Cc10499-hoste.indd

Clinical review: Use of renal replacement therapies in special groups of ICU patients Introduction
Abstract
e use of renal replacement therapy (RRT) in ICU Acute kidney injury (AKI) in ICU patients is typically patients is increasing over the years [1-3]. Th associated with other severe conditions that require may be explained by a higher number of ICU patients special attention when renal replacement therapy with older age and increased comorbidity, as well as by a (RRT) is performed. RRT includes a wide range of decrease of exclusion criteria for RRT, such as in special techniques, each with specifi c characteristics and groups of ICU patients – for example, those with haemo- implications for use in ICU patients. In the present review we discuss a wide range of conditions that RRT encompasses a broad range of techniques (Table 1). can occur in ICU patients who have AKI, and the A distinction can be made based on duration (inter mit- implications this has for RRT. Patients at increased risk tent, continuous), membrane permeability (high fl ux, low for bleeding should be treated without anticoagulation fl ux), diff usion (haemodialysis) or convection (haemo fi l- or with regional citrate anticoagulation. In patients who tra tion) or a combination of these (haemo diafi ltration), are haemodynamically unstable, continuous therapies and equipment used (machine for regular haemodialysis, are most often employed. These therapies allow slow single-pass batch system or machines that are specifi cally removal of volume and guarantee a stable blood pH. In patients with cerebral oedema, continuous therapy is recommended in order to prevent decreased cerebral Examples of continuous techniques include continuous blood fl ow, which will lead to cerebral ischemia. haemodialysis, continuous venovenous haemofi ltration Continuous therapy will also prevent sudden change (CVVH) and continuous venovenous haemodialfi ltration in serum osmolality with aggravation of cerebral (CVVHDF). Intermittent therapies include haemodialysis oedema. Patients with hyponatraemia, as in liver (HD) with varying duration, ranging from short (2 to failure or decompensated heart failure, require extra 4  hours) to long (6 to 12  hours) as in sustained low- attention because a rapid increase of serum sodium ciency daily dialysis (or hybrid therapy, as it alterna- concentration can lead to irreversible brain damage through osmotic myelinolysis. Finally, in patients with performed with a classic dialysis machine, and its dialysis severe lactic acidosis, RRT can be used as a bridging characteristics are intermediate between classic HD and therapy, awaiting correction of the underlying cause. CRRT. Blood fl ow and dialysate fl ows are decreased and Especially in ICU patients who have severe AKI, the treatment time is increased up to 6 to 12  hours per treatment with RRT requires balancing the pros and is treatment allows better haemo dynamic toler- cons of diff erent options and modalities. Exact and ance and some hours per day off -machine, while the specifi c guidelines for RRT in these patients are not dialysis dose is maintained [4]. Intermittent haemodia fi l- available for most clinical situations. In the present tration can be applied as well, at least if online ultrapure article we provide an update on the existing evidence.
Peritoneal dialysis (PD) is very seldom used for the treatment of acute kidney injury (AKI) in ICU patients. Data on the use of this modality are scarce (only 240 adult patients were studied in three randomised studies) and come from developing countries. An initial report on Department of Intensive Care Medicine, ICU, 2-K12C, Ghent University Hospital, PD demonstrated increased mortality of patients Full list of author information is available at the end of the article randomised to this RRT modality in the setting of AKI [5]. However, two other studies (one of which was published twice) found that the outcome of high-volume Table 1. Renal replacement therapy treatment modality options
Modalities
Diff usion or haemodialysis versus convection or haemofi ltration versus a combination or haemodiafi ltration Single-pass batch system versus regular haemodialysis machine versus continuous renal replacement therapy machine No anticoagulation versus heparin (low molecular weight or unfractionated) versus citrate versus less frequently used strategies (for example, prostacyclin or argatroban) and continuous PD was comparable with daily HD or bloodline, binds ionised calcium and hence blocks CVVHDF in AKI [6-8]. An important limita tion of these last studies is that they excluded patients who died on on the dialysate fl ow and/or ultrafi ltration rate. Citrate day  1 after randomisation, thereby rendering a more may enter the systemic circulation and can induce hypo-favourable outcome.
calcaemia, resulting in cardiac problems. Calcium substi- Each of these RRT modalities has specifi c charac teristics tution with a systemic calcium infusion is therefore with implications for their use in specifi c ICU patient groups.
nearly always necessary, especially in CRRT. Th of citrate per minute is greater in dialysis compared with Patients with increased risk for haemorrhage
CVVH. Citrate anticoagulation is therefore theoretically Most modalities of RRT, with the exception of PD, need safer in intermittent haemodialysis compared with CVVH, anticoagulation to prevent clotting in the extracorporeal whereby less citrate is removed. Many diff erent citrate circuit, thereby increasing the circuit life and dose of schemes are used for predilution and postdilution CVVH, RRT and containing the cost for replacement of a new continuous venovenous haemodialysis (CVVHD) and CVVHDF, and we would like the reader to refer to specifi c lants is the increased risk for haemorrhage. Th e most texts on this topic [11,12]. Haemodialysis with citrate can frequently used anticoagulant in ICU patients is unfrac- be performed with calcium-containing dialysate or with ated heparin, followed by low molecular weight calcium-free dialysate, preferred for short and long heparin and regional citrate anticoagulation [9].
sessions, respectively. In continuous modalities, and when using calcium-free dialysate, ionised calcium must be No anticoagulation
measured rigorously and calcium needs to be re-infused In most patients, a 2-hour dialysis session can be per- to prevent hypocalcaemia. Citrate accumulation can be formed without anticoagulation; but in patients with monitored by the total to ionised calcium ratio. When thrombocytopaenia and coagulation disorders, even this is above 2.5, the citrate dose should be decreased and longer sessions up to CRRT can be performed without calcium reinfused. In patients with reduced citrate e use of heparin-coated membranes facilitates metabolism, such as in liver failure and in patients with this session extension. Intermittent fl ushing with saline pre-existing hypocalcaemia and/or hypomagnesaemia, can also postpone clotting. If haemofi ltration is applied, predilution is preferred to prevent haemoconcentration Regional citrate anticoagulation is increasingly used, in the circuit. In addition, in postdilution mode, the and is currently recommended by the Kidney Disease: fi ltration fraction – calculated as effl Improving Global Outcomes consensus group as the fl ow rate – should be below 20 to 25%.
