| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Contrast Media |
1 From the Departments of Radiology (M.R.P., H.E.E., Y.W.), Medicine (J.B.), Pathology (R.W.L.), and Vascular Surgery (K.C.K., H.L.B.), Weill Medical College of Cornell University, 416 E 55th St, New York, NY 10021; and the Rogosin Institute, New York, NY (J.B.). Received December 7, 2001; revision requested February 18, 2002; final revision received January 22, 2003; accepted January 27. Address correspondence to M.R.P. (e-mail: map2008@med.cornell.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Eight hundred ninety-six inpatients with available serum calcium data obtained before and after gadodiamide-enhanced MR imaging were identified. Changes in serum calcium measurements following gadodiamide administration in 1,049 MR imaging examinations performed in these patients were correlated with gadodiamide dose, renal function, and time between gadodiamide administration and phlebotomy.
RESULTS: Following 42 gadodiamide-enhanced examinations, serum calcium measurements spuriously decreased by more than 2 mg/dL (0.5 mmol/L), resulting in laboratory reports of "critical" hypocalcemia (ie, calcium level < 6 mg/dL [1.5 mmol/L]) in 25 examinations. These reduced calcium measurements were correlated with serum creatinine level (r = 0.39, P < .001), gadodiamide dose (r = 0.37, P < .001), and time between gadodiamide injection and phlebotomy (r = -0.28, P < .001). Spurious reductions in calcium measurements after administration of 0.1 mmol of gadodiamide per kilogram of body weight were greater in patients with renal insufficiency (0.6 mg/dL [0.15 mmol/L] ± 0.5 [0.125, SD]) than in those with normal renal function (0.14 mg/dL [0.035 mmol/L] ± 0.4 [0.1]) (P < .001). After administration of more than 0.2 mmol/kg of gadodiamide, spurious calcium measurement decreases were greater in patients with renal insufficiency (2.4 mg/dL [0.6 mmol/L] ± 3.6 [0.9]) than in those with normal renal function (0.4 mg/dL [0.1 mmol/L] ± 0.7 [0.175]) (P < .001). Patients with renal insufficiency had spuriously low calcium measurements up to 4
days after gadodiamide administration. Seven patients were inappropriately treated with intravenous calcium and eleven with oral calcium in response to false-positive laboratory reports of critical hypocalcemia. No patient had characteristic symptoms of hypocalcemia or injuries attributed to the inappropriate medical treatment.
CONCLUSION: Gadodiamide administration causes spurious hypocalcemia, particularly at doses of 0.2 mmol/kg or higher and in patients with renal insufficiency.
© RSNA, 2003
Index terms: Calcium • Contrast media • Contrast media, effects • Magnetic resonance (MR), contrast media, **.121432
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
In patients with normal kidney function, gadodiamide administered at the standard dose, 0.1 mmol per kilogram of body weight, has a small effect on colorimetric calcium measurements. This is because gadodiamide is rapidly excreted by the kidneys, with an elimination half-life of 77 minutes (package insert, Nycomed Amersham) (3). In patients with renal insufficiency, however, the gadodiamide elimination half-life is prolonged (4), and this is expected to interfere in laboratory determinations of serum calcium levels for a longer time following magnetic resonance (MR) imaging. Following the administration of higher doses of gadodiamide (up to 0.3 mmol/kg), which are used primarily for MR angiography, there may be greater interference in laboratory calcium measurements and for longer periods.
The purpose of this retrospective study was to evaluate the prevalence of spurious hypocalcemia after gadodiamide-enhanced MR imaging in a typical hospital-based patient population and to determine the associated clinical and laboratory factors.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
To characterize the typical variation in serum calcium level in the patient population, the two consecutive serum calcium levels measured within 15 days before the gadodiamide administration were recorded. The mean and range of differences in these consecutive pre–gadodiamide injection serum calcium measurements were calculated and served as control values for determining whether the change in calcium measurement after gadodiamide injection was significantly different from typical variations.
In our laboratory, the range of normal serum calcium levels is 8.5–10.5 mg/dL (2.125–2.625 mmol/L). When the serum calcium measurement decreases to less than 6 mg/dL (1.5 mmol/L), it is considered "critical"—that is, life threatening—and the referring physician is immediately telephoned or paged and informed of the results. Because the difference in consecutive pre–gadodiamide injection calcium measurements never exceeded 1.9 mg/dL (0.475 mmol/L), we further investigated the data to identify those patients who had either a calcium level decrease of more than 2 mg/dL (0.5 mmol/L) or a calcium level decrease to less than 6 mg/dL (1.5 mmol/L) to determine whether (a) the decrease in calcium level was spurious, (b) there were any associated symptoms of hypocalcemia, and (c) any treatment was implemented as a result of the low serum calcium measurement. This investigation included reviews of medical charts, electronic discharge summaries, and laboratory records, and interviews with primary physicians.
