Hypocalcemia

Contrast-enhanced magnetic resonance imaging has become a routine diagnostic imaging procedure. Reports in the literature document that 2 of the 4 available gadolinium-based magnetic resonance imaging contrast agents, gadodiamide (Omniscan) and gadoversetamide (OptiMARK), are less stable and readily undergo dechelation. In vitro, this dechelation can result in interference with the most common laboratory methods used to measure total plasma or serum calcium. The result of total calcium measurement soon after contrast-enhanced magnetic resonance imaging with these interfering contrast agents is a spurious lowering of the total calcium level. This low calcium measurement may result in a value consistent with hypocalcemia and can persist in patients with renal insufficiency and in patients receiving higher doses of contrast agent. Alternatively, a clinically significant elevated calcium level may be overlooked because of the artificially lowered value, two of the available gadolinium-based contrast agents, gadoteridol (ProHance) and gadopentetate dime-glumine (Magnevist), have not been to shown to interfere with total calcium measurement. A clinical practice algorithm for the laboratorian, the radiologist, and the clinician is presented to minimize the occurrence and consequences of a spuriously lowered total calcium level due to Omniscan- or OptiMARK-enhanced magnetic resonance imaging.

(Arch Pathol Lab Med. 2004;128:1151-1156)

During the past decade, use of contrast agents to enhance magnetic resonance (MR) images has provided substantial clinical advancement in diagnostic imaging, with improved lesion detection and characterization. In addition, pathophysiologic data can be obtained from contrast-enhanced MR imaging (MRI).1 Contrast-enhanced MRI is now a common medical procedure, used extensively in hospitals and outpatient facilities for the diagnosis of different neoplastic and inflammatory disorders. Recent survey data indicate that 20% to 25% of all MR studies use contrast enhancement.1 All of the approved intravenous contrast agents have equivalent abilities to enhance MR images and possess well-characterized safety profiles with a low incidence of adverse events.1 Recently, the observation has been made that 2 of the 4 MRI contrast agents approved for use in the United States interfere with the laboratory measurement of total calcium.2 In this review article, we discuss the mechanism of interference with this colorimetric assay and the potential clinical repercussions of this interference. In addition, an algorithm is provided for minimizing or preventing the possibility of this interference.

MRI CONTRAST AGENTS

It has been known since the late 1970s that paramagnetic metal ions such as gadolinium improve the MRI signal, but the toxicity of these uncomplexed metal ions has limited their use. During the early 1980s, use of a chelate to bind the metal ion allowed efforts with these agents to progress. The chelated metal ion could be safely excreted at rates approaching 100%.3 Results of the use of paramagnetic metal ion chelate were first presented at a national meeting in 1982, demonstrating enhanced imaging of the kidneys and urinary bladder.3 The first gadolinium chelate approved for use in the United States was gadopentetate dimeglumine (Magnevist; Schering AG, Berlin, Germany) in 1988, followed by gadoteridol (ProHance; Bracco Diagnostics Inc, Princeton, NJ) in 1992, gadodiamide (Omniscan; Amersham Health, Buckinghamshire, United Kingdom) in 1993, and gadoversetamide (OptiMARK; Mallinckrodt Medical Inc, St Louis, Mo) in 1999.1

Of the 4 agents approved for use in the United States, 1 is an ionic linear molecule (Magnevist), 1 is a nonionic macrocyclic ring structure (ProHance), and 2 are nonionic linear structures (Omniscan and OptiMARK) (Figure 1).3 Although these agents cannot be differentiated based on efficacy or adverse reactions, differences in chelate stability are evident in vitro and in vivo. These differences are primarily because of 2 features of these agents: their thermodynamic stability and their kinetic inertia. These 2 features together define the overall stability of the chelates and ultimately the dechelation rate (ie, rate of release of free gadolinium) of the less stable agents.4 The relevant thermodynamic and kinetic characteristics of the agents are shown in Table 1. The macrocyclic (or ring) chelates that are available demonstrate superior stability as evidenced by their higher thermodynamic stability constant and lower dissociation rate.3 In vivo stability has been evaluated in healthy human volunteers by analyzing displacement of zinc with free gadolinium and subsequent measurement of zinc excreted in the urine. As predicted by in vitro stability data, the ring chelate ProHance resulted in the lowest zinc excretion, Magnevist resulted in an intermediate amount, and Omniscan resulted in the highest level of zinc excretion.5

INTERFERENCE WITH LABORATORY MEASUREMENT OF TOTAL CALCIUM

Calcium is the main extracellular divalent cation in the body. Most of the calcium in the body (99%) is found in bones, with the remaining 1% found in extracellular spaces and within cells. Calcium is essential for many important processes, including neuronal excitability and neurotransmitter release, muscle contraction, membrane integrity, and coagulation of blood.6 In addition, calcium functions as a second messenger for many hormones.6 The most common analytic methods for measurement of total calcium use automated colorimetric assays; according to the 2003 College of American Pathologists7 chemistry survey, 82'/O of chemistry laboratories use these techniques. Total calcium contains approximately 50% unbound calcium and 50% bound calcium, most of which is complexed to albumin." However, various factors can alter total plasma or serum calcium measurements, including diet and serum albumin concentration.8

In addition to total calcium, ionized (or free) calcium can be estimated or directly measured; estimated ionized calcium is derived from the total serum calcium plus albumin (or total protein) values, whereas direct measurement of ionized calcium is performed using ion-selective electrode equipment.1' Estimates of ionized calcium are not accurate in critically ill patients as these patients frequently experience physiologic changes that affect calcium homeostasis.10 Therefore, direct measurement of ionized calcium is becoming the preferred method used to assay calcium levels, particularly in the acute care setting in which a rapid, accurate result is especially important." According to the 2003 College of American Pathologists7 survey, approximately 13% of laboratories performing calcium measurements have methods available for direct ionized calcium measurement. Most blood gas laboratories, nearpatient laboratories with whole-blood analyzers, and several point-of-care testing instruments used in the operating room, intensive care unit, and emergency department can measure ionized calcium along with blood gases as part of a rapid response test cluster.11,12

In 2003, Prince and colleagues2 found laboratory results indicating a "life-threatening" hypocalcemia in a patient after an MRI in which contrast enhancement with Omniscan was used; nevertheless, the patient had no clinical symptoms of hypocalcemia, suggesting the possibility of interference with a laboratory measurement. In fact, this phenomenon had been reported much earlier in the literature.11,14 In 1995, Normann and colleagues13 had reported that Omniscan interferes with the colorimetric determination of serum calcium, and that the extent of interference correlates with the concentration of Omniscan injected. They postulated that the gadolinium complex was undergoing an acid-catalyzed dissociation. They further postulated that the free ligand was then able to compete with o-cresol-phthalein (OCP) for the calcium in the sample. Normann and colleagues11 confirmed that actual serum calcium levels were not affected by calcium measurement using an ion-selective electrode or inductively coupled plasma-atomic emission spectroscopy.

Lin and colleagues14 expanded on this work in 1999 by comparing the interaction of 2 linear contrast agents (gadodiamide and gadopentetate dimeglumine) and 1 cyclic (Gd-DOTA) contrast agent with the OCP method of calcium measurement. They found that only gadodiamide was associated with a strong interference in the assay resulting in extremely low apparent calcium values. However, Lin and colleagues refuted the mechanism proposed by Normann et al13 because they observed interference in the absence of acidic conditions. Instead, Lin and colleagues proposed that the decomplexation phenomenon observed with gadodiamide was likely related to the complete displacement of gadolinium from gadodiamide by OCP (Figure 2). The gadodiamide ligand would then be free to compete with the OCP for binding to any calcium in the sample.14

All 4 approved MRI contrast agents have since been tested in vitro with pooled human serum for interference with the OCP calcium measurement technique. Only Omniscan and OptiMARK interfered with the assay; ProHance and Magnevist did not (Figure 3).2 In 2002, the product labeling for Omniscan was changed to state that this agent interferes with colorimetric serum calcium determinations.15

Further investigations of the prevalence and clinical implications of this interference phenomenon were conducted for almost 2 years at New York-Presbyterian Hospital, New York, NY.2 Prince et al2 examined 1049 patients who had received an MRI with the contrast agent Omniscan and had an available calcium measurement before and after the imaging procedure. Significant decreases in serum calcium measurements were found in patients between the precontrast injection and the measurement after Omniscan administration (Table 2). Forty-two patients (4.0%) had calcium measurements that fell by more than 2.0 mg/dL (>0.50 mmol/L), and 25 patients (2.4%) had calcium levels that decreased to less than 6.0 mg/dL (

None of these patients were noted to have exhibited symptoms of hypocalcemia; nevertheless, several received inappropriate treatment.2 Eleven patients were given oral calcium and 7 patients were given intravenous calcium. None of these patients had adverse outcomes that were attributed to the calcium treatments, although the calcium administration complicated treatment in 3 of the patients, 2 with seizure disorders and 1 in a coma due to encephalitis.2

Other reports of spurious hypocalcemia exist in the literature. Doorenbos and colleagues16 observed 7 cases of spurious hypocalcemia after Omniscan-enhanced MRI. The authors identified the first case after a laboratory report of hypocalcemia that was unaccompanied by clinical symptoms in the patient, and 6 other cases were subsequently confirmed. In another recent report, 11 patients with cirrhosis were found to have spurious hypocalcemia after MRI of the liver.17 Because hypocalcemia may be secondary to other complications such as malnutrition and vitamin D deficiency in patients with cirrhosis, the authors recommended that interference from MRI contrast agents be considered in the differential diagnosis of hypocalcemia for patients with cirrhosis.17

Recently, interference of the gadolinium-based MRI contrast agents has been evaluated in vitro in serum assays using multiple analyzers.18 In addition to confirming the negative interference of Omniscan and OptiMARK with colorimetric assays for total calcium, clinically significant positive and negative interference by these 2 agents was detected in the assays for serum angiotensinconverting enzyme, zinc, magnesium, and total iron-binding capacity. Magnevist was also found to negatively interfere with iron assays.18

All 4 agents interfered with a manual colorimetric zinc assay, but to different degrees ranging from 85% interference with Omniscan to 13% interference with ProHance.18 This interference with zinc measurement may be related to the amount of free ligand or calcium chelate that is added to the various contrast agent formulations to decrease the release of gadolinium ions.19 This amount of free ligand or calcium chelate varies widely among the different contrast agents: Omniscan is formulated with 50fold greater calcium chelate (5% Na[Ca-DTPA-BMA]) than ProHance (0.1% Ca[Ca-HP-DOSA]2).19 As mentioned earlier, to varying degrees depending on kinetic inertia, the different chelates are capable of transmetallation with zinc." Therefore, calcium chelate in Omniscan may more readily undergo transmetallation in vitro and scavenge the available zinc, leading to an apparent lowering of the zinc level.

Proctor et al18 concluded that Omniscan and OptiMARK are more likely to produce clinically significant interference because of their effect on total calcium measurement. Notably, all 4 agents interfered minimally with an ionized calcium assay (5% interference by Omniscan and OptiMARK, and 2% interference by Magnevist and ProHance), although only one analyzer was tested. Measurement of total magnesium, a divalent cation with properties similar to those of calcium, was associated with interference by Omniscan and OptiMARK with certain analyzers, but ionized magnesium was not included in the study.18

CLINICAL SIGNIFICANCE OF SPURIOUSLY LOW TOTAL CALCIUM MEASUREMENTS

In addition to the risk of treating a nonexistent hypocalcemia, it is important to consider the risk of missing a true hypercalcemia. Hypercalcemia can be the hallmark for many serious disorders that often coexist in a patient requiring an MRI. Hypercalcemia is common, with 10% to 20% of patients with cancer developing hypercalcemia at some point during their disease20; in addition, if present at initial presentation, hypercalcemia may trigger a clinical workup that ultimately results in a diagnosis of cancer.20 Primary hyperparathyroidism is another common cause of hypercalcemia, occurring in 75 per 100 000 hospitalized patients.20

Aside from the clinical repercussions of spurious hypocalcemia, economic costs must be considered as well. Costs are incurred associated with the handling of a critical or "stat" value, within and outside the laboratory. Critical values, recognized as life-threatening levels, are associated with a specific clinical management response and are typically handled differently from other abnormal results: attempts are made by laboratory personnel to identify and directly contact the ordering clinician. Not all laboratories have dedicated personnel to handle such contact, and staff often must be removed from their normal workstation.

An important area of economic consideration is the potential for liability resulting from inappropriate patient treatment. Diagnoses of disorders in which calcium levels are particularly relevant may be clouded by a spurious hypocalcemia measurement or a missed incidence of hypercalcemia. Inappropriate treatment of a patient with oral calcium is less likely to have serious clinical implications; however, intravenous calcium can be dangerous and is no longer considered an option except in limited situations, such as severe hypocalcemic tetany and laryngospasm." The possibility of doing harm to a patient based on a hypocalcemia measurement that is not appropriately attributed to the use of Omniscan or OptiMARK exists, and the potential clinical significance must be recognized by clinicians, laboratorians, and those responsible for risk management.

PATIENT POPULATIONSAT INCREASED RISK

Patient populations at increased risk for spuriously lowered total calcium measurements after contrast-enhanced MRI with interfering contrast agents have been identified (Table 3).2 Individuals at increased risk include patients whose blood is monitored frequently, such as critically ill patients in intensive care units and patients on hemodialysis. Frequent blood sampling increases the likelihood that a calcium measurement will be made while the apparent serum calcium level is still low. As noted earlier, direct measurement of ionized calcium in the acute care setting is recommended, and this assay seems to be associated with only minimal interference by any of the 4 available MRI contrast agents.18

In patients with normal renal function, measured calcium levels are free of interference within 12 hours; however, in patients with renal insufficiency or other causes for reduced excretion of the interfering MRI contrast agent, spuriously low calcium levels can persist for more than 24 hours.2 Prince and colleagues2 also showed that, as expected, patients receiving higher doses of contrast agent (≥0.2 mmol/kg of body weight) had the largest errors in their serum calcium measurement, and the use of higher doses of contrast agents for some MR applications is increasing.

Certain diagnoses can be confounded by a finding of spurious hypocalcemia. These include different neurologic disorders such as seizure and neuromuscular irritability. Patients with diabetes mellitus, a history of renal insufficiency, hypoalbuminemia, arrhythmia, and cardiac contraction problems are at increased risk for a spuriously low total calcium measurement after injection with Omniscan or OptiMARK.2

CLINICAL ALGORITHM

An algorithm has been developed to address the issue of spuriously low calcium measurements after the administration of MRI contrast agents (Figure 4). The foundation of the algorithm is the education of laboratorians, radiologists, and clinicians. Increased awareness of the causes and potential results of spuriously lowered calcium measurements is reflected in the algorithm, as well as the consideration of choosing an MRI contrast agent that does not interfere with calcium measurement, particularly for higher-risk populations.

Clinical chemists and laboratory personnel need to be aware that this interference exists and may result in a false-positive critical value. If the MRI contrast agent used by a radiology department can cause spuriously low total calcium measurements, then the chemistry laboratory must address this potential interference and have appropriate policies in place. A method to directly measure ionized calcium needs to be available, particularly for the critically ill. Protocols need to be available and staff properly trained to perform such alternative methods in a clinically relevant time frame.

Automated warning systems can be used when a critical hypocalcemia measurement is detected; however, this tactic does not address the possibility of missing a true hypercalcemia. Resorting to accompanying all calcium measurements with an alert about interference caused by contrast-enhanced MRI would not serve to improve overall communication between the laboratory and clinicians. It is not a generalizable practice given the large number of well-documented interferences with the measurement of many analytes. Practice alternatives that remove the likelihood of interference constitute the preferred approach. An educational initiative is recommended to bring laboratorians and radiologists together to ensure that all involved understand the cause and potential risks associated with a spuriously low total calcium result.

Clinicians should also be educated on the potential for interference in a calcium measurement after administration of an MRI contrast agent. Clinicians need to become familiar with the method that is typically being used in their institution to measure calcium, as well as the availability of alternative methods. Particularly when a high dose of contrast agent is used, notification of clinicians is prudent because high doses are associated with a larger apparent decrease in total calcium.2

Clinicians and radiologists need to be aware of the key conditions that are relevant, such as renal dysfunction. The patient's baseline calcium can be measured and then subsequently tracked over time to observe the excretion trend. Alternatively, a model has been developed to predict, based on the glomerular filtration rate of the patient, the minimum length of time to wait before blood collection to avoid a spuriously low total calcium result when a colorimetric method is used.15

Institutions as a whole, including laboratory personnel, clinicians, radiologists, risk management officers, and other administrators, need to assess the risks regarding patient outcome that are being incurred when choosing a contrast agent that causes interference in the laboratory measurement of a common electrolyte. The issue of whether particular patient populations should never receive such an interfering agent because of the increased risks associated with a spurious hypocalcemia is probably best assessed by individual institutions.

CONCLUSIONS

Of the 4 MRI contrast agents available in the United States, 2 of them, Omniscan and OptiMARK, interfere with the measurement of total calcium by colorimetric assays. Patients at increased risk for inaccurate readings include those with renal insufficiency and patients receiving higher doses of contrast agents. In institutions using contrast agents that have been shown to cause spuriously low calcium measurements, recognizing the potential for diagnostic errors, including spurious hypocalcemia and missed hypercalcemia, is important. Preventing serious consequences of spurious hypocalcemia, such as mistreatment with intravenous calcium, requires vigilance and coordination of effort among laboratorians, radiologists, and clinicians. Direct measurement of ionized calcium levels should be available for critically ill patients and others who may require immediate or frequent blood sampling. Individual institutions should assess the potential risk for these diagnostic errors based on the contrast agent in use, the calcium assay used in the laboratory, and the laboratory's capability to provide alternative methods with adequate response times.

References

1. Muroff LR. MRI contrast: current agents and issues. Appl Radial. 2001; 30(suppl):5-7.

2. Prince MR, Erel HE, Lent RW, et al. Gadodiamide administration causes spurious hypocalcemia. Radiology. 2003;227:639-646.

