Background and definition of the decision problem

Health problem

Acute kidney injury is a common and serious complication that typically occurs in the context of an acute critical illness or during a postoperative period. It is associated with prolonged hospital stay, severe morbidity and increased mortality.1,2 Delayed identification of AKI contributes to worse outcomes.3

To pre-empt or avoid lasting consequences of AKI, early detection may be beneficial. Traditionally, AKI diagnosis relies on a rise in serum creatinine levels and/or fall in urine output. Despite its widespread use in the monitoring of kidney health and disease, creatinine is an imperfect marker of kidney function because its level in the blood is not solely dependent on kidney function, and changes in creatinine lag behind reduction in kidney function in AKI.4 When kidney function suddenly falls, even if a reduction in renal excretion occurs instantly, it can take hours or sometimes days for the level of creatinine to rise in the blood sufficiently for AKI to be diagnosed according to current international definitions.5 Moreover, in response to stress, or even kidney damage, the kidneys have reserve capacity and can compensate so that kidney function is maintained. For this reason, in some clinical settings, significant kidney damage can occur without AKI being apparent from changes in blood creatinine. In other settings, such as during a temporary reduction in blood flow to the kidneys, rises in creatinine and a reduction in urine can occur, even when no significant damage has occurred. These limitations related to the use of creatinine assessment have led to the search for novel biomarkers that may detect kidney damage or stress earlier and more reliably.

Biomarker tests for AKI include the NGAL test, which can be measured using a sample of urine or blood.6 NGAL is released from neutrophils and is induced by inflammation, indicating tubular injury.4 One limitation of NGAL is that it is produced throughout the body, making it difficult to distinguish systemic inflammation from localised renal inflammation.4 Novel NGAL tests include the ARCHITECT and Alinity i urine NGAL assays, the BioPorto NGAL plasma test and the BioPorto NGAL urine test.

Another biomarker for AKI is the NephroCheck test, which is a combination of two urinary biomarkers: the tissue inhibitor of metalloproteinase-2 (TIMP-2) and the insulin-like growth factor-binding protein 7 (IGFBP7). Both TIMP-2 and IGFBP7 are cell-cycle arrest proteins that are released into urine as markers of cellular stress in the early phase of tubular cell injury due to a variety of insults (e.g. toxins, drugs, oxidative stress and inflammation), which leads to AKI.7 The US Food and Drug Administration has approved these combined biomarkers to assess the risk of AKI in critically ill patients.4

These novel biomarkers have been developed to detect early damage or stress in the kidneys. If reliable use of these biomarkers can be demonstrated, they may enable earlier identification of AKI, and, therefore, early management of those with a modifiable disease course, with potential for downstream benefits in patients’ clinical outcomes. If demonstrated, the ability of these novel biomarkers for early detection of AKI could have the potential to improve current AKI management by enabling timely measures that could prevent progression to more severe kidney injury, as well as informing decisions about the ‘step-down’ of low-risk patients to a lower level of hospital care, thereby reducing the use of hospital resources.

The purpose of this assessment was to review the current evidence on the diagnostic accuracy, prognostic accuracy, impact on clinical outcomes and cost-effectiveness of novel biomarkers (i.e. the NephroCheck test, ARCHITECT and Alinity i urine NGAL assays, BioPorto NGAL plasma test and BioPorto NGAL urine test) for the assessment of AKI in critically ill patients who are considered for critical care admission.

Aetiology, pathology and prognosis

Acute kidney injury ranges from minor loss of kidney function to complete kidney failure. In current practice, reduced kidney function is identified by elevated serum creatinine levels and/or reduced urine output.