In Vitro Studies
We assessed the potential interference in the OCP colorimetric assay for serum calcium measurement that is associated with the four gadolinium chelates that are currently commercially available in the United States: gadopentetate dimeglumine (Magnevist; Berlex, Wayne, NJ), gadodiamide, gadoversetamide (OptiMARK; Mallinckrodt, St Louis, Mo), and gadoteridol (Prohance; Bracco, Princeton, NJ). Serum samples from 80 patients were pooled to form a sufficient quantity of homogenous human serum for this experiment. With use of the standard micropipette technique, 0.5 mmol of the gadolinium chelate per milliliter was diluted with distilled water (1:10), and then 4-, 10-, 20-, 30-, 40-, 60-, and 100-µL quantities of the diluted agent were added to 1-mL aliquots of the pooled human serum to produce gadolinium concentrations of 0.20, 0.50, 0.98, 1.46, 1.92, 2.83, and 4.55 mmol, respectively. According to Normann et al (1), dilutions of 0.5–2.5 mmol correspond to the expected serum gadodiamide concentrations in the body 20 minutes after intravenous doses of 0.1–0.5 mmol/kg. Separate aliquots of pooled serum were similarly diluted with normal saline for use as control samples.
These dilutions were thoroughly mixed and processed in the routine fashion by hospital chemistry laboratory personnel (R.W.L.) by using the standard OCP reagent technique with an automated analyzer (Hitachi 747-100; Roche Boehringer Mannheim, Indianapolis, Ind). This is the same instrument that is used for routine serum calcium determinations. The calibration of this instrument is checked daily, and it varied by less than ±3% during the 2 years of this study. The analyzer used 10 µL of the gadolinium chelate–serum sample, mixed the sample with an ethanolamine buffer, and then mixed it with OCP and 8-hydroxyquinoline. Optical absorbance was measured at 600 nm to determine the calcium levels. This is the same automated method used for routine measurements of serum calcium levels in actual clinical practice.
Statistical Analysis
Statistical calculations were performed with computer software (SPSS, version 11; SPSS, Chicago, Ill) and by treating each gadolinium-enhanced MR imaging examination as an independent event. The statistical significance of changes in serum calcium measurement following gadodiamide administration was first analyzed in the subset of patients for whom two pre–gadodiamide injection serum calcium measurements and a post–gadodiamide injection measurement were available. The significance of the difference between the change in calcium levels between the pre– and post–gadodiamide injection measurements and the change in calcium levels between the two successive pre–gadodiamide injection measurements was calculated by using the paired Student t test. Pearson correlation coefficients were calculated for assessment of the effects of patient age, serum creatinine level, gadodiamide dose, and time between gadodiamide injection and phlebotomy on the measured change in serum calcium level. A multivariate analysis was performed by using stepwise regression to determine how the serum calcium measurement error varied with each of the significantly correlated factors. We plotted the effect of time between gadodiamide injection and phlebotomy, controlling for gadodiamide dose and renal function.
The statistical significance of the effect of renal insufficiency (defined on the basis of serum creatinine level) on the magnitude of the calcium level decrease after gadodiamide administration was calculated by using the Student t test while controlling for the effect of gadodiamide dose. Similarly, the significance of the effect of dose was calculated with the Student t test while controlling for renal function.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Lower serum calcium measurements were observed following 613 of the 1,049 gadodiamide-enhanced MR imaging examinations, as compared with 88 examinations with no change and 348 with an increase after gadodiamide injection (Table 2). In 165 of these 1,049 examinations, the serum calcium measurement decreased from a normal to hypocalcemic (<8.5 mg/dL) level. A more than 2 mg/dL (0.5 mmol/L) decrease, which is considered well outside the range of normal calcium level variations, was observed in 42 (4%) of the 1,049 examinations. The serum calcium measurement obtained after gadodiamide injection was in the hypocalcemic range in all of these patients. In 25 of these 42 examinations, the serum calcium measurements obtained after gadodiamide injection were lower than 6 mg/dL (1.5 mmol/L) and thus in the critical range according to our laboratory criterion. A typical case is illustrated in Figure 1. In one patient, MR imaging was performed three times, and each time, the immediate post–gadodiamide injection serum calcium measurement decreased to less than 6 mg/dL (1.5 mmol/L). None of these patients was noted in the hospital chart to have developed characteristic symptoms of hypocalcemia. Review of the serum calcium measurements obtained immediately preceding and following the first post–gadodiamide injection measurement revealed this decrease to be an anomalous decrease in an otherwise flat baseline, similar to the decrease illustrated in Figure 1b. For these reasons, we believe that all 25 of these cases of critical hypocalcemia represented spurious measurements caused by the gadodiamide injection.