preferred anticoagulant for CRRT both in patients with If no anticoagulation is allowed, catheter locks used to and without increased bleeding risk (level of evidence 2B fi ll up dialysis catheters between dialysis session should and 2C, indicating a suggestion based on low quality of be heparin-free as leakage of heparin through the side evidence) (data not yet published; M Schetz, personal holes is demonstrated. For that purpose, citrate- communi cation). Several studies have demonstrated the containing solutions can be used as a catheter lock.
feasibility of this technique with diff erent protocols. In the absence of contraindications, such as diverti cu- Regional citrate anticoagulation resulted in an increased litis or recent abdominal surgery, PD in theory can be an fi lter life in some studies, less bleeding complications, alternative RRT modality in patients at high risk for and in one study was even associated with increased bleeding.
survival when compared with low molecular weight heparin, although this could not be confi rmed in another Regional anticoagulation with citrate or with
study where unfractionated heparin was the comparator heparin/protamine
Regional anticoagulation with citrate is based on its Although regional anticoagulation was originally binding with calcium. Citrate, infused into the aff erent described with unfractionated heparin and protamine [18,19], its use has decreased in parallel with the increas- that is adjusted for severity of illness (measured by either ing popularity of citrate. Protamine has several side 2.15  – 0.06  × Acute Physiology and Chronic Health eff ects such as anaphylaxis, hypotension, cardiac depres- Evaluation II score, or by 2.06 – 0.03 × Simplifi ed Acute sion, leukopaenia and thrombocytopaenia. Further, there Physiology Score II) and liver function, assessed by the is risk for a rebound anticoagulant eff ect, due to the indocyanine green disappearance rate (–0.35  + 0.08  × shorter half-life of protamine compared with heparin. indocyanine green disappearance rate) [26]. For ICU Regional anticoagulation with heparin–protamine is patients treated with predilution intermittent venovenous therefore no longer recommended [20].
haemodialysis, recommendations are a loading dose of 75  μg/kg followed by a continuous infusion of 0.4 to Other anticoagulation strategies
0.6 μg/kg/minute until 20 minutes before termination of Prostaglandins and the synthetic protease inhibitor RRT [27].
nafamostat mesilate inhibit platelet aggregation and Despite clinical use of argatroban in RRT, it should be adhesion. In CRRT, prostaglandin I and prostaglandin E mentioned that this is an off -label use for this drug.
administered in a fi xed dose resulted in less fi lter clotting Fondaparinux is a synthetic heparin analogue that can and less bleeding complications in a small prospective be used in patients with heparin-induced thrombo cyto- randomised study (n = 50 patients) [21]. Prostaglandin I paenia type II. In the absence of renal function, a single has also been successfully used in patients with combined intravenous dose of 2.5  mg can maintain dialysis circuit AKI and acute liver failure, who were at increased risk for patency provided that low-fl ux membranes are used [28]. bleeding [22]. Prostaglandin induces vasodilation and Owing to its renal clearance, therapeutic anti-factor Xa therefore hypotension, however, and may also lead to activity is still demonstrated 48  hours after adminis-increased intracranial hypertension and decreased tration of 2.5 mg fondaparinux. Removal of fondaparinux cerebral perfusion pressure [23]. Th is enhanced by the use of high-fl ux membranes. In ICU the cost of this anticoagulation strategy hindered its patients, caution is recommended for this anticoagulant widespread introduction and use.
considering its long half-life and the high bleeding risk, and/or the frequent need for surgical intervention or Patients with heparin-induced thrombocytopaenia type II
invasive procedures. Also, heparin-induced thrombo- When heparin-induced thrombocytopaenia type II is cyto paenia is an off -label indication for fondaparinux.
suspected or confi rmed, the administration of un Lepirudin, or recombinant hirudin, is a direct thrombin tion ated heparin and low molecular weight heparin is inhibitor and can also be used in patients with heparin- contraindicated. In that respect, catheter locks, rinsing induced thrombocytopaenia type II. In dialysis patients, a solutions, dialyser membranes, catheters, and so forth, single intravenous dose of 0.08  mg/kg is recommended must also be heparin free. Besides regional citrate anti- [29]. Because of its renal clearance, this dose results in coagulation (or prostacyclin), the following anticoagu- sustained anticoagulation. Dialyser clearance depends on lants could be used in patients with heparin-induced the membrane characteristics. High-fl ux membranes allow thrombocytopaenia type II.
fi ltration of lepirudin, with the highest sieving coeffi Argatroban, a direct thrombin inhibitor, is mainly for polysulfone (0.97) and lesser sieving coeffi hepatically metabolised with a short half-life of polymethylmethacrylate and polyarylethersulfone (0.75 35  minutes in patients with end-stage kidney disease and 0.73, respectively). Low-fl ux membranes do not fi lter (ESKD). Th e activated clotting time and the activated lepirudin [30]. Dependent on the dialysis frequency, the partial thromboplastin time can be used for monitoring. dialysis membrane, and the activated partial Only minor extracorporeal clearance with high-fl ux thromboplastin time, measured before the start of the membranes is demonstrated. Argatroban, due to its short dialysis session, a reduced dose should be used from the half-life, is a safe anticoagulant in patients with renal second dialysis. Because of its prolonged half-life (52 failure without hepatic impairment. In patients with hours), lepirudin is not an anticoagulant of choice in the multiple organ failure, however, one-tenth of the usual dose without an initial bolus is recommended, depending on hepatic function [24].
