Automated Contrast Injection and Targeted Renal Therapy: Strategies to Prevent Contrast-Induced Nephropathy and to Treat Renal Insufficiency

It is currently estimated that in the United States there are 15–18 million patients with peripheral arterial disease (PAD), and 18–20 million with diabetes mellitus (DM).^1,2 These incidences are increasing, along with the number of PAD patients revascularized with percutaneous peripheral interventions (PPI).^2,3 Several factors are likely to increase the number of PPIs performed yearly and therefore, patient contrast exposure. The rapid adoption of multidetector computed tomography angiography (MDCTA) as a noninvasive modality in PAD diagnosis, treatment and follow-up will likely increase contrast exposure. A large number of patients with significant symptomatic and asymptomatic PAD who are currently undiagnosed and/or untreated will likely be diagnosed by MDCTA and become candidates for PPI. Additionally, PAD patients still require multiple PPIs and reinterventions secondary to the diffuse location of the disease and our current technology. These unique clinical dynamics of PAD underscore a need for contrast optimization and a better understanding of the role of prevention and treatment of contrast-induced nephropathy (CIN) in patients with PAD.

The definition of CIN varies, but in clinical trials CIN has been defined as an increase in serum creatinine (Cr) > 0.5 mg/dl or > 25% of baseline between 24–120 hours after contrast exposure, with peak levels reported between 3–5 days.^4–6 The pathogenesis of CIN is complex and remains incompletely defined. The proposed mechanism of developing CIN results in a “vicious cycle” of events that culminates in critical renal medullary hypoxia and cell necrosis. This cycle includes a direct renal cellular cytotoxic effect, resulting in increased toxic oxygen-free radicals; intense renal medullary vasoconstriction and hypoxia mediated by multiple vasoconstrictors, including adenosine, vasopressin and prostaglandin E2;^7,8 acute increase in renal osmolality, requiring increased renal cellular oxygen consumption; acute reduction of renal blood flow with endothelial dysfunction and decreased nitric oxide production resulting in renal medullary hypercoaguability, hyperviscosity, and worsening ischemia, leading to acute tubular necrosis (ATN). Once initiated, there are few therapeutic options to interrupt this cycle.

McCullough et al. described the clinical impact in percutaneous coronary intervention (PCI) and identified the significant increase in morbidity and mortality associated with CIN.⁹ CIN is the third leading cause of hospital-acquired acute renal failure (ARF). McCullough reported a 14% incidence of CIN in 1,826 PCI patients with a 7.1% in-hospital mortality in patients developing CIN not requiring dialysis, and 35.7% if requiring dialysis (p < 0.001).9 With dialysis, the patient has < 20% two-year survival.⁹ The in-hospital mortality was 0.7% without CIN. Gruberg et al. reported a 37% incidence of CIN (7.3% requiring dialysis) in 440 PCI patients with baseline renal insufficiency (RI) (Cr ? 1.8mg/dl) with three times higher in-hospital mortality (14.9% versus 4.9%) and two times higher 1-year mortality (37.7% versus 19.4%) in patients with CIN.¹⁰ Likewise, Levey et al. noted a 5.5-fold risk for increased mortality in patients undergoing diagnostic studies requiring contrast exposure who developed CIN.¹¹ Clearly, the clinical impact of CIN in the PCI patient population is significant and unappreciated, and CIN potentially stands to have an even greater clinical impact on the treatment of PAD as compared to PCI.

The incidence and impact of CIN in PPI remains almost totally unknown and unexplored. Mehran et al. identified CIN predictors in PCI, including diabetes (DM), age > 75 years, female gender, contrast volume, Cr clearance (CRCL), congestive heart failure (CHF), hypotension, preprocedure renal insufficiency (RI) and anemia, and validated a CIN risk score prediction model.¹² The incidences of these CIN predictors are usually greater in the older PPI versus PCI population. The individual incidences of DM and preprocedural RI in PCI trials was approximately 20%, but the incidence of DM and RI have been reported at a 50–80% incidence in PPI, especially for critical limb ischemia (CLI).¹³ This becomes significant when considering the combination of DM and preprocedure RI, shown by Parfey et al. to increase the incidence of CIN during PCI to 50%.¹⁴

Additionally, there are several significant clinical and periprocedural differences during PPI versus PCI that increase the risk of CIN. These include complex, longer PPI case durations with higher contrast use; higher rates of multiple procedures and secondary reinterventions; overall higher complication rates in PPI; more frequent MDCTA use, therefore, more contrast exposure; and a higher incidence of renal artery stenosis (RAS). A typical PAD patient would be a frail > 80-year-old female weighing 90 pounds, with CLI, hematocrit of 27.5%, serum Cr of 1.9 mg/dL and a calculated CrCL of < 30 mL/min. This typical patient is certainly at extreme risk for developing CIN. Strategies to prevent CIN are not only warranted, but are essential in treating patients with PAD.

