Managing AKI Just Got More Stressful

Tejas P. Desai, MD


October 29, 2015

Furosemide Stress Test and Biomarkers for the Prediction of AKI Severity

Koyner JL, Davison DL, Brasha-Mitchell E, et al
J Am Soc Nephrol. 2015;26:2023-2031


The prevalence of acute kidney injury (AKI) is growing among both hospitalized and ambulatory patients. In the past few years, our understanding of AKI, including the factors that cause it, has expanded. Today nephrologists, primary care providers, and emergency department physicians are keenly aware of identifying and, whenever possible, mitigating factors that can result in AKI.

Despite these efforts, AKI remains one of the most common reasons for an inpatient nephrology consultation and continues to carry a high risk for morbidity and mortality. Once AKI has developed, many healthcare providers are perplexed at the variability in the rate at which patients can recover (or whether they recover at all). Naturally, this mystery complicates both intra- and post-hospitalization care and raises the stress level for both patient and provider.

The Study: Part 1

Arming healthcare providers with reliable information regarding the severity of AKI helps them make better, more timely patient care decisions. Investigators from across the United States have teamed up to publish two papers on the use of furosemide to identify severe AKI.

First, the furosemide stress test (FST) was analyzed in a cohort study in 2013.[1] Euvolemic patients with Acute Kidney Injury Network (AKIN) stage 1 AKI were given a single dose of intravenous furosemide at 1.0 mg/kg (if furosemide-naive) or 1.5 mg/kg (if not naive). Patients who produced less than 200 mL of urine in the subsequent 6-8 hours after furosemide infusion had a statistically greater likelihood of progressing to AKIN stage 3 AKI than their high-urine-output counterparts. With an area under the curve (AUC) between 0.82 and 0.87, the investigators suggested that the FST would be a reliable way to determine the severity of a patient's AKI.

Despite these robust statistical results, providers have been slow to adopt the FST. A common explanation has been that high-dose intravenous furosemide infusions pose greater risks to patients then measuring novel biomarkers, such as urinary neutrophil gelatinase-associated lipocalin (NGAL), tissue inhibitor of metalloproteinase 1 (TIMP-1), and others, even though these biomarkers are still being studied and are not available for routine clinical use.

The Study: Part 2

In the 2015 study, the investigators took a new approach. They compared the predictive utility of the FST against many of the newer biomarkers currently being studied.

Critically ill patients who were euvolemic (or could achieve euvolemia through saline infusions) with AKIN stage 1 AKI were given the same FST regimen as in the 2013 study. Just before administration of furosemide, a battery of biomarkers was measured in each study patient. These included urinary TIMP-2, insulin-like growth factor binding protein 7 (IGFBP-7), kidney injury molecule 1 (KIM-1), interleukin 18 (IL-18), NGAL, creatinine, albumin/creatinine ratio (ACR), uromodulin, and plasma NGAL. The primary endpoint was the same as that in the 2013 study: the likelihood of progression to AKIN stage 3 AKI.

The Results

As in the 2013 study, the FST performed well as a predictor of developing AKIN stage 3 AKI. Using the 200-mL urine output level as a cutoff, the AUC for the FST was 0.87. Not one of the novel biomarkers achieved an AUC comparable to the FST. The closest was the plasma NGAL, reaching an AUC of 0.75.

Perhaps most surprising, the conventional measures of AKI severity performed the worst. Tests that providers rely on heavily, such as the urinary ACR, fractional excretion of sodium, and urine creatinine, significantly underperformed compared with the FST (AUC < 0.57 for any of the three).

When the investigators combined the FST with the new novel biomarkers, the AUCs approached that of FST alone. Plasma NGAL, which performed the best of all the novel biomarkers, had an AUC of 0.86 when pooled with the FST. Uromodulin, which performed the worst among the novel biomarkers, saw a dramatic increase in AUC when combined with the FST, from 0.54 to 0.85. And the dismally performing conventional markers were lifted by the FST to respectable AUCs of 0.83-0.84.

What This Means, and Where We Go From Here

Data from the 2013 and 2015 investigations suggest that the FST is a legitimate diagnostic tool in predicting the severity of AKI. While we wait for further study of a number of novel biomarkers, the FST is available now and can help providers demystify an important component of AKI. In fact, given its stellar performance against conventional tools, perhaps all providers who care for patients with AKI should understand how to execute and interpret an FST.

Justifiable doubts may remain as to the risk posed in implementing this test, a particular concern of mine that the 2015 investigation did not appropriately mitigate. Does the safety of measuring plasma NGAL (AUC, 0.75) justify its use instead of the furosemide stress test? Is the risk to the patient from a high dose of intravenous furosemide worth the improvement in predictive power? Tell us what you think in the comments section below.



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: