Practical Therapeutics

Drugs 40 (5): 713-721 , 1990 0012-6667/90/0011-0713/$04.50/0 © Adis International Limited All rights reserved. DRUG03403

The Clinical Application of Radiopharmaceuticals Norman E. Leeds Department of Radiology, Beth Israel Medical Center and Mt Sinai School of Medicine, New York, New York, USA

Contents

Summary ............... .. ....................................... .. ....... ....................................... ........................... 713 I. Clinical Indications .............................................................................. .... ........... .. ............... 715 2. Magnetic Resonance Imaging ........................................... .............................................. ..... 717 2.1 Choice of Paramagnetic Substances ..... .......................................... .. .......... .. ................ 718 2.2 Clinical Effectiveness .. ................................................... .................... ............................ 718 3. Clinical Treatment Recommendations ............ ..... .. .. .......... ............................................ .... 720

Summary

This article highlights the choices and the arguments in the selection of appropriate contrast materials in radiological examinations - nonionic versus ionic contrast material - and aims to assist the physician in decision-making. Various authors have raised questions concerning the proposed advantages of nonionic contrast material. However, studies in low risk patients have shown more complications with the use of ionic contrast than nonionic contrast materials; this is the important group of patients since in high risk patients nonionics are used almost exclusively. The important factor that increases the controversy is cost, which is significant since nonionic agents cost 10 to 15 times more than ionic agents in the USA. Thus, costbenefit considerations are important because price sensitivity and cost may determine fund availability for equipment or materials that also may be necessary or important in improving patient care. In magnetic resonance imaging (MRI), as in computed tomography (CT), the use of contrast material has improved diagnostic accuracy and the ability to reveal lesions not otherwise easily detected in brain and spinal cord imaging. These include separating scar from disc, meningitis, meningeal spread of tumour, tumour seeding, small metastases, intracanalicular tumours, separating major mass from oedema, determining bulk tumour size and ability to demonstrate blood vessels so dynamic circulatory changes may be revealed.

Intravascular contrast examinations are often performed within hospitals and in outpatient centres; approximately 8 million to 10 million contrast examinations are currently performed each

year (Steinberg et al. 1988). Aside from myelography, these examinations are performed using solutions containing meglumine, meglumine-sodium, or sodium derivatives, diatrizoate or iothal-

Drugs 40 (5) 1990

714

Table I. Osmolality of currently used contrast agents (reproduced in part from Fisher 1986) Agent

Percentage in solution (mg iodine/ml fluid)

Higher osmolality intravascular contrast media lodamide meglumine 24 Diatrizoate meglumine 30 Diatrizoate meglumine 30 lothalamate meglumine 60 Diatrizoate meglumine 60 Diatrizoate meglumine 60 Diatrizoate sodium 8% 60 meglumine 52% lothalamate sodium 54.3 Diatrizoate sodium 10% 76 meglumine 66% Lower osmolality intravascular contrast media lohexol 38.8 lopamidol 40.8 lohexol 51.8 lopamidol 61 lohexol 64.7 loxaglate 58.9 sodium 19.6% meglumine 39.3% lohexol 75.5 lopamidol 76

amic acid, which are ionic high osmolality contrast agents (HOCA), although in some cases low osmolality contrast agents (LOCA), which may contain ionic or non ionic materials, are used. Almen (1969) reported the development of a low osmolality nonionic contrast agent, and in the 1970s metrizamide became available for myelography, gradually replacing iophendylate pantopaque. The reasons for the success of metrizamide despite a marked increase in cost were the absence of developing arachnoiditis (which occurred in up to 25% of cases with iophendylate), the ability to be imaged with computed tomography (CT), and the fact that it did not need to be removed. The removal of iophendylate was time consuming, often difficult, and frequently required a second or third lumbar puncture for complete removal. In 1985, 3 new low osmolality contrast agents became available following Food and Drug Administration (FDA) approval: ioxaglate (,Hex-

Trade name

Iodine (gIL)

