Acute Kidney Injury

A spectrum of acute glomerular dysfunction characterised by a ↓ UO and ↑ serum creatinine and urea, and defined by the KDIGO classification as an:

Diagnosis is more complex in the critically ill due to lack of steady-state conditions, variable rate of creatinine production, and variable rate of elimination.

On Acute tubular necrosis:

  • The term ATN is often used to describe the intrinsic renal injury that occurs following a persistent pre-renal insult
    • This is probably not correct, as the histology of tubular necrosis is not confirmed in this population
    • The classical pattern is one of an:
      • Oliguric renal failure of 1-2 weeks
      • Non-oliguric renal failure of 10-14 days
        Likely due to restoration of glomerular filtration without recovery of tubular concentration.
  • In animal models, necrosis of the PCT (and occasionally DCT) occurs when blood flow is no longer adequate to maintain tubular perfusion and is believed to be a form of ischaemia-reperfusion injury
  • ‘True’ ATN may occur following arterial occlusion (such as AAA rupture or dissection)
The KDIGO Classification
Stage Creatinine Urine Output
1
  • 1.5-1.9× baseline in 7 days
  • ⩾26.5 μmol/L ↑ in Cr in 48 hours
  • ⩽0.5ml/kg/hr for 6-12 hours
2
  • 2-2.9× baseline
  • ⩽0.5ml/kg/hr for ⩾12 hours
3
  • 3× baseline
  • Cr ⩾354μmol/L
  • Initiation of RRT
  • eGFR ⩽35ml/min/1.73m2 in children
  • ⩽0.3ml/kg/hr for ⩾24 hours
  • Anuria > 12 hours

Noting that KDIGO is an imperfect definition: not all AKI will fit into this criteria, and not all cases fitting the criteria are true AKI.

Epidemiology and Risk Factors

Risk factors:

  • Occurs in:
    • 7% of hospital inpatients
    • Up to 25% of ICU patients
  • Significant morbidity and mortality
    • Mortality of patients requiring inpatient dialysis is 50-75%
  • Risk factors
    • Patient factors
      • Age
      • Pre-exiting CKD
      • Dehydration
      • Female
      • Black race
      • Elderly
    • Disease factors
      • COAD
      • Obesity
      • Sepsis
      • CCF
      • IABP
      • DM
      • Cancer
      • Anaemia
      • Hypertension
      • Vascular disease
      • Critical illness
      • Shock
      • Burns
      • Nephrotoxic drugs
      • Toxins
    • Perioperative
      • Trauma
      • Major surgery
        • Cardiac surgery
          • ↑ With CPB

Pathophysiology

Pathophysiology varies by aetiology, which are generally classified into:

  • Pre-renal
    40-70% of AKI.
  • Intra-renal
    10-50% of AKI.
  • Post-renal
    10% of AKI in ICU patients, but the most common cause in the general population.

Most common causes in the critically ill are:

  • Sepsis
  • Heart failure
  • HDx instability
  • Hypovolaemia
  • Nephrotoxins

Pre-Renal

Renal dysfunction due to extra-renal factors, resulting in decreased perfusion. Rapid correction of the systemic cause usually results in rapid renal recovery.

Causes can be divided into:

  • Hypovolaemia
    • Haemorrhage
    • Vomiting/Diarrhoea
    • Inappropriate diuresis
    • Burns
  • Hypotensive/low flow
    • Shock states
    • AAA
    • Renal artery stenosis
    • Hepatorenal syndrome
  • Impaired autoregulation
    • NSAIDS
    • ACE-Is/ARBs
  • Oedema
    • Cardiac failure
    • Cirrhosis
    • Nephrotic syndrome

Intra-Renal

Disease of the renal parenchyma, with structural changes visible on biopsy. Causes include:

  • Glomerular disease
    • Inflammatory
      • Post-infective GN
      • Henoch-Schonlein Purpura
      • SLE
      • Antineutrophil cytoplasmic antibody-associated GN
      • Antiglomerular basement membrane disease
  • Interstitial nephritis
    • Drugs
      • NSAID
      • Aminoglycosides
      • Radiological contrast
        It is theorised that osmotic stress and direct tubular toxic effects lead to renal tubular injury, and may cause acute tubular necrosis.
        • Typically, acute contrast nephropathy is benign, with creatinine returning to baseline within 10-14 days
        • There is significant uncertainty as to whether contrast media do cause acute kidney injury
          If this risk is present, it is probably only relevant in patients who have:
          • Impaired renal function
          • Arterial contrast
        • Rehydration and volume correction are effective in preventing a rise in creatinine
    • Infiltrative disease
      • Lymphoma
      • Granulomatous disease
        • Sarcoidosis
        • Tuberculosis
      • Infection
        • Pyelonephritis
  • Tubular injury
    Typically due to continuation of pre-renal causes.
    • Ischaemia
    • Toxins
      • Drugs
      • Myoglobin
      • Heavy metals
    • Metabolic
      • Hypercalcaemia
      • Immunoglobulins
    • Crystals
      • Urate
      • Oxalate
  • Vascular
    • Vasculitis
    • Cryoglobulinaemia
    • Polyarteritis Nodosa
    • Thrombotic microangiopathy
    • Cholesterol emboli
    • Renal artery/renal vein thrombosis

Post-renal

Divided into:

  • Intraluminal
    • Urinary retention
    • Stone
    • Clot
  • Intramural
    • Urethral stricture
    • Prostatic hypertrophy
    • Tumour
  • Extramural
    • Malignancy

Clinical Manifestations

Diagnostic Approach and DDx

  1. Is this acute or chronic?
    • Previous creatinine measurements
    • Chronicity suggested by:
      • Symptom duration
      • Nocturia
      • Anaemia
      • Hyperphosphataemia
      • Hypocalcaemia
  2. Is there ↓ renal perfusion?
    • Volume status?
    • Cardiac insufficiency?
    • Renal vasoconstriction
  3. Is there obstruction?
    • Ultrasound
      • Bladder
      • Renal tract
    • Palpable bladder
  4. Is there other evidence or concern for renal parenchymal disease?
    • Rashes
    • Arthralgia
    • Myalgia
    • Haematuria
  5. Is there vascular occlusion?

Investigations

Laboratory:

  • Blood
    • Creatinine
      • Normal daily variation in SCr is ~10%
      • Measurements are affected by:
        • Volume state (as it is measured as a concentration)
      • Ideally, results from the same lab should be used
    • Urea
    • Autoimmune profile
      ANCA, ANA, C3, C4. Perform in patients with:
      • Pulmonary-Renal Syndrome
      • Haemoptysis
      • Haemolysis
      • Thrombocytopaenia
  • Urine
    • Sodium and osmolality
      • Healthy kidneys respond to hypovolaemia/hypoperfusion by ↑ urinary osmolarity and ↓ sodium, urea, and water excretion
        This response may be impaired in disease, as well as due to drugs and other therapies (such as CPB).
      • Low urine osmolality may indicate impaired tubular function
      • High osmolality and low sodium with uraemia and oliguria suggests intact tubular function
    • Renal tubular cells
      Tubular disease.
    • Red cell casts/dysmorphic red cells
      Indicate glomerular disease.
    • Oxalate crystals
      Ethylene glycol poisoning.
    • Urate crystals
      Tumour lysis syndrome.
    • Urinary light chains

Imaging:

  • Urinary tract ultrasound
    For diagnosis of obstruction.

Other:

  • Intrabdominal pressure measurement
    • >20mmHg suggests abdominal compartment syndrome contributing to AKI
  • Renal biopsy

Management

Preventative

  • Maintain renal perfusion
  • Avoid nephrotoxins
  • Remove obstruction

Specific therapy:

CRRT is covered in detail under Continuous Renal Replacement.