There are many causes of acute kidney injury,8 including the following:

• pre renal – reduced oxygen delivery to the kidneys, caused by:
  • low blood volume (after bleeding, excessive vomiting or diarrhoea, and severe dehydration)
  • reduced blood flow from the heart (potentially caused by sepsis or heart/liver failure)
  • damage to blood vessels, which can be caused by inflammation or blockages in the kidneys
  • medications that affect blood flow to the kidneys
• intrinsic/renal – damage to the kidney potentially caused by drugs, infections or contrast agents
• post renal – a blockage preventing drainage from the kidneys (potentially caused by an enlarged prostate, a tumour in the pelvis or kidney stones).

Incidence and/or prevalence

Major surgery is a significant risk factor for the development of AKI.4 In general, incidence of postoperative AKI depends on the type of surgery. Rates of AKI after cardiac surgery have been reported to range from 8% to 40%, depending on the patient populations.4 Recent meta-analyses have reported a pooled incidence of AKI in patients admitted to intensive care after abdominal surgery of 13.4% [95% confidence interval (CI) 10.9% to 16.4%],9 and pooled incidences of AKI after major trauma of 24% (95% CI 20% to 29%)8 and 21% (95% CI 16.5% to 24.9%).10

The incidence of AKI for all major, non-cardiac surgery patients and trauma patients can be as high as 50% (e.g. liver transplant patients). In a retrospective cohort of > 27,000 patients, the incidence of AKI, defined according to the Risk, Injury, Failure, Loss of kidney function and End-stage disease (RIFLE) criteria, was 37%.11,12

Impact of the health problem

People with AKI have higher mortality and longer hospital stays.1,2 In addition, AKI is associated with a higher risk of developing chronic kidney disease (CKD) and a need for long-term dialysis. The risk of CKD increases with the increased severity of AKI. More severe AKI has also been associated with increased mortality, length of hospital stay and use of intensive care services, in addition to a reduced chance of renal recovery.1,2 People with more severe AKI (and a greater loss of renal function) are more likely to need temporary RRT.

Measurement of disease

Several tools are available for determining the stage of AKI. A summary of staging system13 for AKI in adults based on the RIFLE,14 Acute Kidney Injury Network (AKIN)15 and Kidney Disease: Improving Global Outcomes (KDIGO)5 systems is presented in Table 1. A patient’s AKI should be staged by the criteria, and a classification of stage 1 or higher indicates AKI.

Description of the technologies under assessment

The NephroCheck test, the ARCHITECT and Alinity i urine NGAL assays, and the BioPorto urine and plasma NGAL tests may help to assess AKI in critically ill people who are considered for admission to critical care in hospital. These tests may be able to detect kidney injury earlier than the methods currently used for monitoring kidney function.

The NephroCheck test

The NephroCheck test measures urine levels of two biomarkers, TIMP-2 and IGFBP7, to assess the risk of moderate to severe AKI (defined as per KDIGO guidelines) in the subsequent 12 hours. The test result must be used in conjunction with clinical evaluation and the results of other tests, such as serum creatinine levels and urine output.

The concentrations of TIMP-2 and IGFBP7 are used to calculate an AKIRisk^® Score (Astute Medical, Inc.) [the concentrations of each (ng/ml) multiplied together and divided by 1000]. A score of ≤ 0.3 indicates a low risk of developing moderate to severe AKI within 12 hours of assessment, whereas a score of > 0.3 indicates a high risk of developing moderate to severe AKI within 12 hours of assessment.6

When used with the Astute140^® Meter (Astute Medical, Inc.), the NephroCheck test system consists of the following components:

• Astute140 Meter kit (a benchtop analyser)
• Astute140 Electronic Quality Control device
• NephroCheck test kit (includes a single-use NephroCheck test cartridge and reagents)
• NephroCheck Liquid Control kit
• NephroCheck Calibration Verification kit.

A fresh or thawed urine sample (mixed with reagent) is added to a single-use test cartridge, which is then inserted into an Astute140 Meter for incubation and result calculation. Preparation takes 3–5 minutes and the results of a NephroCheck test are available in ≈ 20 minutes. In the NHS, the Astute140 Meter would be used in a laboratory and not at the point of care.

