| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Editorials |
1 From the Department of Radiology, University of California Davis Medical Center, 4860 Y St, Sacramento, CA 95817 (R.W.K.); and Patient Research Centre, Health Sciences Centre, St John's, Newfoundland, Canada (B.J.B.). Received August 15, 2006; revision requested October 19; revision received November 25; final version accepted December 14. Address correspondence to R.W.K. (e-mail: richard.katzberg{at}ucdmc.ucdavis.edu).
Contrast material (CM)-induced nephropathy (CIN) is the sudden rapid deterioration of renal function that results from parenteral CM administration. This renal function deterioration has been variably defined as a minimum increase from baseline serum creatinine (SCr) values of 25%–50% or an absolute increase in SCr of 0.5–1.0 mg/dL (1,2). The incidence of CIN has been reported to range from less than 1% to greater than 30%. This wide variation in incidence is attributed to factors that include a lack of consensus in definitions, assessments based on SCr levels rather than more direct measures of kidney function, differing patient populations such as inpatients versus outpatients, wide variability in CM doses, variation in the completeness or timing of patient follow-up, and likely variation in the patient's hydration state. In addition, there has been a paucity of comparisons of the nephrotoxicity caused by CM administered by using different routes (eg, intraarterial vs intravenous) (1–3).
CIN is not common in patients with normal preexisting renal function; rather, it is more frequent in patients with renal impairment, especially when the impairment is due to diabetic nephropathy (4). Even when both of these risk factors have been included, it has been difficult to establish an animal model reflective of the conditions in humans (5). This has hindered efforts to investigate the pathogenesis of CIN and led some to question whether CIN even exists (6).
While there are numerous data on the incidence of CIN following cardiac angiography and intervention, the incidence of CIN after intravenous CM administration, particularly in high-risk patients, is less well established (7). Even less is known about the relative effect of using one CM versus another in terms of CIN following intravenous injection.
In this commentary, we assess evidence from clinical trials of the intravenous administration of CM, evidence derived from personal experience, and literature survey information to gain perspectives on the relative risk of CIN following intraarterial versus intravenous CM administration.
 |
REVIEW OF CLINICAL STUDIES |
|---|
 TOPINTRODUCTION REVIEW OF CLINICAL STUDIES IS THERE A THRESHOLD...COMPARISONS OF CIN IN...CIN: INTRAVENOUS VERSUS...THE IMPORTANCE OF OSMOLALITY...References |
|---|
Cramer et al (9) assessed SCr levels before and 2 days after CM-enhanced brain computed tomography (CT) in 193 patients and after nonenhanced brain CT in 233 control patients. A high-osmolality CM (HOCM) (with four to seven times the osmolality of plasma) was administered (60–350 mL). Renal dysfunction following CT, defined as an increase in SCr to greater than 106.1 µmol/L (1.2 mg/dL) and to a level at least 50% higher than the baseline value, developed in four (2.1%) patients in whom CM was infused and in three (1.3%) who received no CM. The difference in renal dysfunction was not significant (P 
.05). In the Cramer et al study (9), CIN occurred in none of the 19 patients with preexisting renal insufficiency who received HOCM and in two (4.3%) of the 46 patients with preexisting renal insufficiency who received no CM. Note that to be counted as a case of CIN in their study, one had to have an at least 50% increase in SCr, which is a more stringent criterion than that used by many other authors. In addition, there may have been a lack of comparability between the control patients and those who received CM because the decision to perform unenhanced CT in them may have been based partly on the perceived risk of CIN.
