low dosage prevents the development of ARF inpatients

• What few trials there are which seem to show benefit, especially forradiological contrast medium induced ARF include vigorous fluidloading as part of their protocol

• Perioperative use of dopamine has been widespread, e.g cardiac andaortic surgery but this is not supported by scientific evidence

• Dopamine does increase urine output, mainly by a natriuretic or

diuretic effect but also by a general increase in cardiac output (Howeverdobutamine does this better due to its more predictable inotropic properties)

• A recent, large, well conducted study failed to demonstrate any benefit

of low dose dopamine in critically ill patients at risk of renal failure.12Many have in the past used renal dose dopamine for two reasons:

1 It may be renal protective, they just haven’t been able to prove it

2 It may not do any good but at least it does not do any harm

Never mind this latter unusual justification for giving any drug, there is an ing opinion that dopamine is potentially harmful due to

increas-• tachycardia and arrhythmias,

• reversible, dose related depression of pituitary release of prolactin anddepression of indices of cell mediated immunity,13

• the diuresis may be harmful by giving false sense of security and byexacerbating hypovolaemia,

• a recent randomised study on the use of dopamine for X-ray contrastmedium induced renal failure demonstrated a harmful effect on theseverity of renal failure, prolonging the course,14

• dopamine seems to have worse effects on the splanchnic circulationcompared to a pure vasoconstrictor, noradrenaline

E X P E R I M E N TA L D R U G S TO P R E V E N T R E N A L F A I L U R E

Many agents have been studied in animal models as either prevention or treatment

of ARF Less of these agents have made it to clinical trials and none have cingly been shown to be of clinical benefit:

convin-• Atrial natriuretic peptide (ANP), endothelin antagonists and calciumchannel blockers have shown most promise but further studies are needed

• Concerns exist with some of these agents, e.g calcium channel blockersthat interaction with anasesthetics agents may produce more hypoten-sion which may mitigate against any renal protective effect

• Other therapies suggested by animal trials, e.g thyroxine have not beensubstantiated in patient studies

• Hypotensive complications of these agents may limit their usefulness

• Studies have on occasion been poorly controlled with, e.g frequentuncontrolled concomitant use of diuretics and dopamine

• At present all these experimental drugs remain just that, experimental

E S TA B L I S H E D A R F

Once ARF has occurred there are no known pharmacological interventionswhich will reverse the process – including diuretics The kidney will eitherrecover or it will not (though interesting new evidence suggests that the method

of supportive therapy may influence this as discussed below) Thus the mainstay ofmanagement is supportive measures both in the acute stages and later if the ARFbecomes ‘chronic’

A full discussion of renal replacement techniques is beyond the scope of this textbut some interesting results from recent studies will be highlighted especially asthey offer some advice as to whether there are any differences in outcome betweendifferent techniques of renal support

Indications for renal support include:

• Uraemia, a blood urea⬎ 50 is an arbitrary but common guide torequirement for support Uraemic complications may occur withhigher blood urea including pericarditis and GI bleeding Support shouldperhaps be instituted earlier in malnourished patients with reducedmuscle mass whose urea and creatinine will rise slower

• Severe acidosis.

• Hyperkalaemia.

• Pulmonary oedema secondary to fluid overload in oliguric patient.

In modern critical care practice any of the above should be sufficient to triggerrenal support unless there are definite contraindications to its provision.

Recent studies examining the role of early versus late intervention in terms ofproviding renal support are of interest For example:

• Early initiation of continuous venovenous haemofiltration (CVVH) in

a small study of patients with septic shock has been shown to improvehaemodynamic and metabolic responses and improve survival.15

• Early haemofiltration is associated with better than predicted survival inARF after cardiac surgery.16

• Despite similarities in injury severity and risk of developing renal ure, survival was increased in the early filtration group in a retrospectivereview of 100 trauma patients.17

fail-Thus, many ICUs institute support at an early stage before biochemical pensation occurs

decom-Continuous versus intermittent replacement therapies

A majority of modern ICUs offer continuous renal support usually CVVH Somestill offer only intermittent dialysis with some units offering both or neither.The advantages of CVVH include:

• ‘Gentler’ The fluid shifts are gradual and continuous leading to greaterhaemodynamic stability

• Gradual removal of solutes avoiding the potential for disequilibriumsyndrome

• Experimental role in removing inflammatory mediators in sepsis andpancreatitis

Recent studies provide evidence that CVVH is better than intermittent dialysis:

• The rapid fluid shifts and haemodynamic upset associated with dialysisare associated with further ischaemic insults to the kidney in animals

• Creatinine clearance falls after dialysis! (but does not after CVVH) –presumably because of these shifts.18

• CT studies of the brain demonstrate changes after dialysis consistentwith increased brain water which are not present after CVVH.19

• A RCT has shown that intrinsic renal function among survivors ofARF recovers to a predetermined level in over 90% of patients treatedwith CVVH but in only 59% of patients treated with dialysis (i.e more

patients will require chronic renal support when ARF is managed withintermittent dialysis).20

• Several studies support the observation that the duration of ARF is lesswhen the patient is managed with CVVH than with dialysis

• Meta-analysis of (mainly) retrospective studies show improved survivalwith continuous techniques Large RCTs are awaited

High volume haemofiltration?

The volume of ultrafiltrate may be important Large volumes of ultrafiltrate (high volume or ‘aggressive’ CVVH) increase creatinine clearance and mayremove greater quantities of inflammatory mediators The only large RCT of over

400 patients demonstrated improved survival with high volumes of ultrafiltratecompared to lower volumes.21The authors recommend ultrafiltrate volumes of atleast 35 ml/kg/h

O U TC O M E O F A R F

• ‘Medical’ renal failure has a better outcome than ‘surgical’ ARF orARF in the critically ill patient One study from 1997 found a 16%mortality in the ‘nephrotoxic’ group compared to a 30% mortality inthe ‘ischaemic’ group.22

• Many studies have found the mortality of ARF in the ICU to beapproximately 40–70% When ARF is associated with the need forIPPV (i.e part of the multiple organ failure syndrome) the mortality is

of the order of 70%

• Mortality is high when the onset of ARF is late in the course of a ical illness, i.e a recent French multicentre trial of over 1000 patientsfound that mortality was worse in patients developing ARF in ICUcompared to patients admitted to ICU with ARF.23

crit-• There is cause for optimism, however, from the studies quoted abovecomparing the relatively newer technique of CVVH with dialysis Inaddition comparisons of different cohorts treated for ARF at the MayoClinic have been published:24

Patients Survival (%)