Patients with severe haemodynamic instability
erent argatroban dosing regimens have been Haemodynamically unstable patients should be treated described for diff erent RRT modalities. A proposed dose carefully, which can be achieved either by CVVH(D), in chronic haemodialysis patients is a 250  μg/kg bolus ciency daily dialysis or continuous HD. followed by continuous infusion of 2 μg/kg/minute [25]. In many haemodynamically unstable patients, HD can be In ICU patients treated with CRRT, it is recommended to performed when specifi c precautions are taken [31,32]. administer a loading dose of 100  μg/kg argatroban Th ese precau tions include: less aggressive ultrafi ltration, followed by a maintenance infusion rate (μg/kg/minute) increasing the treatment time in CRRT and daily treatment in intermittent HD, eventually using blood considered as an alternative. Concerns on this modality volume measure ments to guide ultrafi ltration; increasing cacy and the eff ects on intra-abdominal dialysate sodium and calcium concentrations to respec- e intra-abdominal pressure may increase tively 145 mmol/l and 1.5 mmol/l; adapting the dialysate secon dary to infusion of PD fl uid in the abdominal cavity, temperature to obtain isothermic dialysis; connecting which may have diverse eff ects on intracranial pressure aff erent and eff erent bloodlines simultaneously at the and cerebral perfusion [40,41]. Increased abdominal start of the procedure; using low blood fl ow (<150  ml/ pressure may translate into increased intracranial minute) and low dialysate fl ows; using biocompatible pressure. Further, intra-abdominal hypertension may also membranes; and using (ultra)pure water [31].
decrease systemic preload and increase afterload, leading An argument in favour of CRRT in haemodynamically to lower blood pressure and decreased cerebral perfusion unstable patients is that solute control is more constant. is factor may be of particular benefi t in patients with severe lactic acidosis, where RRT may help to stabilise Patients with hyponatraemia
the haemodynamic status by correction of blood pH. In When chronic hyponatraemia is corrected rapidly, patients with severe lactic acidosis, intermittent therapy patients may develop osmotic demyelination syndrome will only off er a temporal improvement of blood pH – [42,43], a condition with most irreversible brain damage. between treatments there will be recurrent acidosis, with erefore, in asymptomatic patients with chronic its untoward haemodynamic consequences.
hyponatraemia, the sodium concentration should be In summary, patients with severe haemodynamic increased slowly, with a maximum increase of 10 to instability are best treated with CRRT in order to allow 12  mmol/l during the fi rst 24  hours and of 18  mmol/l removal of extravascular fl uid and to maintain a stable during the fi rst 48  hours of treatment [44]. High serum urea concentration may protect the brain against the A more elaborate discussion on the choice between development of osmotic demyelination [45,46]. An CRRT and HD is presented in another publication in this explanation for this may be the slow diff usion of urea over the blood–brain barrier. Dialysis will decrease serum urea rapidly, but urea that has accumulated in Patients with intracranial hypertension or cerebral
brain cells will decrease only slowly. Th create a blood–brain gradient of urea, and will lead to Patients with cerebral oedema or intracranial hyper ten- accumulation of water in brain cells and a slower decline sion have decreased or absent autoregulation of cerebral of cell volume, which is the mechanism that leads to blood fl ow. A decrease in systemic blood pressure, as may occur during RRT, will therefore lead to decreased CVVH is recom mended in these patients.
cerebral blood fl ow and to cerebral ischaemia, which will Some authors have recommended the use of replace- consequently lead to more oedema [34]. Continuous ment fl uid with reduced sodium concentration. Th arteriovenous haemofi ltration has been proven to better be established by adding sterile water to the replacement maintain cerebral blood fl ow in patients with acute liver fl uid bag [47]. Diluting replacement fl uid will also result failure and cerebral oedema compared with intermittent in decreased potassium and bicarbonate concentrations, ese fi ndings have been extrapolated to and therefore may induce hypokalaemia and acidosis. ment recommendations for patients with other Th ese patients therefore need frequent follow-up of causes of cerebral oedema and to newer modalities of sodium and potassium concentrations and blood pH. During treatment, increasing plasma sodium concen- trations may require less diluted replacement fl uid; that this therapy will seldom be complicated with acute patients may also have need for additional potassium or and important tonicity changes of the systemic bicarbonate infusion. Because this procedure is poten tially circulation. Because intermittent HD is a very effi error prone, and may lead to contamination of sterile RRT modality, these changes may occur after a session of replacement fl uid, an alternative strategy is intravenous HD. A decrease in serum osmolality will subsequently infusion of hypotonic fl uid (for example, 5% glucose).
lead to water uptake by the cells, and to development of cellular oedema [39]. Intracranial pressure may therefore Patients with high serum urea concentration
When serum urea is high (typically >175 mg/dl) patients If a patient is also at increased risk for intracranial are at risk for developing dialysis disequilibrium syn- haemor rhage, such as after traumatic brain injury, RRT drome, a neurological condition characterised by nausea should be administered without anticoagulation or with e exact mechanism of dialysis disequili- brium is uncertain. A sudden decrease of serum osmo lality and slower decrease of (brain) cell osmolality A small prospective randomised study suggests that with resultant increase of cellular water content is the treatment with artifi cial liver support may improve most probable cause. A change in intracellular pH and accu mulation of organic osmolites are also mentioned as powered (n  =  13 patients), however, and the control a possible cause for this condition. Preventive measures popu lation was not treated with vasopressin analogues, include initiation RRT with ultrafi ltration followed by which is currently the standard of care. A large pros-dialysis, which will increase plasma osmolarity and so pective randomised study comparing the standard of care prevent development of cerebral oedema and dialysis for liver dialysis with that of the Prometheus liver dialysis disequilibrium. Decreasing the dose of dialysis will also system (Fresenius Medical Care, Bad Homburg, Germany) prevent important changes of urea concentration, and so in patients with acute on chronic liver failure recently found in subanalysis that patients with hepatorenal shortening the fi rst RRT session when intermittent syndrome treated with Prometheus liver dialysis had modalities are used (2  hours), decreasing blood fl ow (<200 ml/minute), using a small less effi as an abstract, so we can only discuss preliminary results.
using low-dose CVVH (for example, ultrafi ltration rate ere are no data that support the use of one specifi c <20 ml/kg/hour). A urea reduction ratio of 0.4 to 0.45 or RRT modality above another for hepatorenal syndrome, a urea clearance over time of the dialysis session although generally less aggressive modalities such as corrected for volume of distribution (Kt/V) of 0.6 to 0.7 has been proposed. Administration of osmotic agents used most.
such as mannitol (1  g/kg intravenously per dialysis session) may also be of help. Finally, an increased dialysate sodium concentration (143 to 146  mmol/l) is Coagulation abnormalities
also recommended in patients at risk [48-50].