The role of CIN and perioperative ARF during vascular and cardiac surgical procedures likewise remains poorly defined. Strategies for contrast optimization and CIN prophylaxis will become increasingly important to the surgeon as increasing numbers of surgeons acquire catheter-based skills. Interestingly, contrast administration < 48 hours before coronary artery bypass (CABG) is a known, independent, preoperative predictor of worsening RI, but no guidelines exist for optimal renal protection in CABG performed within 72 hours of angiography.^12,15 In fact, the association and incidence of CIN during cardiac surgery is unknown, and likely underestimated and a major contributing factor to a significant number of patients with worsening RI and ARF after cardiac surgery yearly. Loc et al. has shown that the 1-year mortalities after CABG are significantly higher (p < 0.0001) in patients with mild RI (11.1%) versus no RI (3.8%).¹⁵ The clinical implications become significant when considering that approximately 18% of the U.S. population (> 50 million) has some degree of RI and 14% of the approximately 750,000 patients undergoing cardiac surgical procedures have preoperative RI, and therefore are at risk for perioperative ARF.^15–17

A heightened awareness of CIN and RI will become increasingly important in the treatment of aortic aneurysmal disease with endovascular aneurysm repair (EVAR) now available for thoracic (TAA) and abdominal aortic aneurysms (AAA). It is estimated that 10% of every male in the U.S. > 70 years old is harboring a AAA, and that there are approximately 1,000,000 AAAs that remain untreated.^17,18 Approximately 30% of those 100,000 treated yearly undergo EVAR (30,000), and therefore will require contrast exposure.^17,18 With the widespread acceptance of MDCTA, and AAA screening expected to be reimbursed in certain patient populations, it is anticipated that an increasing number of AAAs will be diagnosed yearly, and therefore treated with EVAR. Interestingly, the FDA has recently approved a novel, miniature implantable device capable of remote, radio-frequency monitoring of the AAA (or even TAA) sac pressure after EVAR. The EndoSure Wireless AAA Pressure Sensor (CardioMEMS, Inc., Atlanta, Georgia) would have the potential to significantly decrease the number of contrast studies currently recommended for EVAR follow up.

The incidence of RI after catheter-based AAA EVAR is also associated with preoperative RI and high post-procedure mortality.^19,20 ARF is reported after 2–6% of infrarenal open AAA repairs, and is significantly higher in TAA.^19–21 Worsening RI after EVAR is the third most commonly experienced morbidity and few reports exist implicating CIN as a prevalent etiology.^19–21 Worsening RI post-EVAR has been reported from 6–39%.^19–21 Carpenter et al. reported a 20% incidence of preop RI in 98 EVAR cases with an average volume of intraoperative contrast use of 152 cc (35–420 cc), underscoring the potential for intraoperative-induced CIN.²² Permanent RI was reported at 16% in this series, despite the liberal use of MRA and gadolinium. With the rapid adoption of MDCTA for the treatment of patients with TAA, AAA and PAD, the additional 75-mL to 125-mL of contrast volume required will further mandate the surgeon to develop strategies to minimize CIN.

Multiple strategies have been proposed for the prevention of CIN in PCI but few have strong supporting data. Randomized trials have demonstrated the importance of 0.9% saline infusions before and after contrast exposure in decreasing CIN.^23,24 The infusion rates have varied between 1–3 mL/kg/hr for 6-12 hours after contrast exposure. Isotonic sodium bicarbonate solution started one hour before contrast exposure was recently associated with a 2% versus 17% incidence of CIN in a high-risk patient population when randomized to 0.9% saline.²⁵ Further trials are being structured to validate the role of serum and urine alkalinization in the prevention of CIN. The role of N-acetylcysteine (NAC) in CIN prevention remains uncertain, with inconsistent results reported with regards to both the route and dose of administration.²⁶ Interestingly, Barret and Parfay, recently reporting a review of strategies to reduce CIN, stated that NAC and IV sodium bicarbonate were “not generally recommended unless efficacy was confirmed by further trials.”²⁶ Most cath lab protocols now incorporate some form or combination of 0.9% saline, sodium bicarbonate and NAC for CIN prevention, especially in the high-risk patient, despite a lack of validating data.