Heller et al (10) examined 292 inpatients who received HOCM and 405 patients who did not receive CM. They also examined a group of patients who received a low-osmolality CM (LOCM), which has two to two and a half times the osmolality of plasma. Patients in the no-contrast-material group were selected on the basis of increased risk of CIN. As such, the low rate of CIN in the HOCM group was partly due to negative selection (selective use in low-risk patients) and the high rate of CIN in the LOCM group was partly due to positive selection (selective use in high-risk patients). Renal impairment was defined as a maximal increase in SCr of 50% or more or of greater than 0.5 mg/dL from the baseline value on at least one of the subsequent 4 days. Such outcomes are open to ascertainment bias, as more frequent monitoring in the higher-risk subjects will reveal more cases. CIN was seen in 12 (4%) of the 292 patients given HOCM and in 16 (4%) of the 405 patients who did not receive CM. An additional 187 patients received LOCM, and 23 (12%) of these subjects developed CIN.
In the Heller et al study (10), an acute increase in SCr was seen in seven (10%) of 68 patients with preexisting renal insufficiency who received CM versus in six (7%) of 88 with preexisting renal insufficiency who did not. It is possible that in any series of ill patients, some will experience renal failure as a coincident event or as an adverse reaction to medication intake or some other nephrotoxic event. There is also the possibility of random variation or "background noise" in the SCr. Another interesting observation from the Heller et al study was that blood transfusion was an important independent predictor of acute renal failure. This finding highlights the importance of prerenal factors, such as hypotension due to blood loss, and the distinct possibility that CM-induced hypovolemia secondary to nonspecific osmotic diuresis could contribute to CIN in some cases.
Studies of Patients Given Intravenous LOCM
A number of reports (Tables 1 and 2) have been focused on whether SCr levels significantly (P < .05) increase following CM injection. Dalla Palma et al (11) examined eight patients with SCr levels of between 2.2 and 7.6 mg/dL who received 300 mL of iopamidol (Isovue; Bracco Diagnostics, Princeton, NJ) in a drip infusion. In these subjects, there was no evidence of significant (P < .05) postcontrast alterations in SCr values. Levorstad et al (12), Rankin and Eng (13), McClennan et al (14), Spataro et al (15), Cochran et al (16), and Newhouse et al (20) all reported the average increase in SCr levels among patients without preexisting renal insufficiency who received an intravenous LOCM. A total of 599 patients are reported on in these studies, none of which revealed a substantial increase in average SCr level following CM administration.
|
|
Harris et al (18) examined 51 patients with renal insufficiency (SCr levels 1.4–2.4 mg/dL) who underwent contrast-enhanced CT with the nonionic monomer iohexol; a greater than 50% increase in SCr was the criterion used to identify CIN. Only one (2%) of the 51 subjects developed CIN (Table 1). The authors concluded that there were no cases of clinically important CIN in their investigation.
To determine the rates of CIN associated with HOCM (diatrizoate meglumine, Hypaque; GE Healthcare) and LOCM (iohexol), Moore et al (19) conducted a randomized double-blind clinical trial involving patients who underwent diagnostic angiocardiography (n = 430) or contrast-enhanced CT (n = 499). The frequency of nephrotoxicity overall was similar between patients who received LOCM and those who received HOCM: 13 (2.7%) of 479 patients versus 13 (2.9%) of 450 patients, respectively (P = .87) (4.4% [10 of 229] versus 4.0% [eight of 201] patients, respectively, in angiocardiography group [P = .84]; 1.2% [three of 250] versus 2.0% [five of 249] patients, respectively, in body CT group [P = .35]). Overall, only 26 (2.8%) of the 929 patients had preexisting renal insufficiency (SCr 1.5 mg/dL [133 µmol/L]), but these subjects were not categorized into intraarterial versus intravenous examination groups. Multivariate logistic regression analysis of risk factors revealed angiocardiography to be independently associated with increased risk of CIN.
Lundqvist et al (21) assessed the possibility of CIN following excretory urography in 63 patients with baseline renal insufficiency who received iohexol (50 mL, 300 mg of iodine per milliliter). Creatinine clearance was determined at 24 hours and on day 7 after CM administration. Creatinine clearance decreased by more than 25% in nine (14%) patients and increased by more than 25% in 12 (19%). These data illustrate the degree of noise inherent in these measures of kidney function, which are prone to considerable random error.