Patients with liver failure have decreased production of coagulation factors II, V, VII, IX, X and XI, thrombo- Patients with acute or acute on chronic liver failure
cytopaenia and decreased thrombocyte function [55,56]. ere are several aspects that deserve special attention in is will lead to decreased coagulation. Many patients patients with liver failure and AKI.
with severe liver failure can therefore undergo RRT without anticoagulation. However, decreased production Encephalopathy and cerebral oedema
of protein C, protein S and antithrombin III, increased Patients with acute liver failure and encephalopathy are concentrations of factor VIII, von Willebrand factor and very likely to have cerebral oedema, especially when the heparin cofactor II, and a decreased concentration of time between occurrence of jaundice and encephalopathy plasminogen will lead to increased coagulation, despite is short; for example, as in fulminant liver failure, defi ned abnormal coagulation tests. Coagulation tests can there- by a time delay between icterus and encephalopathy of fore be misleading and, despite abnormal tests, fi lter clotting may occur. Monitoring the coagulation status risk for increased intracranial pressure and brain stem with func tional haemostasis monitoring devices such as herniation. One should therefore apply the principles for thrombo elastography or Sonoclot (Sienco Inc., Arvada, prevention of deterioration of intracranial pressure as Colorado, USA), which measure the individual contri-described above. In summary, CRRT with special bution of coagulation by platelets and coagulation attention for haemo dynamic stability and maintenance of factors, and fi brinolysis may be of help to better under- cerebral perfusion pressure is warranted.
stand the coagulation status of these patients.
Hyponatraemia
Patients with liver failure often experience chronic Patients with severe lactic acidosis
hyponatraemia. Hence, special attention should be given
e treatment for lactic acidosis should be aimed at to a slow increase of serum sodium, as discussed above.
correcting the cause. However, as severe acidosis may in itself lead to profound systemic hypotension, RRT is Hepatorenal syndrome
sometimes used as a bridge until correction of the Patients with acute and acute on chronic liver failure may underlying cause, especially in patents who also have AKI develop AKI as a consequence of hepatorenal syndrome. [57]. RRT may correct blood pH, by removing lactic acid Especially in patients with acute on chronic liver failure and through administration of bicarbonate from the this is an end stage of the process of water retention and dialysate. Con tinuous modalities are preferred to prevent increased catecholamine stress, characterised by hypo- dialytic rebound, and dialysis is more effi removing small molecules as lactate [58,59].
Patients with congestive heart failure
fomepizole. At present, the most frequently encountered Patients with decompensated heart failure and diuretic intoxications treated with RRT are caused by lithium, t from isolated methanol, ethylene glycol and iodine. Rare intoxications ultrafi ltration [60,61]. When these patients are haemo- that should in special indications be treated with haemo- dynamically unstable, the ultrafi ltration rate should be dialysis include valproic acid, isoniazid and metformin limited by increasing the treatment time. Th e mainstay for treatment of salicylate intoxication can be achieved by CRRT or hybrid therapy. PD is also remains alkalinisation of blood and urine. In patients used for fl uid removal in patients with congestive heart with severe salicylate intoxication and in patients uid removal is less presenting with severe fl uid overload and/or pulmonary predictable com pared with extracorporeal ultrafi ltration or cerebral oedema, however, HD should be used to eff ectively remove salicylate from blood [68,69].
Special attention should be given to patients with With the introduction of high-fl ux membranes and congestive heart failure and (chronic) hyponatraemia (see haemo diafi ltration, haemoperfusion with charcoal cart- the discussion on hyponatraemia above).
e lack of residual endogenous clearance, sometimes Patients with burn injury
caused by the intoxication itself, will accelerate the On average, 3% of patients with severe burn injury are decision to start dialysis. Examples include AKI in treated with RRT for AKI [64]. Patients with burn injury patients with lithium, iodine or ethylene glycol intoxi- and AKI often have large wounds and repetitive surgical cations. If the molecular weight of a toxin is low and interventions with increased risk for bleeding.
protein binding is limited, dialyser clearance is expected Another issue in burn patients is accumulation of to be high. Th iodine. Burn patients treated with topical povidone– body clearance depends on the compartmental behaviour. iodine may have elevated iodine concentrations, secon- If the distribution is multicompartmental, with slow dary to increased absorption in combination with equili bration between the diff erent compart ments, reb-hampered renal excretion. Th is can lead to metabolic ound can ensue. In these cases, long dialysis times are acidosis, AKI, and heart conduction abnormalities, necessary. Because of this latter reason and because of eventually leading to heart block. Th the availability of alternative treatment (digoxin-specifi c of iodine is 253; hence the dialyser clearance during high- Fab fragments), digoxin intoxications are seldom treated fl ux dialysis is comparable with that of small solutes such as urea. Intercompartmental clearance is low, however, so Our preference is to apply haemodialysis (mostly high long treatment times are mandatory [65]. Protracted fl ux) with high blood fl ow (250 to 300  ml/minute) and high-fl ux haemodialysis or haemodiafi ltration, with high high dialysate fl ow (500 ml/minute) and a long treatment blood and dialysate fl ow, is therefore the treatment of time or even continuously in order to optimise both dialyser and body clearance. For larger solutes, protracted or continuous online haemodiafi ltration is more effi Patients with accidental hypothermia
CVVH or CVVHD results in much less dialyser clearance Consensus exists about cardiopulmonary bypass as the because of absent or limited dialysate fl ow. Patients with treatment of choice in cases of severe accidental hypo- lithium intoxication without renal failure should thermia with cardiac arrest. In settings where cardio- preferably not be treated with a single-pass batch system, pulmonary bypass is not available, or in haemo- as premature mixing of dialysate has been documented in dynamically stable patients, HD is a valuable alternative the absence of uraemic solutes [70]. In these patients, [66]. Setting dialysate fl ow and temperature to their haemodialysis with a conventional haemodialysis maximum optimises the effi ciency of warming. Blood machine is recommended.
fl ow and the membrane surface should also be optimised. HD can be started without exogenous anticoagulation Chronic haemodialysis patients admitted to
because hypothermia is associated with coagulopathy.