Intuitively, increasing renal blood flow and renal medullary vasodilatation would seem protective against CIN, since the proposed etiology of injury in CIN is an acute toxic injury induced by severe medullary vasoconstriction and critical cellular hypoxia. Fenoldopam (FEN) (Corlopam, Abbott Laboratories, Abbott Park, Illinois) is a short-acting, selective dopamine-1 agonist and vasodilator that is the only agent shown to increase both renal cortical and medullary blood flow.²⁷ The initial favorable clinical reports of systemic IV-FEN administration in reducing CIN in PCI were not reduplicated in the randomized CONTRAST trial.²⁸

Unfortunately, IV-FEN has a first-pass renal metabolism and can cause systemic hypotension at mild to moderate systemic doses. Therefore, it has been theorized that the CONTRAST trial results were secondary to an inability to deliver therapeutic doses directly to the renal medulla. Direct high-dose intrarenal (IR) infusion of FEN, in concept, has the potential to deliver selective high-dose renal vasodilatation and increased medullary blood flow without systemic hypotension, with the potential to reduce CIN in both the cath lab and surgical suites.

Targeted renal therapy (TRT) is a novel technique for direct IR infusion of therapeutic doses of FEN to prevent CIN and potentially treat a wide variety of other clinical scenarios, all associated with worsening RI. TRT is delivered by the Benephit PV Infusion System (FlowMedica, Inc., Fremont, California). It is available in an 8-Fr dual port and 5-Fr single port system that includes an introducer sheath with low-profile, atraumatic bifurcated infusion arms with tips that can be easily positioned into both renal arteries for continuous infusions of fluids (Figure 1). The Benephit PV Infusion System has been granted 510-k approval and is currently commercially available in 40-cm, 105-cm, and 140-cm lengths. We have used both the femoral and brachial artery approach in a wide variety of both percutaneous endovascular and surgical cases in patients with RI at risk for CIN or worsening RI. Trials are being initiated with RI patients at high risk of worsening renal function during CABG, cardiac valve surgery, AAA (open and EVAR), PCI and the endovascular treatments of PVD.

The early data and trials with TRT have been encouraging. The completed FEN-01 study validated the feasibility, safety, and efficacy of IR-FEN delivery and validated several key hypotheses about IR-FEN.²⁹ Thirty-three patients with mild RI were randomized 2:1 to receive IV-FEN, then IR-FEN versus placebo during PCI. Inulin was used to measure GFR, and serum FEN levels were also measured. In concept, the IR-FEN infusion maximized the favorable hemodynamic benefits of FEN by “targeted” infusion of FEN into the renal arteries. Systemic serum FEN levels were lower with IR-FEN compared to IV-FEN due to first pass renal metabolism and excretion of FEN. Administration of IR-FEN compared to IV-FEN (0.2 mcg/kg/min) resulted in 30% lower serum levels of FEN. Additionally, there was 45% less reduction in systemic blood pressure with IR-FEN compared to placebo (12 ± 3 mmHg versus 23 ± 3 mmHg, p < 001). Importantly, GFR was not significantly altered by IV-FEN, but was significantly increased by 25% during IR-FEN (p < 0.001 compared to baseline). The increase in GFR after IR-FEN was significantly greater than seen with IV-FEN (25% versus 5%, p < 0.001). GFR, measured two hours after the procedure, remained significantly elevated (25% increase) in IR-FEN patients, while GFR in placebo patients declined by 14%. This 25% prolonged increase in GFR with TRT compared to a 14% decrease with placebo has significant clinical implications in not only preventing CIN, but also in multiple clinical scenarios that would benefit from improved GFR.

Additional, non-randomized clinical data are being accumulated in high-risk patients in the ongoing Be-Rite Registry.³⁰ Thus far, mean baseline Cr is 2.3 with 26% of patients having baseline Cr > 2.5. TRT was performed with FEN infusions of 0.2–0.4 mcq/kg/min. Two-thirds (67%) of patients experienced no increase in Cr, while 28% actually showed a decrease in Cr. Only 5% of these high-risk patients developed an increase in Cr.

The recently begun multicenter PATRICIA Trial (Peripheral Angiographers’ Targeted Renal Infusion for Contrast Injury Avoidance) evaluates the use of TRT to deliver IR-FEN (0.4 mcg/kg/min) in patients at high risk for CIN (serum Cr > 2.0 mg/dL or CrCL < 45 mL/min) undergoing PPI. In this 50-patient pilot trial, both the treatment and control arm patients will receive IV saline hydration, IV sodium bicarbonate and oral NAC. Serum Cr and CrCL will be evaluated at 24 and 72 hrs to determine the incidence of CIN. This trial should be enlightening, especially in light of a recent study by Srodon et al. reporting significant renal impairment in 46% of patients undergoing PPI.³¹

Case Study
A 69-year-old male was transferred from an outside facility for urgent CABG with unstable angina, CHF, recent pulmonary edema and chronic RI. The serum Cr was 2.6 mg/dL with a calculated CrCL of 38 cc/min. Coronary angiography, utilizing 100 cc contrast 12 hours prior to transfer, revealed a 80–90% left main stenosis and 90% proximal right coronary artery stenosis. TRT was not considered by the cardiologist at that facility. Further history included a left renal stent 6 months prior. He was taken urgently to the cath lab for TRT under a high-risk CABG-ARF protocol to be followed by longer-term perioperative TRT infusion.
A 5-Fr femoral artery access was obtained and the Benephit introducer sheath was positioned, utilizing the renal stent as reference. A limited, 8-cc “contrast puff” revealed a totally occluded left RA and a 95% right RAS (Figure 2A). An immediate decision was made to revascularize the left RA first, followed by right RA PTA/stenting with minimal contrast use and peri- and post-procedural TRT. The occluded left RA was recannalized and restented with a balloon-expandable ostial stent and a distal self-expanding stent (Figures 2B and 2C).

Uncomplicated right RA PTA/stenting was immediately performed with a total contrast use of 30 cc. The Benephit bifurcated catheter was then easily placed and IR-FEN 0.4 mcq/kg/min was infused (Figures 2D and-2F). The patient stabilized on a high dose of NTG and anticoagulation, allowing avoidance of emergent CABG. TRT was continued for 24 hours and the serum Cr decreased to 1.9 mg/dL at 72 hours. The patient was stabilized for 1 week, and TRT was restarted immediately before elective CABG and continued preoperative for 8 hours. The patient developed no device-related complications and was discharged on the 8th postoperative day with a serum Cr of 1.7 mg/dL.

Discussion
Randomized CABG-TRT trials are being organized. This case not only demonstrates the benefits of TRT in CIN prevention, but also in a surgical clinical scenario, where the outcomes could potentially be benefited by a 25% increase of GFR.

Techniques directed towards contrast injection and optimization in the cath lab during PCI and diagnostic coronary angiography have been described, but not during PPI. Recent advancements in contrast delivery systems have reported benefits.³² Contrast injection with a manual stopcock-manifold system has been the standard technique during PCI and PPI. The ACIST Injection System (ACIST Medical Systems, Eden Prairie, Minnesota) is a new automatic injection system that allows online hemodynamic monitoring and control of the contrast injection rate and the amount to be delivered (Figure 3A).³² Previous studies comparing manual injection to an automatic injector have shown safety, reliability, predictability and equivalent image quality.^33–35

The ACIST Injection System is a software-controlled, variable rate, self-purging syringe injector connected to an automated manifold without stopcocks that has almost completely removed the risk of injecting air bubbles (Figure 2B). It supplies contrast media to a catheter at a user-determined, variable flow rate that can be varied instantaneously and continuously. The user can control the flow rate of contrast from the injector to the catheter with a user-actuated hand controller. By operating this hand controller, the user can vary the flow rate of the contrast from the injector, and therefore the volume of contrast or saline flush delivered to the patient on each injection. In a randomized trial of 453 PCI patients, Anne et al. reported a significant reduction in the total volume of contrast (the amount wasted and the amount injected) and the total volume delivered to the patient with the ACIST Injection System as compared to manual injection.³⁵

These benefits with the ACIST Injection System in PCI will likely be magnified in the endovascular treatments of PVD. As noted earlier, the higher incidence of DM and chronic RI in the PVD patients mandate strict contrast optimization and conservation. Since totally converting our interventional labs to the ACIST Injection System, we have found the hand-controlled injections to allow optimal control over the rate and amount of each injection, therefore injecting less volume to our patients as compared to manual injection. Catheter tip pressures are also continuously displayed until the injection is triggered.