Carraro et al (22) compared the effects on renal function between a nonionic dimer (iodixanol, Visipaque; GE Healthcare) and a nonionic monomer (iopromide, Ultravist; Schering, Berlin, Germany) in patients with mild to moderate renal insufficiency who were scheduled to undergo excretory urography. Sixty-four consecutive subjects with SCr values of between 135 and 265 µmol/L were recruited, with CIN defined as an increase in SCr concentration of 50% or greater over baseline 24 hours following the examination. For both CM groups, the SCr level decreased during the observation period but without significant differences (P .05) (SCr level 150.3 µmol/L before iodixanol administration, 146.7 µmol/L 24 hours after; SCr level 149.4 µmol/L before iopromide administration, 135.3 µmol/L 24 hours after). Transient CIN developed in only one patient, who had received iodixanol.
Tepel et al (23) performed a prospective study involving 83 patients with chronic renal insufficiency (mean SCr level, 2.4 mg/dL ± 1.3 [standard deviation]) who underwent CT with iopromide, an LOCM. Patients were randomly assigned to receive intravenously N-acetylcysteine (600 mg orally twice) with 0.45% saline before and after CM administration or to receive placebo with saline. Our interest in this report is in the subjects who did not receive N-acetylcysteine. Nine (21%) of these 42 subjects had a 0.5 mg/dL increase in SCr 48 hours after CM administration. The authors stated that repeated measurements during the week before CM administration revealed "only minor" changes in SCr level (mean variation, 0.1 mg/dL ± 0.3 [9 µmol/L ± 26]; P = .12).
Lufft et al (24) compared the incidence of CIN following intravenous CT angiography with that following intraarterial subtraction angiography for suspected renal artery stenosis. Since the CM was injected directly into the renal artery for digital subtraction angiography, these cases are not considered a part of the intravenous CM database. CIN, defined as a greater than 25% or greater than 0.5 mg/dL increase in SCr level compared with the baseline value, developed in one (4%) of 24 subjects with a baseline SCr level of greater than 1.5 mg/dL or an inulin clearance of less than 75 mL/min who received intravenous CM.
Garcia-Ruiz et al (25) performed a prospective study involving 50 patients with chronic renal insufficiency (SCr greater than 1.58 mg/dL) who underwent spiral CT angiography with iopromide, a nonionic LOCM. CIN was defined as an increase of 20% or more from the baseline SCr level within 72 hours after CM administration. Only two (4%) patients experienced an increase in SCr level of 20%. Renal function returned to baseline within 7 days in these two patients. The patients had been encouraged to drink 1 L of water 12 hours before and 2 L of water more than 24 hours following the procedure. The authors concluded that in patients with chronic renal insufficiency, spiral CT angiography performed with iopromide and prophylactic oral hydration carries a low risk of CIN.
Kolehmainen and Soiva (28) compared the renal tolerance of iobitridol (Xenetix; Guerbet) and iodixanol in patients who presented with severe renal impairment and underwent cranial or whole-body CT. Fifty patients (25 per CM group) were enrolled in a double-blind randomized fashion. Mean baseline SCr levels were 242.0 µmol/L ± 192 and 229.7 µmol/L ± 141 in the iobitridol and iodixanol groups, respectively. Absolute and relative variations from the baseline value were not significantly (P .05) different between the two groups. Four patients (16%) in each group had a greater than 0.5 mg/dL (44 µmol/L) increase in SCr level. The authors concluded that there was no significant difference in renal safety between iobitridol and iodixanol.