the ICU
Chronic haemodialysis patients represent a minority of
Patients with intoxications
ICU patients. In a tertiary care centre in the USA, 3.7% of e use of RRT in the treatment of intoxicated patients ICU patients were chronic haemodialysis patients [71]. has decreased over recent years. Reasons for this decline Other studies found that 11 to 12% of all patients treated are diverse. Some intoxications, such as with paraquat with RRT in the ICU are chronic dialysis patients [72,73]. and theophyllin, are nowadays less frequent in the Th e main reasons for ICU admission are sepsis and western world. Other intoxications, such as with cardiovascular complications, which are com parable with methanol and ethylene glycol, are often treated with AKI patients [72,73]. Instead of the regular outpatient dialysis schedule of 4 hours of dialysis three times weekly, high-volume haemofi ltration – 6 to 8 hours of CVVH at RRT during the ICU stay should be adapted towards the 85  ml/kg/hour followed by 16 to 18  hours of CVVH at current needs of the patient. Daily, protracted dialysis 35  ml/kg/hour [77], and ultrafi ltration of 35  l during a can easily be performed through the arteriovenous 4-hour period followed by conventional CVVH [78] – fi stula. Staff should be extremely vigilant in keeping this was associated with better outcomes than compared with ey should therefore be historic controls [77,78]. In the High Volume on Intensive advised against using the arm with the arteriovenous Care (IVOIRE) study, septic study patients were random-fi stula for blood sampling, arterial line insertion or non- ised to high-volume haemofi ltration (70  ml/kg/hour for invasive blood pressure measurement. Needling of this 96  hours) or to standard-dose CVVH (35  ml/kg/hour). arteriovenous fi stula should only be performed by an Th e study was recently stopped after interim analysis, but experienced dialysis nurse. A temporary dialysis catheter was not yet published at the time of this review. Th is preferred when a continuous technique is applied.
results will inform us on the exact place of this technique for treatment of sepsis patients.
Renal replacement therapy during surgery
Occasionally, haemodialysis is performed during surgery.
Polymyxin-B haemoperfusion
Because liver transplantation in patients with renal Polymyxin-B bound to a membrane is a compound that impair ment can be associated with severe hyperkalaemia, ciently binds endotoxin, a component released from especially during reperfusion, intraoperative dialysis the cell membrane of Gram-negative bacteria. Endotoxin should be considered. Unstable patients with acidosis is a crucial component in the infl ammatory cascade and persistent electrolyte disturbances may also benefi t ing Gram-negative infection. Several smaller from a continuation of the dialysis in the operation room. studies that were bundled in a meta-analysis and a recent Special attention should be paid to the temperature of the small prospective randomised study in patients with intra- dialysate, to avoid cooling of the patient. If no water abdominal sepsis demonstrated that binding endotoxin treatment is available in the operation room, haemo- with polymyxin-B haemoperfusion may improve dialysis can be performed with a single-pass batch haemodynamics and organ function at 72 hours, and may system. Anticoagulation is mostly contraindicated, or in improve short-term (28-day) mortality in sepsis patients the case of liver transplantation is not needed. If [79,80]. Th ese results are promising, but the dataset is necessary, a heparin-coated membrane can be used [74]. still too small to adopt this therapy in routine practice.
During liver transplantation, citrate may accumulate when regional citrate anticoagulation is used. Th e citrate Coupled plasma fi ltration and absorption combined with
load is often already elevated in these patients due to the haemodialysis.
adminis tration of fresh frozen plasma and packed cells. Coupled plasma fi ltration and absorption also targets We therefore do not recommend citrate anticoagulation diminishing the infl ammatory process, by removing infl ammatory mediators through absorption. For this technique, blood is fi ltered by a plasma fi lter. Plasma that Extracorporeal therapy in sepsis
is produced by the plasma fi lter is run through an Continuous venovenous haemofi ltration
Two prospective randomised studies evaluated the use of subsequently fi ltered by a high-permeable polysulfone standard CVVH compared with the standard of care in haemofi lter. Results from a small pilot crossover study in 10 patients with severe septic shock demonstrated that infl ammatory mediators were removed at a constant rate in the CVVH intervention group, and that the infl am ma- tory process could therefore be halted. Neither study compared with CVVHDF [81]. Responsiveness of white could demonstrate that CVVH led to decreased serum blood cells to stimulation with lipopolysaccharide was concentrations of mediators, or to improvement of organ also normalised after treatment with coupled plasma dysfunction. In fact, the most recent study was fi ltration and absorption. A prospective randomised discontinued after an interim analysis on 76 patients study on this technique was concluded in Italy recently, demonstrated more organ dysfunction and more severe but the results have not yet been published.
organ dysfunction in patients randomised to CVVH compared with the standard of care [76].
Conclusions
Especially in ICU patients who have severe AKI,
High-volume haemofi ltration
treatment with RRT requires balancing the pros and cons Observational data by Ratanarat and colleagues and by of diff erent options and modalities. Special groups of Honore and colleagues demonstrated that short-term ICU patients that deserve extra consideration are, for example, patients with coagulation abnormalities or with 15. Kutsogiannis DJ, Gibney RT, Stollery D, Gao J: Regional citrate versus increased risk for bleeding, such as after surgery, patients systemic heparin anticoagulation for continuous renal replacement in critically ill patients. Kidney Int 2005, 67:2361-2367.
with severe haemodynamic instability, liver failure 16. Oudemans-van Straaten HM, Bosman RJ, Koopmans M, van der Voort PH, patients and patients with hyponatraemia. Exact and Wester JP, van der Spoel JI, Dijksman LM, Zandstra DF: Citrate specifi c guidelines for RRT in these patients are not anticoagulation for continuous venovenous hemofi ltration. Crit Care Med 2009, 37:545-552.
available for most clinical situations.
17. Hetzel GR, Schmitz M, Wissing H, Ries W, Schott G, Heering PJ, Isgro F, Kribben Abbreviations
A, Himmele R, Grabensee B: Regional citrate versus systemic heparin for AKI, acute kidney injury; CVVH, continuous venovenous haemofi ltration; anticoagulation in critically ill patients on continuous venovenous CVVHD, continuous venovenous haemodialysis; CVVHDF, continuous haemofi ltration: a prospective randomized multicentre trial. Nephrol Dial venovenous haemodiafi ltration; CRRT, continuous renal replacement therapy; ESKD, end-stage kidney disease; HD, haemodialysis; PD, peritoneal dialysis; RRT, 18. Darby JPJ, Sorensen RJ, O’Brien TF, Teschan PE: Effi monitoring regional heparinization and hemodialysis. N Engl J Med 1960, 262:654-657.
