Authors are listed in alphabetic order. *Denotes group facilitator.

Acute Dialysis Quality Initiative Fifth International Consensus Conference

Workgroup 2

Fluids for Prevention and Management of Acute Kidney Injury

Jorge Cerda

Lewis J. Kaplan

John A. Kellum*

Mitra K. Nadim

Paul M. Palevsky

Introduction

Fluids are the only known method of attenuating renal injury. Furthermore, fluid prescriptions whether for

hydration, resuscitation or renal replacement therapy, must be tailored to the fluid and electrolyte,

cardiovascular status and residual renal function of the patient. Different fluids have significantly different

effects both on volume expansion as well as electrolyte and acid-base balance. While controversial,

different fluids may even influence renal function differently. This ADQI workgroup has focused on fluids

for prevention and management of acute kidney injury. We have reviewed the evidence and have made

recommendations for clinical practice and future studies.

Fluids for Prevention of AKI

Prevention of AKI has significantly relied on fluid administration with or without an identified effective

circulating volume deficit in concert with avoidance of nephrotoxic medications. Fluid administration

serves to expand plasma volume, increase renal blood flow, and presumably glomerular filtration rate.

Increased renal blood flow augments renal parenchymal oxygen delivery to support cellular metabolism

while enhanced tubular filtration flushes potentially toxic elements out from the renal tubules. While these

benefits of fluid administration should reduce or ameliorate AKI, some components of commonly used

fluids may initiate or worsen existing AKI. This section will explore fluid prescription benefits and risks

with regard to AKI, including those derived from crystalloids, colloids, and hemoglobin-based oxygen

carriers.

ADQI

2

Do different crystalloid solutions pose a risk of AKI?

i. Anion composition

Hyperchloremic vs. Balanced lactated solutions. Large volume crystalloid resuscitation can result in

significant alterations in electrolyte and acid-base balance in patients. While perhaps of limited concern in

healthy subjects, these effects may create life- threatening abnormalities in the critically ill and injured. An

intra-operative comparison of 0.9% saline (NS) and lactated ringer’s solution (LR) for abdominal aortic

reconstruction demonstrated hyperchloremic metabolic acidosis (HCMA) occurred with NS much more

often than with LR, and NS resuscitated patients required greater amounts of blood component therapy.[1]

However, no differences in renal function were identified. In another small, related study, patients

undergoing renal transplantation were randomized to receive either NS (n=26) or LR solution (n=25) in a

double-blind fashion with a primary outcome measure of creatinine at day 3 post-transplant.[2]

Unexpectedly, the authors found a markedly increased incidence of hyperkalemia (> 6 meq/L) in 5 of the

NS group but it none of the LR group. Significant metabolic acidosis was noted only in the NS treated

patients, 8 of whom received some form of therapy to raise the pH. Due to these significant differences

between the groups, the trial was halted early for safety reasons.

Although hyperchloremia has been associated with impaired organ perfusion and function, including

diminished renal blood flow and longer time to diuresis in comparison to lactated Ringer’s solution [3, 4],

no significant differences in clinical outcomes have been reported. Acidosis has also been demonstrated to

impair coagulation. Importantly, the effect of acidosis on coagulation parallels that of hypothermia.[5] Of

note, both LR and NS have been identified as potent triggers of immune cell activation, although the

significance of these observations remains uncertain.[6]

Hyperchloremic vs. Balanced solutions with non-lactate buffer. Despite the availability of bicarbonate and

acetate buffered crystalloid solutions for renal replacement therapy. These agents have not been used for

maintenance fluids or resuscitation in any controlled trials.

Summary: There is no evidence that anion composition of currently available crystalloids increase the risk

of AKI in humans. Large volume administration of saline leads to HCMA. Some animal data suggesting

that HCMA can alter renal blood flow (RBF) but no data are available in humans. Compared to LR,

delayed diuresis has been observed with NS. Recommendations for clinical practice: Other than the need to

be aware of potential side effects seen with large volume crystalloid resuscitation, no recommendations can

be given concerning choice of crystalloids at this time. Recommendations for future research: Adequately

powered randomized trials will be needed to assess differences in risk for AKI form various fluids. Given

that these trials will need to be very large, the likelihood that they will be conducted seems low. As an

alternative to randomized trials, large observational studies might provide important information on the

risks, if any, of various crystalloids. These studies should include assessment of clinical consequences

related to volume requirements, metabolic acidosis, electrolyte abnormalities, and infectious complications

in addition to risk of AKI.

3

ii. Tonicity (Saline, Mannitol)

Hypertonic Saline. Hypertonic saline (HTS) has been evaluated as a possible fluid for resuscitation in

injury and sepsis, because of its purported hemodynamic, endothelial and immunomodulatory effects.[7, 8]

Data are available that support potential benefits of hypertonic saline infusion in various aspects of the

pathophysiology of sepsis, including tissue hypoperfusion, decreased oxygen consumption, endothelial

dysfunction, cardiac depression, and the presence of a broad array of proinflammatory cytokines and

various oxidant species as well as in the pathophysiology of traumatic brain injury associated increases in

cerebral parenchymal volume.[8] Despite the purported advantages of HTS in ex-vivo studies, no durable

outcome benefit has been identified with HTS use as a plasma volume expander even in a large multi-

institutional prehospital trial specifically addressing traumatic brain injury.[9] HTS solutions are commonly

utilized to correct abnormalities of water balance. However, due to the hypertonic nature of these solutions,

they may also serve as plasma volume expanders with the added benefit of drawing fluid into the vascular

compartment on the basis of water movement along a concentration gradient. HTS is also commonly

coupled with dextran to support plasma volume expansion. Accordingly, HTS-dextran has been utilized for

plasma volume expansion during surgery. One such study compared a single 250 cc intra-operative dose of

7.5% NS/6% dextran-70 versus 250 cc of a hydroxyethylstarch (HES) in NS in patients undergoing aortic

aneurysm surgery.[10] Both groups were resuscitated with NS as the concomitant crystalloid solution. The

authors identified no difference in acid-base balance between the two groups due to the large amounts of

NS and relative small volumes of test agents given. Importantly, no differences in renal function were

noted with either regimen for 24 hours post-operatively. A small study in burn patients compared those

resuscitated with HTS (1991-1993) to patients resuscitated with LR in an earlier time frame (1986-1988

versus a later time frame (1993-1994). The authors found a four-fold increase in renal failure and a

doubling of mortality with HTS compared to LR resuscitated patients.[11]

A recent meta analysis (Cochrane review) from 2002 comparing isotonic crystalloids with hypertonic

solutions included 5 studies in trauma patients and demonstrated no benefit of HTS on outcome.[12] A

2004 Cochrane review of the issue[13] did not generate enough data to establish whether hypertonic

crystalloid is better than isotonic and near-isotonic crystalloid for the resuscitation of patients with trauma,

burns, or those undergoing surgery. However, the confidence intervals were wide and did not exclude

clinically significant differences. Further trials which clearly state the type and amount of fluid used and

that are large enough to detect a clinically important difference are needed. Small volumes of HTS do not

appear to cause, or increase the risk, of AKI with isolated or sporadic use, but detailed studies and reporting

are lacking. By contrast, HTS as a primary resuscitation fluid for burns appears to result in increased risk of

AKI.

Hypertonic mannitol solutions. HT-mannitol may be harmful because of the potential to induce rapid

intravascular volume expansion, leading to pulmonary edema and resulting in hyperoncotic kidney injury

in those without a severe plasma volume deficit.[14-16] Moreover, mannitol-based volume expansion may

induce AKI due to excessive osmotic diuresis in patients with hypovolemia.

4

Summary: HTS as sole resuscitation fluid is associated with ARF and excess mortality in burns. Small

volume HTS appears safe with isolated or sporadic use and was not associated with AKI in available

studies–detailed studies and reporting are lacking. HT-mannitol may induce AKI with excessive osmotic

diuresis and untoward hyperoncotic. Recommendations for clinical practice: HTS should be used with

caution in patients at risk for AKI and should be avoided in patients with burns. HTS is indicated for

treatment of severe hyponatremia in appropriate clinical settings. Use of HTS for plasma volume expansion

is not recommended. HT-mannitol is not recommended for fluid resuscitation. Recommendations for future

research: Given the immunomodulatory properties of HTS this agent should be specifically studied to

define a patient population who may benefit from HTS administration. Specifically, timing and dose should

be clearly defined and explored to clarify appropriate use. Pairing HTS with other fluids such as dextran,

starches, or HBOC’s should be explored as a means of maximizing potential benefits of both fluids in

specific circumstances.