The test can also be run on the VITROS^® 3600 Immunodiagnostic System (Ortho-Clinical Diagnostics Inc., Raritan, NJ, USA) and on the VITROS^® 5600 Integrated System (Ortho-Clinical Diagnostics Inc.) clinical chemistry analysers. All systems generate a single numerical result (the AKIRisk Score).

For surgical patients, it is recommended that the NephroCheck test is administered 2–4 hours after surgery. As NephroCheck exhibits a characteristic rise and fall after various exposures, a second administration of the test within the first 24 hours may be considered in patients with an ongoing risk of developing AKI.

In the UK, the NephroCheck test is marketed for people aged > 21 years.

Neutrophil gelatinase-associated lipocalin assays

ARCHITECT and Alinity i urine neutrophil gelatinase-associated lipocalin assays

The ARCHITECT urine NGAL assay is a chemiluminescent microparticle immunoassay for the quantitative determination of NGAL in human urine. NGAL can be used as a marker of kidney injury.

The ARCHITECT urine NGAL assay might be used as follows:

• for early detection of AKI
• to provide a measure of the severity of AKI
• to predict the requirement for RRT
• to help differentiate AKI from CKD and dehydration.

For diagnostic purposes, the test results should be used in conjunction with clinical assessment and the results of any other testing that has been undertaken (including serum creatinine levels and urine output). In addition, if the NGAL results are inconsistent with clinical assessment and other test results, additional testing can be undertaken to confirm the NGAL results.

The test could be used daily until a diagnosis is made or treatment for AKI is initiated.

The expected range for the assay (for people without kidney injury) is ≤ 131.7 ng/ml, based on the 95th percentile from specimens of non-hospitalised donors, but results from individual laboratories may vary and the manufacturer recommends that each laboratory should determine its own reference range based on the particular locale and population characteristics. The test has no age restrictions in use.

The assay is run on the ARCHITECT system (i1000SR, i2000, i2000SR, ci4100, ci8200 or ci16200) (Abbott Laboratories) in a laboratory. The throughput of the system is up to 200 tests per hour, and the time to first result is 36 minutes.

In addition to the ARCHITECT Urine NGAL Reagent Kit, the following materials are also needed:

• ARCHITECT Urine NGAL Calibrators
• ARCHITECT Urine NGAL Controls or other control material
• ARCHITECT i pre-trigger solution
• ARCHITECT i trigger solution
• ARCHITECT i wash buffer
• ARCHITECT i reaction vessels
• ARCHITECT i sample cups
• ARCHITECT i septum
• ARCHITECT i replacement caps.

The Abbott NGAL assay is also available for use on the Alinity i immunoassay system. The reagents for the Alinity i and ARCHITECT NGAL assays are the same.

The BioPorto neutrophil gelatinase-associated lipocalin test (using urine or plasma)

The BioPorto NGAL test is a particle-enhanced turbidimetric immunoassay for the quantitative determination of NGAL in human urine, ethylenediaminetetraacetic acid (EDTA) plasma and heparin plasma on automated clinical chemistry analysers. NGAL measurements may be useful in pre-empting the diagnosis of AKI, which may lead to acute renal failure. Urinary NGAL can serve as an early marker of AKI after cardiopulmonary bypass surgery, and both urinary and plasma levels of NGAL provide an early indication of acute renal injury in unselected patients in intensive care.

The NGAL test is intended to be used alongside monitoring of serum creatinine levels and urine output (rather than as a standalone test), and the significance of any raised NGAL level should be interpreted in the light of a patient’s clinical features.

The NGAL test can be administered as a single measurement, but also as a serial measurement, to detect any further development of AKI during hospitalisation or any improvement in the clinical condition. For patients admitted to intensive care, the test can be used to predict stage 2/3 AKI or as a negative predictive marker to rule out the presence of AKI.

To indicate the presence of AKI, the NGAL concentration in an isolated sample of urine and/or EDTA plasma should exceed 250 ng/ml. This threshold has been chosen to minimise the risk of an unacceptably high proportion of false-positive results.