|
|
|
According to the physician notes regarding the cases of unexpected acute hypocalcemia, the possibility of an association between calcium level decrease and gadolinium-enhanced MR imaging was considered for only two patients. In most of the patients, the clinicians found no supporting clinical evidence of hypocalcemia. In one patient with unexplained seizures, however, the spurious hypocalcemia was initially hypothesized to be a possible cause of the seizures. In seven patients, treatment with intravenously administered calcium gluconate was initiated. Eleven patients were treated with orally administered calcium. No patient had complications that were attributed to the calcium treatments, although the treatment of one patient, who was in a coma due to encephalitis, and of two other patients with seizure disorders became more difficult. Of the 48 patients with spurious calcium level decreases either of more than 2 mg/dL or to less than 6 mg/dL, 24 had concomitant hypoalbuminemia, which further complicated the interpretation of spurious serum calcium measurement changes.
It was not possible to assess the extent to which medical procedures or other aspects of the patients’ work-ups and treatments were delayed. However, a sense of urgency to manage the spuriously low serum calcium level was reflected in the fact that all orders for intravenously administered calcium were coded "stat."
In Table 2, the characteristics of the examinations in which the serum calcium measurements decreased following gadodiamide injection are listed and compared with those of the examinations in which there was either no change or an increase in the calcium measurement. These data suggest that older age, higher gadodiamide dose, shorter time between gadodiamide injection and phlebotomy, and renal insufficiency all were associated with larger errors in measured serum calcium levels. Stepwise linear regression analysis revealed statistically significant correlations between calcium level decrease and the following factors: gadodiamide dose, serum creatinine level, and gadodiamide injection–phlebotomy interval (Table 3). After accounting for the effects of the other significant correlations, we observed no statistically meaningful correlation between calcium measurement decrease and either sex or age.
|
|
|
|
In Vitro Studies
Investigation of the gadolinium chelates that were directly added to pooled human serum revealed that the administration of neither gadoteridol nor gadopentetate dimeglumine had an effect on the calcium measurement in human serum, and both agents produced results that were identical to those produced by the control agent (saline) at all concentrations tested (0.20–4.55 mmol). Both gadodiamide and gadoversetamide interfered with the colorimetric measurement of serum calcium and to a comparable degree. These in vitro study data are illustrated in Figure 3.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Although the administration of gadodiamide previously has been reported to interfere in colorimetric serum calcium measurements (1,2), this interference is not widely known and is not included in U.S. drug labeling or in the American Association of Clinical Chemists references of known clinical laboratory interferences (10). Because of the lack of awareness of this phenomenon, 18 patients (approximately one patient per month) in this study received improper treatment on the basis of laboratory reports of spurious hypocalcemia. Although no patient injury could be traced to the erroneous calcium measurements, even though seven patients received calcium gluconate intravenously, this laboratory "artifact" is a potentially important cause of unnecessary and potentially dangerous medical interventions. There is also a substantial institutional cost associated with spurious laboratory reports of critical conditions. Such reports consume physician and staff time for emergency assessment. Spurious results also lead to additional patient examinations and may delay the diagnosis and management of the patient’s condition.
These consequences are likely to be a problem at most institutions because of the widespread use of Arsenazo III dye and OCP colorimetric assay techniques. According to results of the 2001 College of American Pathologists chemistry survey, 4,518 (87%) of 5,220 chemistry laboratories used these automated colorimetric techniques (11).