Do different crystalloid solutions have different effects on mitigating the risk of AKI?

i. Anion composition (Chloride vs Bicarbonate)

Radiocontrast nephropathy (RCN). Reducing the concentration of chloride (e.g. use of bicarbonate) in

solutions for the prevention of RCN has been recently tried in a small (n=119 patients), single-centered,

randomized trial, in which plasma volume expansion with sodium bicarbonate based solutions was shown

to be superior to NS for prevention of RCN.[17] It is unclear if the putative benefit from bicarbonate-based

plasma volume expansion derives from amelioration or prevention of HCMA at the same time as contrast

administration, or an undescribed interaction of contrast material with the anion. A larger trial (n=326

consecutive patients) of medium to high risk patients compared three different fluid regimens for the

prevention of radiocontrast nephropathy: 1) saline + n-acetyl cysteine (NAC) (n=111), 2) sodium

bicarbonate + NAC (n=108), and 3) saline + NAC + ascorbic acid (n=107).[18] In this study, 1.9% of

bicarbonate + NAC treated patients sustained RCN (25% or greater creatinine increase at 48 hours post-

contrast exposure), while RCN was identified in 9.9% of those receiving NS + NAC and 10.3% of those

receiving NSS/+ NAC+ ascorbic acid. While the results of these two studies suggest that isotonic

bicarbonate may provide greater benefit than isotonic saline, either with or without NAC, neither study can

be considered to be conclusive. Thus, confirmation of these results in a large, multicenter RCT is needed.

Rhabdomyolysis. There is good evidence that rhabdomyolysis leads to AKI although the exact mechanisms

are still in dispute. In addition to hypovolemia that is present to some degree in all patients with

rhabdomyolysis (based on the antecedent cause), the myoglobin that is released into the circulation as a

result of muscle damage may also cause renal cellular damage via direct renal cytotoxicity, renal

vasoconstriction and renal tubular obstruction.[19, 20] Therefore, prevention of AKI in this setting involves

vigorous plasma volume expansion to maintain renal perfusion pressure and dilute myoglobin and other

toxins. Although there is limited evidence, clinical practice also commonly involves urinary alkalinization

with bicarbonate aimed at preventing tubular precipitation of myoglobin. Correction of hypovolemia is

5

essential to prevent AKI.[21] Plasma volume should be aggressively expanded to restore and maintain

effective circulating volume and to maintain a urine flow rate of 100 to 150 cc/hr.[22] While the

administration of sodium bicarbonate together with crystalloid plasma volume expansion is widely

recommended, the extent to which this interventions provides additional benefit to volume expansion with

crystalloid solution alone remains uncertain. If used, bicarbonate containing solutions should be isotonic

and care should be exercised to avoid metabolic alkalosis or electrolyte abnormalities (e.g. hypokalemia).

While randomized controlled trials are lacking, the available evidence suggests that bicarbonate have no

benefit over and above aggressive fluid resuscitation.[23-25] Knottenbelt has demonstrated that large-

volume infusion of crystalloid alone creates a solute diuresis sufficient to alkalinize the urine.[21]

Summary: The value of reducing chloride (i.e. using bicarbonate) in solutions for the prevention of RCN is

presently uncertain although risks are small in most patients. Bicarbonate solutions may be of value in

rhabdomyolysis but more evidence is necessary. Recommendations for clinical practice: The primary

recommendation for fluids to prevent AKI in the setting of radio-contrast exposure and rhabdomyolysis is

to ensure that they are given. Exact volumes or anion composition cannot be determined based on available

evidence, although bicarbonate appears safe in high risk groups. Consider isotonic bicarbonate in

preference to NS in patients without metabolic alkalosis. Recommendations for future research:

Randomized clinical trials to separate the effects of bicarbonate versus saline and other interventions

(NAC, tonicity of contrast) in preventing RCN in high risk patients with and without underlying kidney

disease. Randomized controlled trials are needed, to ascertain the value of bicarbonate vs. NS in the setting

of rhabdomyolysis, possibly with and without mannitol.

ii. Tonicity (Hypotonic vs isotonic, saline, mannitol)

Isotonic vs. hypotonic saline. A large, prospective, randomized, controlled trial in patients undergoing

cardiac catheterization found plasma volume expansion with NS to be superior to 0.45% saline.[26]

However, the incidence of RCN was lower than previously reported in comparable patient cohorts (0.7% in

NS and 2% in the 0.45% saline group).

Mannitol. Although studies in animals have shown mannitol to help protect the kidney against ischemic

injury, human studies fail to demonstrate the efficacy of mannitol in preventing AKI. Although

prophylactic mannitol has also been promoted in patients considered to be at high risk for AKI, such as

those undergoing vascular (aortic aneurysm) surgery, cardiac surgery, or patients developing obstructive

jaundice, several small RCTs have found no reduction in the incidence of AKI with mannitol

administration or over-hydration.[27-30] In small studies of patients undergoing kidney transplantation,

mannitol administration appears to have salutary effects with regard to AKI. In these studies, 250 ml of

20% mannitol given immediately before vessel unclamping reduced the incidence of AKI, as determined

by a decrease need for post-transplant dialysis.[31-38] However, no durable outcome difference at 3

months was found in comparison to patients who did not receive mannitol.[38]

6

Mannitol results in a higher incidence of radiocontrast-induced nephrotoxicity as compared with saline

plasma volume expansion alone in either diabetic or non-diabetic patients.[39, 40] Solomon et al found that

25 g of mannitol prior to contrast administration plus plasma volume expansion with saline was not

associated with any reduction in risk compared with saline alone; instead, there was a trend toward

harm.[39] A forced diuresis regimen which included intravenous crystalloid, mannitol, furosemide and

low-dose dopamine similarly exerted no effect on the overall incidence of contrast-induced nephropathy.

[40] The trial design allowed independent evaluation of the effects of mannitol and the results demonstrated

no additive benefit. In patients with both diabetes and chronic kidney disease receiving radiocontrast agent,

mannitol increased the incidence of nephrotoxicity.[41, 42] Conversely, the use of mannitol has been

advocated in the setting of rhabdomyolysis to create an osmotic diuresis [43, 44], vasodilatation of renal

vasculature[45] and free-radical scavenging.[46, 47] However, available evidence suggests that mannitol

offers no benefit over and above aggressive fluid resuscitation.[23-25] Furthermore, mannitol can be

harmful if urine output cannot be maintained.

Summary: Isotonic fluids are superior to hypotonic fluids for plasma volume expansion prior to contrast

material with regard to AKI risk mitigation. Mannitol is of unproven benefit with rhabdomyolysis or renal

transplant to decrease AKI incidence, and should not be routinely utilized in clinical practice. Mannitol

may increase the risk of RCN associated AKI in diabetic patients, and should not be used in this patient

population. Recommendations for clinical practice: Use isotonic IV fluids to reduce the risk of AKI from

radiocontrast and rhabdomyolysis. Mannitol should be avoided in patients with AKI; its role in preventing

AKI in the setting of rhabdomyolysis is currently uncertain and it should be avoided in the setting of RCN.

Recommendations for future research: Further clarification regarding the duration and volume of isotonic

IV fluids before and after radiocontrast agent administration to guide clinical practice.

Do different colloid solutions pose a risk of AKI?

Albumin. Human albumin has been widely used as the colloid for the treatment of hypovolemia in critically

ill patients. The SAFE study which was a large multicenter RCT (n=6997) demonstrated no difference in

renal outcomes (including incidence of AKI, need for RRT or duration of RRT), mortality, LOS in the ICU,

or duration of mechanical ventilation between patients managed with 4% albumin or normal saline for fluid

resuscitation.[48] The authors concluded that albumin is a safe as NS for resuscitation of the critically ill.

Importantly the study used a preparation of albumin in NS and few patients in the SAFE trial received large

volumes fluid resuscitation. Whether specific subgroups of critically ill patients, including those with AKI,

fare better with albumin is unknown.

