Clinical practice guidelines for the provision of renal service in Hong Kong: General Nephrology

About one-third of the acute kidney injury (AKI) burden occurs in the perioperative context, and the incidence of AKI continues to rise. While mortality rates in AKI have improved, the figures remain significant. Crucial factors that determine the prognosis include timing of onset, severity and duration of injury, recovery status and recurrence. AKI is associated with an increased hospital mortality, risk of hospital readmissions and risk of chronic kidney disease (CKD). 1.5–1.9 times baseline OR ≥26.5 μmol/L increase 3.0 times baseline OR Increase to ≥354 μmol/L OR Initiation of dialysis OR In patients <18 years, decrease in estimated glomerular filtration rate (eGFR) to <35 mL/min per 1.73 m2 (Schwartz formula) <0.3 mL/kg/h for ≥24 h OR Anuria for ≥12 h The cause of AKI should try to be determined. (Not Graded) Generally, discontinuation of nephrotoxic agents whenever possible, maintenance of volume status and perfusion pressure, functional haemodynamic monitoring and that of SCr and urine output, avoiding hyperglycaemia and resort to alternatives avoiding radio-contrast procedures should all be considered for patients at high risk of or have developed AKI. Patients with AKI require diagnostic workup, and stage 2/3 patients require change in drug dosing as renal impairment advances and consideration for RRT. It is recommended that patients be stratified for risk of AKI according to their susceptibilities and exposures (R), and managed accordingly in order to reduce such risk. (Not Graded) Exposures that may cause AKI include sepsis, critical illness, circulatory shock, burns, trauma, cardiac surgery (especially with cardiopulmonary bypass), major non-cardiac surgery, nephrotoxic drugs, radio-contrast media, poisonous plants and animals. Even with these exposures, the risk of AKI would vary between different patient groups and in different clinical context. For patients at increased risk for AKI, monitor their SCr and urine output to detect (and stage severity) AKI, at individualized frequency and duration based on patient risk and clinical course. (Not Graded) The use of SCr and urine output would be meaningful when these are monitored regularly at a defined fashion, but clinical practice would usually be dictated by clinical judgment based on the clinical setting and indication, with a tendency for high-risk and critically ill patients being monitored more frequently. The availability of time-dependent biomarkers would be helpful. Evaluate patients 2 months after AKI to look for return to baseline, or development of new CKD or worsening of pre-existing CKD. (R) There is strong association of AKI with subsequent development of CKD and end-stage renal disease (ESRD).10, 11 Prediction tools have been tested to help identify patients with AKI who are at high risks of CKD.10, 12-14 For initial fluid replacement in hypovolemic (not haemorrhagic) patients at risk for, or with, AKI, isotonic crystalloids rather than albumin or starches are recommended. (R) Recent data from multi-centre, open-label trial continues to support this recommendation.15-17 The Acute Dialysis Quality Initiative group recently concluded on the evidence for harm with hetastarch (hydoxyethyl starch) or albumin administration in traumatic brain injury cases.18 While 0.9% saline may result in a chloride-induced tubule-glomerular feedback–mediated vasoconstriction and metabolic acidosis compared with more physiologically balanced and buffered crystalloids, supporting evidence is mainly derived from post hoc analyses of large patient data sets, rather than prospective, controlled trials.19 Vasopressors are recommended in volume-resuscitated patients with vasomotor shock at risk for, or with, AKI. (R) Current clinical data are insufficient to conclude on the best vasoactive agent in preventing AKI.20-23 Protocoled therapies with specific physiological goals (haemodynamic and tissue oxygenation targets) are suggested to reduce perioperative AKI in high-risk patients. (D) The results from few multi-centre trials that looked at protocoled resuscitation with or without an oximetric central venous oxygen saturation monitoring (early goal-directed therapy) failed to suggest survival benefit in patients with septic shock who have received timely antibiotics and usual fluid resuscitation.24-26 In high-risk patients in the perioperative setting, while studies using different protocoled therapies with specific physiological goals (haemodynamic and tissue oxygenation targets)27 have been shown to significantly reduce post-operative AKI, there is no evidence to support the identification of the best regime. In patients with stress hyperglycaemia, the target plasma glucose for insulin therapy is suggested to be 6.1–8.3 mmol/L, and to avoid hypoglycaemia. (D) It is important to avoid the danger of potentially serious hypoglycaemia. While the target blood glucose between 6.1 and 8.3 mmol/L have not been directly studied in randomized controlled trial (RCT), they are interpolated from the comparisons tested in the trials.28, 29 An energy intake of 20–30 kcal/kg per day is suggested, and enteral feeding is preferred. (D) Though the optimal energy intake has not been well determined, data from both retrospective and randomized trials in AKI patients support a total energy intake of at least 20, but not to exceed 25–30 kcal/kg per day.30, 31 Studies have suggested that enteral feeding is associated with improved outcome and survival in ICU patients.32-34 Protein intake is suggested to be 0.8–1.0 g/kg per day in non-catabolic patients not on dialysis, 1.0–1.5 g/kg per day in dialysis patients and up to 1.7 g/kg per day in hypercatabolic patients or patients on continuous renal replacement therapy (CRRT). (D) Since malnutrition is associated with increased mortality in