Is Computed Tomography with Contrast or Magnetic Resonance Imaging with Contrast Preferred in Patients with Chronic Kidney Disease?

This is a clinical question that keeps coming up on nephrology rounds. Radiocontrast-induced nephropathy is a real concern in patients with risk factors for this complication. The most important risk factor is the presence and degree of underlying chronic kidney disease (CKD). In the past, magnetic resonance imaging (MRI) was generally chosen as an alternative to computed tomography (CT) when kidney disease was present, as it was thought that MRI contrast agents, including gadolinium, were safe in CKD. This all changed with the discovery of nephrogenic systemic fibrosis (NSF) in the late 1990s and the probable role of gadolinium in its pathogenesis, as gadolinium deposits have been found in tissue specimens from patients with NSF. NSF is relatively rare but serious disease, with an unremitting course and high morbidity and mortality (Mendoza et al., 2006).

RADIOCONTRAST NEPHROPATHY

Epidemiology

Radiocontrast nephropathy is a generally reversible form of acute kidney injury (AKI) that occurs soon after the administration of radiocontrast media (Rudnick et al., 1994). As the pathogenesis is difficult to study in
humans, most of our understanding of the mechanisms of radiocontrast nephropathy derives from animal models. AKI results from a combination of renal vasoconstriction resulting in medullary hypoxia (possibly mediated by alterations in nitric oxide, endothelin, and/or adenosine) and direct cytotoxic effects of the contrast agents (Persson et al., 2005). Inhibitors of both nitric oxide and prostaglandins cause marked ischemia and tubular necrosis in the medullary thick ascending limb of the loop of Henle in rats exposed to radiocontrast (Agmon et al., 1994). The outer medulla appears particularly susceptible to contrast-induced reduction in renal blood flow because of the relative hypoxic conditions in this nephron segment (Heyman et al., 2005). Direct tubular injury is associated with generation of oxygen free radicals, which has led to numerous studies evaluating a possible apparent protective effect of N-acetylcysteine (NAC) as well as trials of bicarbonate therapy to prevent nephropathy.

Incidence

The reported incidence of radiocontrast-induced nephropathy varies widely, depending on the presence or absence of patient risk factors, the most important being CKD. Other patient risk factors include age, diabetes, decreased renal blood flow from any cause (such as hypovolemia, hypotension, heart failure, cirrhosis), and possibly multiple myeloma (Table 14.1). The most important modifiable risk factors are dose of contrast and the presence of hypovolemia. In patients with no risk factors, for example, a young nondiabetic patient, the risk of radiocontrast nephropathy is negligible (<1%), but very high-risk patients, for example, an older diabetic patient with severe CKD, the risk may be 30% or greater.

TABLE 14.1 Risk Factors for Contrast-Induced Nephropathy

Age

CKD

Diabetes, if CKD present

Decreased renal blood flow from any cause (hypovolemia, hypotension, heart failure, cirrhosis, non-steroidal anti-inflammatory drugs)

Multiple myeloma?

Hyperglycemia (without diabetes)

CKD and Diabetes

The increase in risk with CKD is likely linearly related to the glomerular filtration rate (GFR). Among patients with CKD, diabetic patients are at higher risk for radiocontrast nephropathy compared with nondiabetic patients. In one prospective study, in patients with CKD (serum creatinine >1.7 mg/dL [150 µmol/L]) undergoing CT scan with contrast, diabetic patients had a greater than twofold higher incidence of radiocontrast nephropathy (8.8% vs. 4.0%, respectively) (Parfrey et al., 1989). Similarly, in an analysis of data from a randomized trial that included 250 patients with serum creatinine more than 1.5 mg/dL (133 µmol/L) who received iohexol during percutaneous coronary interventions, diabetic patients had a much higher incidence than controls (33% vs. 12%, respectively); however, diabetes did not increase the risk in patients without CKD (Rudnick et al., 1995).

Hyperglycemia

The degree of hyperglycemia may increase the risk for radiocontrast nephropathy in nondiabetic patients. In a study of 6,358 patients undergoing angiography following myocardial infarction, the adjusted risk of radiocontrast nephropathy increased as glucose levels increased in nondiabetic patients, with greater than twofold risk when blood glucose was more than 200 mg/dL (Stolker et al., 2010). Interestingly, an association between hyperglycemia and risk of contrast nephropathy was not seen in diabetic patients.

Type of Radiocontrast Agent

It is clear that first-generation hyperosmolal ionic contrast agents increase the risk of AKI compared to nonionic low-osmolal or iso-osmolal agents (Rudnick et al., 1995). However, it is less clear whether there is any difference between low-osmolal and iso-osmolal agents, the latter of which is considerably more expensive. This is discussed further later in the chapter.

Specific Radiologic Procedure

The risk of radiocontrast nephropathy in patients undergoing percutaneous angiography, particularly coronary angiography, is substantially higher than that with contrast CT scans, though the dose used is generally higher as well (Weisbord et al., 2008). The incidence of AKI may also be higher for patients who receive contrast for a nonelective versus elective contrast CT (Mitchell et al., 2010). In a prospective study of 633 outpatients who underwent contrast CT in the emergency department of a tertiary care center, of whom only 2.4% had eGFR less than 60 mL/min/1.73 m², 11% developed AKI (defined as an increase in serum
creatinine >0.5 mg/dL [44 µmol/L] or >25% within 2-7 days after contrast administration). Six patients (1%) developed severe AKI (increase in serum creatinine ≥3 mg/dL or need for dialysis). The relatively high incidence of AKI in this study may be due to lack of adjustment for comorbidities such as hypotension, hyperglycemia, or volume depletion, but does point to the possibility of increased risk in an emergency room population.

Clinical Features

Clinical manifestations such as oliguria or an increase in the serum creatinine are generally observed within 24 to 48 hours after contrast exposure. The majority of patients are nonoliguric, so the severity of AKI is usually determined by the extent of increase in the serum creatinine. Unlike other types of AKI, radiocontrast nephropathy is generally characterized by relatively rapid recovery of renal function, with the serum creatinine starting to decline within 3 to 7 days (Rudnick et al., 1994). The urinary sediment may show classic findings of acute tubular necrosis (ATN), that is, pigmented “muddy-brown” granular casts. If contrast is still present in the urine, it will cause a markedly increased specific gravity (˜1.050) because of the heavy weight of iodine (the effect of contrast on osmolality is less marked). The fractional sodium excretion (FENa) is often less than 1% in patients with radiocontrast nephropathy, particularly in its early phase. This may indicate intrarenal vasoconstriction and/or be a reflection of the nonoliguric state, leading to a lower urinary sodium concentration.

The differential diagnosis includes other causes of renal tubular injury (ischemic or toxic), acute interstitial nephritis, renal atheroemboli (after angiography), and prerenal failure. Renal atheroembolism after angiography should be suspected if there are other embolic lesions (such as blue toes or livedo reticularis), transient eosinophilia and/or hypocomplementemia, delayed development of AKI, or failure to recover renal function.

Radiocontrast nephropathy generally has a good prognosis. Most patients recover renal function without need for dialysis. However, some patients, especially diabetic patients with severe underlying CKD (i.e., eGFR <30 mL/min/1.73 m²), may require dialysis, which may be permanent in a small percentage of cases.

PREVENTION OF RADIOCONTRAST NEPHROPATHY

Type and Amount of Contrast Agent

The renal toxicity of contrast agents appears to be related to their osmolality. Ionic high-osmolal first-generation agents were associated
with a high rate of nephrotoxicity and are no longer employed. Second-generation agents, such as iohexol, are nonionic monomers with a lower osmolality than high-osmolal radiocontrast media; however, they still have an increased osmolality compared with plasma. In addition, there is an ionic low-osmolal contrast agent (ioxaglate). The newest nonionic contrast agents are iso-osmolal, being dimers with an osmolality of approximately 290 mmol/kg (iodixanol, the first such agent, is available in the United States).

A variety of preventive measures may reduce the risk of contrast nephropathy (Asif and Epstein, 2004; Pannu et al., 2006). The first consideration is to avoid contrast entirely by use of an alternate imaging method such as ultrasonography, MRI, or CT scanning without radiocontrast agents. If contrast is essential, use of lower doses of contrast (Cigarroa et al., 1989; Marenzi et al., 2009) and avoidance of repetitive studies that are closely spaced (within 48-72 hours) decrease the risk. The relationship between dose and toxicity is demonstrated by the observation that very small amounts of radiocontrast (<10 mL) can be safely used in patients with very severe CKD for examination of poorly maturing arteriovenous fistulae (Kian et al., 2006). Reduction of contrast dose during coronary angiography can be achieved by avoiding ventriculography and the use of endovascular ultrasound and during peripheral vascular contrast procedures with the use of carbon dioxide angiography (usually in combination with small amounts of iodinated contrast).

Only gold members can continue reading. Log In or Register to continue