Starches. Some studies have shown that patients treated with certain HES solutions (High molecular weight

(HMW)-HES and HES with a high C2-C6 molar substitution ratio) may suffer from renal dysfunction.[49,

50] Histological studies have shown reversible tubular cell edema causing obstruction and medullary

ischemia.[51] Boldt recently reviewed the renal complications of HES use. Study limitations included a

lack of fixed endpoints for intravascular volume therapy, crude measures of kidney function, and lack of

7

control groups in those receiving crystalloid infusions.[52] Additionally, HES has been noted to accumulate

in renal tubular cells and is believed to be a cause of AKI.[52] However, appropriately controlled studies

are lacking. Use of HES has been associated with AKI in post-cadaveric renal transplant patients.[50, 53]

In an observational cohort study, Legendre reported an 80% rate of osmotic nephrosis-like lesions

(vacuolization of proximal tubular cells) in transplanted kidneys after administration of HES with a mid-

range MW (200kD) and high molar substitution ratio (0.62) to brain dead organ donors.[54] The lesions

had no negative effect on graft function or serum creatinine at 3 and 6 months post transplantation. Similar

lesions have been described with other substances including mannitol and dextran. Infusion of hyperoncotic

colloids likely induces hyperviscosity and stasis of tubular fluid, especially when HES concentration is high

(>10%). Large amounts (>2000 ml or > 20 cc/kg bw/day) of HES preparations with LMW or MMW (e.g.

HES 130/0.4 or HES 200/0.5) and a low MS (0.4, 0.5) have been used safely in humans as have HWM-

HES with high molar substitution (670/0.75).[55, 56]

Younes studied HES in NS compared to NS in trauma patients with hemorrhagic shock and documented

that less HES volume was required to achieve an identical blood pressure compared to the NS resuscitated

patients.[57] These data are consistent with that from Boldt who resuscitated 30 trauma and 30 septic

patients with a more concentrated starch (10%) of lower molecular weight (100kDa) and lower molar

substitution ratio (0.3) that that used in the US (6%, 670 kDa/0.75) versus albumin and found that the HES

group evidenced better splanchnic (septic patients only) and systemic hemodynamic indices (septic and

trauma patients).[58] Finally, intra-op evidence of improved microcirculation as judged by improved tissue

oxygenation was seen when comparing HES (130/0.4) to LR resuscitation during major surgery.[59] Thus,

starch resuscitation may have some benefits in restoring effective circulating volume and supporting

microcirculatory flow, but the data are difficult to interpret because of multiple different molecular weights

and molar substitution ratios, as well as different resuscitation targets and patient populations.

Worse serum creatinine was noted in septic patients resuscitated with HES compared to gelatin.[49] This

study used a low molecular weight starch with high molar substitution ratio (200 kDa/0.6-0.66) in NS, and

found that serum creatinine was higher on days 6 and 7 compared to gelatin. However, baseline serum

creatinine was also slightly higher in the HES group (NS) and differences between groups may not have

been clinically significant nor did they persist (values were identical at 14 days). Also, survival was

identical for both groups. By contrast, intra-operative infusion of 500 cc of 6% HES in a balanced

electrolyte solution containing lactate (Hextend) has been associated with improved urine output, systemic

perfusion, gut mucosal perfusion, and the absence of acidosis or hypocalcemia requiring correction in

elderly patients undergoing major surgery with an anticipated blood loss exceeding 500ml; no differences

in post-op renal function (or dysfunction) was identified.[60] This study compared Hextend to HES in NS

(Hespan). Similarly, the prospective (Hextend vs Hespan) Phase III US FDA registration trial for Hextend

utilized quite large volumes of Hextend (up to 5L) without identifying diminution in renal function.[61]

Interestingly, patients receiving Hextend demonstrated better clotting ability as interrogated by

thromboelastography that correlated with reduced intra-operative blood loss and component therapy.

8

Additionally, a retrospective comparison of cardiac surgery patients treated with a low molecular weight

(200 kDa) and degree of molar substitution (0.5) starch vs. gelatin, or both[62] found no differences in

post-operative serum creatinine. Thus, it may be that sepsis induces a sensitivity to starch resuscitation that

is not evident in other conditions like trauma, surgery, and cardiopulmonary bypass that are also

accompanied by a substantial inflammatory and cytokine response. In contrast, a rodent model of sepsis

demonstrated improved survival with Hextend resuscitation compared to 0.9% NS or LR despite identical

resuscitation endpoints.[63]

Gelatins. In contrast to the controversies that surround HES use, there is little negative data regarding

gelatin infusions. The Cochrane analysis did not identify a significant benefit for gelatin resuscitation but

neither did it identify a detrimental effect.[64] Gelatin resuscitation has been utilized to great effect in

trauma, surgery and burns. However, its utility is limited by its relatively short half-life and relative low

affect on colloid oncotic pressure. There is no data that gelatin infusion causes renal injury, or that its use

with AKI should be restricted. Although gelatin is widely used in the UK, Australia and parts of Europe, it

has never been approved for use in the US. Limited data are available comparing it to other colloids.

Colloids do contain “unmeasured anions” and thus will increase the anion (or strong ion) gap, although this

effect is of unknown significance.

Summary: Albumin in saline appears to be equivalent to 0.9% saline in terms of risk of AKI. The risks of

AKI associated with HES, if any, appear to vary depending on the size, concentration, molar substitution

and possibly the diluents’ composition used in these preparations. Current evidence is inconclusive to judge

the risk of AKI attributable to HES. Gelatin does not appear to result in a measurable risk of AKI; however

studies evaluating this endpoint have been limited. Recommendations for clinical practice: HES should be

used with caution in patients with sepsis and renal transplantation due to limited evidence suggesting risk

of AKI in these groups. However, current data do not preclude their use in these settings. HES appears safe

for trauma, burns, and large blood loss emergency and elective surgery. Neither albumin nor gelatin

appears to increase the risk of AKI. Insufficient evidence exists to recommend one colloid preparation over

another in terms of risk for AKI. Recommendations for future research: HES of different concentrations,

molecular weights, and degrees of molar substitution ratios in different diluents (based on chloride

concentration) should be studied; possibly in sepsis and non-septic models prior to conducting a

prospective randomized trial in humans in a multi-center fashion. Gelatin should be studied in a fashion

similar to starch, but with additional comparators to include different crystalloid solutions of different

chloride concentrations. Compound agents like hypertonic HES should have their investigation deferred

until the optimal HES construction has been defined with regard to concentration, molecular weight, and

degree of molar substitution ratio, and diluents’ electrolyte composition. HES administration with different

concentrations, molecular weights, and degrees of molar substitution ratios, in different diluents should be

specifically administered in an animal model to determine the incidence and prevalence of renal tubular

incorporation of starch with both hypovolemia and hypervolemia, hemorrhagic shock, and sepsis as well as

the incidence and prevalence of AKI in each of those settings.

9

Do different colloid solutions have different effects on mitigating the risk of AKI?

i. Cirrhosis

Albumin. Daily administration of albumin has been shown to be beneficial in reducing the risk of AKI in

patients with cirrhosis and spontaneous bacterial peritonitis, although the exact mechanism underlying this

effect remains opaque.[65] In a multicenter, randomized trial by Sort et al,[66] the addition of an

intravenous infusion of human serum albumin to standard antibiotic treatment in patients with spontaneous

bacterial peritonitis reduced the rates of renal impairment and mortality; whether the effect occurs

exclusively on the basis of expanded effective circulating volume alone is not known. Although large-

volume paracentesis (LVP) has been used in the treatment of ascites, the need for plasma volume expansion

after therapeutic paracentesis remains controversial. While hemodynamic studies in patients with ascites

have shown that a significant reduction in intravascular volume and renal impairment occurs more

commonly when ascitic fluid is totally mobilized without prior or concomitant plasma volume expansion,

[65, 67-69] other studies of LVP without the administration of any plasma volume expanders document no

increase in renal complications. [70-72] Based on the International Ascites Club guidelines, 6-8 grams of

albumin is recommended to be infused per liter of ascites fluid removed for paracentesis volumes greater

than 5-6 liters.[17]

