Acute Kidney Injury – Indications for Dialysis

What are the indications for Renal Replacement Therapy (RRT) with dialysis in Acute Kidney Injury (AKI)?

The common indications:

• Uremia
• Life threatening volume overload refractory to medical treatment
• Life threatening hyperkalemia refractory to medical treatment
• Life threatening metabolic acidosis refractory to medical treatment
• Life threatening intoxication with a dialyzable drug

This leaves the following questions:

• What about oliguria? Anuria?
• Does initiating RRT earlier improve prognosis/ outcomes
• What is life threatening? What is uremia?

There have been several studies that help provide clarity for practical management / decision making in the Intensive Care Unit (ICU) setting with stage 2 or 3 acute kidney injury (AKI).   3 of the 4 studies required stage 3 AKI. Practically this means:

• Creatinine either doubled (Stage 2 AKI) or tripled (Stage 3 AKI) or increased by at least 0.3 g/dl to a level > 4 g/dl (Stage 3 AKI).

Or

• Urine output was < 0.5 ml/kg/hr for 12 hours (Stage 2 AKI) or < 0.3ml/kg/hr for 24 hrs (Stage 3 AKI) or there was anuria for 12 hours (Stage 3 AKI).

The Studies

AKIKI – Artificial Kidney Initiation in Kidney Injury

Initiation Strategies for Renal-Replacement Therapy in the Intensive Care Unit | NEJM

• Randomized trial of patients with ventilator and/or vasopressor requirement
  • Early – within 6 hrs of diagnosis of stage 3 acute kidney injury.
  • Delayed – Emergent indication (see table) or oliguria for > 72 hrs

• Outcomes:
  • No difference in 60-day mortality (48.5 vs 49.7%)
  • Delayed dialysis group:
    • 49% did not require Renal Replacement Therapy (RRT)
    • Diuresis returned earlier
    • Less catheter bloodstream infections ( 5 vs 10%)

IDEAL-ICU

Timing of Renal-Replacement Therapy in Patients with Acute Kidney Injury and Sepsis | NEJM

• Randomized trial of patients with RIFLE Failure stage (Stage 3 AKI) and early vasopressor dependent septic shock (within 48 hours of vasopressor)
  • Early RRT: within 12 hours of diagnosis
  • Delayed RRT: Emergent indication (see table) or no recovery after 48 hrs
• Outcomes:
  • Death at 90 days – no difference (58% vs 54%)
  • 38% delayed did not require RRT
    • 29% recovered
    • 8% death
    • 2% other
  • 17% required emergent RRT
  • More patients developed hyperkalemia in the delayed group ( 4% vs 0%) and there was a trend toward increased metabolic acidosis (17% vs 9% p = 0.07). Otherwise no significant difference in adverse events.

STARRT-AKI

Timing of Initiation of Renal-Replacement Therapy in Acute Kidney Injury | NEJM

• Randomized controlled trial of critically ill (ICU) patients with stage 2 or 3 acute kidney injury (AKI)
• Accelerated RRT initiation
  • Within 12 hours of AKI diagnosis
• Standard RRT initiation
  • Clinical judgment (Discouraged unless potassium > 6 meq/L, pH < 7.20, serum HCO3 < 12 mmol/L or PaO2/FIO2 < 200)
  • AKI persisted for > 72 hours
• Outcomes
  • No difference in mortality (death at 90 days) – 43.9% vs 43.7%
  • Higher dependence on RRT for survivors (>90 days) in accelerated initiation group (10.4% vs 6.0%)
  • More adverse events in the accelerated group (23% vs 16.5%).
    • Mainly hypotension/ hypophosphatemia.
    • No difference in severe adverse events (1.0% vs 0.5%)

AKIKI 2

Comparison of two delayed strategies for renal replacement therapy initiation for severe acute kidney injury (AKIKI 2): a multicentre, open-label, randomised, controlled trial – The Lancet

• Randomized controlled trial of patients in the delayed arm of AKIKI trial who did not require an urgent indication for RRT prior to endpoint (oliguria for 72 hrs or BUN > 112 mg/dl)
• Delayed
  • Initiate RRT
• More delayed
  • Delay RRT initiation until:
    • Urgent indication (see table)
    • BUN > 140 mg/dl
  • Outcomes
    • No difference in primary outcome (days alive without requiring RRT)
    • Trend toward increased 60 day mortality
      • 44 vs 55% (p= 0.071)
      • Increased Hazard Ratio (multivariate analysis) for death (1.65)
    • No difference in complications

My Perspective

• What about oliguria? Anuria?

  • If you have oliguria or anuria long enough you will develop a life threatening indication (volume overload/ hyperkalemia) as there will be an inability to remove fluid and potassium.
  • Oliguria by itself is not an indication to start RRT (for at least up to 72 hrs).

• Does initiating RRT earlier improve prognosis/ outcomes

  • Allowing 48-72 hours for AKI to recover is not associated with worse outcomes.
  • But, delaying dialysis until BUN is > 140 mg/dl (in patients with AKI) may be associated with increased mortality.

• What is life threatening?

  • Potassium:
    • > 6-6.5 meq/L
  • Metabolic acidosis:
    • pH < 7.15 – 7.20 and
      • Depressed pCO2 or an inability to further ventilate or
      • HCO3 < 18 or Base deficit > 5
    • HCO3 < 12 mmol/L
  • Fluid overload/pulmonary edema.
    • FIO2 >50% or PaO2/FiO2 ratio < 200 despite diuretics

• What is uremia?

  • Findings severe enough to warrant dialysis initiation:
    • Pericardial friction rub
    • Unexplained myoclonus or seizures
  • Other potential uremic findings (ie mental status changes, nausea, bleeding) are too nonspecific (if BUN is < 112-140 mg/dl).
  • BUN elevation (if not associated with clinical findings is azotemia not uremia)
    • Somewhere between 112-140 mg/dl. This is in patients with Acute Kidney Injury, not necessarily an indication for dialysis in patients without AKI.

Summary

Delaying the initiation of dialysis for 48-72 hrs (in the absence of a potential life threatening indication) in patients with acute kidney injury is not associated with an increase in mortality. A significant percentage of these patients (approximately ⅓  –  ½)  won’t end up needing dialysis and will be able to avoid the potential complications associated with this invasive procedure.  The above studies provide some practical guidelines as to what constitutes a life threatening indication.  As always, clinical judgment should be used in each specific case.