Contrast-induced acute kidney injury

Epidemiology

CI-AKI is often mentioned as the most common iatrogenic cause of AKI and overall the third most common cause of AKI in the hospital setting based on old and outdated literature. ^, The overall incidence of AKI requiring dialysis has been gradually increasing, but this appears largely driven by the increasing comorbidity in the population rather than more contrast use. The actual incidence of AKI after contrast-enhanced imaging has been reported to vary from as low as 1% to as high as 30% and depends on the nature of the contrast administered and the underlying risk factors in the patient population. In addition, these estimates are clouded by the fact that AKI after contrast administration can often be caused by other etiologies (e.g., acute tubular necrosis from cardiogenic shock in a patient with acute coronary syndrome undergoing coronary angiography or severe sepsis in a patient undergoing a contrast-enhanced computed tomography [CT] scan, or atheroembolic renal disease), ^, leading some researchers to use the term contrast-associated AKI or postcontrast AKI .

The risk of CI-AKI varies with the route of administration. Typically, after elective coronary angiography, about 10%–15% of patients may develop AKI (as defined by the rise in creatinine), although the incidence of severe AKI requiring dialysis is much lower, at less than 1%. The incidence of AKI after intravenous administration of contrast (as with contrast-enhanced CT scans) is believed to be much lower. A prospective study reported this incidence at 2.5% overall, with the risk sequentially increasing as the underlying baseline kidney function decreases. However, other studies have questioned whether the true incidence of contrast nephropathy after intravenous contrast is this high, given that there are underlying fluctuations in serum creatinine levels. In the last decade, large epidemiologic studies that incorporated propensity-matched controls who did not receive contrast made a reasonable case that there is little additional increase in AKI after intravenous contrast administration when adjusted for the underlying baseline risk of AKI. However, despite the fact that confounding can be accounted for in these analyses, selection bias, in addition to residual confounding cannot be excluded. As an example, patients perceived to be at high risk of AKI according to the treating clinician may not receive contrast and hence may be overrepresented in the controls in such studies. As an example, in one such analysis, contrast administration seemed to be associated with a lower risk of AKI, which is an implausible result merely reflecting underlying selection bias, as mentioned earlier. Nevertheless, the totality of evidence suggests that the risk of true CI-AKI is low in most patients, especially after intravenous contrast and with preserved kidney function.

The most important underlying risk factor for the development of CI-AKI is compromised baseline kidney function. The incidence is less than 2% in the unselected general population but has been reported to be as high as 20%–30% with the addition of decreased kidney function and other risk factors. The incidence of CI-AKI increases in a graded manner as the severity of the underlying kidney function worsens. Among patients undergoing percutaneous coronary intervention, the risk of CI-AKI has been reported to be just under 20%, with a baseline glomerular filtration rate (GFR) of less than 60 mL/min/1.73 m ² , compared with less than 1% in unselected populations. These numbers mostly describe AKI, not necessarily CI-AKI. With IV contrast, there are mixed data on the difference in AKI with lower GFR. ^, Among critically ill patients, one propensity score–matched study suggests that intravenous contrast (CT) was associated with higher AKI and need for dialysis among patients with GFR <45 mL/min/1.73 m ² (odds ratio [OR] 2.72, 95% confidence interval [CI] 1.14–6.46), though with a very low absolute risk increase of 4.2%. However, no such effect was seen in another study in critically ill patients with sepsis.

Other risk factors for CI-AKI described in the literature include diabetes, older age, cirrhosis, proteinuria, and other comorbid conditions, including congestive heart failure, hypotension, volume depletion, and concomitant administration of other nephrotoxic agents (e.g., nonsteroidal antiinflammatory drugs). It is unclear if some of these risk factors constitute an additional risk after adjusting for the underlying GFR. In addition, some of these risk factors, such as hypotension and congestive heart failure, are risk factors for AKI, mostly from ischemic acute tubular necrosis (ATN), irrespective of contrast administration. Risk scores that include these factors identify patients at risk of AKI from ATN, not necessarily CI-AKI, and hence should be interpreted with caution.

Metformin is often considered to be a risk factor for CI-AKI (because the package insert advises withholding it for at least 48 hours before contrast imaging), but this is untrue. In patients who develop AKI, there is a higher incidence of metformin-associated lactic acidosis, which is associated with a high fatality rate. However, metformin-associated lactic acidosis only occurs in a fraction of those patients who are not only taking metformin but also either have unstable kidney function at baseline or do develop severe CI-AKI and in whom the recognition of kidney failure does not result in discontinuation of the metformin. ^,

Among patients who are already dialysis dependent, either on peritoneal dialysis (in which they often do have significant residual kidney function) or on hemodialysis, contrast imaging does not cause a decline in the residual kidney function. There is no role for early or intensive dialysis to remove contrast material in these patients.