Starch. In a pilot study comparing the use of albumin and HES (in normal saline) in patients with cirrhosis

and spontaneous bacterial peritonitis, renal outcome was similar between the two groups.[73] Although

albumin is the most frequently used plasma volume expander for LVP, other plasma volume expanders

including Dextran 70,[74] 5% synthetic polymerized gelatin (hemaccel),[75] HES (in NS),[76] have also

been shown to be as effective as albumin in the prevention of complications related to large-volume

paracentesis. In one study, Cabrera et al. removed a mean of 8 L of ascitic fluid from 14 patients, with

replacement only by intravenous normal saline solution.[70] None of the patients had any clinical

complications or alterations in kidney function tests.

ii. Renal Transplant

Albumin. Several observational studies in kidney transplant recipients suggest that volume expansion with

albumin reduces the onset and the extent of post-transplant oliguria, and AKI while improving 1-year graft

and patient survival. Unfortunately controlled clinical studies are still lacking.[77-79] In one study,

although higher intra-operative albumin dose (1.2-1.6 g/kg bodyweight) was associated with better

outcomes in comparison to those receiving lower albumin dose (0-0.4 g/kg), the possible effects of

concomitantly administered mannitol, furosemide and crystalloid plasma volume expansion on outcomes

cannot be ignored.[77]

Starch. Comparison of intra-operative albumin and dextran-40 in 17 renal transplant recipients from a

living related donor failed to show any difference between the two plasma volume expanding agents with

regard to renal function.[80]

10

Summary: Volume expansion appears to be useful in avoiding renal dysfunction following LVP and

possibly in the setting of bacterial peritonitis even with small volume removal. It is unknown whether

different colloid preparations are more or less effective for volume expansion in patients with cirrhosis.

Data are inconclusive comparing different colloids in renal transplant. Recommendations for clinical

practice: No recommendations can be made for selecting between colloid preparations for mitigating risk of

AKI following LVP or renal transplant. Recommendations for future research: Randomized controlled

trials comparing HES and albumin for LVP and most renal transplant are needed.

How potassium containing fluids should be used in patients with AKI?

Blood Products. Packed red blood cell (PRBC) transfusion will deliver a measurable amount of potassium

to the blood volume of a recipient. The amount of potassium transfused increases with the age of banked

blood, and is least with fresh whole blood transfusion, as sporadically occurs in military scenarios with type

specific “buddy” transfusion. As banked blood ages, the integrity of the red blood cell plasma membrane is

compromised leading to extracellular translocation of normally intracellular potassium. While the amount

of potassium infused is measurable, hyperkalemia requiring therapy is rare with normal plasma potassium,

and is virtually nonexistent with hypokalemic states. Patients at greatest risk for transfusion associated

hyperkalemia are those receiving component transfusion on a massive transfusion protocol for

exsanguinating hemorrhage from a variety of clinical conditions.[81] Despite this theoretic risk, the major

risk with massive transfusion is hyperkalemia and hypocalcemia, is readily identified with transfusion of

frozen PRBC.[82] A recent study of 131 penetrating trauma victims found that, by multivariable logistic

regression analysis, independent risk factors for hyperkalemia were an emergency-room plasma potassium

level of 4.0 mmol/L or higher (RR 3.40, 95% CI 1.17-9.84, p=0.024), and transfusion of 20 or more units

of cell- or plasma-based transfusion products (RR 14.66, 95% CI 4.68-45.97 p < 0.001).[83] Hyperkalemia

risk may also be ameliorated by preinfusion washing methods.[84] Some risk may be potentially

ameliorated by incorporating activated factor VII into massive transfusion protocols as a means of reducing

total transfused units of PRBC; there is no data at present to support this speculation. Since the potassium

load with components other than PRBC is minimal, transfusion associated hyperkalemia is not a clinically

relevant issue when transfusing fresh frozen plasma, platelets, or cryoprecipitate.

Lactated Ringer’s solution. Lactated Ringer’s solution is formulated with a small amount of potassium,

and, according to conventional wisdom, has been commonly avoided as a resuscitation fluid in patients

with AKI or CKD. Certainly, in patients with a normal of low plasma potassium, LR is safe with regard to

hyperkalemia induction. This condition is exceedingly rare in the trauma patient population for whom LR

is the predominant resuscitation fluid in the US. However, in the renal transplant patient population, large

quantity plasma volume expansion with NSS was compared to expansion using lactated Ringer’s.[2] In this

study, only NSS was associated with hyperkalemia (K+ > 6 mEq/L) and metabolic acidosis that required

therapy. Moreover, this study was terminated early due to safety concerns in the NSS arm. One may

11

speculate that this observation reflects diminutions in RBF and GFR that accompany hyperchloremic

metabolic acidosis, but the mechanism underpinning these observations has not been fully elucidated.

Summary: Transfusion associated hyperkalemia is rare, but should be assessed when treating a patient

using a massive transfusion protocol. Lactated Ringer’s is safe for plasma volume expansion in those at risk

for AKI. NSS should be used with caution in the renal transplant patient due to the risk of hyperkalemia.

Recommendations for clinical practice: No change in transfusion practice is recommended as transfusion

associated hyperkalemia is rare. LR does not carry a risk of hyperkalemia.

What is the risk of AKI associated with blood substitutes?

Hemoglobin-based oxygen carriers (HBOC). Over the last two decades animal and human trials have

explored the use of HBOC for both plasma volume expansion and restoration of oxygen carrying capacity

in malaria, orthopedic surgery, cardiac surgery, and hemorrhagic shock from injury. Since the HBOCs that

are currently under clinical investigation are formulated in a normal to low potassium concentration diluent,

there is little to no risk of infusion associated hyperkalemia. Due to the colloidal nature of the products, and

the known vasoactivity of the products, resuscitation with an HBOC utilizes less total volume compared to

crystalloids and therefore, delivers less potassium. HBOC vasoactivity appears to be related to the

tetrameric composition of the product, and is reduced with current products (< 3% tetrameric Hgb)

compared to the 1st generation progenitor product, Baxter’s ill-fated DCLHB (> 30% tetrameric Hgb).

Vasoactivity is principally related to nitric oxide scavenging, although not exclusively so and has been

chiefly explored in the systemic circulation (systolic and mean arterial pressures) as well as the

cardiopulmonary circuit (pulmonary artery pressures and cardiac output).[85]

There are two main US competitor products, neither of which has currently garnered FDA approval for

licensing: Nothfield’s Polyheme and Biopure’s Hemopure (HBOC-201). Both products have undergone

extensive preclinical testing in predominantly porcine models of hemorrhagic shock, a condition associated

with AKI.[86] HBOC-201 does not demonstrate increased rates of AKI compared to standard crystalloid

plasma volume expansion even in prolonged low-volume resuscitation conditions.[87] HBOC-201

controlled and uncontrolled hemorrhagic shock preclinical data also includes histopathologic and electron

microscopic analysis of renal parenchyma and tubules documenting the absence of renal injury in treated

pigs.[88] Similarly, both products have undergone clinical trials without an increased incidence of AKI

compared to crystalloid plasma volume expansion.[89-93] These trials included patients with sickle cell

disease, patients undergoing elective vascular surgery, and trauma patients. Each of these (except the sickle

cell disease trial) trials include patients in RIFLE classification R as well as many in the I category. There

is no evidence of increased AKI occurrence in these trials supporting the potential use of HBOC in plasma

deficit conditions from which patients are at risk of AKI. The reader should note that Northfield has

recently completed a pre-hospital and in-hospital hemorrhagic shock trial, the results of which are not yet

publicly available, but will shed additional information on the use of HBOC in at-risk patients for AKI.

12

Biopure also manufactures a veterinary product, Oxypure, whose use is unassociated with renal injury in a

variety of mammalian species.

Summary: No definitive data exists on the use of HBOCs and abrogation or induction of AKI at present.

Recommendations for clinical practice: No recommendation can be made with regard to HBOCs at present.

Recommendations for future research: Randomized controlled trials comparing HBOC plasma volume

expansion and oxygen carrying capacity to crystalloid and colloid plasma volume expansion in RIFLE “R”,

“I” and “F” patients in the pre-hospital and in-hospital setting.

Fluids for Renal Replacement Therapy (RRT)

This section will review and make recommendations on the use of fluids for RRT.

What should be the composition of replacement fluid and dialysate?

General considerations. Although Peritoneal Dialysis is still used in different areas of the world, it is not

the preferred modality of renal replacement therapy (RRT) in AKI. Therefore, the composition of PD

solutions will not be further discussed in this review. Given the technical differences between intermittent

Hemodialysis (IHD) and continuous renal replacement therapies (CRRT), different considerations apply to

fluid composition in each modality. Finally, although other extracorporeal techniques (MARS, Plasma

exchange) are used in AKI, given that they are generally not used as RRT, they will not be discussed in this

document.

