Acute kidney injury revisited


Acute kidney injury (AKI) is defined as a reduction of renal function over a short period of time (typically within 7 days or less) as measured by a rising serum creatinine level or by diminished urine output. The term AKI is synonymous with “acute renal failure,” but the former has become the preferred clinical terminology.

For many years, there had been much confusion and controversy around the diagnostic standard for AKI. As of 2002, at least 35 different definitions of acute renal failure had been published in the medical literature. Several conflicting diagnostic standards have been proposed, such as RIFLE in 2004 and AKIN in 2007, but none have been generally accepted as definitive.

Photo by Thinkstock
Photo by Thinkstock

Finally, in 2012, the Kidney Disease: Improving Global Outcomes (KDIGO) Clinical Practice Guideline for Acute Kidney Injury was published and endorsed by the National Kidney Foundation. KDIGO is now the authoritative, professional consensus standard for the diagnosis of AKI, superseding and replacing other criteria, including RIFLE and AKIN. The KDIGO definition of AKI encompasses all causes: prerenal (most commonly due to dehydration), renal (including such things as acute tubular necrosis [ATN] and glomerulonephritis), and postrenal (typically urinary obstruction).

KDIGO defines AKI as the occurrence of any 1 of the following:

  • an increase in serum creatinine level by 0.3 mg/dL or more, measured prospectively by at least 2 separate levels obtained within 48 hours, or
  • an increase in serum creatinine level to 1.5 times baseline or greater, which is known or presumed to have occurred within the prior 7 days, or
  • a urine volume of less than 0.5 mL/kg/h for 6 hours or longer.

Absolute creatinine levels are irrelevant to the diagnosis of AKI; only the relative changes specific to each individual patient matter. For example, a patient who has an initial creatinine level of 0.5 mg/dL that increases to 1.0 mg/dL has AKI. Even though 1.0 mg/dL is within the so-called “normal” reference range, it is not normal for this patient, whose baseline is 0.5 mg/dL. In this case, the patient has actually lost 50% of his renal function, indisputably a serious situation. Alternatively, a patient with chronic kidney disease (CKD) who has a baseline creatinine level of 3.0 mg/dL that increases to 4.0 mg/dL over 5 days does not have AKI because the increased level was not 1.5 times the baseline.

KDIGO also provides advice for the proper interpretation and application of these criteria. The criterion of a 0.3-mg/dL increase is strictly defined as the difference between 2 measurements obtained within a 48-hour period. No presumption of a pre-existing baseline level may be made. The criterion of 1.5 times baseline is more broadly interpreted, allowing for reasoned clinical judgement. It can be applied prospectively if a patient has an increase in the creatinine level of 1.5 times within 7 days. For example, a patient admitted with heart failure and a creatinine level of 1.2 mg/dL is given IV furosemide. A subsequent creatinine level 3 days later is 2.0 mg/dL. This represents AKI, since 2.0 is more than 1.5 times the measured baseline of 1.2 (1.2 × 1.5 = 1.8).

How is the criterion of 1.5 times baseline used to retrospectively diagnose AKI? Take the common situation of a patient who has had an acute illness within the preceding 7 days and is admitted with an elevated creatinine level of 2.0 mg/dL and unknown baseline. This could be AKI, if baseline creatinine level was 1.3 mg/dL or less, or it could be CKD with a chronically elevated creatinine level.

The question will be answered as the patient is hydrated. If the creatinine level remains elevated near 2.0 mg/dL, then the patient has CKD. If the creatinine decreases, we can compare subsequent creatinine levels retrospectively to the admission level. Because KDIGO directs us to consider the lowest creatinine level reached during admission as a baseline (for example, if it dropped to 1.2 mg/dL in this case), we can then retrospectively compare this “baseline” of 1.2 mg/dL to the admission level of 2.0 mg/dL to see if the patient met the diagnostic criteria for AKI on admission. Once again, this does represent AKI since the admission creatinine level of 2.0 mg/dL, presumed to have occurred during the preceding acute illness of 7 days or less, is more than 1.5 times the baseline of 1.2 mg/dL.

Finally, KDIGO notes that in patients without CKD or particularly serious intercurrent illnesses, a known creatinine level from 6 months or even 1 year earlier may be considered the baseline.

One of 4 diagnostic terms must be documented to assign the correct code for AKI. AKI is the preferred professional term, but “acute renal failure,” “acute kidney failure,” or “acute renal injury” will result in the same code as AKI. Any other terminology, including “renal insufficiency” and “pre-renal azotemia,” will result in the assignment of incorrect codes that do not reflect the seriousness or significance of the patient's condition. The same is true if the word “acute” is omitted.

It is also particularly important to identify ATN when it is the confirmed or suspected cause of AKI. ATN is the cause of about one-third of AKI among inpatients and is assigned a higher level of severity. ATN may result from progression of prerenal AKI, commonly caused by volume depletion, dehydration, or edematous states. ATN is frequently associated with hypotension, especially when severe or prolonged as with sepsis, severe hemorrhage, and “shock” states, and it is the usual cause of AKI associated with many medications. IV contrast-induced AKI is nearly always due to ATN.

Even though the word “necrosis” has been retained, the current clinical concept of ATN encompasses transient tubular dysfunction lasting 72 hours or more without structural damage or necrosis. Urinalysis is typically unremarkable. Only in extremely severe cases with necrosis can renal epithelial cells and “muddy brown” casts be identified.

Functional ATN can be recognized clinically by the response to effective fluid resuscitation and hydration. With ATN, the elevated creatinine level is expected to require at least 72 hours to recover toward baseline; with “prerenal” AKI, particularly volume depletion or dehydration, the creatinine level typically returns to baseline in about 24 to 48 hours. If measured, urinary sodium concentration is typically above 40 mEq/L, and the fractional excretion of sodium is usually (but not always) above 2%.

KDIGO criteria are the current, authoritative consensus standards for the diagnosis of AKI, requiring any 1 of the following: a prospective increase in serum creatinine level by 0.3 mg/dL or more within 48 hours; an increase in serum creatinine level to 1.5 times baseline or more, applied prospectively or retrospectively, and known or presumed to have occurred within a 7-day period; or a urine volume of less than 0.5 mL/kg/h for 6 hours or more.

AKI is the preferred professional term, but “acute renal failure,” “acute kidney failure,” or “acute renal injury” will result in the same code as AKI. Otherwise, the correct code will not be assigned, underreporting the severity of illness connected with AKI. ATN is a common inpatient cause of AKI and has a distinct code and greater severity of illness than a simple diagnosis of AKI. ATN as the cause of AKI may be distinguished clinically from prerenal AKI when it takes 72 hours or longer following initiation of treatment for the creatinine level to recover toward baseline. Contrast-induced AKI is almost always due to ATN.