Becker and Reiser (26) studied iodixanol used for multidetector CT angiography in patients with renal impairment (SCr level, 1.5–6.0 mg/dL). SCr levels were measured on days 3 and 7 after the intravenous administration of 100 mL of iodixanol-270 at 5 mL/sec. CIN was defined as an increase in SCr of greater than 0.5 mg/dL above baseline on day 3. Nine patients (9%) developed CIN. Seven of these subjects recovered completely by day 7, and the remaining two subjects had elevated SCr levels on day 7 but did not develop renal impairment requiring dialysis during their hospital stay.
Barrett et al (27) compared the effects of a nonionic monomer, iopamidol-370, and a nonionic dimer, iodixanol-320, on renal function in patients with renal impairment who underwent contrast-enhanced multidetector CT in a multicenter, double-blind, randomized, parallel-group study. One hundred fifty-three patients with stable, moderate, or severe chronic renal disease (SCr level 1.5 mg/dL and/or creatinine clearance 10–59 mL/min) were enrolled, and CIN was defined as an absolute increase in SCr level of 0.5 mg/dL or greater. The two study groups were comparable with regard to all baseline characteristics. An absolute increase in SCr of 0.5 mg/dL or greater was observed in two (3%) of 76 patients who received iodixanol-320 and in none of the 77 patients who received iopamidol-370 (95% confidence interval: –6.2%, 1.0%; P = .2). The authors concluded that the rate of CIN was low and not significantly different between patients with moderate to severe chronic kidney disease who received a nonionic monomer and those who received a nonionic dimer.
Given the multiple differing criteria and clinical settings for CIN, it is not possible to render precise true rates of CIN with intravenous use. Criteria for CIN have seemingly become more sensitive (ie, less stringent) over time. The earlier criterion used by Cramer et al (9) and Heller et al (10) was an at least 50% or greater increase in SCr. Further examination of the data from patients at increased risk presented in Table 2 revealed that only one (1.2%) of 83 subjects who received a nonionic monomer potentially met this criterion and that no subjects required dialysis. Even given the possible selection bias among the control subjects reported by Cramer et al (9) and Heller et al (10), the risk of CIN appears low.
Further attention is directed to two studies described in Table 2. The first is the Lundqvist et al (21) investigation, in which more patients had improved renal function rather than decreased renal function following CM administration. It is conceivable that this result simply represents random variation, as noted earlier for the Cramer et al (9) and Heller et al (10) studies. The second study of note is the Tepel et al (23) investigation, which appears to be an outlier in that 21% of the subjects who received intravenous CM with a hydration regimen developed CIN. It is possible that these patients had a greater level of renal insufficiency than did the subjects in the other studies and/or that they had unstable baseline values.
|
IS THERE A THRESHOLD OF RENAL INSUFFICIENCY FOR HIGHER RISK? |
|---|
|
TOPINTRODUCTIONREVIEW OF CLINICAL STUDIESIS THERE A THRESHOLD...COMPARISONS OF CIN IN...CIN: INTRAVENOUS VERSUS...THE IMPORTANCE OF OSMOLALITY...References |
|---|
0.5 mg/dL). These data are interesting and suggest a possible trend; however, the database was too small to make firm conclusions. |
COMPARISONS OF CIN IN AT-RISK PATIENTS WHO RECEIVED LOCM VERSUS NONIONIC ISO-OSMOLAR CM |
|---|
|
TOPINTRODUCTIONREVIEW OF CLINICAL STUDIESIS THERE A THRESHOLD...COMPARISONS OF CIN IN...CIN: INTRAVENOUS VERSUS...THE IMPORTANCE OF OSMOLALITY...References |
|---|
|
CIN: INTRAVENOUS VERSUS INTRAARTERIAL CM ADMINISTRATION |
|---|
|
TOPINTRODUCTIONREVIEW OF CLINICAL STUDIESIS THERE A THRESHOLD...COMPARISONS OF CIN IN...CIN: INTRAVENOUS VERSUS...THE IMPORTANCE OF OSMOLALITY...References |
|---|
Moore et al (19) reported a higher CIN rate after angiocardiography (4.4%) than after CT (1.2%) with administration of LOCM in patients who predominantly did not have renal impairment and found that the procedure type itself was a significant (P < .05) independent risk factor of CIN. On the other hand, Barrett et al (30) reported results for patients with renal compromise by using multiple linear regression models, the results of which failed to show the route of CM administration as a predictive factor of CIN.