Fluids for IHD and IHDF. Higher dialysate sodium concentrations are associated with increased

hemodynamic stability in IHD.[94-96] In IHDF, whether the replacement fluid is infused pre- or post-filter

has an impact on the relationship between sodium concentration and sodium balance. During dialysis, the

majority of sodium movement is due to ultrafiltration, Sodium can also potentially move by diffusion. As a

result of a variety of physicochemical factors, including the difference between plasma sodium

concentration and the true concentration of sodium in plasma water and the Gibbs-Donnan effect, plasma

sodium concentration generally must be at least 4 mMol/L greater than dialysate sodium in order to diffuse

down a concentration gradient.[97] During hemofiltration, sodium moves with plasma water across the

hemofilter membrane, but due to protein coating of the membrane retarding sodium movement, the amount

of sodium sieved is < 1.0. Thus, when the same sodium concentration is used in dialysate and replacement

solutions, convective therapies tend to result in a greater positive sodium balance than pure dialysis,[98, 99]

possibly accounting for the greater cardiovascular stability with convective techniques are compared with

pure dialytic techniques for the same net rate of ultrafiltration.[100] Commercially available hemofiltration

replacement fluids vary in sodium concentration from 138-155 mMol/l.[101] Studies have suggested that a

positive sodium balance occurs when plasma sodium is less than 8 mMol/l higher than the replacement

fluid.[98] During hemofiltration the overall sodium balance is also affected by the site of infusion of

replacement fluid, as post-dilution results in greater sodium retention than mid-dilution, and mid-dilution

13

greater sodium retention than predilution, due to changes in protein polarization and hematocrit as blood

flows through the hemofilter.

Requirements for chloride, potassium, magnesium are variable in different clinical situations. Maintenance

of normoglycemia in the critically ill patient has been associated with improved outcomes.[102-104]

Removal of glucose from the dialysate has been shown to create risk of severe hypoglycemia. Trace

elements, including water soluble metals, micronutrients, aminoacids and folate are lost during

dialysis.[105, 106]

Summary and Recommendations for clinical practice: In hemodynamically unstable patients, consider

using increased sodium (up to 150-155 mMol/L) dialysate concentration. When the same sodium

concentration is used in dialysate and replacement solutions, convective therapies tend to result in a greater

positive sodium balance than pure dialysis,[98, 99] possibly accounting for the greater cardiovascular

stability with convective techniques are compared with pure dialytic techniques for the same net rate of

ultrafiltration.[100] Chloride, Potassium, Magnesium and other anion needs must be adjusted to each

individual situation. Glucose. Utilize physiological glucose concentration in dialysis fluids. Replace trace

elements, aminoacids, micronutrients and vitamins as necessary.

Fluids for CRRT. Existing clinical practice based on available evidence and opinion demonstrates that

sodium is generally kept at an isonatric (physiologic) concentration except when special prescriptions are

used in combination with some citrate anticoagulation protocols. Chloride, potassium, magnesium and

anion needs are variable in different clinical situations. Hypophosphoremia due to increased clearance and

intracellular shifts due to “refeeding” are common in CRRT, and may place patients at risk of

complications including rhabdomyolysis.[107] Maintenance of normoglycemia has been shown to be

associated with lesser mortality in the critically ill patient.[102-104] Trace elements, including water

soluble metals, micronutrients, aminoacids and folate are lost during CRRT.[105, 106]

Summary and Recommendations for clinical practice: Physiologic concentrations of sodium should be used

except when using citrate anticoagulation. In the latter circumstances, adjustments may be necessary given

the variable contents of sodium in different citrate solutions. Chloride, potassium, magnesium anions

should be present in replacement fluid and or dialysate in concentrations tailored to patient need and

anticoagulation procedures. Calcium should be absent from the replacement fluid and dialysate when

citrate is used as an anticoagulant. To avoid hypophosphoremia, phosphate should be provided either as a

supplement in the CRRT fluids (replacement fluid or dialysate) or as a nutritional supplement.[107] To

avoid hyperglycemia, Glucose can be either absent or present at physiological concentration in dialysis

fluid. Loss of trace elements (water soluble metals, micronutrients, aminoacids, and folate) must be

appropriately replaced.

14

What metabolizable anion should be used and under what circumstances?

Both lactate and bicarbonate ions have been used in replacement fluid and dialysate for CRRT.

Historically, lactate has been preferentially used as a buffer due to the instability of bicarbonate-based

solutions when stored over prolonged periods of time. This problem has recently been overcome, allowing

commercial availability of bicarbonate-based fluids. Controlled (though not all randomized) trials have

suggested that lactate and bicarbonate buffered solutions have a similar efficacy for correction of metabolic

acidosis during CRRT.[96, 108-112] However, recent studies [113, 114] showed better control of metabolic

acidosis with bicarbonate as compared to lactate. Barenbrock [114] also showed better hemodynamic

tolerance of therapy with bicarbonate. Neither study showed a difference in mortality, but neither was

adequately powered. Blood levels of lactate are generally higher when lactate is used as a buffer and may

confuse the clinical interpretation of these measurements. Whether this hyperlactatemia is associated with

morbidity is not clear. Depending on tissue redox status and substrate availability, lactate is either

metabolized back to pyruvate and into the citric acid cycle, resulting in proton buffering, or into glucose by

gluconeogenesis. Potential concerns with excessive lactate accumulation are hemodynamic compromise

[111], increased urea generation [109, 115] and cerebral dysfunction.[115] Hyperlactatemia may develop in

situations of impaired lactate clearance including liver failure and tissue hypoperfusion. This

hyperlactatemia can be expected to be more pronounced if lactate-buffered solutions are used during high-

volume hemofiltration. Accumulation of the D-isomer of lactate may also be a concern as the D-isomer

constitutes 50% of the total lactate contents of racemic mixes. As D-lactate is non metabolizable by

humans, it may accumulate, contributing to severely elevated lactatemia and associated with neurologic

impairment.

In the IHD literature, acetate has been shown to be associated with impaired myocardial contractility and

decreased cardiac function. This anion has been rarely used as a buffer in CRRT. Citrate, used primarily for

its anticoagulant properties, serves as an effective buffer. Scant evidence is available on the use of citrate

exclusively as a buffer in CRRT. Importantly, citrate metabolism is often impaired in liver failure or

muscle hypoperfusion, both situations posing risk of hypercitratemia when citrate is utilized.

Hypercitratemia carries the risk of decreased ionized extracellular calcium concentration. Importantly,

blood products contain citrate as an anticoagulant; massive blood or plasma product transfusions are

associated with high citrate loads, which accumulate when citrate is simultaneously used as an

anticoagulant and/or a buffer. Low concentrations of citrate are present in some commercial dialysate

solutions for IHD. Complications of citrate toxicity have not been associated with these agents.

Summary: Both lactate and bicarbonate are able to correct metabolic acidosis in most CRRT patients;

however, correction of acidosis may not be as efficient with lactate as with equimolar bicarbonate.

Worsening hyperlactatemia has been noted when lactate was used in patients with lactic acidosis or liver

failure. The clinical relevance of this finding is unknown. Citrate used as an anticoagulant has also been

effectively used as a buffer in CRRT. Recommendations for clinical practice: There is consensus that

bicarbonate is the preferred dialysis buffer in IHD. For CRRT, bicarbonate is an effective buffer and is

15

currently the preferred organic buffer in commercially manufactured solutions. Lactate-buffered solutions

are safe and efficacious in the majority of patients, but these solutions may be hazardous whenever lactate

clearance is impaired, such as in liver failure patients or patients with severe tissue hypoperfusion. D-

lactate should be removed from lactate-containing solutions, which should consists almost exclusively of

L-Lactate. There is insufficient data to evaluate the use of acetate-buffered solutions in CRRT. The

metabolism of sodium citrate used for regional anticoagulation during CRRT, generates three moles of

bicarbonate per mole of citrate and functions as an efficacious organic buffer. Use of citrate in the setting of

decreased citrate clearance or when patients receive large doses of citrate during massive transfusions

should be done with individualized adjustment of citrate dose and with close monitoring of plasma ionized

calcium levels. Recommendations for future research: Further studies comparing lactate and bicarbonate in

high volume hemofiltration are required. Optimal acid-base management of patients with permissive

hypercapnia requires further investigation.

Should the buffer concentration be higher in patients with permissive hypercapnia?

In patients with ARDS/ALI on lung protecting ventilator strategies, the resulting respiratory acidosis can be

partially or completely compensated by elevation of plasma bicarbonate with CRRT. The discussion of to

what level should acidemia be corrected is beyond the scope of this discussion.

How should fluids be compounded?

Assuring correct electrolyte formulation by repeat testing as needed, given that serious errors have been

reported when local pharmacies compound without regular monitoring of composition.[116-118]

Recommendations for clinical practice: When dialysis fluids are compounded locally, appropriate

monitoring of RRT fluid composition must be assured.

References

1. Waters, J.H., et al., Normal saline versus lactated Ringer's solution for intraoperative fluid management in patients undergoing abdominal aortic aneurysm repair: an outcome study. Anesth Analg, 2001. 93(4): p. 817-22.

2. O'Malley, C.M., et al., A randomized, double-blind comparison of lactated Ringer's solution and 0.9% NaCl during renal transplantation. Anesth Analg, 2005. 100(5): p. 1518-24, table of contents.

3. Wilcox, C.S., Regulation of renal blood flow by plasma chloride. J Clin Invest, 1983. 71(3): p. 726-35.

4. Williams, E.L., et al., The effect of intravenous lactated Ringer's solution versus 0.9% sodium chloride solution on serum osmolality in human volunteers. Anesth Analg, 1999. 88(5): p. 999-1003.

5. Engstrom, M., et al., Acidosis impairs the coagulation: A thromboelastographic study. J Trauma, 2006. 61(3): p. 624-8.

16

6. Stanton, K., et al., Human polymorphonuclear cell death after exposure to resuscitation fluids in vitro: apoptosis versus necrosis. J Trauma, 2003. 54(6): p. 1065-74; discussion 1075-6.

7. Oliveira, R.P., et al., Acute haemodynamic effects of a hypertonic saline/dextran solution in stable patients with severe sepsis. Intensive Care Med, 2002. 28(11): p. 1574-81.

8. Oliveira, R.P., et al., Clinical review: Hypertonic saline resuscitation in sepsis. Crit Care, 2002. 6(5): p. 418-23.

9. Cooper, D.J., et al., Prehospital hypertonic saline resuscitation of patients with hypotension and severe traumatic brain injury: a randomized controlled trial. Jama, 2004. 291(11): p. 1350-7.

10. Bruegger, D., et al., Effect of hypertonic saline dextran on acid-base balance in patients undergoing surgery of abdominal aortic aneurysm. Crit Care Med, 2005. 33(3): p. 556-63.

11. Huang, P.P., et al., Hypertonic sodium resuscitation is associated with renal failure and death. Ann Surg, 1995. 221(5): p. 543-54; discussion 554-7.

12. Bunn, F., P. Alderson, and V. Hawkins, Colloid solutions for fluid resuscitation. Cochrane Database Syst Rev, 2003(1): p. CD001319.

13. Alderson, P., et al., Human albumin solution for resuscitation and volume expansion in critically ill patients. Cochrane Database Syst Rev, 2004(4): p. CD001208.

14. Perez-Perez, A.J., et al., Acute renal failure following massive mannitol infusion. Am J Nephrol, 2002. 22(5-6): p. 573-5.

15. Visweswaran, P., E.K. Massin, and T.D. Dubose, Jr., Mannitol-induced acute renal failure. J Am Soc Nephrol, 1997. 8(6): p. 1028-33.

16. Biesenbach, G., et al., Severe tubulopathy and kidney graft rupture after coadministration of mannitol and ciclosporin. Nephron, 1992. 62(1): p. 93-6.

17. Merten, G.J., et al., Prevention of contrast-induced nephropathy with sodium bicarbonate: a randomized controlled trial. Jama, 2004. 291(19): p. 2328-34.

18. Briguori, C., et al., Renal Insufficiency Following Contrast Media Administration Trial (REMEDIAL): a randomized comparison of 3 preventive strategies. Circulation, 2007. 115(10): p. 1211-7.

19. Better, O.S., I. Rubinstein, and D.N. Reis, Muscle crush compartment syndrome: fulminant local edema with threatening systemic effects. Kidney Int, 2003. 63(3): p. 1155-7.

20. Slater, M.S. and R.J. Mullins, Rhabdomyolysis and myoglobinuric renal failure in trauma and surgical patients: a review. J Am Coll Surg, 1998. 186(6): p. 693-716.

21. Knottenbelt, J.D., Traumatic rhabdomyolysis from severe beating--experience of volume diuresis in 200 patients. J Trauma, 1994. 37(2): p. 214-9.

22. Better, O.S. and J.H. Stein, Early management of shock and prophylaxis of acute renal failure in traumatic rhabdomyolysis. N Engl J Med, 1990. 322(12): p. 825-9.

23. Brown, C.V., et al., Preventing renal failure in patients with rhabdomyolysis: do bicarbonate and mannitol make a difference? J Trauma, 2004. 56(6): p. 1191-6.

24. Homsi, E., et al., Prophylaxis of acute renal failure in patients with rhabdomyolysis. Ren Fail, 1997. 19(2): p. 283-8.

25. Girbes, A.R., Prevention of acute renal failure: role of vaso-active drugs, mannitol and diuretics. Int J Artif Organs, 2004. 27(12): p. 1049-53.

26. Mueller, C., et al., Prevention of contrast media-associated nephropathy: randomized comparison of 2 hydration regimens in 1620 patients undergoing coronary angioplasty. Arch Intern Med, 2002. 162(3): p. 329-36.

17

27. Ip-Yam, P.C., et al., Renal function and proteinuria after cardiopulmonary bypass: the effects of temperature and mannitol. Anesth Analg, 1994. 78(5): p. 842-7.

28. Beall, A.C., Jr., et al., Mannitol-Induced Osmotic Diuresis During Renal Artery Occlusion. Ann Surg, 1965. 161: p. 46-52.

29. Gubern, J.M., et al., A randomized trial on the effect of mannitol on postoperative renal function in patients with obstructive jaundice. Surgery, 1988. 103(1): p. 39-44.

30. Gubern, J.M., F. Martinez-Rodenas, and A. Sitges-Serra, Use of mannitol as a measure to prevent postoperative renal failure in patients with obstructive jaundice. Am J Surg, 1990. 159(4): p. 444-5.

31. van Valenberg, P.L., et al., Mannitol as an indispensable constituent of an intraoperative hydration protocol for the prevention of acute renal failure after renal cadaveric transplantation. Transplantation, 1987. 44(6): p. 784-8.

32. Tiggeler, R.G., et al., Prevention of acute tubular necrosis in cadaveric kidney transplantation by the combined use of mannitol and moderate hydration. Ann Surg, 1985. 201(2): p. 246-51.

33. Porras, I., et al., A multivariate analysis of the risk factors for posttransplant renal failure: beneficial effect of a flush solution with mannitol. Transplant Proc, 1992. 24(1): p. 52-3.

34. Richards, K.F., et al., Mannitol reduces ATN in cadaveric allografts. Transplant Proc, 1989. 21(1 Pt 2): p. 1228-9.

35. Grino, J.M., et al., Flush solution with mannitol in the prevention of post-transplant renal failure. Transplant Proc, 1987. 19(5): p. 4140-2.

36. Hoitsma, A.J., et al., Important role for mannitol in the prevention of acute renal failure after cadaveric kidney transplantation. Transplant Proc, 1987. 19(1 Pt 3): p. 2063-4.

37. Lauzurica, R., et al., Hydration and mannitol reduce the need for dialysis in cadaveric kidney transplant recipients treated with CyA. Transplant Proc, 1992. 24(1): p. 46-7.

38. Weimar, W., et al., A controlled study on the effect of mannitol on immediate renal function after cadaver donor kidney transplantation. Transplantation, 1983. 35(1): p. 99-101.

39. Solomon, R., et al., Effects of saline, mannitol, and furosemide to prevent acute decreases in renal function induced by radiocontrast agents. N Engl J Med, 1994. 331(21): p. 1416-20.