Competing interests
19. Teschan P, Baxter C, O’Brian T, Freyhof J, Hall W: Prophylactic hemodialysis in AD declares that she has no competing interests. EAJH advised both Gambro the treatment of actue renal failure. Ann Intern Med 1960, 53:992-1016.
and Fresenius in an advisary board, and received a fee for this.
20. European Best Practice Guidelines Expert Group on Hemodialysis, European Renal Association: European Best Practice Guidelines. Section V. Chronic Author details
intermittent haemodialysis and prevention of clotting in the extracorporal 1Department of Intensive Care Medicine, ICU, 2-K12C, Ghent University Hospital, system. Nephrol Dial Transplant 2002, 17(Suppl 7):63-71.
De Pintelaan 185, 9000 Gent, Belgium. 2Research Foundation Flanders, Ghent 21. Kozek-Langenecker SA, Spiss CK, Gamsjager T, Domenig C, Zimpfer M: University Hospital, De Pintelaan 185, 9000 Gent, Belgium. 3Nephrology Section, Anticoagulation with prostaglandins and unfractionated heparin during Ghent University Hospital, De Pintelaan 185, 9000 Gent, Belgium.
continuous venovenous haemofi ltration: a randomized controlled trial. Wien Klin Wochenschr 2002, 114:96-101.
22. Davenport A, Will EJ, Davison AM: Comparison of the use of standard heparin and prostacyclin anticoagulation in spontaneous and pump- References
driven extracorporeal circuits in patients with combined acute renal and Hoste EAJ, Schurgers M: Epidemiology of AKI: how big is the problem? hepatic failure. Nephron 1994, 66:431-437.
Crit Care Med 2008, 36:S1-S4.
23. Davenport A, Will EJ, Davison AM: The eff ect of prostacyclin on intracranial Xue JL, Daniels F, Star RA, Kimmel PL, Eggers PW, Molitoris BA, Himmelfarb J, pressure in patients with acute hepatic and renal failure. Clin Nephrol 1991, Collins AJ: Incidence and mortality of acute renal failure in Medicare benefi ciaries, 1992 to 2001. J Am Soc Nephrol 2006, 17:1135-1142.
24. Beiderlinden M, Treschan TA, Gorlinger K, Peters J: Argatroban Waikar SS, Curhan GC, Wald R, McCarthy EP, Chertow GM: Declining anticoagulation in critically ill patients. Ann Pharmacother 2007, 41:749-754.
mortality in patients with acute renal failure, 1988 to 2002. J Am Soc 25. Murray PT, Reddy BV, Grossman EJ, Hammes MS, Trevino S, Ferrell J, Tang I, Hursting MJ, Shamp TR, Swan SK: A prospective comparison of three Kielstein JT, Kretschmer U, Ernst T, Hafer C, Bahr MJ, Haller H, Fliser D: Effi argatroban treatment regimens during hemodialysis in end-stage renal and cardiovascular tolerability of extended dialysis in critically ill patients: disease. Kidney Int 2004, 66:2446-2453.
a randomized controlled study. Am J Kidney Dis 2004, 43:342-349.
26. Link A, Girndt M, Selejan S, Mathes A, Bohm M, Rensing H: Argatroban for Phu NH, Hien TT, Mai NT, Chau TT, Chuong LV, Loc PP, Winearls C, Farrar J, anticoagulation in continuous renal replacement therapy. Crit Care Med White N, Day N: Hemofi ltration and peritoneal dialysis in infection- associated acute renal failure in Vietnam. N Engl J Med 2002, 347:895-902.
27. Sun X, Chen Y, Xiao Q, Wang Y, Zhou J, Ma Z, Xiang J, Chen X: Eff ects of Gabriel DP, Caramori JT, Martim LC, Barretti P, Balbi AL: High volume argatroban as an anticoagulant for intermittent veno-venous peritoneal dialysis vs daily hemodialysis: a randomized, controlled trial in hemofi ltration (IVVH) in patients at high risk of bleeding. Nephrol Dial patients with acute kidney injury. Kidney Int Suppl 2008, 73:S87-S93.
Transplant 2011, 26:2954-2959.
Gabriel DP, Caramori JT, Martin LC, Barretti P, Balbi AL: Continuous peritoneal 28. Sombolos KI, Fragia TK, Gionanlis LC, Veneti PE, Bamichas GI, Fragidis SK, dialysis compared with daily hemodialysis in patients with acute kidney Georgoulis IE, Natse TA: Use of fondaparinux as an anticoagulant during injury. Perit Dial Int 2009, 29(Suppl 2):S62-S71.
hemodialysis: a preliminary study. Int J Clin Pharmacol Ther 2008, George J, Varma S, Kumar S, Thomas J, Gopi S, Pisharody R: Comparative study of continuous venovenous hemodiafi ltration and peritoneal dialysis 29. Vanholder R, Dhondt A: Recombinant hirudin: clinical pharmacology and in critically ill patients with acute kidney injury: a pilot study. Perit Dial Int potential applications in nephrology. BioDrugs 1999, 11:417-429.
30. Benz K, Nauck MA, Bohler J, Fischer KG: Hemofi ltration of recombinant Ricci Z, Ronco C, D’Amico G, De Felice R, Rossi S, Bolgan I, Bonello M, hirudin by diff erent hemodialyzer membranes: implications for clinical Zamperetti N, Petras D, Salvatori G, Dan M, Piccinni P: Practice patterns in the use. Clin J Am Soc Nephrol 2007, 2:470-476.
management of acute renal failure in the critically ill patient: an 31. Schortgen F, Soubrier N, Delclaux C, Thuong M, Girou E, Brun-Buisson C, international survey. Nephrol Dial Transplant 2006, 21:690-696.
Lemaire F, Brochard L: Hemodynamic tolerance of intermittent 10. Pinnick RV, Wiegmann TB, Diederich DA: Regional citrate anticoagulation for hemodialysis in critically ill patients. Usefulness of practice guidelines. hemodialysis in the patient at high risk for bleeding. N Engl J Med 1983, Am J Respir Crit Care Med 2000, 162:197-202.