critically ill patients, nutritional protein administration is not recommended to be restricted as a means to attenuate the rise in serum urea level when renal function declines. On the other hand, there is little evidence that hypercatabolism can be overcome by increasing protein intake to higher than physiological levels.35, 36 During RRT, nutritional support should include replacement for the losses during the procedures, especially with modalities associated with high filtration rates, including CRRT, sustained low efficiency dialysis (SLED) or peritoneal dialysis (PD).37 The use of diuretics to prevent (R) or treat (D) AKI is not recommended, except in the presence of volume overload or acute decompensated heart failure (ADHF). Though diuretics theoretically may reduce renal tubular oxygen consumption and attenuate intra-tubular obstruction, the benefits of its use has remained contentious in AKI prevention. In the setting of cardiac surgery, a double-blind RCT has demonstrated a higher rate of AKI being associated with the use of furosemide.38, 39 Studies have suggested that diuretics given to treat post-operative AKI is best avoided.40, 41 This also applies to patients on CRRT.42, 43 In patients with ADHF, there were no significant differences in symptom relief or renal safety when diuretic therapy was administered by bolus compared with continuous infusion.44 Although high-dose therapy (2.5 times the usual oral dose given intravenously) improved diuresis, with a trend towards improved symptom relief, there was associated increase in renal adverse events compared with low-dose therapy (usual oral dose given intravenously).45 Individual careful clinical judgment is needed. The use of low-dose dopamine to prevent or treat AKI is not recommended. (R) The early positive results in the use of dopamine for renal protection in the critically ill have been opposed by quality RCT trial and meta-analysis.45-47 The use of either dopamine or synthetic natriuretic peptide on top of standard therapy in ADHF is also not associated with enhanced pulmonary decongestion or improved renal function.48 This is also true in the setting of AKI after cardiac surgery.49, 50 The use of fenoldopam to prevent or treat AKI is not suggested. (D) Fenoldopam mesylate is a dopamine type-1 receptor agonist with similar haemodynamic renal effects as low-dose dopamine, without systemic α- or β-adrenergic stimulation. While early data suggests a lower incidence of AKI was associated with the use of fenoldopam, data from adequately powered multi-centre trials with clinically significant endpoints do not support recommending fenoldopam to either prevent or treat AKI,51, 52 noting in particular the concern of the associated hypotension. The use of atrial natriuretic peptide to prevent or treat AKI is not suggested. (D) Studies demonstrating benefits of recombinant human atrial natriuretic peptide in reduced need for dialysis and improved dialysis-free survival after cardiac surgery53 or solid organ transplantation tend to be underpowered, its routine use for the prevention or treatment of AKI cannot be recommended, the concern of hypotension aside.53-56 The use of off-pump coronary artery bypass graft in order to prevent post-operative AKI is not suggested. (D) While the conclusion from systematic review and meta-analysis of studies looking at off-pump surgery compared with on-pump surgery in cardiopulmonary bypass suggest a 43% reduction in the risk of post-operative AKI, it has been cautioned that the definitions of AKI were variable and that the RCT were associated with lower than data is to a for that include fluid and clinical context. (Not Graded) The optimal timing of of has to be determined, and is a clinical In results have been with different definitions of early of dialysis being while for for patients with CKD may not be for AKI, the best timing for may be RCT when there are the of the target patients and the of therapy at the There is a trend to in the more critically The use of diuretics to kidney function recovery is not suggested. (D) RCT has the of by continuous at a rate of on top of continuous While patients a significantly increased volume and compared to the there were no differences in the need for or renal recovery during ICU or hospital The use of during RRT, except with is (R) Studies have shown that treatment increased and increased and The of include low administration and monitoring and availability of include and risk of from low and increased risk of from studies in chronic with concluded that both are in of and in Recent concluded that the risk of compared with improved especially in patients with increased that for monitoring are in to the risk of there are studies improved renal recovery and hospital survival associated with the use of Patients with failure may have the in by low and high metabolic continuous and are therapies in AKI patients. (Not Graded) and should be as modalities of RRT, as by the of data either in of hospital or ICU mortality, of and renal recovery in and would the clinical between both modalities would be based on the clinical status of the such as and clinical of the patients such as of diagnostic or of is the suggested for both patients and patients with (D) CRRT, rather than in a significantly higher and a lower of is well in the is and may have a when other of are not but data from studies are in patients with brain injury or brain may perfusion as a result of hypotension or by and volume and of per for is recommended (R), and the should be monitored RCT looking at the dose of in AKI failed to in mortality or renal recovery when the dialysis dose was either by a higher or by a serum urea level It is recommended to of for in AKI. dialysis be in order to fluid in hypercatabolic or in the presence of or fluid has been shown to be an risk for mortality in AKI An volume of mL/kg/h for is