Osmolality (mOsm/kg)

Renovue-Dip Hypaque 30% Reno-M-Dip Conray Hypaque 60% Reno-M-60 Renografin-60

111 141 141 282 282 282 292

433 633 566 1400 1415 1500 1420

Conray-325 Hypaque-76

325 370

1700 2016

Omnipaque Isovue-M-200 Omnipaque Isovuc-300 Omnipaque Hexabrix

180 200 240 300 300 320

411 413 504 616 709 600

Omnipaque Isovue-370

350 370

862 796

abrix'), which is not non ionic but has low osmolality, and 2 nonionic agents, iohexol and iopamidol. These non ionic contrast agents have replaced metrizamide for myelography because they are more stable, have lower morbidity and are less costly than metrizamide (Elkin et al. 1986; Witwer et al. 1984). The use of low osmolality agents leads to an apparent reduction in chemotoxicity (in comparison with high osmolality agents) [Fisher 1986]; this is the damage to the endothelial lining of the blood vessel wall that may occur at the site of injection as a consequence of the osmolality of the contrast agent, as well as vascular wall damage by other direct effects of contrast material. A variety of chemical changes may result, but the actual inciting mechanism is only speculative (Dawson 1985). What is this osmolality (table I; Ansell 1970) that accounts for the chemotoxicity of HOCA? The osmolality of a solution is dependent on the num-

715

Clinical Use of Radiopharmaceuticals

Table II. Aetiology of contrast media reactions Allergy Anaphylaxis Cardiovascular collapse Anxiety induced by the examination Hyperosmolality Nephrotoxicity

ber of particles in solution, and inversely on the molecular weight of the compound (McClennan 1987). Osmolality results in altered physiological changes, which contribute to recognised side effects (tables II and III); these include hypervolaemia, cellular diuresis, urinary diuresis, altered vascular and glomerular endothelial permeability, vasodilation causing decreased vascular resistance, hypotension and pain, cardiac conduction defects such as sinus bradycardia, and permeability of the bloodbrain barrier (Brasch 1988). Because of their composition, nonionic LOCA (i.e. iopamidol and iohexol) do not dissociate and therefore maintain their iodine concentrations of 3 iodine atoms to I particle in solution (3: 1) [McClennan 1987]. The ionic LOCA ioxaglate is composed of 6 iodine atoms to 2 particles in solution (6: 2; 3: I). Therefore, since these agents have low osmolality, chemotoxicity is reduced. The chemotoxic effects of osmolality were first recognised by Almen (1969), who was instrumental in the development of low osmolar contrast media (LOCM). High osmolar contrast media (HOCM) dissociate in solution into an anion and cation. The anion contains the iodine and represents 50% of the compound molecular weight; thus, with these compounds the number of particles in solution and molecular weight result in an effective doubling of osmolality (McCiennan 1987). A variety of contrast reactions have been described (table II). In addition to the effects previously commented upon, the pathophysiological effects caused by the high osmolality and chemotoxicity of HOCM also include increased histamine release and elevated IgE (complement activation). Acetylcholinesterase inhibition is also reduced with the use of LOCM (Brasch 1988; Daw-

son & Edgerton 1983; Howell & Dawson 1985; Lasser 1985).

1. Clinical Indications A significant benefit of LOCM is almost no pain or heat, which provides a significant advantage during brachial or femoral angiography. These contrast materials are also advantangeous in the very young or older patient in whom the effects of chemotoxicity and osmolality would lead to increased patient risk. In the patient experiencing a prior contrast reaction or who has an allergic history, these agents should also be used. It has also been observed that during invasive examinations of the heart, including coronary artery examination, an increase in myocardial sensitivity occurs and potential c~rdiac conduction defects are equal, and unusual, with HOCM or LOCM (Kinnison et al. 1989). Another group of patients who benefit from LOCM agents are those with hypertension and cardiac or myocardial dysfunction. Although renal effects are reported to be less with LOCM, there is currently not enough evidence to prove the need to use LOCM in patients with diabetes or renal dysfunction unless the patient has a significant reduction in glomerular filtration rate.

Table III. Contrast medical reactions Mild Nausea Vomiting Headache Pruritus Hives

Severe Respiratory difficulty Cardiovascular problems hypotension congestive heart failure pulmonary oedema arrythmia shock Seizure Renal shutdown Death

716

A disadvantage of LOCM in intravascular catheter studies is the increase in the potential for clot formation compared to HOCM, which tend to inhibit the formation of clots (Ing et al. 1989). Since LOCM seem to have such significant advantages over HOCM, why have they not replaced HOCM except in certain high-risk patients or procedures? The reason is the significant difference in cost in the US; LOCM cost 10 to 15 times more than HOCM. The question of whether they are worth this significant added expense is a very difficult, or almost impossible, question to answer since if cost were not a factor then LOCM would have replaced HOCM in almost all instances. It is the cost that leads to careful analysis and studies of complication rates and side effects of these 2 very different types of contrast agents. In fact, rather than being used in almost 100% of cases, LOCM are being utilised in 20 to 50% of patients in many institutions, although some have switched over exclusively to LOCM. 'Who will pay for this significant increase in cost?' is the major question. Should the patient, hospital, insurance company or government bear the cost? Until this question is resolved, there will be variations in usage and studies will be performed to investigate the indications and need for use of LOCM. McClennan (1987), in an excellent paper on the subject, feels the cost factor results in circular reasoning and that if the LOCM are truly better they should be used, ignoring the chants of 'liquid gold', 'cost effective', or 'cost! benefit'. White and Halden (1986), who use the term 'liquid gold', highlight effectively the significant impact of cost on the Radiology Department. While McCiennan's thinking is attractive, the cost effects highlighted by White and Halden cannot be ignored. The economic impact of exclusive LOCM use would be significant; large departments which currently spend $50,000 on contrast agents would see this cost explode to $750,000 or more for contrast media alone. Are they worth these significant cost differentials? A recent review comparing randomised controlled trials of LOCM versus HOCM by Kinnison et al. (1989) revealed that no significant differences occurred regarding nephrotoxicity, frequency of

Drugs 40 (5) 1990

nausea, vomiting, urticaria, bronchospasm, laboratory test abnormalities or neurological events; cardiovascular abnormalities, however, were substantially greater with HOCM. The authors point out that to substantiate these variations studies of larger numbers of patients are necessary, as well as stratifying low and high risk patients. Powe et al. (1988), in a prospective study of 795 consecutive patients using intravascular contrast in examinations for cardiac catheterisation, peripheral angiography, and CT of head and body, examined the costs related to complications using LOCM and HOCM. Although the complications were higher with HOCM, when they took into consideration the costs of caring for complications as well as variance in the cost of contrast media, LOCM was still significantly more costly. Recent large studies in Australia (Palmer 1988) and Japan (Katayama et al. 1988) have shown certain advantages of LOCM, particularly in low risk groups. Wolf et al. (1989) reported on adverse effects of ionic and nonionic contrast agents and noted that the use of ionic contrast media resulted in higher adverse effects (4.17% compared to 0.69%), and were similar to the Australian report (Palmer 1988). In a recent commentary on the use of ionic versus non ionic contrast media by Wolf et al. (1989), Lasser and Berry (1989) state that a true randomised study would be of benefit, but they wonder if it is possible! They also state, 'non ionic contrast media apparently confer protection, but a precise unbiased quantification of degree of protection is lacking.' A review of medical, economic, legal and policy issues concerning use of LOCM by Jacobson and Rosenquist (1988) predicts an incremental cost of I billion dollars per year if implementation of LOCM occurs. The authors suggest that this total replacement would not be cost effective, and the use of LOCM should be limited to high risk patients. The debate goes on, and physicians will have to make decisions in view of significant cost differentials until the variation in protection and safety can be verified.

