Postcontrast Acute Kidney Injury After Transarterial Chemoembolization

Intravascular (IV) administration of contrast media (CM) is commonly utilized to enhance the capabilities of imaging modalities commonplace in both diagnostic and therapeutic interventions. There are numerous iatrogenic reactions associated with CM administration that have been described in literature, ¹ broadly categorized as allergic and nonallergic. Postcontrast acute kidney injury (PC-AKI) is a subset of nonallergic CM reactions and is especially important in patients with an underlying medical condition such as chronic kidney disease (CKD). This may be exacerbated by IV administration of CM during long interventional procedures such as angiography. In this article, a case of PC-AKI is presented followed by a discussion of the topic. In addition, guidelines for prevention and treatment of PC-AKI are presented.

Case Presentation

A 58-year-old man with a past medical history of CKD and hepatitis C–induced cirrhosis complicated by hepatocellular carcinoma (HCC) presented to interventional radiology (IR) clinic for discussion of liver-directed therapy. The patient had a history of esophageal varices and hydrothorax status post transjugular intrahepatic portosystemic shunt placement 6 years previously. On clinical presentation, he did not have encephalopathy, and had a good performance status (Eastern Cooperative Oncology Group performance status 0). Laboratory analysis and physical examination revealed compensated liver function (Child–Pugh class A) and Barcelona Clinic Liver Cancer (BCLC) stage B disease. His baseline serum creatinine was 2.09 mg/dL, and his baseline glomerular filtration rate (GFR) was 35 mL/minute. A magnetic resonance imaging (MRI) study of the abdomen demonstrated a 2.2-cm observation in segment 8 of the liver with characteristic arterial enhancement and delayed washout ( Fig. 1 ) classified as Liver-Imaging Reporting and Data System 5 (LI-RADS 5). The patient was scheduled for a conventional transarterial chemoembolization (cTACE).

Fig. 1.

Axial precontrast T1-weighted ( a ) and T2-weighted ( b ) magnetic resonance images demonstrate a 2.2 × 2.1 cm slightly hyperintense observation in segment 8 of the liver (arrows). The observation shows early arterial enhancement (arrows) ( c ), enhancing capsule and nonperipheral washout (arrows) ( d ) on axial fat-saturated T1-weighted images in the early arterial ( c ) and delayed ( d ) phases, which are characteristics of hepatocellular carcinoma.

The patient returned to IR approximately 3 weeks later to undergo treatment of the segment 8 lesion. He was pretreated with cefazolin 2 g IV before the procedure. Arterial access was obtained into the right common femoral artery using standard ultrasound guidance and a 5-Fr vascular sheath (Pinnacle; Terumo, Somerset, NJ) was placed. The preprocedure MRI demonstrated a replaced right hepatic artery; therefore, a reverse curve catheter (Sos Omni Selective 2; AngioDynamics, Latham, NY) was used to select the superior mesenteric artery. A microcatheter (Progreat 2.8 French, Terumo, Japan) and a 0.018-inch wire (Fathom, Boston Scientific, Naidich, MA) were used to subselect and perform sequential angiograms of the right hepatic artery branches, confirming a segment 8 hypervascular tumor ( Fig. 2a ). Chemotherapy emulsion—consisting of 20 mL of a 2:1 mixture of ethiodized oil (Lipiodol Ultra Fluide; Guerbet, Villepinte, France) and 10 mL aqueous chemotherapy solution comprising doxorubicin 50 mg and mitomycin C 10 mg—was injected. Intermittent angiograms were obtained. Embolization using one vial of 100 to 300 μm microspheres (Embospheres; Merit Medical, South Jordan, UT) was performed. The postembolization angiography demonstrated dense staining of the treated right hepatic lobe tumor and minimal residual vascularity. All the catheters and wires were removed and access site hemostasis was obtained using a vascular closure device (MynxGrip, Access Closure Inc., Santa Clara, CA). A total of 60 mL of intra-arterial Omnipaque-300 (GE Healthcare, Wauwatosa, WI) was used during the procedure.

Fig. 2.

Digital subtraction images of the superior mesenteric artery angiography ( a ) demonstrate a replaced right hepatic artery with a subtle tumor blush in segment 8 of the liver (arrows). The post–conventional transarterial chemoembolization unenhanced CT of the liver ( b ) confirms dense ethiodized oil staining of the hepatic segment 8 lesion (arrows). A transjugular intrahepatic portosystemic shunt stent is incidentally noted (arrowhead).

The patient tolerated the procedure well, and per institutional protocol an immediate postprocedure unenhanced computed tomography (CT) of the abdomen confirmed ethiodized oil staining of the hepatic segment 8 lesion ( Fig. 2b ). One day after the procedure, routine laboratory evaluation demonstrated acute worsening of the patient's kidney function, with an increase in serum creatinine to 2.9 mg/dL and a potassium level of 6.0 mmol/L. The patient was diagnosed with PC-AKI, and was started on IV 0.9% normal saline 100 mL/hour and 650 mg IV of sodium bicarbonate twice a day. The hyperkalemia was treated with insulin therapy. The patient responded to the treatments and was discharged after 6 days with a creatinine of 2.14 mg/dL and a potassium of 4.8 mmol/L ( Fig. 3 ). His serum creatinine peaked at 3.0 mg/dL on postprocedure day 3. The patient was closely followed up as an outpatient; at his 1-month follow-up visit after cTACE, a partial response of the segment 8 lesion was demonstrated and he was scheduled for further liver-directed therapies.

Fig. 3.

Serum creatinine ( a ), potassium ( b ), and estimated GFR ( c ) levels at baseline, which deteriorated on day 1 (D-1) following conventional transarterial chemoembolization and recovered gradually in response to treatments. K, potassium; GFR, glomerular filtration rate.

Discussion

Definition: Postcontrast Acute Kidney Injury versus Contrast-Induced Nephropathy

Postcontrast acute kidney injury is a term used to define manifestations of acute kidney injury following IV use of CM; the term is often used interchangeably with contrast-associated acute kidney injury (CA-AKI). There are multiple definitions available within the current literature. PC-AKI is a correlative diagnosis, and the criteria for PC-AKI can be met without direct evidence that CM is the primary underlying precipitator. ¹ The American College of Radiology (ACR) defines PC-AKI as the deterioration of renal function within 48 hours of the administration of IV iodinated CM. The ACR and European Society of Urogenital Radiology (ESUR) have both released recommendations regarding diagnostic criteria for PC-AKI. The ACR recommends use of the diagnostic criteria set by the International Society of Nephrology in the Kidney Disease-Improving Global Outcomes (KDIGO) document. The ACR guidelines define PC-AKI as, “An absolute serum creatinine increase of at least 0.3 mg/dL within 48 hours (of contrast administration); a percentage increase in serum creatinine of at least 50% (1.5-fold above baseline) which is known or presumed to have occurred in the past 7 days; or reduction in urine output to 0.5 mL/kg/h for at least 6 hours.” ¹ The ESUR defines PC-AKI similarly as, “…an increase in serum creatinine > 0.3 mg/dl (or > 26.5 µmol/l), or > 1.5 times baseline, within 48–72 hours of intravascular administration of a contrast agent.” ² Although both definitions are correlative and utilize the KDIGO guidelines, the ESUR definition allows for diagnosis to be made up to 72 hours after CM administration. Additionally, changes in urine output are not included. ^{3 4 5}

Contrast-induced nephropathy (CIN) is a subset of PC-AKI, referring to a rapid deterioration of renal function following IV CM which, in contrast to PC-AKI, is a causative diagnosis. The underlying pathophysiology of CIN is not clearly understood, nor is it well differentiated from PC-AKI. CIN is a diagnosis of exclusion and other potential causes of acute renal impairment must be ruled out prior to diagnosis. ³ Based on the ESUR guideline, a threshold increase in serum creatinine within 3 days following CM administration is additionally required to diagnose CIN. ² CIN is now believed to occur very rarely and only within specific predisposed populations. ¹ PCI-AKI and CIN were previously used interchangeably in the literature, which attributed to the previously overestimated incidence of CIN.

Mechanism and Pathophysiology

The presentation of acute kidney injury following IV CM administration is very well-documented. However, recent studies have suggested that CM use alone may pose a much smaller risk of developing PC-AKI and CIN in patients who are not previously predisposed. ⁶ The exact mechanism of CIN is not clearly defined, although there are several hypotheses that have helped elucidate the seemingly multifactorial mechanism in which CM causes nephrotoxicity. Direct tubular toxicity has long been proposed to serve a central role in the manifestation of PC-AKI and CIN. ⁷ Disruption of renal homeostasis at the cellular level, tubular cell necrosis and apoptosis, and structural damage and enzymatic dysregulation of renal vasculature and parenchyma due to accumulation of free iodine in CM have all been proposed as possible causes of nephrotoxicity. ^{7 8} In addition, CM can precipitate renal injury through induction of hypoperfusion and hypoxia, ⁹ which may be induced by prolonged vasoconstriction of the renal vasculature following a brief vasodilatory phase. Conversely, CM has also been described to promote vasodilation of the extrarenal vasculature following an initial vasoconstriction. The combination of the two hemodynamic changes is thought to result in renal hypoperfusion, ischemia, and reperfusion injury of the kidneys. ¹⁰

There has been growing evidence that endothelial dysfunction potentiates the hypoxic renal injury induced by CM. This may occur through an imbalance of physiologic mediators such as endothelin, nitric oxide, and prostaglandins. Endothelin has been linked to the vasoconstrictive effects that CM has on the kidneys. ^{9 11} Multiple studies have linked CM administration with increasing levels of urine and plasma endothelin. ^{12 13} Hyperglycemia has been associated with increased circulating levels of endothelin, which may explain higher incidences of PC-AKI in cases of chronic renal failure caused by diabetes. ^{14 15} Nitric oxide is another vascular mediator serving a major role in maintaining adequate perfusion throughout the renal vasculature, CM may promote vasoconstriction and renal hypoxia through disruption of nitric oxide metabolism. ^{16 17} Inactivation of available nitric oxide and inhibition of nitric oxide synthase have both been proposed. ¹⁸ Prostaglandin E (PGE) and prostacyclin especially have been well described in the literature as serving a major role in maintaining renal perfusion and glomerular filtration. ¹⁹ Like nitric oxide, it has been further hypothesized that CM may downregulate production of prostaglandins, although the mechanism is unclear. ⁹ However, the benefits associated with administration of prostaglandins and discontinuation of nonsteroidal anti-inflammatory drugs (NSAIDs) prior to CM use are unclear.

The nephrotoxic effects of CM osmolality in high-risk patients have remained controversial within the literature. Multiple studies report that high osmolar CM (HOCM) use increases the risk of PC-AKI when compared with low-osmolar CM (LOCM). ²⁰ Enhanced diuresis and perfusion reduction both have previously been linked to variations in osmolality of CM. ²¹ There are mixed results regarding the additional theoretical risk that HOCM holds in the general population; however, diabetic patients appear to be at much greater risk of AKI when exposed to HOCM. This may be partially attributable to the increased adenosine sensitivity of the kidneys that has been observed among diabetics. ^{14 22}

Predisposing Factors and Prophylactic Measures

Predisposing factors for PC-AKI include diabetes mellitus, dehydration, cardiovascular disease, diuretic use, advanced age, multiple myeloma, hypertension, and hyperuricemia. ¹ Preexisting severe renal insufficiency is the most important risk factor for PC-AKI. ^{3 6} It has been reported that an estimated GFR (eGFR) of less than 30 mL/min prior to IV contrast administration is a significant risk factor for PC-AKI; patients with eGFR 30 to 44 mL/min have an increased risk that bordered on significance; and patients with eGFR greater than 45 mL/min do not have an increased risk for PC-AKI. ^{1 6} CM may exacerbate some of the underlying diseases through non–allergic-like reactions, including severe CKD, AKI, severe hyperthyroidism, congestive heart failure, cardiac arrhythmias, and myasthenia gravis. ¹

In high-risk patients, prophylaxis against PC-AKI can be taken to ensure that patients receive contrast necessary to perform diagnostic studies and/or interventional procedures while decreasing the risk for an adverse effect on the kidneys. The major prophylactic measure against PC-AKI is IV volume expansion. Isotonic fluids, such as 0.9% saline, are generally preferred over other fluids. ^{3 23 24} Sodium bicarbonate is another IV agent that has been studied for the prevention of PC-AKI. While the answer is unclear, some randomized clinical trials have suggested that sodium bicarbonate may be superior to 0.9% normal saline in preventing PC-AKI. ²⁵ Theoretically, alkalinization of the renal tubular fluid following the IV administration of sodium bicarbonate may inhibit free-radical formation, which is one of the proposed mechanisms of CIN. However, a systematic review of 23 trials by Zoungas et al concluded that the effectiveness of sodium bicarbonate in preventing CIN is uncertain. ²⁶

There is variability in the best combination of volume and duration of fluid administration for the prevention of PC-AKI in the literature. One consistent feature among the protocols is fluid administration before and after the procedure for which CM is used. The ACR manual on CM states that one protocol is the administration of 0.9% saline at 100 mL/hour for 6 to 12 hours before and 4 to 12 hours after contrast administration. ³ Since the need for prolonged IV fluid administration presents a logistical challenge to PC-AKI risk mitigation for high-risk patients undergoing outpatient imaging studies or procedures, volume expansion via the oral route has also been investigated. A systematic review of studies comparing oral and IV hydration concluded that the oral route may be similarly effective, but adequately powered trials are needed to fully investigate this question. ²⁷ It is worth mentioning that the current recommendation of required perioperative fasting is 6 hours for solid foods and 2 hours for clear liquids, ²⁸ and it may be important to encourage high-risk patients scheduled for outpatient interventional procedures to stop drinking for only 2 hours before the procedure to prevent dehydration.

N-acetylcysteine (NAC) is another agent that has been investigated for the prevention of PC-AKI/CIN. One meta-analysis of 86 randomized controlled trial showed a clinically and statistically significant benefit in patients receiving low-dose NAC plus IV saline, when compared with IV saline alone. However, the strength of this evidence was deemed to be low. ²⁹ This is consistent with the ACR guidelines that state that NAC cannot be considered a substitute for patient screening and volume expansion. ³ Fewer trials have been performed to investigate statins as a potential prophylactic agent used to reduce the incidence of PC-AKI, but meta-analyses of studies performed in patients undergoing coronary angiography have demonstrated that statin therapy may be effective in reducing the incidence of PC-AKI. ^{30 31} Other agents, such as mannitol, furosemide, and theophylline, have been discredited for the prevention of PC-AKI. ³

Conclusion

PC-AKI is usually multifactorial and more common than CIN. Risk of PC-AKI in high-risk patients can be mitigated by considering non–contrast-enhanced studies. In high-risk populations, IV volume expansion with normal saline administration before and after the contrast-enhanced study remains the main prophylaxis against PC-AKI.