Meta-Analysis Questions Strength of Link Between AKI and Mortality

Contrast-induced acute kidney injury (AKI) is linked to heightened mortality risk after angiography, but that relationship may be much weaker than has been previously suggested. New findings, from a meta-analysis published online January 15, 2013, ahead of print in Circulation: Cardiovascular Interventions, suggest that baseline differences may instead predispose certain patients to both AKI and death.

Matthew T. James, MD, PhD, of Foothills Medical Centre (Calgary, Canada), and colleagues pooled 39 observational studies published between 1990 and 2011 that compared outcomes between patients with and without AKI after coronary angiography.

Adjustment Lessens Effect

Among the 34 studies that reported mortality data, 33 showed increased risk of death in conjunction with AKI. However, the effect size varied. Studies without adjustment (n = 11) showed a much stronger association between AKI and death (RR 8.19; 95 CI 4.30-15.60) than did those (n = 23) that took potential confounders into account (adjusted RR 2.39; 95% CI 1.98-2.90).

Meta-regression analyses of the studies reporting adjusted data found only 1 explanation for the heterogeneity in mortality risk with AKI: those with follow-up durations only lasting until hospital discharge or 30 days reported greater risk compared with those offering follow-up of at least 6 months. Risk did not vary by definition of AKI, loss to follow-up, or study distributions of age, diabetes, baseline chronic kidney disease, STEMI, or whether angiography was diagnostic or interventional.

All of the 14 studies reporting cardiovascular events showed an increased mortality risk associated with AKI (crude RR 2.59; 95% CI 1.05-6.27 and adjusted RR 1.98; 95% CI 1.52-2.59).

Three studies reporting on progression to end-stage renal disease also found an effect (crude RR 15.26; 95% 1.86-125.01 and adjusted RR 6.95; 95% CI 2.51-19.26). Long-term worsening of kidney function occurred at rates ranging from 0 to 0.2% in patients without AKI and 0.2% to 4.5% in those with AKI.

In addition, 10 studies with unadjusted data found hospital stays ranged from 0.5 to 8.3 days longer in patients with AKI, while the 1 study with adjusted data reported an extra 1.6 days of hospitalization.

Winnowing Down the Confounders

Peter A. McCullough, MD, MPH, of St. John Providence Health System (Warren, MI), told TCTMD in a telephone interview that the association between AKI and mortality “is very heavily confounded. The patients who have changes in renal function after catheterization and angioplasty are definitely older, sicker, and more likely to have diabetes and baseline kidney disease going into the procedure.” Even so, the relationship has consistently been observed, he added.

Also in a telephone interview with TCTMD, Dr. James said that the investigators “were struck by the large degree of variability in risk associated with contrast-induced kidney injury,” much of which could be explained “by the degree to which studies adjusted for confounders.”

The remaining association, “is not insignificant,” he noted. “It’s still around a twofold increased risk even when studies adjust for these features.” The fact that these studies were observational, he added, means that there is the possibility that residual confounding or bias might reduce the effect of AKI even further.

According to Dr. McCullough, the current meta-analysis represents probably the largest one to date on the topic of AKI and outcome. Importantly, it “gives us more insight into the development of end-stage renal disease,” he commented. “We always worry that if we have a patient with kidney disease and we do an elective cardiac procedure, that we’re really pushing them down the road to end-stage renal disease. In fact, that’s the case.” Certainly some would have progressed even without AKI, he added, but patients need to be informed of the risk.

Gains Despite No ‘Home Runs’

“Fortunately, the rates of kidney injury are coming down,” Dr. McCullough explained, “because we are using less dye and using dye that’s probably safer. We’re doing more approaches through the radial artery, so there’s less opportunity for catheters to brush off atherosclerotic material into the kidneys when we do the catheter exchanges. . . . In [patients] who are older and sicker, we tend to avoid dehydration and give them IV fluids.”

“But our big home runs—like protective medicine, prophylactic agents, and special catheters—all those things have failed,” he noted, reporting that there is progress toward development of dyes with less or no renal toxicity.

Another idea when managing at-risk patients, Dr. James added, is to “stop any medications that could be toxic to the kidneys.”

However, “we don’t have any studies right now telling us what we can do in the long term to prevent these long-term outcomes. Most of us would think that medications to prevent chronic kidney disease progression in patients who have residual impairment of kidney function, [such as] ACE inhibitors in people with proteinuria, would be important,” while patients with known cardiovascular disease could be protected from future events by medical therapy such as statins, antiplatelet agents, and beta-blockers, Dr. James suggested. “But we really don’t have evidence from trials yet to say that this is going to change the association [between acute damage and long-term risk].”

 

---

Source:
James MT, Samuel SM, Manning, MA, et al. Contrast-induced acute kidney injury and risk of adverse clinical outcomes after coronary angiography: A systematic review and meta-analysis. Circ Cardiovasc Interv. 2013;Epub ahead of print.

 

 

Related Stories:

• Acute Kidney Injury After Angiography Increases Long-Term Complications
• Contrast-induced Acute Kidney Injury Remains Major Safety Issue
• Contrast-Induced Nephropathy Linked to Long-Term Adverse Events