Frequent point-of-care creatinine measurement in the critically ill allows early prediction of AKI

Acute kidney injury (AKI), which is a sudden reduction in kidney function, is a common complication of many conditions that warrant admission to intensive care, such as trauma, sepsis and major surgery. Around 50 % of intensive care patients develop AKI during their critical illness and this complication is independently associated with increased morbidity and mortality; in the long term, survival following AKI is associated with increased risk of chronic kidney disease (CKD). Prompt diagnosis of AKI is required to ensure most favorable outcome for these critically ill patients.

AKI is defined by a rise in serum creatinine concentration and/or reduction in urine output, and diagnosis is confirmed if any of the following three criteria are satisfied:

Daily monitoring of serum creatinine has long been a standard of care for early detection of AKI among the critically ill. Now, more frequent (every few hours) rapid creatinine measurement has been made logistically possible by the availability of creatinine assays on blood gas analyzers and other point-of-care platforms sited in intensive care units. Could more frequent point-of-care serum creatinine measurement be used to detect deteriorating renal function earlier than currently used daily creatinine measurement? That is the main proposition tested in this recently published prospective observational study conducted at an Australian tertiary intensive care unit. Study authors sought to test their hypothesis that small short-term serum creatinine increase is an early marker of AKI. They also examined the wider clinical significance of the small short-term creatinine increase they identified in a significant proportion of study patients.

The study population comprised 387 critically ill adult patients admitted to intensive care during a 6-month period in 2018. A qualification of entry to the study was that patients had no evidence or history of reduced renal function (AKI or CKD) at the time of admission to intensive care.

At this Australian intensive care study site, serum creatinine assay is available on the blood gas analyzer within the unit. Whenever clinical staff deemed blood gas analysis was warranted (typically every 3-4 hours), each study patient’s serum creatinine was recorded and the first occurrence of an episode of short-term creatinine increase noted. (For the purposes of the study a short-term increase was defined as a rise in serum creatinine >1 µmol/L/hr between two sequential blood gas analyses 3-4 hours apart).

Of the 387 study patients, 279 (72.1 %) developed an episode of short-term increase in serum creatinine. Median short-term increase for these 279 patients was 7 µmol/L (interquartile range-IQR 5-10 µmol/L) which occurred on average 5 hours (IQR 2-10 hrs) after admission to intensive care. These episodes lasted for a median time of 3 hours (IQR 2-4 hrs).

Of all 387 study patients, 212 (54.8 %) developed AKI during their stay in ICU. Median time from ICU admission to diagnosis of AKI was 15 hours. Those who had a short-term small creatinine increase episode were significantly more likely to develop AKI than those who did not; of the 279 who developed an episode of short-term increase in serum creatinine, 185 (66.3 %) subsequently developed AKI, whereas of the 108 patients who did not have such an episode, just 27 (25 %) developed AKI.

Statistical analysis revealed that short-term episodes of slight increase in serum creatinine, occurring on average 10 hours before development of AKI, is a highly sensitive but poorly specific early marker of AKI.

When compared with those 108 study patients who did not develop an early serum creatinine rise episode, the 279 patients who did were significantly more acutely ill by several measures: APACHE II and III scores, need for mechanical ventilation, length of stay in ICU, and mortality rate. The authors were able to conclude that small creatinine increase episodes are markers not only of AKI but also illness severity.

They suggest that refinement of the frequent serum creatinine measurement approach adopted for this study along with simultaneous assessment of urine output could improve early prediction of AKI. This could lead ultimately to earlier treatment of AKI and thereby better outcome for many critically ill patients.