40. Stevens, M.A., et al., A prospective randomized trial of prevention measures in patients at high risk for contrast nephropathy: results of the P.R.I.N.C.E. Study. Prevention of Radiocontrast Induced Nephropathy Clinical Evaluation. J Am Coll Cardiol, 1999. 33(2): p. 403-11.

41. Anto, H.R., et al., Infusion intravenous pyelography and renal function. Effect of hypertonic mannitol in patients with chronic renal insufficiency. Arch Intern Med, 1981. 141(12): p. 1652-6.

42. Weisberg, L.S., P.B. Kurnik, and B.R. Kurnik, Risk of radiocontrast nephropathy in patients with and without diabetes mellitus. Kidney Int, 1994. 45(1): p. 259-65.

43. Abassi, Z.A., A. Hoffman, and O.S. Better, Acute renal failure complicating muscle crush injury. Semin Nephrol, 1998. 18(5): p. 558-65.

44. Malinoski, D.J., M.S. Slater, and R.J. Mullins, Crush injury and rhabdomyolysis. Crit Care Clin, 2004. 20(1): p. 171-92.

45. Johnston, P.A., et al., Prostaglandins mediate the vasodilatory effect of mannitol in the hypoperfused rat kidney. J Clin Invest, 1981. 68(1): p. 127-33.

46. Tay, M., et al., The inhibitory action of oxygen radical scavengers on proteinuria and glomerular heparan sulphate loss in the isolated perfused kidney. Biochem Int, 1990. 20(4): p. 767-78.

18

47. Bratell, S., et al., Effects of oxygen free radical scavengers, xanthine oxidase inhibition and calcium entry-blockers on leakage of albumin after ischaemia. An experimental study in rabbit kidneys. Acta Physiol Scand, 1988. 134(1): p. 35-41.

48. Finfer, S., et al., A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med, 2004. 350(22): p. 2247-56.

49. Schortgen, F., et al., Effects of hydroxyethylstarch and gelatin on renal function in severe sepsis: a multicentre randomised study. Lancet, 2001. 357(9260): p. 911-6.

50. Cittanova, M.L., et al., Effect of hydroxyethylstarch in brain-dead kidney donors on renal function in kidney-transplant recipients. Lancet, 1996. 348(9042): p. 1620-2.

51. Boldt, J. and S. Suttner, Plasma substitutes. Minerva Anestesiol, 2005. 71(12): p. 741-58.

52. Boldt, J. and H.J. Priebe, Intravascular volume replacement therapy with synthetic colloids: is there an influence on renal function? Anesth Analg, 2003. 96(2): p. 376-82, table of contents.

53. Riou, B. and M.L. Cittanova, An international review of HES. Intensive Care Med, 1999. 25(11): p. 1340-1.

54. Legendre, C., et al., Hydroxyethylstarch and osmotic-nephrosis-like lesions in kidney transplantation. Lancet, 1993. 342(8865): p. 248-9.

55. Vogt, N.H., et al., Large-dose administration of 6% hydroxyethyl starch 200/0.5 total hip arthroplasty: plasma homeostasis, hemostasis, and renal function compared to use of 5% human albumin. Anesth Analg, 1996. 83(2): p. 262-8.

56. Jungheinrich, C., et al., Volume efficacy and reduced influence on measures of coagulation using hydroxyethyl starch 130/0.4 (6%) with an optimised in vivo molecular weight in orthopaedic surgery : a randomised, double-blind study. Drugs R D, 2004. 5(1): p. 1-9.

57. Younes, R.N., et al., Use of pentastarch solution in the treatment of patients with hemorrhagic hypovolemia: randomized phase II study in the emergency room. World J Surg, 1998. 22(1): p. 2-5.

58. Boldt, J., et al., Influence of different volume therapy regimens on regulators of the circulation in the critically ill. Br J Anaesth, 1996. 77(4): p. 480-7.

59. Lang, K., et al., Colloids versus crystalloids and tissue oxygen tension in patients undergoing major abdominal surgery. Anesth Analg, 2001. 93(2): p. 405-9 , 3rd contents page.

60. Wilkes, N.J., et al., The effects of balanced versus saline-based hetastarch and crystalloid solutions on acid-base and electrolyte status and gastric mucosal perfusion in elderly surgical patients. Anesth Analg, 2001. 93(4): p. 811-6.

61. Gan, T.J., et al., Hextend, a physiologically balanced plasma expander for large volume use in major surgery: a randomized phase III clinical trial. Hextend Study Group. Anesth Analg, 1999. 88(5): p. 992-8.

62. Wiesen, P., et al., Effect of hydroxyethylstarch on renal function in cardiac surgery: a large scale retrospective study. Acta Anaesthesiol Belg, 2005. 56(3): p. 257-63.

63. Kellum, J.A., Fluid resuscitation and hyperchloremic acidosis in experimental sepsis: improved short-term survival and acid-base balance with Hextend compared with saline. Crit Care Med, 2002. 30(2): p. 300-5.

64. Roberts, I., et al., Colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Database Syst Rev, 2004(4): p. CD000567.

65. Gines, P., et al., Randomized comparative study of therapeutic paracentesis with and without intravenous albumin in cirrhosis. Gastroenterology, 1988. 94(6): p. 1493-502.

66. Sort, P., et al., Effect of intravenous albumin on renal impairment and mortality in patients with cirrhosis and spontaneous bacterial peritonitis. N Engl J Med, 1999. 341(6): p. 403-9.

19

67. Luca, A., et al., Beneficial effects of intravenous albumin infusion on the hemodynamic and humoral changes after total paracentesis. Hepatology, 1995. 22(3): p. 753-8.

68. Gines, A., et al., Randomized trial comparing albumin, dextran 70, and polygeline in cirrhotic patients with ascites treated by paracentesis. Gastroenterology, 1996. 111(4): p. 1002-10.

69. Pozzi, M., et al., Time course of circulatory and humoral effects of rapid total paracentesis in cirrhotic patients with tense, refractory ascites. Gastroenterology, 1994. 106(3): p. 709-19.

70. Cabrera, J., et al., Large-volume paracentesis and intravenous saline: effects on the renin-angiotensin system. Hepatology, 1991. 14(6): p. 1025-8.

71. Pinto, P.C., J. Amerian, and T.B. Reynolds, Large-volume paracentesis in nonedematous patients with tense ascites: its effect on intravascular volume. Hepatology, 1988. 8(2): p. 207-10.

72. Kao, H.W., et al., The effect of large volume paracentesis on plasma volume--a cause of hypovolemia? Hepatology, 1985. 5(3): p. 403-7.

73. Fernandez, J., et al., A randomized unblinded pilot study comparing albumin versus hydroxyethyl starch in spontaneous bacterial peritonitis. Hepatology, 2005. 42(3): p. 627-34.

74. Fassio, E., et al., Paracentesis with Dextran 70 vs. paracentesis with albumin in cirrhosis with tense ascites. Results of a randomized study. J Hepatol, 1992. 14(2-3): p. 310-6.

75. Salerno, F., et al., Randomized comparative study of hemaccel vs. albumin infusion after total paracentesis in cirrhotic patients with refractory ascites. Hepatology, 1991. 13(4): p. 707-13.

76. Altman, C., et al., Randomized comparative multicenter study of hydroxyethyl starch versus albumin as a plasma expander in cirrhotic patients with tense ascites treated with paracentesis. Eur J Gastroenterol Hepatol, 1998. 10(1): p. 5-10.

77. Dawidson, I.J., et al., Intraoperative albumin administration affects the outcome of cadaver renal transplantation. Transplantation, 1992. 53(4): p. 774-82.

78. Dawidson, I., et al., Intraoperative albumin improves the outcome of cadaver renal transplantation. Transplant Proc, 1987. 19(5): p. 4137-9.

79. Dawidson, I., et al., The effect of intraoperative fluid management on the incidence of acute tubular necrosis. Transplant Proc, 1987. 19(1 Pt 3): p. 2056-7.

80. Dawidson, I., et al., Intravascular volumes and colloid dynamics in relation to fluid management in living related kidney donors and recipients. Crit Care Med, 1987. 15(7): p. 631-6.

81. Cinat, M.E., et al., Improved survival following massive transfusion in patients who have undergone trauma. Arch Surg, 1999. 134(9): p. 964-8; discussion 968-70.