Pathophysiology

Decreased renal blood flow, tubular cell damage, and tubular obstruction are the most commonly described pathways to AKI occurring after contrast administration. Coronary angiography and left ventriculography have been shown to cause a decrease in renal blood flow measured directly using renal artery catheterization. In addition, animal data suggest differential vasoconstriction of afferent more than efferent arterioles, causing a direct effect on decreasing GFR. The vasoconstriction also occurs in the renal medulla via decreased blood flow in the vasa recta. Tissue hypoxia then results in free radical release, leading to oxidant damage to the tubular cells from reactive oxygen species. Contrast agent uptake in tubular cells has been reported to be via the brush border enzyme dipeptidase-1, and CI-AKI may be related to volume depletion and dependent on resident renal phagocytes, interleukin-1, and leucocyte recruitment. Tubular filtration of relatively higher osmolar contrast media also results in osmotic diuresis, increasing medullary oxygen consumption and exacerbating the medullary hypoxia. Additionally, reabsorption of water leaves a high concentration of viscous contrast material in the tubules, which can result in intratubular physical obstruction.

Clinical features and diagnosis

Patients with CI-AKI are generally asymptomatic but have an acute rise in serum creatinine concentration 24–72 hours after the administration of the contrast agent. Kidney failure is usually nonoliguric in mild cases, but it may be oliguric, especially if there is significant preexisting renal impairment. Clinically significant deterioration is unlikely if the serum creatinine concentration does not increase by more than 0.5 mg/dL within 24 hours. To make an unequivocal diagnosis of CI-AKI, other potential causes of AKI must be ruled out. Prerenal factors, atheroembolic disease, and other nephrotoxic insults should be excluded. ^, The relatively rapid onset and typical course may help differentiate CI-AKI from other causes of AKI. Urinalysis may be unremarkable or may show granular casts, tubular cells, or small amounts of proteinuria. Fractional excretion of sodium is usually low and is unhelpful in differentiating CI-AKI from prerenal, volume-responsive causes of AKI. Because this is a clinical diagnosis, the criteria in Table 102.1 may be helpful to determine if an AKI episode should be attributed to contrast.

Prognosis

Usually, the natural course of CI-AKI is peak creatinine (i.e., reflecting the lowest point of kidney function) occurring between 24 and 72 hours after contrast administration followed by relatively rapid improvement over the next few days, back to baseline serum creatinine levels by 7–14 days. Overall, less than 1% of patients with CI-AKI will develop kidney failure that requires dialysis, and a smaller proportion of such patients (estimated at 10%–50%) will remain dialysis-dependent. The minority that do remain dialysis-dependent consist of a mixture of patients who had true CI-AKI along with those who had atheroembolic disease or other causes of AKI that often occur in the population. ^, Nevertheless, despite the fact that most patients with CI-AKI recover, a large body of literature has emerged showing that an episode of AKI is associated with poor long-term outcomes, with a faster decline in kidney function and higher rates of subsequent RRT requirement, in addition to higher rates of hospitalization for heart failure and all-cause mortality. ^,

Although the association of CI-AKI with adverse clinical outcomes has been clearly and consistently shown, it is not yet known whether CI-AKI is on the causal pathway to these outcomes or if it is merely a marker of patients who are at high risk of these events. If the latter is true, CI-AKI may indeed be a less important health issue. Future trials using a variety of interventions with different mechanisms of action showing parallel diminution in CI-AKI and adverse events are required to establish more robust evidence for causality.

Prevention

The most effective method of preventing CI-AKI is to not give iodinated contrast unless absolutely essential, especially in patients at high risk, such as those with advanced kidney disease. Unfortunately, the risks with contrast-enhanced magnetic resonance imaging in this patient population (i.e., the risk of nephrogenic systemic fibrosis from gadolinium) also limit the imaging options, though these risks are much lower with the currently used group II gadolinium compounds. Given the elective nature of the nephrotoxic insult that allows for attempting prophylaxis, many different interventions have been tested for CI-AKI prevention. Many of the studies are contradictory, and the numbers of systematic reviews and meta-analyses are also quite high, so the reader needs to look at the entire body of literature in order to interpret the data.

Fluid administration