This machine learning system accurately predicted which patients would develop acute kidney injury or need dialysis. And they did so 41 hours before the patients actually developed it! Imagine the ramifications not just for AKI but for predicting rapid response, need for ICU care, and more.
In patients admitted to the ICU, use of balanced fluids resulted in a lower rate of major acute kidney events (MAKE) at 30 days compared to normal saline (14.3% vs 15.4%). This is a NNT of 94 to avoid one MAKE.
In non-critically ill patients that received IV fluids in the ED, there was a lower incidence of major adverse kidney events in the balanced crystalloid group compared to saline (4.7% vs 5.6%) with a NNT of 111. There was no difference in terms of hospital-free days between the groups.
In patients with renal disease (GFR <60) presenting with possible acute coronary syndrome, high-sensitivity troponin (hs-TN) can still be used to identify those who are low-risk. But the overall percentage deemed low risk was much lower, as was specificity, in patients with renal impairment.
There are good data to support contrast-associated acute kidney injury (CA-AKI). With the use of low osmolar contrast agents, the risk may be lower. However, none of the myth-busting studies are good enough to exclude the risk of CA-AKI, especially in the most vulnerable patients.