Clinical Use of Radiopharmaceuticals

2. Magnetic Resonance Imaging Magnetic resonance imaging (MRI) provides unexcelled imaging of the central nervous system, musculoskeletal system and spine. The reason for this is the high signal to noise ratio, particularly of superconductive magnets at high field strength (I.5T) [15 000 gauss]. (The signal reflects the features that contribute to image formation, while noise refers to the inherent factors that cause degrading of the image.) Excellent images are possible because of excellent contrast and spatial resolution. The image reflects a summation of T I and T 2 relaxation times, proton density and flowing blood (TI and T2 refer to the relaxation times of the protons that significantly contribute to image intensity. T I is the time required for tissue recovery following an applied radiofrequency pulse to a patient within a magnetic field. T2 is the time required for tissue protons to lose uniform orientation or to decay in a magnetic field after the application of a radiofrequency pulse). The quality of the images with this noninvasive technique, without radiation, has been acknowledged universally. Again, however, as in CT, the question has been raised concerning the usefulness of contrast agents to add diagnostic information and improve specificity. The reason for this is that although MRI provides excellent images, and sensitivity or lesion detection is high, specificity or lesion characterisation is low. It was felt that a contrast agent might improve lesion characterisation. The search for contrast materials began soon after MRI became practical. Excellent reviews on this subject have been published by both Brasch (1983) and Runge et al. (1983). Paramagnetic agents (table IV) cause a shortening of T I and T 2 relaxation times by changing the magnetic environment and thus influencing neighbouring hydrogen nuclei (Brasch 1983; Runge et al. 1983). In addition, these agents have their own magnetic moment (the magnetic effect of a paramagnetic substance). The alignment of these magnetic moments is random except when the substance is

717

Table IV. Paramagnetic substances (taken from Brasch 1983; Runge et al. 1983) Unpaired electrons Nitric oxide (NO) Nitrogen dioxide (N02) Paired electrons with parallel spins Molecular oxygen (02) Metallic ions Transitional metal series titanium (Ti 3+) iron (Fe 2+. Fe 3+) vanadium (V4+) cobalt (Co3+) chromium (Cr2+, Cr3+) nickel (Ni 2+) manganese (Mn2+, Mn 3+) copper (Cu 2+) Lanthanide series praseodynium (Pr3+) gadolinium (Gd 3+) europium (Eu 3+) dysprosium (Dy3+) Actinide series protactinium (Pa 4+) Nitroxide stable free radicals (NSFR) Pyrrolidine derivatives Piperidine derivatives

placed in a magnetic field, when they become appropriately aligned because of their magnetic properties (Runge et al. 1983). The local magnetic field that results from the effects of the paramagnetic substance causes shortening of T I and T 2 relaxation times of neighbouring hydrogen nuclei, a process that has been labelled 'proton relaxation enhancement' (PRE) [Brasch 1983]. Radiographic contrast agents, particularly iodinated materials, are effective because of an associated high electron density that produces its contrast effect by the absorption of the generated xrays. The effect of paramagnetic agents effective during MRI is caused by their altering local magnetic environment and producing magnetic susceptible effects (Brasch 1983; Runge et al. 1983). Paramagnetic materials have one or more unpaired protons, neutrons or electrons. An unpaired particle is effective since it will not have a similar

718

Drugs 40 (5) 1990

particle with a spin in the opposite direction that would cancel it. The significant feature that makes the paramagnetic particle effective is the strength of its magnetic moment. The magnetic moment of unpaired electrons is much greater than that of protons or neutrons (Brasch 1983). Paramagnetic substances then cause their effects by inducing changes in local magnetic field and cause a reduction in T I and T 2 relaxation times. These effects are induced by the concentration of paramagnetic substance and the square of their magnetic moment (Brasch 1983). 2.1 Choice of Paramagnetic Substances A number of paramagnetic elements have been examined (table IV). The substance with the highest magnetic moment is gadolinium (Od), from the lanthanide series, with 7 unpaired electrons. However, although it is an excellent paramagnetic substance with a high T I relaxivity of 9 mmol/sec, gadolinium forms an insoluble salt and is extremely toxic, with intracellular deposition. (Relaxivity effects refer to the shortening of relaxation times, particularly T I, and are measured as signal change/ mol.) The combination of gadolinium with an appropriate chelate - DTPA - _makes it more stable, prevents intracellular deposition and facilitates rapid and complete excretion while still maintaining a high T I relaxivity of 4.5 mmol/sec (Runge et al. 1983). In clinical trials Od DTPA demonstrated excellent T I relaxivity and biodistribution, was appropriately metabolised [with excretion primarily (90 to 95%) in urine and 5 to 10% in faeces], and had high stability and acceptable toxicity (Weinmann et al. 1984). A number of minor reactions have been observed, for example headache, nausea, urticaria, pruritus, hypotension and, rarely, seizures, but no serious side effects have as yet been reported with the dosages recommended. An important point is that T I effects (relaxivity) are similar up to a magnetic field strength of I.OT. After 1.0T a decrease in relaxivity, and thus effectively less T I shortening, has been demonstrated. Chromium and manganese have also been stud-

ied as chelates, but they are not as stable or safe as gadolinium. Od DTPA is administered intravenously, at a dose of 0. 1 mmol/kg. The drug received FDA approval in 1988 for brain studies, since the safety and effectiveness of this drug were demonstrated during appropriate clinical trials (Russell et al. 1987). Od DTPA is distributed throughout the intravascular space of the brain and then may accumulate in portions of the brain parenchyma when disruption of the blood-brain barrier occurs, similar to intravenous contrast administration during CT examinations. Od DTPA uptake in the brain is therefore related to administered dose, time of examination, pathophysiological changes within the brain, and field strength of magnet. Other paramagnetic contrast agents are being investigated or are undergoing clinical trials. These include Od-DOT A, which would have similar applications to Od DTPA. In addition, various ironcontaining compounds are being investigated to examine the gastrointestinal tract. Other substances are being examined which will aid in evaluating liver and kidneys. Recently, perfluorooctylbromide has been developed as a new contrast agent (Mattrey 1989). This agent may be given orally or emulsified for intravenous administration. The portion imaged is the _19F nucleus, which is the next best nucleus after hydrogen for MR imaging. These compounds are being investigated in imaging vascular pool, liver, spleen and macrophage collections. 2.2 Clinical Effectiveness Clinical trials of Od DTPA have been completed to determine its safety and clinical efficacy, and the combination has received FDA approval for use in the brain. A paper by Kilgore et al. (1986) highlighted the normal appearance of Od DTPA within the brain on T I images after intravenous administration. In the normal patient, Od DTPA is distributed through the intravascular compartment without crossing the blood-brain barrier. Blood vessels and

Clinical Use of Radiopharmaceuticals

cerebrospinal fluid are not affected. In a majority of cases meningeal enhancement is seen because of its normal vascularity. The pituitary gland and infundibulum show a marked and homogeneous increase in signal intensity; the authors state an increase of 60 to 80%. The cavernous sinus is of equal or greater intensity than the pituitary gland. In addition, an increase in intensity in the intracavernous segments of the cranial nerves III to VI, the gasserian ganglion as well as the second and third divisions of the fifth cranial nerve, is revealed (Kilgore et al. 1986). The mucosa of paranasal sinuses and nasopharynx also demonstrated an increase in signal intensity. The choroid plexus in the lateral and fourth ventricle increased an average of 70%. The retinal choroid becomes enhanced, but the optic nerves and intrinsic eye muscles are not affected. It is important to recognise which normal structures enhance after intravenous administration of Gd DTPA, so that normal accumulations can be distinguished from abnormal accumulations, to improve diagnostic accuracy (Kilgore et al. 1986). Gd DTPA may be useful in identifying time of appearance of an abnormality as well as time of emptying; duration of contrast opacification may also prove useful. Russell et al. (1989) reported the results of a multicentre randomised study of the use of Gd DTP A for examinations of the brain. The following pertinent results were reported following infusion: enhancement was observed in 43/57 patients (75%); improvement in diagnostic capability in 37/ 57 patients. In 9;17 patients improvement in diagnosis was made; in 43 patients with enhancement oflesion, 16 (32%) had a change in diagnosis, while in 10 patients (23%) an increase was noted in number oflesions detected. In addition, the authors reported minimal clinical aberrations following intravenous injection, for example minor alterations of blood pressure, pulse and serum iron levels postinfusion. Wolf (I 989) examined the complications in 4000 cases studied with Gd DTPA and observed adverse reactions in 3% of patients. The 2 most serious complications were seizures and vomiting, which