3. Runge VM. Safety of magnetic resonance contrast media. Top Magn Reson Imaging. 2001:12:309-314.

4. Tweedle MF, Hagan JJ, Kumar K, Mantha S, Chang CA. Reaction of gadolinium chelates with endogenously available ions. Magn Reson Imaging. 1991;9: 409-415.

5. Puttagunta NR, Gibby WA, Smith GT. Human in vivo comparative study of zinc and copper transmetallation after administration of magnetic resonance imaging contrast agents. Invest Radiol. 1996:31:739-742.

6. Marcus R. Agents affecting calcification and bone turnover. In: Hardman JG, Limbird LE, Gilman AG, eds. Goodman & Gilman's The Pharmacological Basis of Therapeutics. 10th cd. New York, NY: McGraw-Hill; 2001:1715-1773.

7. College of American Pathologists. Surveys 2003 (Chemistry): Participant Summary. Northfield, III: College of American Pathologists: 2003:23, 59.

8. Ravel R, ed. Serum electrolytes and protein-calorie malnutrition. In: Clinical Laboratory Medicine. 6th ed. St Louis, Mo: Mosby-Year Book: 1995:406-435.

9. Kost GJ, Jammal MA, Ward RE, Safwat AM. Monitoring of ionized calcium during human hepatic transplantation. Am J CUn Pathol. 1986:86:61-70.

10. Holloway PAH. Point-of-care testing in intensive care. In: Kost GJ, ed. Principles and Practice ot Point-ot-Care Testing. Philadelphia, Pa: Lippincott Williams & Wilkins: 2002:133-156.

11. Kost GJ, Bullock J, Despotis GJ. On-site and near-patient testing in the operating room. In: Kost GJ, ed. Principles & Practice of Point-of-Care Testing. Philadelphia, Pa: Lippincott Williams & Wilkins: 2002:119-132.

12. Kost GJ. The hybrid laboratory, therapeutic turnaround time, critical limits, performance maps, and knowledge optimization. In: Kost GJ, ed. Principles and Practice of Point-of-Care Testing. Philadelphia, Pa: Lippincott Williams & Wilkins: 2002:13-25.

13. Normann PT, Froysa A, Svaland M. Interference of gadodiamide injection (OMNlSCAN®) on the colorimetric determination of serum calcium. Scand I CUn Lab Invest. 1995:55:421-426.

14. Lin I, Idee JM, Port M, et al. Interference of magnetic resonance imaging contrast agents with the serum calcium measurement technique using colorimetric reagents. J Pharm Biomed Anal. 1999:21:931-943.

15. Kang HP, Scott MG, Joe BN, Narra V, Heiken J, Parvin CA. Model for predicting the impact of gadolinium on plasma calcium measured by the o-cresolphthalein method. Clin Chem. 2004:50:741-746.

16. Doorenbos CJ, Ozyilmaz A, van Wijnen M. Severe pseudohypocalcemia after gadolinium-enhanced magnetic resonance angiography. N Engl I Med. 2003:349:817-818.

17. Kefalas CH, Murray NGB, Aguanno JJ, et al. Pseudohypocalcemia after magnetic resonance imaging with gadolinium in patients with cirrhosis. Liver Transpl. 2004:10:136-140.

18. Proctor KAS, Rao LV, Roberts WL. Gadolinium magnetic resonance contrast agents produce analytic interference in multiple serum assays. Am J Clin Pathol. 2004:121:282-292.

19. Corot C, Idee |M, Hentsch AM, et al. Structure-activity relationship of macrocyclic and linear gadolinium chelates: investigation of transmetallation effect on the zinc-dependent metallopeptidasc angiotensin-converting enzyme. J Magn Reson Imaging. 1998:8:695-702.

20. Hemphill RR, Hypercalcemia. Emedicine [serial online]. 2002. Available at: http://www.emedicine.com/emerg/topic260.htm. Accessed February 9, 2004.

Jane Emerson, MD, PhD; Gerald Kost, MD, PhD

Accepted for publication June 11, 2004.

From the Department of Pathology, UCI Medical Center, University of California, Irvine, Orange (Dr Emerson); and Department of Pathology and Laboratory Medicine, University of California, Davis (Dr Kost).

The authors have no relevant financial interest in the products or companies described in this article.

Reprints: Jane Emerson, MD, PhD, Department of Pathology, University of California, Irvine, UCI Medical Center, 101 The City Dr, Orange, CA 92868 (e-mail: jemerson@uci.edu).

Copyright College of American Pathologists Oct 2004
Provided by ProQuest Information and Learning Company. All rights Reserved