The SPLIT trial was the long awaited randomized control trial that pitted normal saline against the balanced solution du jour, Plasma Lyte-148. I have never before dedicated to the statistical plan for a study before the wait up to SPLIT:
And let’s face it the SPLIT trial did not go the way a lot of us wanted or expected. Everybody that was paying attention to the discussion had joined #TeamBalancedSolution. If you want to get a feel for the sense of inevitability of balanced solutions re-read the description from NephMadness 2014 or check out this editorial in KI.
SPLIT went the other way: it showed no adverse affects from saline and no advantage to balanced solutions. Much of the blogging about SPLT was fairly critical. Much of the criticism focused on the patient population (relatively low risk) and a lot of the criticism focused on the relatively low volume of fluid given to the participants. PulmCrit’s discussion on the volume of fluid is typical of much of the criticism:
How does this data apply to other situations? A broader interpretation of the study is that administration of 1-2 liters of normal saline would not increase the risk of renal failure compared to plasmalyte. This is not particularly controversial. Even the most ardent supporters of balanced crystalloid would probably agree that fluid selection doesn’t make a big difference at a volume of 1-2 liters. The proposed mechanism of nephrotoxicity due to saline is induction of a hyperchloremic metabolic acidosis, which tends to occur with larger volumes of fluid.
Shaw, et al. did a beautiful retrospective analysis of 0.9% saline versus Plasma-Lyte with propensity scoring. They showed a litany of problems with 0.9% saline:
- In-hospital mortality was 5.6% in the saline group and 2.9% in the balanced group (P < 0.001)
- One or more major complications occurred in 33.7% of the saline group and 23% of the balanced group (P < 0.001)
- Balanced fluid was associated with fewer:
- complications (odds ratio 0.79; 95% confidence interval 0.66-0.97).
- Postoperative infections (P = 0.006)
- Renal failure requiring dialysis (P < 0.001)
- Blood transfusion (P < 0.001)
- Electrolyte disturbance (P = 0.046)
- Acidosis investigation (P < 0.001)
- Acidosis intervention (P = 0.02)
How much fluid was needed to provide all of this hazard? About 2 liters of saline and 1.6 liters of Plasma-Lyte:
Yunos’s found the use of saline compared to Hartman’s (Australian for Ringer’s lactate):
- Mean increase in creatinine while in the ICU was 22.6 μmol/L vs 14.8 μmol/L (P = 0.03)
- The incidence of injury and failure class of RIFLE-defined AKI was 14% vs 8.4% (P <.001)
- The use of acute dialysis was 10% vs 6.3% (P = .005).
How much 0.9% saline had to be infused to get this disaster? 3.2 liters.
Another highly referenced article in the saline versus balanced solution cannon is Chowdhury’s randomized controled trial of healthy volunteers that used MRI imaging to measure renal blood flow following saline compared to plasmalyte 142 infusions. They found a significant reduction in mean renal artery flow velocity (P = 0.045) and renal cortical tissue perfusion (P = 0.008) from baseline with saline, but not Plasma-Lyte 148.
The volume of fluid needed to demonstrate decreased perfusion? Two liters.
After reading and digesting SPLIT, I’m still on #TeamBalancedSolution but my certainty has been shaken because to my mind SPLIT is the best done study with real patient oriented outcomes and it was convincingly negative.
I look forward to a spirited discussion in #NephJC on January 12th at 9pm EST and January 13th at 8 pm GMT.