32. Vinsonneau C, Camus C, Combes A, Costa de Beauregard MA, Klouche K, 11. Davenport A, Tolwani A: Citrate anticoagulation for continuous renal Boulain T, Pallot J-L, Chiche J-D, Taupin P, Landais P: Continuous venovenous replacement therapy (CRRT) in patients with acute kidney injury admitted haemodiafi ltration versus intermittent haemodialysis for acute renal to the intensive care unit. NDT Plus 2009, 2:439-447.
failure in patients with multiple-organ dysfunction syndrome: a 12. Oudemans-van Straaten HM: Citrate anticoagulation for continuous renal multicentre randomised trial. The Lancet 2006, 368:379-385.
replacement therapy in the critically ill. Blood Purif 2010, 29:191-196.
33. Vanholder R, Van Biesen W, Hoste E, Lameire N: Pro/Con debate: Continuous 13. Bagshaw SM, Laupland KB, Boiteau PJ, Godinez-Luna T: Is regional citrate versus intermittent dialysisfor acute kindey injury: a never-ending story superior to systemic heparin anticoagulation for continuous renal yet approaching the fi nish? Crit Care 2011, 15:204.
replacement therapy? A prospective observational study in an adult 34. Davenport A, Will EJ, Losowsky MS: Rebound surges of intracranial pressure regional critical care system. J Crit Care 2005, 20:155-161.
as a consequence of forced ultrafi ltration used to control intracranial 14. Monchi M, Berghmans D, Ledoux D, Canivet JL, Dubois B, Damas P: Citrate vs. pressure in patients with severe hepatorenal failure. Am J Kidney Dis 1989, heparin for anticoagulation in continuous venovenous hemofi ltration: a prospective randomized study. Intensive Care Med 2004, 30:260-265.
35. Davenport A, Will EJ, Davidson AM: Improved cardiovascular stability during continuous modes of renal replacement therapy in critically ill lactic acidosis [abstract]. Critical Care 2008, 12(Suppl 2):P479.
patients with acute hepatic and renal failure. Crit Care Med 1993, 60. Costanzo MR, Saltzberg M, O’Sullivan J, Sobotka P: Early ultrafi ltration in patients with decompensated heart failure and diuretic resistance. J Am 36. Davenport A, Will EJ, Losowsky MS, Swindells S: Continuous arteriovenous Coll Cardiol 2005, 46:2047-2051.
haemofi ltration in patients with hepatic encephalopathy and renal failure 61. Costanzo MR, Ronco C: Extracorporeal fl uid removal in heart failure [case report]. Br Med J (Clin Res Ed) 1987, 295:1028.
patients. Contrib Nephrol 2010, 165:236-243.
37. Stravitz RT, Kramer AH, Davern T, Shaikh AO, Caldwell SH, Mehta RL, Blei AT, 62. Sanchez JE, Ortega T, Rodriguez C, Diaz-Molina B, Martin M, Garcia-Cueto C, Fontana RJ, McGuire BM, Rossaro L, Smith AD, Lee WM: Intensive care of cacy of peritoneal ultrafi ltration in the patients with acute liver failure: recommendations of the U.S. Acute Liver treatment of refractory congestive heart failure. Nephrol Dial Transplant Failure Study Group. Crit Care Med 2007, 35:2498-2508.
Anticoagulation options for patients with heparin-induced 63. Cnossen N, Kooman JP, Konings CJ, van Dantzig JM, van der Sande FM, thrombocytopenia requiring renal support in the intensive care unit. Leunissen K: Peritoneal dialysis in patients with congestive heart failure. Contrib Nephrol 2007, 156:259-266.
Nephrol Dial Transplant 2006, 21(Suppl 2):ii63-ii66.
Practical guidance for dialyzing a hemodialysis patient 64. Brusselaers N, Monstrey S, Colpaert K, Decruyenaere J, Blot SI, Hoste EA: following acute brain injury. Hemodial Int 2008, 12:307-312.
Outcome of acute kidney injury in severe burns: a systematic review and 40. De laet I, Citerio G, Malbrain ML: The infl uence of intraabdominal meta-analysis. Intensive Care Med 2010, 36:915-925.
hypertension on the central nervous system: current insights and clinical 65. Eloot S, Dhondt A, Hoste E, Verstraete A, De Waele J, Colpaert K, Hoeksema H, recommendations, is it all in the head? Acta Clin Belg Suppl 2007, 1:89-97.
Tromp F, Vanholder R: How to remove accumulated iodine in burn-injured 41. Hunter JD, Damani Z: Intra-abdominal hypertension and the abdominal patients. Nephrol Dial Transplant 2010, 25:1614-1620.
compartment syndrome. Anaesthesia 2004, 59:899-907.
66. Caluwe R, Vanholder R, Dhondt A: Hemodialysis as a treatment of severe 42. Huang WY, Weng WC, Peng TI, Ro LS, Yang CW, Chen KH: Central pontine accidental hypothermia. Artif Organs 2010, 34:237-239.
and extrapontine myelinolysis after rapid correction of hyponatremia by 67. Temmerman W, Dhondt A, Vandewoude K: Acute isoniazid intoxication: hemodialysis in a uremic patient. Ren Fail 2007, 29:635-638.
seizures, acidosis and coma. Acta Clin Belg 1999, 54:211-216.
43. Tarhan NC, Agildere AM, Benli US, Ozdemir FN, Aytekin C, Can U: Osmotic 68. Cohen DL, Post J, Ferroggiaro AA, Perrone J, Foster MH: Chronic salicylism demyelination syndrome in end-stage renal disease after recent resulting in noncardiogenic pulmonary edema requiring hemodialysis. hemodialysis: MRI of the brain. AJR Am J Roentgenol 2004, 182:809-816.
Am J Kidney Dis 2000, 36:E20.
44. Verbalis JG, Goldsmith SR, Greenberg A, Schrier RW, Sterns RH: Hyponatremia 69. Glisson JK, Vesa TS, Bowling MR: Current management of salicylate-induced treatment guidelines 2007: expert panel recommendations. Am J Med pulmonary edema. South Med J 2011, 104:225-232.