719

occurred in 1/1000 patients. The author reported one death, which does not appear to be related to the use of the contrast agent. Contrast is useful in revealing small intracanalicular lesions within the internal auditory canal or in clearly delineating extraaxial lesions that are isointense, such as meningiomas. Spin echo proton density and T2W images (T2 weighted images; these reflect the emphasis of the image being on the T2 relaxation time) will reveal the presence of most intra parenchymal lesions. Gd DTPA may be useful in some cases in identifying the full extent of the mass by separating the well defined lesion from the surrounding oedema. Sze et al. (1988b) studied 75 patients with metastases, utilising CT, MRI and MRI with Gd DTPA. CT and MRI were equivalent in 49 of 75 cases (65%); in the remaining 26, neither was superior. The findings were discordant, with CT superior in 8 cases and MRI advantageous in 9 cases. Gd DTPA-enhanced MRI images provide findings similar to contrast-enhanced CT since Gd DTPA acts in an analogous fashion to intravenous contrast and also has the advantages ofMRI. Punctate (small) lesions and metastases at the base or on the periphery (meningeal surface) may be recognised as well, which otherwise could be lost sight of within the bone artifact or so-called 'Houndsfield artifacts'. Contrast-enhanced MRI is therefore the examination of choice as it has the advantages of both examinations and thus provides maximum information in the workup for metastatic lesions. MRI with Gd DTPA is advantageous when examining the spine after disc surgery. The epidural fibrosis that may occur postoperatively will enhance immediately, and after contrast administration, therefore disc can be distinguished from scar in many cases (Ross et al. 1989). In intramedullary neoplasms or recurrent tumours with cyst or syrinx, the contrast material should highlight the neoplasm and permit accurate localisation. In leptomeningeal lesions, Gd DTPA often revealed the presence and extent of lesions when the nonenhanced scan failed to reveal any abnormality (Sze et al. 1988c).

720

Drugs 40 (5) 1990

Gd DTPA will reveal findings in some, but not all, lesions in arachnoiditis; in fact, many will show no enhancement after contrast administration. In these instances, other findings that should be searched for include clumping of nerve roots and inhomogeneity of cerebrospinal fluid. In intradural-extramedullary lesions, the tumour will be highlighted and may be recognised more easily than on noncontrast MRI (Sze et al. 1988a). Finally, Sze et al. (I988c), at the seventh meeting of SMRM in 1988, came to the following conclusions concerning the usefulness ofGd DTPA in the examination of the spinal canal: '( I) Differentiating disc herniation from epidural tumor; (2) Indicating regions of more active tumor for biopsy; (3) Outlining areas of cord compression when necessary; (4) Suggesting response to therapy.'

3. Clinical Treatment Recommendations The selection of a contrast agent depends on the clinical examination to be performed. Other variables include whether nonionic agents are exclusively used or if a policy exists as to the use of nonionics based upon certain clinical indicators because of their excess cost. If in doubt, a non ionic contrast agent should be used. The following is a summary of the appropriate types of agent for particular examinations. 1. CT: either contrast material may be used. 2. Dynamic CT: use nonionic contrast material for patient comfort, since less heat is generated. 3. Intravenous pyelography and cystography: either contrast material may be used. 4. Pulmonary angiography: because of heat and potential effects on the heart, use non ionic contrast material. 5. Cardiac, coronary and aortic arch angiography: use nonionic contrast material because of potential cardiac arrythmias. 6. Abdominal angiography: use ionic contrast material because of reduced formation of blood clots. 7. Brachial angiography and femoral angiography to study pelvis and peripheral vessels: use

nonionic contrast material because of improved patient comfort. 8. Cerebral angiography: use non ionic contrast' material for patient comfort, but if a prolonged study is expected then use ionic contrast material since a reduction in clot formation occurs. 9. In myelography nonionic contrast material is used exclusively because of reductions in arachnoiditis and because removal is unnecessary. 10. In digital venous angiographic studies nonionics are preferred because of improved patient comfort. 11. In peripheral venous studies either contrast material may be used. 12. In paediatric patients below I year of age and in patients above 65 or with proven cardiac abnormality or hypertension nonionics should be used for intravenous examinations. In most examinations, therefore, as a consequence of increased patient comfort and reduction in serious complications, nonionics would be the contrast material of choice. If price were not a factor (10 to 15 times ionic), then nonionics would be preferred except in prolonged arterial examinations where an increased potential for clot formation has been demonstrated with non ionic contrast material. In MRI, the use of a contrast agent would be recommended in the following instances: I. The evaluation of the brain for metastases. 2. The separation of gross tumour from the surroundings, in planning surgery in intracranial neoplasms. 3. The visualisation of a suspected meningioma. 4. The delineation and diagnosis of brain abscess. 5. Meningeal or ventricular seeding of neoplasm. 6. Meningeal enhancement in meningitis. 7. Opacification of intracanalicular components of eighth nerve tumours. 8. The enhancement of blood vessel visualisation. 9. To separate scar from disc following disc surgery.

Clinical Use of Radiopharmaceuticals

10. To visualise spinal neoplasms. II. To identify spinal seeding of tumours or infection. 12. To examine cerebral blood flow, blood volume diffusion and perfusion. MRI with a contrast agent may be used in arachnoiditis, but is only about 50% successful. Contrast agents used in MRI have not received FDA approval for use in the body.

References Almen T. Contrast agent design: some aspects on the synthesis of water soluble contrast agents of low osmolality. Journal of Theoretical Biology 24: 216-226. 1969 Ansell G. Adverse reactions to contrast agents: scope of problems. Investigative Radiology 5: 374-384. 1970 Brasch RC. Work in progress; methods of contrast enhancements for MR imaging and potential applications. a subject review. Radiology 147: 781-788.1983 Brasch RC. Decisions about radiographic and MRI contrast media. Proceedings of the 74th annual meeting of the Radiological Society of North America. Chicago. course no. 607. November 1988 Dawson P. Chemotoxicity of contrast media and clinical adverse effects: a review. Investigative Radiology 20: 584-591. 1985 Dawson P. Edgerton D. Contrast media and enzyme inhibition: I. cholinesterase. British Journal of Radiology 56: 653-656. 1983 Elkin CM. Le Van AT. Leeds NE. Tolerance of iohexol. iopamidol and mctrizamide in lumbar myelography. Surgical Neurology 26: 542-546. 1986 Fisher HW. Catalog of intravascular contrast media. Radiology 159: 561-563. 1986 Howell MJ. Dawson P. Contrast agents and enzyme inhibition: II. mechanisms. British Journal of Radiology 58: 845-848. 1985 Ing JJ. Smith DC. Bull BS. Differing mechanisms of clotting inhibition by ionic and nonionic contrast agents. Radiology 172: 345-347. 1989 Jacobson TJ. Rosenquist J. The introduction oflow-osmolar contrast agents in radiology; medical economic legal and public policy issues. Journal of the American Medical Association 260: 1586-1592. 1988 Katayama H. Kozuka T. Takashima T. Matsuura K. Yamaguchi K. Adverse reactions to contrast media: high-osmolality versus low-osmolality media. A scientific exhibit presented at the annual meeting of the Radiological Society of North America. November 1988 Kilgore DP. Breger RK. Daniels DL. et al. Cranial tissue. normal MR appearance after intravenous injection ofGD-DTPA. Radiology 160: 757-761. 1986 Kinnison ML. Powe NR. Steinberg EP. Results of randomized control trials of low versus high osmolality contrast agents. Radiology 170: 381-389. 1989 Lasser EC. Etiology of anaphylactoid responses: the promise of non-ionics. Investigative Radiology I: 579-583. 1985 Lasser EC. Berry Cc. Commentary: non ionic vs ionic contrast media. What does the data tell us? American Journal of Roent-

721

genology 152: 945-946. 1989 Mattrey RE. Perfluorooctylbromide: a new contrast agent for CT. sonography. and MR imaging. American Journal of Roentgenology 152: 247. 1989 McClennan BL. Low-osmolality contrast media. premises and promises. Radiology 162: 1-8. 1987 Palmer FJ. The RACR survey of intravenous contrast media reactions: a preliminary report. Australasian Radiology 32: 8-11. 1988 Powe NR. Steinberg EP. Erickson JE. Moore BD. Smith CR. et al. Contrast medium-induced reactions: economic outcome. Radiology 169: 163-168. 1988 Ross JS. Delamarter H. Alfidi MG. et al. Gadolinium-DTPA enhanced MR imaging of the post-operative lumbar spine: time course and mechanism of enhancement. American Journal of Neuroradiology 10: 37-46. 1989 Runge YM . Clanton JA. Lukehart KM. et al. Paramagnetic agents for contrast; enhanced NMR imaging. a review. American Journal of Roentgenology 141: 1209-1215. 1983 Russell EJ. Geremia GK. Johnson CEo Huckman MS. Ramsey RG. et al. Multiple cerebral metastases: detectability with GdDTPA-enhanced MR imaging. Radiology 165: 609-617. 1987 Russell EJ . Schaible TF. Dillon W. et al. Multicenter double-blind placebo controlled study of gad ope netate meglumine as an MR contrast agent: evaluation in patients with cerebral lesions. American Journal of Neuroradiology 10: 53-63. 1989 Steinberg EP. Anderson GF. Powe NR. Sakin JW. Kinnison ML. et al. Use oflow-osmolality contrast media in a price-sensitive environment. American Journal of Roentgenology 151: 271274. 1988 Sze G. Abramson A. Krol G . Gadolinium-DTPA in the evaluation of intradural extramedullary spinal disease. American Journal of Neuroradiology 9: 153-163. 1988a Sze G. Shin J. Krol G. Intraparenchymal brain metastases: MR imaging versus contrast-enhanced CT. Radiology 168: 187-194. 1988b Sze G. Shin J. Krol G. Temporal and quantitative enhancement of gadolinium-DTPA in spinal tumors: proceedings of the 7th annual meeting of the Society of Magnetic Resonance in Medicine. San Francisco. p. 9. 1988c Weinmann HJ. Brasch RC, Press WR. Wesbey GE. Characteristics of gadolinium-DTPA complex: a potential NMR contrast agent. American Journal of Roentgenology 142: 619-624. 1984 White RI. Halden Jr WJ. Liquid gold: low-osmolality contrast media. Radiology 159: 559-560. 1986 Witwer G. Cacayorin ED. Bernstein AD. Hubballah MY. Yuan HA. et al. Iopamidol a.nd metrizamide for myelography: prospective double blind clinical trial. American Journal of Roentgenology 143: 869-873. 1984 Wolf GL. Current status of MR imagirig agents: special report. Radiology 170: 311. 1989 WolfGL. Arenson RL. Cross AP. A prospective trial of ionic vs non ionic contrast agents in routine clinical practice: comparison of adverse effects. American Journal of Roentgenology 152: 939-944. 1989

Correspondence and reprints: Dr Norman E. Leeds. Director. Department of Radiology. Beth Israel Medical Center. First Avenue at 16th Street. New York, NY 10003, USA.