UMEM Educational Pearls

STARRT-AKITrial

The Standard versus Accelerated initiation of Renal Replacement Therapy in Acute Kidney Injury

The development of acute kidney injury (AKI) in the critical care setting portends a greater morbidity and mortality for patients. Additionally, it places the patient at high risk of complications and requires a greater use of resources. Several studies in the past have examined if the timing of initiation of renal replacement therapy (RRT) would result in a mortality benefit, but have failed to demonstrate consistent outcomes.

The STARRT-AKI trial was a multinational, randomized controlled trial designed to determine if early initiation of RRT in critically ill adult patients with AKI lowered the risk of 90-day mortality. The Kidney Disease Improving Global Outcomes (KDIGO) classification was used to define AKI and over 2900 patients were randomly assigned to two groups over a 4 year period. Exclusion criteria included: recent RRT, a renal transplant within the preceding year, advanced CKD, an overdose necessitating RRT, or a strong suspicion of obstruction or autoimmune/vascular cause of their AKI.

Groups:

• The accelerated strategy group
  • Initiation of RRT within 12 hours of meeting eligibility criteria (AKI based on KDIGO definition)
• The standard strategy group –
  • General goal of withholding RRT unless the patient met the following specific parameters:
  • K+ >6.0,  pH <7.20,  HCO3 <12mmol/L,  moderate ARDS with clinical picture concerning for volume overload, or persistent AKI >72hr after randomization

Outcomes/Results:

• The study’s primary outcome measure was all cause mortality at 90 days
  •  There was no significant difference between the groups
  •  P=0.92 with RR 1.00
• Secondary outcomes evaluated several things including ventilator and vasoactive free days, hospital length of stay, number of days without RRT at 90 days as well as adverse events directly related to RRT
  • Interestingly, at 90 days, the patients in the accelerated strategy group were more likely to have ongoing RRT needs at 10.4% compared to the standard strategy group at 6.0% (not statistically significant).
  •  Overall, no significant difference between the groups when assessed for death in the ICU, major adverse events, or with regard to hospital length of stay.

Take home points:

• This was a well done, well randomized trial from many countries and ICU settings
• No significant mortality benefit between groups at 90 days
• Interestingly, the patients in the accelerated group were more likely to have suffered adverse events related to RRT and were more likely to be dependent on RRT at 90 days
  • It is unclear why this is, but suggestive that early initiation of RRT may compromise the intrinsic healing of the kidney
  • Emphasizes a greater risk for adverse events without clear benefit
• Ultimately, the decision to initiate RRT should be based on the patient’s clinical picture, acid/base status, electrolyte abnormalities, and volume status and NOT on a general trend of their renal indices.

References