Download this article's Factoid in PDF (& PPT for Gold Subscribers)

Over almost a decade, the incidence of acute kidney injury (AKI) decreased in patients with acute myocardial infarction (AMI) despite the rising prevalence of AKI risk factors, according to a new study published in the February 13, 2012, issue of the Archives of Internal Medicine. The findings may reflect increased clinician awareness of the problem, better risk stratification, or greater use of AKI preventive measures, the authors suggest.

Mikhail Kosiborod, MD, of Saint Luke’s Mid America Heart Institute (Kansas City, MO), and colleagues analyzed data from a contemporary registry of 33,249 consecutive hospitalizations and 31,532 AMI patients across 56 US centers from 2000 to 2008. AKI was defined as an absolute increase in creatinine level of at least 0.3 mg/dL or a relative increase of at least 50% during hospitalization.

Over the study period, average patient age increased from 66.5 to 68.6 years (P < 0.001). Other AKI risk factors became more prevalent as well (P ≤ 0.01 for each), including:

• Chronic kidney disease (3.9% to 12.7%)
• Cardiogenic shock (4.3% to 5.7%)
• Diabetes (30.3% to 35.1%)
• Heart failure (29.8% to 32.7%)
• Coronary angiography (59.0% to 70.0%)
• PCI (32.1% to 47.0%)

Despite these increases in risk, AKI incidence lowered from 26.6% in 2000 to 19.7% in 2008 (P < 0.001), representing an absolute decrease of 6.9%. After adjustment for trends in potential confounders and practice pattern changes over time, a 4.4% decline in AKI was seen per year (95% CI 2.0-6.8; P < 0.001).

The decrease in AKI incidence was most pronounced among patients undergoing cardiac catheterization (24.6% to 16.5%; P for trend < 0.001), with only a slight decline for those treated conservatively (29.4% to 27.0%; P for trend = 0.66). After adjustment for hospital site and other confounders, a decrease in AKI was seen in both groups, although the difference remained more substantial in catheterized patients (5.6% for cardiac catheterization; P = 0.001; 3.3% for conservative treatment; P = 0.01).

The study also looked at temporal trends in the use of medications potentially related to AKI development and found that while diuretic use decreased over time (56.4% to 47.0%; P < 0.001), use of N-acetylcysteine (NAC) increased (0.6% to 10.6%; P < 0.001). After adjustment, NAC was the only medication related to AKI risk that increased in use over time (OR 1.19; 95% CI 1.01-1.40; P = 0.04).

Substantial variation in AKI incidence among hospitals was observed, ranging from 10% to 32%, and this remained after multivariate adjustment (median OR 1.26). Decline in AKI incidence was the same regardless of the length of time a hospital participated in the registry or patient length of stay. Also, incidence of severe AKI decreased 5.2% per year (95% CI 1.8-8.4; P < 0.001).

In terms of outcomes, in-hospital mortality decreased over time (19.9% to 13.8%; P = 0.003), persisting after adjustment (OR 0.96 per year; 95% CI 0.93-0.98; P = 0.004).

Processes of Care Important

Dr. Kosiborod told TCTMD in a telephone interview that he was surprised by the magnitude of decline in AKI rates, especially with risk factors increasing over time. While this is a “very positive message,” he said the data only give a “glimpse into what the possible reasons for this [trend] could be.”

It comes down to differences in processes of care, he continued, and the data show this in 3 ways. First, although NAC use has increased substantially over time, “it’s really a surrogate marker for paying greater attention to preventing kidney injury,” he said, adding that previous data suggest it is not a “particularly effective” treatment for preventing AKI. “Typically, when somebody thinks of using N-acetylcysteine, it’s likely that they would use other measures such as preprocedural hydration or thinking about reducing contrast volume and potentially avoiding nephrotoxic medications,” he observed.

On a similar note, Dr. Kosiborod said that although patients undergoing invasive procedures are at greater risk and in fact experience more AKI, clinicians nonetheless have more opportunities for reducing AKI risk in that setting than when giving conservative treatment.

Lastly, the substantial variation in AKI incidence observed among hospitals suggests differences in processes of care. This is “illuminating,” he said, “because that means there are opportunities to improve quality of care and potentially further reduce AKI rates if we learn from hospitals that do it really well and have very low rates of AKI. If we try to understand what they do in addition to standard preventative measures, that can inform practice at other hospitals.”

How Much Variation Is Too Much?

In an accompanying editorial, Raymond K. Hsu, MD, and Chi-yuan Hsu, MD, MSc, both of the University of California, San Francisco (San Francisco, CA), observe that the study was strengthened by the adjustment for demographic variables, comorbidities, and surveillance patterns for AKI as well as sensitivity analyses.

Even so, Hitinder S. Gurm, MD, of the University of Michigan Medical School (Ann Arbor, MI), told TCTMD in a telephone interview that he is concerned about the large variation in AKI incidence among hospitals.

“The difference is a little too extreme to pass the smell test. When [a hospital] has an incidence of less than 10%, I wonder about ascertainment bias,” he said, adding that some of the difference might be due to patient-level variation, although he agreed that clinicians might have something to learn from hospitals with a low AKI incidence.

In response, Dr. Kosiborod maintained that “while some patient bias may persist despite our very careful best efforts to try to minimize confounding, I think most of that variation is likely due to the processes in care and not differences in patients.”

More Data Needed to Learn What Hospitals Do Right

In a telephone interview with TCTMD, Richard Solomon, MD, of the University of Vermont (Burlington, VT), said that the results are “believable” because more and more cardiac cath labs have begun to enact protocols to identify high-risk patients and follow procedures to minimize AKI risk. However, he would like to see these findings confirmed in larger registry studies and potentially look more closely at which processes of care directly influence AKI incidence.

“[This study] shouldn’t be interpreted that we no longer have to worry about AKI,” he said. “The changes reported here are a reflection of the fact that we are more aware and more concerned about AKI, and that’s translated into reducing incidence. We ought to continue that level of heightened awareness.”

Dr. Kosiborod agreed. The study suggests that cardiologists’ efforts to prevent AKI “seem to be producing positive results,” he said, adding, “I think there is a lot more we can do to make sure that these trends continue to go in the right direction.”

2. Hsu RK, Hsu C-Y. Acute kidney injury: Glimpses into epidemiology and opportunities for improvement. Arch Intern Med. 2012;172:253-254.

Related Stories:

• Iodinated Contrast Agents Tied to Potential Thyroid Dysfunction
• Acute Kidney Injury After Angiography Increases Long-Term Complications
• ACT Published: Definitive Results Nix NAC for Preventing Kidney Damage