70. Dhondt A, Verstraete A, Vandewoude K, Segers H, Eloot S, Decruyenaere J, 45. Oo TN, Smith CL, Swan SK: Does uremia protect against the demyelination ciency of the Genius batch hemodialysis system with low associated with correction of hyponatremia during hemodialysis? A case serum solute concentrations: the case of lithium intoxication therapy. Am report and literature review. Semin Dial 2003, 16:68-71.
J Kidney Dis 2005, 46:e95-e99.
46. Soupart A, Penninckx R, Stenuit A, Decaux G: Azotemia (48 h) decreases the 71. Clermont G, Acker CG, Angus DC, Sirio CA, Pinsky MR, Johnson JP: Renal risk of brain damage in rats after correction of chronic hyponatremia. Brain failure in the ICU: comparison of the impact of acute renal failure and end-stage renal disease on ICU outcomes. Kidney Int 2002, 62:986-996.
47. Ostermann M, Dickie H, Tovey L, Treacher D: Management of sodium 72. Uchino S, Morimatsu H, Bellomo R, Silvester W, Cole L: End-stage renal disorders during continuous haemofi ltration. Crit Care 2010, 14:418.
failure patients requiring renal replacement therapy in the intensive Arieff AI: Dialysis disequilibrium syndrome: current concepts on care unit: incidence, clinical features, and outcome. Blood Purif 2003, pathogenesis and prevention. Kidney Int 1994, 45:629-635.
49. Patel N, Dalal P, Panesar M: Dialysis disequilibrium syndrome: a narrative 73. Rocha E, Soares M, Valente C, Nogueira L, Bonomo HJ, Godinho M, Ismael M, review. Semin Dial 2008, 21:493-498.
Valenca RV, Machado JE, Maccariello E: Outcomes of critically ill patients 50. Rodrigo F, Shideman J, McHugh R, Buselmeier T, Kjellstrand C: Osmolality with acute kidney injury and end-stage renal disease requiring renal changes during hemodialysis. Natural history, clinical correlations, and replacement therapy: a case–control study. Nephrol Dial Transplant 2009, infl uence of dialysate glucose and intravenous mannitol. Ann Intern Med 74. Lavaud S, Paris B, Maheut H, Randoux C, Renaux JL, Rieu P, Chanard J: Practical management of acute liver failure in the intensive care Assessment of the heparin-binding AN69 ST hemodialysis membrane: II. unit. Curr Opin Crit Care 2009, 15:163-167.
Clinical studies without heparin administration. ASAIO J 2005, 51:348-351.
52. Gines P, Schrier RW: Renal failure in cirrhosis. N Engl J Med 2009, 75. Cole L, Bellomo R, Hart G, Journois D, Davenport P, Tipping P, Ronco C: A phase II randomized, controlled trial of continuous hemofi ltration in 53. Mitzner SR, Stange J, Klammt S, Risler T, Erley CM, Bader BD, Berger ED, sepsis. Crit Care Med 2002, 30:100-106.
Lauchart W, Peszynski P, Freytag J: Improvement of hepatorenal syndrome 76. Payen D, Mateo J, Cavaillon JM, Fraisse F, Floriot C, Vicaut E: Impact of with extracorporeal albumin dialysis MARS: results of a prospective, continuous venovenous hemofi ltration on organ failure during the early randomized, controlled clinical trial. Liver Transpl 2000, 6:277-286.
phase of severe sepsis: a randomized controlled trial. Crit Care Med 2009, 54. Rifai K, Kribben A, Gerken G, Haag S, Herget-Rosenthal S, Treichel U, Betz C, Sarrazin C, Van Vlierberghe H, Hoste E, Escorsell A, Gines P, Hafer C, 77. Ratanarat R, Brendolan A, Ricci Z, Salvatori G, Nalesso F, Cal M, Cazzavillan S, Schuchmann M, Galle PR, Bernardi M, Caraceni P, Àbeles R, Frieder-Berr Petras D, Bonello M, Bordoni V, Cruz D, Techawathanawanna N, Ronco C: Knotek M, Kozik-Jaromin J: A prospective randomized controlled multi- Pulse high-volume hemofi ltration in critically ill patients: a new approach center study of extracorporeal liver support by fractionated plasma for patients with septic shock. Semin Dial 2006, 19:69-74.
separation and absorption (Prometheus (R)) in patients with acute-on- 78. Honore PM, Jamez J, Wauthier M, Lee PA, Dugernier T, Pirenne B, Hanique G, chronic liver failure (HELIOS study) [abstract]. Hepatology 2010, 52:333A.
Matson JR: Prospective evaluation of short-term, high volume isovolemic Coagulation abnormalities in patients who have liver disease. hemofi ltration on the hemodynamic course and outcome in patients with Clin Liver Dis 2006, 10:665-678, x-xi.
intractable circulatory failure resulting from septic shock. Crit Care Med 56. Wada H, Usui M, Sakuragawa N: Hemostatic abnormalities and liver diseases. Semin Thromb Hemost 2008, 34:772-778.
79. Cruz D, Perazella M, Bellomo R, de Cal M, Polanco N, Corradi V, Lentini P, 57. Peters N, Jay N, Barraud D, Cravoisy A, Nace L, Bollaert PE, Gibot S: Metformin- Nalesso F, Ueno T, Ranieri VM, Ronco C: Eff ectiveness of polymyxin associated lactic acidosis in an intensive care unit. Crit Care 2008, 12:R149.
B-immobilized fi ber column in sepsis: a systematic review. Crit Care 2007, 58. Levraut J, Ciebiera JP, Jambou P, Ichai C, Labib Y, Grimaud D: Eff ect of continuous venovenous hemofi ltration with dialysis on lactate clearance 80. Cruz DN, Antonelli M, Fumagalli R, Foltran F, Brienza N, Donati A, Malcangi V, in critically ill patients. Crit Care Med 1997, 25:58-62.
Petrini F, Volta G, Bobbio Pallavicini FM, Rottoli F, Giunta F, Ronco C: Early use 59. Vuylsteke S, Dhondt A, Danneels C, De Waele J, Blot S, Hoste E: Initiation of of polymyxin B hemoperfusion in abdominal septic shock: the EUPHAS renal replacement therapy in patients with acute kidney injury and severe randomized controlled trial. JAMA 2009, 301:2445-2452.

Source: http://www.gambramedical.ru/assets/files/hotline/08.2013/Obzor%2026.08.13.pdf