The assay can be run on clinical chemistry analyser systems from F. Hoffman-La Roche Ltd (cobas^®, Modular P) (Basel, Switzerland), Siemens Healthineers (ADVIA^®) (Erlangen, Germany), Abbott Laboratories (AEROSET^®, ARCHITECT) and Beckman Coulter Inc. (Olympus AU) (Brea, CA, USA) in a laboratory. The assay takes 10 minutes to run.

In addition to the NGAL Test Reagent Kit, the following materials are also needed:

• NGAL Test Calibrator Kit
• NGAL Test Control Kit
• 0.9% weight by volume (w/v) aqueous sodium chloride solution as zero calibrator
• analyser-specific reagent containers.

At present, the test has no age restrictions on use.

Identification of important subgroups

The primary scope of this assessment was the optimisation of current secondary care of critically ill patients to decide whether or not the use of novel biomarkers would improve detection of AKI and, consequently, the current care pathway. The relevant population considered in this assessment was critically ill people at risk of developing AKI (i.e. those who are having their serum creatinine levels and urine output monitored) who are being assessed for possible admission to critical care. There is variation in intensive care utilisation across the world; in most studies conducted outside the UK, critically ill participants are usually admitted to critical or intensive care. The following patient subgroups have been identified as particularly relevant for the purpose of this assessment:

• type of surgery (e.g. major vascular/cardiac surgery, major non-vascular surgery, trauma, solid organ transplant)
• type of setting [e.g. post-surgery care, cardiac care, intensive or critical care, emergency department (ED)]
• type of sample medium (i.e. urine, blood plasma)
• people with a different underlying risk of AKI (e.g. depending on underlying condition: CKD, sepsis, hip fracture, major trauma, chronic liver disease)
• presence or absence of urinary infection and other inflammatory conditions (tests may perform differently in these populations).

Relevant comparator

Novel biomarkers need to be compared for incremental advantage over standard approaches to measuring kidney function. As discussed previously, AKI diagnosis traditionally relies on a rise in serum creatinine levels and/or fall in urine output. Creatinine has limitations as a biomarker because its concentration depends on the total body muscle mass, which varies between individuals. Some creatinine is also eliminated from the body by mechanisms other than filtering by the kidneys, which can be influenced by a variety of medications, including some commonly used antibiotics. In an illness that causes a sudden fall in kidney function (AKI), there may be a lag ranging from hours to days before creatinine levels in the blood rise to a level sufficient for AKI to be diagnosed according to current international definitions.5 In addition, in response to stress or even kidney damage, the kidneys have reserve capacity and can compensate so that kidney function is maintained. For this reason, in some clinical settings, significant kidney damage can occur without AKI being apparent from changes in blood creatinine. In other settings, such as during a temporary reduction in blood flow to kidneys, rises in creatinine and a reduction in urine can occur even when no significant damage has occurred.

Care pathway

The NICE clinical guideline on AKI16 recommends measuring serum creatinine and comparing it with the baseline for adults, children and young people with acute illness if risk factors for the condition are likely or present. Risk factors include sepsis, hypovolaemia and deteriorating early warning scores (using a paediatric version for children and young people). NHS England and NHS Improvement have endorsed the National Early Warning Score (NEWS) for use in acute and ambulance settings.17 An updated version of the score (NEWS2)17 was published in December 2017. The score should not be used with children (aged < 16 years) or pregnant women.

The NICE guideline16 further recommends monitoring serum creatinine regularly in all adults, children and young people with or at risk of AKI. The guideline development group did not wish to define ‘regularly’ because this would vary according to clinical need, but recognised that daily measurement was typical while in hospital.

An AKI algorithm to help with detection and diagnosis of the condition has been endorsed by NHS England.18 In some hospitals, the algorithm has been integrated into laboratory information management systems to help identify potential cases of AKI from laboratory data in real time.