  • Pharmacological
    • Avoid nephrotoxins
      • Avoid aminoglycosides unless no other alternatives are available
        If used:
        • Administer as single daily dose
        • Monitor levels when used for >48 hours
      • Use lipid formulations of amphotericin B
      • Use azoles or echinocandins as alternatives to amphotericin B if able
      • IV contrast
  • Procedural
    • RRT
      • Indicated for life-threatening:
        • Volume overload
        • Electrolyte disturbance
        • Acidosis
        • Uraemia
          Typically initiated for uraemia prior to onset of uraemic complications.
          • Uraemic pericarditis
          • Encephalopathy
          • Coagulopathy
        • Toxins
      • Consider:
        • Later initiation in single-organ failure as:
          • We are poor at predicting renal recovery
          • There is potential harm from unnecessary early RRT
        • Earlier initiation in multi-organ failure
          • Optimise volume and physiological state
      • Aimed at:
        • Maintaining electrolyte, solute, and acid-base homeostasis
        • Preventing further renal insult
        • Permitting renal recover
        • Facilitate delivery of other supportive measures
      • Ceased when either:
        • Intrinsic kidney function has recovered
          Mean duration ~12 days.
        • Not consistent with goals of care

Not that RRT is indicated by creatinine and urea values alone.

There is no mortality difference between IHD and CRRT, but CRRT is less likely to cause hypotension.

Supportive care:

  • C
    • Maintain intravascular volume
      • Crystalloid > colloid
    • Maintain renal perfusion pressure
      Maintain MAP 65-90mmHg.
      • Consider higher MAP targets in patients with hypertension
      • Vasopressors and fluids
  • G
    • Avoid hyperglycaemia
      Use IV insulin to maintain BSL 6-10 mmol/L.
    • Maintain 20-30 kcal/kg/day of energy intake
      Enteric route if possible.
    • Avoid protein restriction
  • H
    • Correct anaemia

Marginal and Ineffective Therapies

  • Loop diuretics
    Reduce volume but do not affect survival or renal recovery.
    • Theoretically reduce oxygen consumption in loop of Henle and therefore lessen ischaemic injury
      Minimal supportive evidence outweighed by evidence of harm.
    • Appropriate to treat volume overload in the setting of AKI, but not AKI itself
    • Ineffective in reducing frequency or duration of RRT
    • ⩾1g/day of frusemide may cause ototoxicity
  • Mannitol
    Likely no benefit.
  • Dopamine
    ↑ UO but does not improve renal recovery. Some evidence of harm (particularly arrhythmias).
  • Natriuretic peptides (e.g. ANP)
    No benefit.
  • Fenoldopam
    No benefit.
  • (rh)IGF-1
    No benefit.
  • NAC

Anaesthetic Considerations

Major perioperative considerations:

  • Meticulous fluid management
  • Cardiac output and blood pressure control
  • Consideration of dynamic changes in renal clearance

Complications

  • Death
    • Perioperative mortality over 50%
  • C
    • Volume overload
  • F
    • Metabolic acidosis
    • Hyperkalaemia
    • Uraemia
      • Platelet dysfunction
      • Pericarditis
      • Encephalopathy
  • I
    • Sepsis

Key Studies

Early RRT:

  • AKI is common in ICU patients and a high proportion need CRRT

  • Early dialysis may correct the abnormal metabolic and inflammatory milieu before instability occurs, but ↑ risk of procedural complications from an unnecessary intervention

  • ELAIN (2016)

    • 230 German with KDIGO 2 AKI despite resuscitation and some marker of severity
    • Single-centre, block randomised RCT
    • Early (<8 hours of KDIGO 2) vs. delayed (<12 hours of KDIGO 3, or an absolute indication for) RRT
      • Early RRT
        All patients received RRT, median in 6 hours.
      • Delayed RRT
        91% received RRT, median in 25 hours.
      • All received CVVHDF
    • ↓ Mortality with early RRT (39.3% vs. 55%)
    • 47% cardiac surgical population
    • Only 9% not dialysed in the delayed group, so low risk of inappropriate CRRT with early filtration