Microshowers of cholesterol emboli have been shown to occur in up to 50% of percutaneous interventions in which a catheter is passed through the aorta, and they may be a cause of increased risk for CIN (31). Autopsy studies have revealed cholesterol emboli in 30% of patients who died within 6 months after undergoing aortography (32). In age-matched control subjects who had not undergone a previous vascular procedure, the prevalence of cholesterol emboli was 4.3%.
|
THE IMPORTANCE OF OSMOLALITY WITH REGARD TO CIN |
|---|
|
TOPINTRODUCTIONREVIEW OF CLINICAL STUDIESIS THERE A THRESHOLD...COMPARISONS OF CIN IN...CIN: INTRAVENOUS VERSUS...THE IMPORTANCE OF OSMOLALITY...References |
|---|
In conclusion, the data in the literature do not enable a statistically rigorous comparison of the CIN rates following intravenous as opposed to intraarterial or cardiac injection of CM. The incidence, severity, and clinical importance of the kidney injury that may follow the intravenous use of CM appear to be lower than those reported when CM are used for cardiac catheterization and intervention. The extent to which the patient's specific medical condition and other possible mechanisms of renal injury that occur with intraarterial procedures such as reduced hydration, higher renal plasma levels that can occur with aortography, higher CM doses, and cholesterol emboli contribute to this apparent difference remains unclear.
The implications from a clinical practice point of view are that the recommendations for the avoidance of CIN and the use of specific prophylactic regimens derived from studies of intraarterial use do not necessarily apply to the intravenous use of CM. The apparently lower risk of CIN associated with intravenous CM administration in clinical settings such as contrast-enhanced multidetector CT makes it defensible to consider using CM even in patients with greater levels of background risk factors (eg, greater degrees of preexisting chronic renal insufficiency) than one would be comfortable with in the intraarterial setting, as long as there is a clear indication for the diagnostic examination. In addition, given the logistic challenges in the outpatient setting, the use of specific prophylactic measures in the intravenous setting could be limited to those subjects who are at higher levels of background risk than they would be when an intraarterial procedure was planned. From a research perspective, further studies with inclusion of control subjects are clearly needed to assess the true burden of CIN, if any, associated with current intravenously administered CM.
There apparently is insufficient evidence to recommend one specific CM over another for its potential to cause less CIN following intravenous administration in at-risk patients with renal insufficiency. However, CIN rates are apparently so low that a prospective study of literally thousands of subjects would be necessary to demonstrate a clinically relevant difference between agents, and if such a study were performed, the true clinical importance of this difference would have to be questioned.
|
FOOTNOTES |
|---|
|
References |
|---|
|
TOPINTRODUCTIONREVIEW OF CLINICAL STUDIESIS THERE A THRESHOLD...COMPARISONS OF CIN IN...CIN: INTRAVENOUS VERSUS...THE IMPORTANCE OF OSMOLALITY...References |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. J. Kuhn, N. Chen, D. V. Sahani, D. Reimer, E. J. R. van Beek, J. P. Heiken, and G. J. So The PREDICT Study: A Randomized Double-Blind Comparison of Contrast-Induced Nephropathy After Low- or Isoosmolar Contrast Agent Exposure Am. J. Roentgenol., July 1, 2008; 191(1): 151 - 157. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. R. Herts, E. Schneider, E. D. Poggio, N. A. Obuchowski, and M. E. Baker Identifying Outpatients with Renal Insufficiency before Contrast-enhanced CT by Using Estimated Glomerular Filtration Rates versus Serum Creatinine Levels Radiology, July 1, 2008; 248(1): 106 - 113. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