The KDIGO Clinical Practice Guideline for Acute Kidney Injury19 highlights the importance of screening patients who have had an exposure that may cause AKI (e.g. sepsis or trauma) and recommends that high-risk patients continue to be monitored until the risk subsides. The guideline19 states that the frequency of serum creatinine measurements is a matter of clinical judgement, but suggests as a general rule that high-risk inpatients should have serum creatinine measured at least daily and more frequently after an exposure. Critically ill patients should also have urine output monitored.

For adults who are at risk of AKI, the NICE AKI guideline16 also recommends that systems are put in place to recognise and respond to oliguria (urine output < 0.5 ml/kg/hour).

For children and young people who are at risk of AKI, the guideline16 recommends:

• measuring urine output
• recording weight twice daily to determine fluid balance
• measuring urea, creatinine and electrolytes
• considering measuring lactate, blood glucose and blood gases.

Further detail on these recommendations and further recommendations on the ongoing assessment of the condition of patients in hospital can be found in section 1.2 of the NICE clinical guideline on AKI.16

The NICE guideline16 recommends diagnosing AKI in line with the RIFLE14 (or the paediatric-modified RIFLE),20 AKIN15 or KDIGO5 definitions, by using any of the following criteria:

• a rise in serum creatinine of ≥ 26 µmol/l within 48 hours
• a ≥ 50% rise in serum creatinine levels known or presumed to have occurred within the previous 7 days
• a fall in urine output to < 0.5 ml/kg/hour for > 6 hours in adults and for > 8 hours in children and young people
• a ≥ 25% fall in estimated glomerular filtration rate (eGFR) in children and young people within the previous 7 days.

There are no direct therapies for treating AKI. Care focuses on optimising haemodynamics and fluid status, avoiding nephrotoxic treatments and carrying out investigations to identity and resolve the underlying cause as quickly as possible. In general, the goal of care is to prevent any further kidney injury and to stop the worsening of the underlying illness to prevent mortality or renal progression to such a degree that RRT is needed.

The NICE clinical guideline on AKI16 highlights the importance of identifying the cause(s) of AKI and has recommendations on the use of urinalysis and ultrasound for this purpose.

The KDIGO Clinical Practice Guideline for Acute Kidney Injury19 also recommends prompt evaluation of people with AKI to determine the cause. Identifying possible reversible causes of the condition is highlighted as important in reducing the severity of the condition.

The NICE clinical guideline on AKI16 has recommendations on managing AKI (section 1.5), covering removing urological obstruction, pharmacological management, RRT and referral to nephrology services. The KDIGO Clinical Practice Guideline for Acute Kidney Injury19 recommends staging severity of AKI with serum creatinine and urine output, and to manage the condition according to stage and cause. General management principles for people at high risk of AKI (or with the condition) are to:

• discontinue nephrotoxic agents if possible
• monitor volume status and perfusion pressure
• consider functional haemodynamic monitoring
• monitor serum creatinine and urine output
• avoid hyperglycaemia
• consider alternatives to radiocontrast procedures.

Further actions, such as initiating RRT, should be considered at higher stages of AKI only. Dosages of drugs may also need to be adapted because of reduced kidney function. The KDIGO guideline19 also has more detailed guidance on the prevention and treatment of AKI (section 3). This includes haemodynamic monitoring and support, glycaemic control and nutritional support, the use of diuretics and vasodilator therapy.

In UK clinical practice, the NephroCheck test and NGAL assays are likely to be used for the assessment of AKI among people who are considered for admission to critical care, rather than among patients already in critical care. It is worth pointing out that the NephroCheck test, the ARCHITECT and Alinity i urine NGAL assays and the BioPorto plasma and urine NGAL tests would not replace serum creatinine and urine output monitoring, but would be used alongside current monitoring to facilitate earlier detection of kidney injury and prompt adoption of strategies to prevent further progression of kidney disease.