To explain the spurious laboratory result of critical calcium level decrease, Normann et al (1) proposed that gadodiamide undergoes acid-catalyzed dissociation under the acidic conditions of the OCP assay. Free ligand then competes with OCP for the serum calcium, and this phenomenon results in less calcium-OCP complex to detect photometrically. In a subsequent investigation, Lin et al (2) observed this phenomenon under alkaline conditions, and their findings appear to refute the previously suggested acid-catalyzed dissociation theory. In addition, they observed free ligand and large complexes of 2-gadolinium-2-OCP, which indicated that the gadolinium3+ ion transferred from the gadodiamide to the OCP. Lin et al (2) suggested that the lower thermodynamic stability (Table 6) of gadodiamide allows OCP to be "an efficient competitor ligand" that can completely displace the gadolinium3+ ion from gadodiamide. Consequently, calcium in the serum sample binds to the free ligand and thus is unavailable for measurement with the colorimetric reagent. Note that the patient’s actual serum calcium level is not affected, because this phenomenon occurs in the laboratory test tube.
|
Gadoterate meglumine, gadopentetate dimeglumine, and gadoteridol all have higher thermodynamic equilibrium constants compared with gadodiamide (Table 6), and they do not interfere with colorimetric assays for serum calcium measurement (package inserts: Nycomed Amersham, Mallinckrodt, Berlex Laboratories, and Bracco Diagnostics) (17,18). However, ionic contrast materials, including gadopentetate dimeglumine, are known to bind to serum calcium ions (19), which can transiently suppress cardiac function (20–22).
Because the degree of interference in calcium measurement is directly proportional to the concentration of gadodiamide in the blood sample, the factors that cause increases in blood gadodiamide levels result in greater interference in the calorimetric assay for calcium measurement and thus lower apparent serum calcium measurements. In the present study, as expected, the patients who received high-dose gadodiamide (ie, 
0.2 mmol/kg), the patients from whom blood was drawn shortly after the MR imaging examination, and the patients who had renal insufficiency that caused a delay in the excretion of gadodiamide had the largest serum calcium measurement errors. As time passed, more gadodiamide was excreted and the interference in serum calcium measurements gradually ended. In patients with renal insufficiency, it can take several days for the interference to completely stop. The calcium measurements in patients who are undergoing dialysis are not expected to return to baseline levels until after the next dialysis treatment.
One limitation of this retrospective study was the bias that was introduced by treating multiple gadodiamide-enhanced MR examinations performed in the same patient as separate events. This factor may have weighted the results to reflect those in more ill patients who require multiple MR imaging examinations. Another limitation was the inclusion of only those inpatients with available data on serum calcium levels measured before and within 24 hours after gadodiamide injection. Such patients are also likely to be more ill and have renal insufficiency. Thus, the true prevalence of spurious hypocalcemia following gadodiamide administration in the general population may be lower. However, because of the potential for spuriously lower serum calcium measurements in every patient, we recommend that referring physicians be informed of this potential laboratory error. The protocol could call for referring physicians to draw blood before MR imaging or to wait at least a day or two after MR imaging—or longer in patients with renal failure—before drawing blood. Alternatively, the ionized calcium level could be measured with an ion-specific electrode, or the total calcium level could be measured with atomic emission spectroscopy; however, these methods are more expensive and cumbersome. The ionized calcium level must be measured within 2 hours after drawing fresh blood that is stored on ice.
To help evaluate whether a low serum calcium measurement obtained after gadodiamide administration is spurious, it is useful to examine the patient for signs of true hypocalcemia, including neuromuscular irritability and tetany (which typically manifest as peripheral and perioral paresthesias), oral tetany (ie, the Chvostek sign), carpal or pedal spasm (ie, the Erb sign), anxiety, bronchospasm, laryngospasm, and seizure. Coma and death can also occur. Electrocardiography is useful for identifying the cardiac arrhythmias and QT interval lengthening that occur with true hypocalcemia.
At our institution, high-dose (ie, 2.0 mmol/kg) gadolinium-based contrast material is used primarily in patients who undergo MR angiography. These patients often have diabetes and/or atherosclerotic disease, and they frequently have renal insufficiency as well. Both the high dose of contrast agent and renal insufficiency amplify spurious hypocalcemia following gadodiamide administration. Because of the importance of managing electrolyte levels in patients with renal insufficiency, especially those who are undergoing hemodialysis, these patients are more likely to have blood drawn shortly after an MR imaging examination. This factor further increases the frequency of spurious hypocalcemia following MR angiography. Thus, we prefer to avoid administering gadodiamide in patients who undergo MR angiography. We also prefer not to use gadodiamide in patients who are undergoing dialysis, patients who have neurologic disorders—especially seizure or neuromuscular irritability, patients who have diabetes or a history of renal insufficiency, and patients who have hypoalbunemia or arrhythmia. We also avoid administering gadodiamide in patients in the intensive care unit because of the tendency for frequent blood sampling in critically ill patients.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
See also the editorial by Choyke and Knopp in this issue.
