When it comes to getting abundant information from a single laboratory test, few analytes compare with creatinine. Physicians and clinical laboratory professionals use this marker of renal function directly and indirectly for many purposes in diagnosing and managing various conditions. One particular example of creatinine’s versatility involves acute kidney injury (AKI), an often under-recognized disorder due to its silent nature for both patients and clinicians. Despite the advent of newer biomarkers for AKI such as neutrophil gelatinase-associated lipocalin (NGAL), no gold standard AKI diagnostic exists. Clinicians rely mainly on measurements of serum creatinine and urinary output, closely following the former in most hospitalized patients.

A reported AKI incidence of 3.2%–18.3% among hospitalized patients underscores the need to identify this condition promptly. In a hospitalized patient with complex medical issues, a decline in renal function might be overlooked, especially when the individual’s primary illness does not involve the kidneys. Also, pressure to make room for patients awaiting hospital beds could lead to those with elevated creatinine levels being discharged without appropriate referrals for nephrology consultations. In the United Kingdom, the Think Kidneys campaign was launched to highlight the urgency of AKI and to make recommendations for preventing the considerable AKI-related mortality and comorbidity, including progression to chronic kidney disease (CKD). 

A Highly Individualized Marker

While serum creatinine is relatively easy to test, many laboratorians appreciate that creatinine levels are highly individualized, meaning that comparing a single result to the reference interval may not accurately identify an abnormal condition. A patient’s age, sex, muscle mass, and several other factors can influence creatinine levels. Consequently the definition of AKI when using serum creatinine measures is based on elevations relative to a baseline level within a specified period of time. The most recent AKI classification system is the Kidney Disease Improving Global Outcomes clinical practice guideline that also provides a staging system for AKI by following changes in serum creatinine. 

How can we as laboratorians do more to improve AKI diagnosis and management? For starters, we already excel at making it easier for clinicians to interpret creatinine results. One important tool would be our laboratory information systems (LIS) to facilitate reviewing prior creatinine results, calculating the change in value, and assessing the degree of change. This takes some of the guesswork out of documenting possible AKI in the patient’s chart and flagging clinically important results for action. By having the LIS estimate and automatically report the AKI stage and creatinine results coupled to electronic alerts (e-alerts), laboratorians can inform clinicians to begin supportive measures and medication review, with the aim of avoiding nephrotoxicity, or even dialysis. 

Before labs implement an AKI-related e-alert, however, they need to collaborate closely with their nephrology service to develop a computerized algorithm. The U.K. experience with the Think Kidneys campaign raised several questions about divergent approaches to identifying AKI based on serum creatinine levels. For example, which creatinine level among the previous values should be used? What time frame for prior results is valid? What if the current creatinine measurement has no baseline value for comparison? These are just some of the challenges that must be considered that would have a direct impact on positivity rates for AKI. 

The Toronto AKI Initiative

At Toronto General Hospital, our first venture into an e-alert for AKI began in April 2016, and we had to make certain assumptions. We limited the algorithm to the medical inpatient unit and validated our algorithm against in-hospital consults for AKI. While we showed that computerized alerts were effective at identifying 62% of AKI cases 1 day earlier than the recorded referral, our algorithm was not as adept at distinguishing between advanced CKD and AKI. The nephrology service would need to crosscheck against known renal patients with CKD. We are still in the process of assessing this quality improvement initiative, with an eye toward further improving our follow-up process with nephrology so that high-risk patients are not missed before their discharge home.

Will e-alerts for AKI improve patient outcomes? Time will tell. In the U.K., this has been a priority with a mandate that all hospitals implement an e-alert for AKI in support of the Think Kidneys campaign. Due to initial variability in reporting, the U.K. National Health Service devised a standardized algorithm that even needed LIS manufacturers to be on board. In the U.S. the only large randomized controlled trial with an e-alert—conducted at the University of Pennsylvania—found there was no effect on AKI progression, receipt of dialysis, or death (Lancet 2015;385:1966–74). However in an era in which adding newer tests like NGAL to the laboratory menu may have its own challenges, enhancing the value of a tried-and-true creatinine test with some help from the LIS could be the next best answer for detecting AKI. 

Paul M. Yip, PhD, FCACB, DABCC, is a clinical biochemist at the University Health Network and an assistant professor in the department of laboratory medicine and pathobiology at the University of Toronto. +Email: paul.yip@uhn.ca.