In addition to the contrast optimization benefits of the ACIST System, we have found several additional advantages, including overall cost savings for the cath lab and hospital on multiple levels; reduced set-up times and overall procedure times; less radiation exposure; precise recording of contrast volume used; and the use of smaller 4-Fr catheters and the elimination of all vascular closure devices (VCD) for diagnostic cases. We now utilize the Boomerang™ (Cardiva Medical, Mountain View, California) 4–5 Fr and 6–10 Fr vascular access management system to facilitate and limit manual compression (MC) on all PCI, PPI and diagnostic cases. This novel system eliminates any foreign body being left behind, and has greatly decreased our VCD complications, limited MC, increased patient and staff satisfaction, allowed early safe ambulation and facilitated overall throughput within our facility.

A particularly important advantage we have experienced with a combination of the ACIST and the Boomerang system is a decrease in hand, wrist and joint complaints within the cath lab and hospital staff. Hand injections through small sheaths are difficult, painful and can be a source of injury. The use of the automated push-button contrast injector has eliminated this injury. In addition to using smaller sheaths with the ACIST, the Boomerang system converts the existing sheath (4–10 Fr) to an arteriotomy the size of an 18g needle, therefore decreasing the amount and duration of MC required, resulting in less potential staff injury. This is extremely important in PPI, where a higher risk for complications exists.

Carpal tunnel syndrome and multiple other musculo-skeletal injuries are becoming a significant source of concern regarding the issue of “injury and safety in the workplace.” The U.S. Department of Labor and the Occupational Safety and Health Administration (OSHA) has compiled shocking data on the subject of work-related musculo-skeletal injuries. This underappreciated problem needs to have higher priority on the radar screen of all healthcare workers associated with diagnostic or interventional labs. The U.S. Department of Labor states that work-related injuries are “the nation’s most common and costly occupational health problem, affecting hundreds of thousands of American workers and costing more than $20 billion a year in workers’ compensation.”³⁶

Although intravenous contrast injections are less likely to cause CIN, the rapid adoption of MDCTA in the treatment of PAD is a potential source of significant CIN.³⁷ Few protocols exist when attempting to optimize contrast utilization and MDCTA in treating PVD, especially with 64-channel “cardiac” scanners. Most current 64-channel protocols utilize 100–125 cc contrast volume, and all parameters are designed to optimize cardiac and coronary imaging and not the more distant infrainguinal arteries. When considering the challenging imaging delays presented by a patient with infrainguinal disease and the current cardiac-based protocols, it will be necessary to develop 64-channel imaging protocols designed to optimize contrast utilization and imaging of the patient with PAD.

Recently, our group reported our 64-channel infrainguinal validation study with a revised protocol.^38,39 The revised protocol was used for 60 consecutive patients with severe infrainguinal disease. Protocol revisions included the automated trigger being lowered from the mid-chest to 1 cm above the aortic bifurcation in the distal aorta. The automated trigger Hounsfield units were increased from 180H to 250H with a 5-second scan delay. The contrast volume was reduced from 125 cc to 70 cc (Isovue, BRACCO Diagnostics, Princeton, New Jersey) with a 40 cc NS bolus chase. All other parameters remained unchanged. In concept, we were resetting the system to “delay or slow down” to better acquire infrapopliteal images. The resolution quality and post-processing imaging time were improved and validated when compared to 60 matched patients. This protocol has allowed us to minimize and optimize contrast exposure in our PAD patients, especially those with infrainguinal disease (Figures 4A and 4B).

Conclusion
The mortality and morbidity of CIN is significant and likely underappreciated. The potential for increased CIN morbidity during PPI is unknown; therefore, the need for strategies to optimize contrast exposure in the treatment of PAD is likely greater than during PCI. The rapid adoption of MDCTA and the proliferation of endovascular and surgical treatments of PAD underscore the need for optimizing novel strategies designed to preserve renal function. Several such novel strategies, utilizing the ACIST Injection System, TRT, and revised MDCTA protocols, hold promise in optimizing contrast utilization and preventing CIN when treating PAD.