recommended. (R) The multi-centre the of Acute and the of not that a more therapy with was associated with improved patient survival or recovery of renal studies in have shown that usually of the as a result of such as blood reduced efficiency with or It is recommended to a higher dose at 25–30 in order to the recommended In patients to have look for other of AKI (Not Graded) The incidence may be as high as in patients with pre-existing renal impairment or with other risk factors such as heart and administration of nephrotoxic Patients who have a risk for or of SCr is looking for the risk for and for renal impairment in patients for a that or administration of and in patients at increased risk. (Not Graded) While the suggested that the risk of clinically important when the baseline SCr is mmol/L in and mmol/L in to an mL/min per 1.73 there are data that patients with SCr mmol/L are the group at a SCr is not a or a for urine protein may be for pre-existing kidney The risk of to be after compared to venous administration of media, with an incidence of after procedures that risk factors include heart volume the use of haemodynamic use of nephrotoxic drugs, metabolic and large volume or high of the The use of either or media, but not (R), at the dose (Not Graded) in patients at increased risk is recommended. should be for h in patients without risk factors for and for h in with risk AKI after should be the SCr level has to fluid clinically either isotonic or is recommended for patients at increased risk. (R) the of RCT that directly the of in the prevention of AKI, comparisons in trials looking at different when with suggest a large benefit from The would be patients with fluid Even there is no evidence from the to the of the optimal rate and duration of fluid administration in a urine output in the h after the has been associated with reduced rates of crystalloids given would not be in the for this target urine rate an rate of 1.0–1.5 mL/kg/h for h and h after the 0.9% saline has been to be to saline in prevention. For the between and suggests that studies tend to the of there were no effects in of the risk for dialysis, heart failure and total The burden and harm from in the at the or by may the use of this fluid at The multi-centre, double-blind controlled trial isotonic with isotonic with for the has to The fluid to the volume status of patients a when compared with the usual fluid of duration and such trial the in high-risk patients cardiac The other tested include fluid administration with urine or volume suggest using oral with in patients at increased risk. (D) The of on the incidence of is with of the studies being in in the studies it is not that but not all a There is at no evidence that either oral or can including mortality and need for after administration to patients at risk for The benefit of is not or oral has a safety and is studies of with administration have shown reduced incidence of but not that of renal when compared to the of with The use of as an to volume is not suggested. (Not Graded) Although RCT are associated with significant including the on they the of for prevention of including the of CKD patients, the results from Studies to patients with stage would be to support a The use of or for in patients at increased risk is not suggested. (D) While there are data on the use of in the incidence of to either increased or tendency towards being a meta-analysis of studies using blood concluded that such not decrease the incidence of a of AKI in chronic disease by systemic and haemodynamic without kidney The of a rise in SCr and of other of AKI such as and renal There are data that early of treatment with therapy with albumin patient survival and renal The defined CKD as either kidney or mL/min per 1.73 m2 for is defined as or the presence of of including in blood or urine or to the kidney can be the large blood or The of kidney a to the of the kidney and in association with other clinical the of kidney In CKD as of kidney or for more than with for CKD is based on and The of CKD were as The in CKD were as The was also by to treatment as CKD of with or without a kidney including both CKD and CKD CKD of with or without a kidney CKD CKD CKD of with a kidney and of the are risk factors with the that that are with and of CKD in heart with CKD being more in than in and reduced of in to SCr in is and the should be (R) The of in more blood urea and serum than was to be for routine clinical and an was and in clinical The was to in CKD patients, as is when to patients with kidney function than mL/min per 1.73 recommended the to overcome the of the was derived from an and both and with CKD. Even it be in that the of the in and of the of the include the use of an or serum and also to for in the group developed an that both SCr and in and the of the in the the and the and Since is not in the that recommended using the for CKD be and in at increased risk of including with and disease 2 (D) The tools should include blood urine for protein and and of should be for specific for urine in or in Patients with CKD should be to a for and the clinical cannot be the of the patient cannot be or the recommended treatment cannot be (R) The for vary with practice and In the consideration of The of of CKD is at a of factors to be associated with have been shown to and CKD or risk factors may and directly CKD and include and of the In of other metabolic such as blood and acidosis are also For and patients with than h the suggested target is For and patients with h the suggested target is (D) evidence is but not that a blood target of than clinical more than a target of than in with 50 with CKD be with a or (D) Recent studies have on in patients with CKD. Although evidence is these studies suggest that protection during early rather than For by mmol/L not kidney disease in a of patients with CKD in a large randomized using