  • AKIKI (2016)

    • ~620 Frenchpersons with KDIGO stage 3 injury, ventilated or on catecholamine infusions, without a hard indication for dialysis
    • Multicentre (31), block-randomised, unblinded trial
    • 546 patients gives 90% power to detect 15% (!!) ↓ ARR in death from baseline mortality of 55%
    • Early vs. delayed RRT
      • Early
        • RRT within 6 hours
        • 98% received RRT
      • Control
        • RRT if profound metabolic abnormalities, APO, or 72 hours of oligo/anuria
        • 51% received RRT
    • No change in 60 day mortality (48.5% vs 49.7%)
    • Double CLABSI in the early RRT group (10% vs. 5%)
    • Post-hoc analyses:
      • Lowest mortality in those who never received RRT
      • Highest mortality in those who received RRT late

  • STARRT-AKI (2020)

    • ~3000 adults with severe AKI without immediate need for dialysis
    • Open-label, assessor-blinded RCT
    • 90% power for 6% ↓ 90 day mortality, assuming 40% mortality in control group
    • Accelerated vs. standard RRT
      • Accelerated received RRT within 12 hours (mean 6)
      • Standard received RRT unless necessary
      • Modality and anticoagulation were at clinician preference
    • No difference in mortality
    • ↑ In RRT dependence at 90 days (10% vs 6%) in accelerated RRT group

Other:

  • Bellomo et al (2000)
    • 230 adults with ⩾SIRS features, a central line, and a sign of early renal dysfunction; without significant CKD, renal transplant, or recent AKI
    • Multicentre (23), block randomised, double-blind, placebo-control trial
    • 80% power for 20% ↓ in serum creatinine
    • Dopamine vs. placebo
      • Dopamine at 22µg/kg/min
      • Placebo infusion at same rate
      • Continued until death, RRT, ICU discharge, or resolution of SIRS and renal dysfunction
    • No change in serum creatinine
    • No change in secondary outcomes, notably arrhythmia
    • No change in urine output in dopamine group
  • RENAL (2009)
    • ~1,500 Australasian ICU patients requiring CRRT
    • Multi-centre (35) RCT
    • 90% power for 8.5% ARR in mortality from control group mortality of 60%
      This was described as a “conservative” effect.
    • High-intensity (40mL/kg/hr) vs. low-intensity (25mL/kg/hr) CRRT
      • High-intensity received an average of 33.4mL/kg/hr
      • Low-intensity received an average of 22mL/kg/hr
    • No change in mortality (44.7% vs. 44.7%)
    • Significantly ↑ hypophosphataemia in high-intensity group

References

  1. Hilton R. Acute renal failure. BMJ. 2006 Oct 14;333(7572):786-90.
  2. Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney inter., Suppl. 2012; 2: 1–138.
  3. The Royal Australian and New Zealand College of Radiologists. Iodinated Contrast Media Guideline. Sydney: RANZCR; 2016.
  4. Ostermann M, Joannidis M. Acute kidney injury 2016: diagnosis and diagnostic workup. Crit Care [Internet]. 2016 Dec 27 [cited 2018 Jan 31];20(1):299.
  5. The STARRT-AKI Investigators. Timing of Initiation of Renal-Replacement Therapy in Acute Kidney Injury. N Engl J Med. 2020;383(3):240-251. doi:10.1056/NEJMoa2000741
  6. Bellomo R, Chapman M, Finfer S et al. Low-dose dopamine in patients with early renal dysfunction: a placebo-controlled randomised trial. The Lancet. 2000;356(9248):2139-2143. doi:10.1016/S0140-6736(00)03495-4