Quantifying Acute Kidney Injury in Patients on Antibiotics in the Inpatient Setting

Bruce A. Mueller, PharmD, professor of Clinical Pharmacy at the University of Michigan College of Pharmacy, shares how clinicians in the inpatient setting can quantify acute kidney injury in adults on antibiotics.

Interview Transcript (slightly modified for readability):

It’s a difficult question to try to figure out what someone’s renal function is, and if they have acute kidney injury. This is really a problem in the intensive care unit, where you've got a patient who’s maybe not making urine and you’re saying, "Why is that the case? Does this patient have sepsis? Is this patient having a fluid onboard? Do they have acute kidney injury already and if they do, why do they have it?"

Our traditional markers for renal function are generally BUN, blood urea nitrogen, and serum creatinine. Yet, those are pretty terrible markers in the acute setting, but they’re great in the outpatient setting. They’re great when someone’s at steady state. We have all kinds of equations that we use with the serum creatinine; the Cockcroft gault equation and the MDRD, which is what’s used in most hospitals, and so when you look at a patient chart they’ll say the eGFR, estimated GFR, is based on the MDRD equation, but those were based on a steady state. That’s somebody who has had this normal steady amount of renal function—maybe not normal, but a steady amount of renal function,—they’ve been making creatine just as quickly as they’ve been getting rid of creatinine, and these equations do a nice job of making an estimate of what someone’s GFR is. In the acute setting, they don’t work anymore because we’re no longer at steady state.

So, the problem is that you have, all of a sudden, an abrupt cessation of any clearance of eGFR and creatinine is not being cleared anymore. But, the creatinine hasn’t had time to move up at all. So, you may have an anuric patient who's got a normal serum creatinine and if you use those traditional equations, the Cockcroft gault’s, the MDRDs, all those sorts of things, they totally missed the picture of what’s going on with your patient. In fact, your patient has got acute kidney injury, but you don’t see it because you’re applying the wrong equations and there may not be any equation that can work, to be honest.

The problem is, what if this happens really acutely, what if some very traumatic event happens and GFR goes to just about to nothing. Again, serum creatinine hasn’t had a chance to move there’s no equation that’s going to fix that and your best marker might just be urine output. Some people have given a furosemide stress test where you give a dose of furosemide and look for two hours let’s see how much urine gets made, and if no urine’s being made, and the patient is adequately hydrated, you know you’ve got bad acute kidney injury and you need to adjust your doses right now correctly.

The other thing that messes with clinicians is the fact that often, patients who have acute kidney injury in the ICU often have sepsis. They’ve been filled up with a lot of fluid and actually that fluid overload dilutes out the serum creatinine. So, serum creatinine goes down in these patients who’ve been resuscitated, who are anuric, and you look at that serum creatinine and you say, “Gosh it’s only 0.8. It’s a low normal value.” And, yet, the patient is anuric; they’re in acute kidney injury, and you need to adjust your antibiotic does appropriately. Now, maybe one of those adjustments is that with all that fluid on board, you have to give a big loading dose. That might be one of the adjustments, but then how often you give the doses may have to be extended because your patients is in acute kidney injury.