Abbreviation: OCP = ortho-cresolphthalein
Author contributions: Guarantor of integrity of entire study, M.R.P.; study concepts, all authors; study design, M.R.P., H.E.E., R.W.L.; literature research, H.E.E.; clinical studies, H.E.E., J.B., K.C.K., H.L.B.; experimental studies, H.E.E., R.W.L.; data acquisition, M.R.P., H.E.E.; data analysis/interpretation, all authors; statistical analysis, M.R.P., Y.W.; manuscript preparation, M.R.P., Y.W., H.E.E.; manuscript definition of intellectual content, editing, revision/review, and final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  P. M. Colletti Nephrogenic Systemic Fibrosis and Gadolinium: A Perfect Storm Am. J. Roentgenol., October 1, 2008; 191(4): 1150 - 1153. [Full Text] [PDF]
|
|
|
|
 |
|
 G. C. Kane, A. W. Stanson, D. Kalnicka, D. W. Rosenthal, C. U. Lee, S. C. Textor, and V. D. Garovic Comparison between gadolinium and iodine contrast for percutaneous intervention in atherosclerotic renal artery stenosis: clinical outcomes Nephrol. Dial. Transplant., April 1, 2008; 23(4): 1233 - 1240. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. D. Wiginton, B. Kelly, A. Oto, M. Jesse, P. Aristimuno, R. Ernst, and G. Chaljub Gadolinium-Based Contrast Exposure, Nephrogenic Systemic Fibrosis, and Gadolinium Detection in Tissue Am. J. Roentgenol., April 1, 2008; 190(4): 1060 - 1068. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Gandhi, V. R. Narra, J. J. Brown, A. Guo, D. S. Grosu, C. A. Parvin, and M. G. Scott Clinical and Economic Impact of Falsely Decreased Calcium Values Caused by Gadoversetamide Interference Am. J. Roentgenol., March 1, 2008; 190(3): W213 - W217. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Brown, M. R. Hynes, and J. H. Wible Jr. Measurement of Serum Calcium Concentration After Administration of Four Gadolinium-Based Contrast Agents to Human Volunteers Am. J. Roentgenol., December 1, 2007; 189(6): 1539 - 1544. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Broome, M. S. Girguis, P. W. Baron, A. C. Cottrell, I. Kjellin, and G. A. Kirk Gadodiamide-Associated Nephrogenic Systemic Fibrosis: Why Radiologists Should Be Concerned Am. J. Roentgenol., February 1, 2007; 188(2): 586 - 592. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. B. Mark, E. Mazonakis, D. Shapiro, R. J. Spooner, and R. Stuart C. Rodger Pseudohypocalcaemia in an elderly patient with advanced renal failure and renovascular disease Nephrol. Dial. Transplant., July 1, 2005; 20(7): 1499 - 1500. [Full Text] [PDF]
|
|
|
|
|
|
M. Decupere, V. Ooms, R. Oyen, D. R. J. Kuypers, and K. Claes Artifactual hypocalcaemia after intravenous administration of gadodiamide (Omniscan(R)) Nephrol. Dial. Transplant., July 1, 2005; 20(7): 1516 - 1517. [Full Text] [PDF]
|
|
|
|
 |
|
 J. H. Wible Jr and M. R. Hynes Measurement of Serum Calcium Concentration after Administration of Gadoversetamide in Dogs Radiology, October 1, 2004; 233(1): 158 - 164. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. P. Kang, M. G. Scott, B. N. Joe, V. Narra, J. Heiken, and C. A. Parvin Model for Predicting the Impact of Gadolinium on Plasma Calcium Measured by the o-Cresolphthalein Method Clin. Chem., April 1, 2004; 50(4): 741 - 746. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Prince, P. L. Choyke, M. V. Knopp, C. J. Doorenbos, K. Koster, and M. van Wijnen More on Pseudohypocalcemia and Gadolinium-Enhanced MRI N. Engl. J. Med., January 1, 2004; 350(1): 87 - 88. [Full Text] [PDF]
|
|
|
|
|
|
P. L. Choyke and M. V. Knopp Pseudohypocalcemia with MR Imaging Contrast Agents: A Cautionary Tale Radiology, June 1, 2003; 227(3): 627 - 628. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
), phosphorus (
), and magnesium (
) levels measured before and after gadodiamide injection indicates a critical decrease in serum calcium level to 4.2 mg/dL (1.05 mmol/L), according to results of the first blood sampling after gadodiamide injection, with a gradual return to the baseline value during the following 48 hours. However, there is no corresponding elevation in serum phosphorus level, and the patient had no symptoms of hypocalcemia.