82. Rao, T.L., et al., Serum potassium levels following transfusion of frozen erythrocytes. Anesthesiology, 1980. 52(2): p. 170-2.

83. Perkins, R.M., et al., Resuscitative Hyperkalemia in Noncrush Trauma: A Prospective, Observational Study. Clin J Am Soc Nephrol, 2007. 2(2): p. 313-319.

84. Bansal, I., et al., A comparative study of reducing the extracellular potassium concentration in red blood cells by washing and by reduction of additive solution. Transfusion, 2007. 47(2): p. 248-50.

85. Fitzpatrick, C.M., et al., Resuscitation with a blood substitute causes vasoconstriction without nitric oxide scavenging in a model of arterial hemorrhage. J Am Coll Surg, 2004. 199(5): p. 693-701.

86. Gurney, J., et al., A hemoglobin based oxygen carrier, bovine polymerized hemoglobin (HBOC-201) versus Hetastarch (HEX) in an uncontrolled liver injury hemorrhagic shock swine model with delayed evacuation. J Trauma, 2004. 57(4): p. 726-38.

20

87. Fitzpatrick, C.M., et al., Prolonged low-volume resuscitation with HBOC-201 in a large-animal survival model of controlled hemorrhage. J Trauma, 2005. 59(2): p. 273-81; discussion 281-3.

88. Johnson, T., et al., Bovine polymerized hemoglobin (hemoglobin-based oxygen carrier-201) resuscitation in three swine models of hemorrhagic shock with militarily relevant delayed evacuation--effects on histopathology and organ function. Crit Care Med, 2006. 34(5): p. 1464-74.

89. Sprung, J., et al., The use of bovine hemoglobin glutamer-250 (Hemopure) in surgical patients: results of a multicenter, randomized, single-blinded trial. Anesth Analg, 2002. 94(4): p. 799-808, table of contents.

90. Gonzalez, P., et al., A phase I/II study of polymerized bovine hemoglobin in adult patients with sickle cell disease not in crisis at the time of study. J Investig Med, 1997. 45(5): p. 258-64.

91. Kasper, S.M., et al., Effects of a hemoglobin-based oxygen carrier (HBOC-201) on hemodynamics and oxygen transport in patients undergoing preoperative hemodilution for elective abdominal aortic surgery. Anesth Analg, 1996. 83(5): p. 921-7.

92. Johnson, J.L., et al., Resuscitation with a blood substitute abrogates pathologic postinjury neutrophil cytotoxic function. J Trauma, 2001. 50(3): p. 449-55; discussion 456.

93. Gould, S.A., et al., The first randomized trial of human polymerized hemoglobin as a blood substitute in acute trauma and emergent surgery. J Am Coll Surg, 1998. 187(2): p. 113-20; discussion 120-2.

94. Schortgen, F., et al., Hemodynamic tolerance of intermittent hemodialysis in critically ill patients: usefulness of practice guidelines. Am J Respir Crit Care Med, 2000. 162(1): p. 197-202.

95. Paganini, E.P., et al., The effect of sodium and ultrafiltration modelling on plasma volume changes and haemodynamic stability in intensive care patients receiving haemodialysis for acute renal failure: a prospective, stratified, randomized, cross-over study. Nephrol Dial Transplant, 1996. 11 Suppl 8: p. 32-7.

96. Leblanc, M., et al., Bicarbonate dialysate for continuous renal replacement therapy in intensive care unit patients with acute renal failure. Am J Kidney Dis, 1995. 26(6): p. 910-7.

97. Davenport, A., Audit of the effect of dialysate sodium concentration on inter-dialytic weight gains and blood pressure control in chronic haemodialysis patients. Nephron Clin Pract, 2006. 104(3): p. c120-5.

98. Di Filippo, S., et al., Sodium removal during pre-dilution haemofiltration. Nephrol Dial Transplant, 2003. 18 Suppl 7: p. vii31-6; discussion vii57-8.

99. Uchino, S., et al., Solute mass balance during isovolaemic high volume haemofiltration. Intensive Care Med, 2003. 29(9): p. 1541-6.

100. Santoro, A., Cardiovascular dialysis instability and convective therapies. Hemodial Int, 2006. 10 Suppl 1: p. S51-5.

101. Davenport, A., Replacement and dialysate fluids for patients with acute renal failure treated by continuous veno-venous haemofiltration and/or haemodiafiltration. Contrib Nephrol, 2004. 144: p. 317-28.

102. Cariou, A., C. Vinsonneau, and J.F. Dhainaut, Adjunctive therapies in sepsis: an evidence-based review. Crit Care Med, 2004. 32(11 Suppl): p. S562-70.

103. Lewis, K.S., et al., Intensive insulin therapy for critically ill patients. Ann Pharmacother, 2004. 38(7-8): p. 1243-51.

104. Dellinger, R.P., et al., Surviving Sepsis Campaign guidelines for management of severe sepsis and septic shock. Intensive Care Med, 2004. 30(4): p. 536-55.

21

105. Story, D.A., C. Ronco, and R. Bellomo, Trace element and vitamin concentrations and losses in critically ill patients treated with continuous venovenous hemofiltration. Crit Care Med, 1999. 27(1): p. 220-3.

106. Lafrance, J.P. and M. Leblanc, Metabolic, electrolytes, and nutritional concerns in critical illness. Crit Care Clin, 2005. 21(2): p. 305-27.

107. Troyanov, S., et al., Phosphate addition to hemodiafiltration solutions during continuous renal replacement therapy. Intensive Care Med, 2004. 30(8): p. 1662-5.

108. Thomas, A.N., et al., Comparison of lactate and bicarbonate buffered haemofiltration fluids: use in critically ill patients. Nephrol Dial Transplant, 1997. 12(6): p. 1212-7.

109. Kierdorf, H., et al., Continuous venovenous hemofiltration in acute renal failure: is a bicarbonate- or lactate-buffered substitution better? Contrib Nephrol, 1995. 116: p. 38-47.

110. Morgera, S., et al., Comparison of a lactate-versus acetate-based hemofiltration replacement fluid in patients with acute renal failure. Ren Fail, 1997. 19(1): p. 155-64.

111. Heering, P., et al., Acid-base balance and substitution fluid during continuous hemofiltration. Kidney Int Suppl, 1999(72): p. S37-40.

112. Zimmerman, D., et al., Continuous veno-venous haemodialysis with a novel bicarbonate dialysis solution: prospective cross-over comparison with a lactate buffered solution. Nephrol Dial Transplant, 1999. 14(10): p. 2387-91.

113. McLean, A.G., et al., Effects of lactate-buffered and lactate-free dialysate in CAVHD patients with and without liver dysfunction. Kidney Int, 2000. 58(4): p. 1765-72.

114. Barenbrock, M., et al., Effects of bicarbonate- and lactate-buffered replacement fluids on cardiovascular outcome in CVVH patients. Kidney Int, 2000. 58(4): p. 1751-7.

115. Veech, R.L., The untoward effects of the anions of dialysis fluids. Kidney Int, 1988. 34(5): p. 587-97.

116. Johnston, R.V., et al., Responding to tragic error: lessons from Foothills Medical Centre. Cmaj, 2004. 170(11): p. 1659-60.

117. Culley, C.M., et al., Implementing a standardized safety procedure for continuous renal replacement therapy solutions. Am J Health Syst Pharm, 2006. 63(8): p. 756-63.

118. Barletta, J.F., et al., Medication errors and patient complications with continuous renal replacement therapy. Pediatr Nephrol, 2006. 21(6): p. 842-5.

22

Authors:

In alphabetic order:

Jorge Cerda, MD. Division of Nephrology, Albany Medical College and Capital District Renal Physicians Email: cerda@nycap.rr.com

Lewis J. Kaplan, MD Department of Surgery; Section of Trauma, Surgical Critical Care and Surgical Emergencies, Yale University School of Medicine Email: Lewis.Kaplan@yale.edu

John A. Kellum, MD. Department of Critical Care Medicine, University of Pittsburgh School of Medicine. Email: kellumja@ccm.upmc.edu

Mitra K. Nadim, MD. Division of Nephrology, Department of Medicine, University of Southern California Keck School of Medicine Email: nadim@usc.edu

Paul M. Palevsky, MD. VA Pittsburgh Healthcare System. University Drive Division, Pittsburgh, PA. Email: palevsky@pitt.edu