analyses of the no significant of on the rate of change in more to have more effects than high-dose in patients with who have renal reduction should be during CKD when plasma exceed or (D) The major can be associated with high and high the of and and low and low or and The that to disease to at of CKD. of and reduced of and reduced have been patients with mL/min or or are in (D) of with oral or or can reduce the of and may be to treat clinical of such as improved of disease as by of serum agents can be considered when or its is or (Not Graded) There is data on the use of with of and may reduce rates of and The use of may be associated with development of disease when are There is insufficient evidence to support the routine use of agents in the of CKD in either or (Not Graded) There is also insufficient evidence to suggest that is and more than treatment without in of the of adverse and The recommended of protein intake at g/kg per day in patients with mL/min per 1.73 m2 (D) There is insufficient evidence to for or routine of protein to the CKD. lower protein intake of g/kg per day can be considered for patients with lower and to and of (D) Individual is after the risks and There is that the of CKD compared with is recommended, this has to be balanced the risk of in such as (R) the metabolic of is a cause of target organ including kidney disease and studies have the of in preventing or the of treatment of hyperglycaemia with it an increased risk of may be considered as a (D) The use of the of has also been shown to the of It be in that such as in the of of receptor or is in not recommended to adverse (D) Although the level of is a significant these have been to significant effects such as hypotension or The have a lower incidence of (Not Graded) The that may be potentially with an or is the that has receptor than and receptor than in with a lower incidence of than in a dose to renal function as a agent should be (R) of is for this of patients. should not be in patients with mL/min per 1.73 and is not recommended for high risk patients with mL/min per 1.73 and AKI. It should be with at the dose in lower risk patients with mL/min per 1.73 Patients with mL/min have risk. (D) of is associated with increased risk of in patients with significant renal function The is by of the and other and the is should not be in with a mL/min per 1.73 (R) CKD is a risk for acute from consumption of oral should not be in with low is for a include in the clinical is and is in of the to the of the tissue to a specific with a level of and to of the of acute or chronic The for for at least on the kidney during therapy or a should be considered it may a change in There is no systematic evidence to support for when or it is There is a of urine albumin or protein is the to glomerular protein the standard as it the of to the and Although this is to to or the of urine may its or on a urine or a urine is a to urine It is in clinical practice and in clinical trials the is to and is not by urine by in the there is a between and protein but is in patients with high protein and should not be in the clinical setting h urine is is defined as per and should be an of the therapy for blood The for blood is not but suggest that should be the treatment There are data to support a lower target of there is The of treatment is diuretics and per diuretics with or administration are usually preferred. in of the diuretic may be of and by bolus or may be to an a diuretic with a diuretic or with is an oral may be with diuretics to treat diuretic but are of patients may AKI as a result a with diuretic such as may is cause of AKI to a The risk of events significantly as serum albumin as a of or can the risk. low-dose is be at (R) have been including failure and in patients who received and In or patients are at high risk of they are to or or or high dose or for suggests a of 2 is the (R) The suggested for the initial of change disease is at a dose of This initial high dose should be for a of is or for a of of is not In patients who should be a of up to months For or of or for is recommended. (R) or for the duration are for patients who cannot high dose may also be For of can be (D) also for other of the and kidney for the of or other The recommended treatment is given at a dose of (D) The initial high dose of should be for a of and for up to as or has been In patients who should be a of up to months For of or can be for (D) there is to should be for at least may be an in patients who are to For and and patients, treatment with a that target may be in that results have been treatment can be for patients with clinical of (D) therapy should of a of and or can be an to Patients who do not well to either of these can be to the other that is to and For patients with significant of low and blood for and for is recommended has been to at least in up to of patients in Patients with months of and mL/min per 1.73 m2 can be with a of There are results with in While the no the demonstrated effects an was in the that retrospective of the suggested that may also be in patients with lower using an of also demonstrated significant effects after months of There is insufficient evidence to support the use of the and treatment should of and (R) and is an for patients in is or This from initial of of from trials and retrospective in patients with and disease in high rates reduction of and of randomized trials for of new or to the of for this is for patients with renal evidence of pulmonary or with (R) In a large trial of patients with a new of systemic by renal and SCr plasma was associated with a reduction in risk for to of from 43% to at months when compared with patient survival and adverse rates were similar in both treatment should of and (R) should be as early as to reduce the of renal Patients who are on have a renal prognosis and therapy has to be would to for support for the