Contrast Agents in Pediatric Neuroimaging

Contrast Agents in Pediatric Neuroimaging

Sylvester Chuang1

From the Hospital for Sick Children, Toronto, Ontario

During the last two decades the development which substitutions are made at the acid group. of contrast agents for myelography, angiography, Examples of common contrast agents used in computed tomography (CT) and magnetic reso­ North America are listed in Table 1. nance imaging (MR) has been remarkable. What Recent developments have all centered on the follows is a review of contrast agents used in nonionic contrast agents, because these are less neuroradiology-specifically, the ionic and non­ toxic than the ionics (1); however, the cost of the ionic contrast agents for radiographic/CT proce­ nonionics is significantly higher than that of the dures, and the paramagnetic contrast agents used ionics. Despite the debate regarding the cost in MR studies. Topics covered are the indications effectiveness of the nonionics (2-9), we routinely for and use of contrast agents, doses and com­ use nonionic contrast agents for all radiologic parisons, complications and cost benefits, and investigations at the Hospital for Sick Children in what the future of contrast media might hold. Toronto. We do so because of the lower compli­ cation rate, although there is no convincing evi­ Iodinated Contrast Agents dence of any decrease in the mortality rate. Nevertheless, when there is extravasation of con­ The basic structure of iodinated contrast trast into the tissue on IV injection, the morbidity agents is the benzene ring (Fig. 1), with its at­ is lower. tached iodine atoms. The remaining components Elimination of contrast agents is dependent on are the acid group and organic substitutes which organic substitutions. If one of the organic sub­ influence excretion and toxicity. The hydrogen stituents is a hydrogen atom, then the compound atoms in the acid group can be replaced by a is excreted by the biliary system. If none of the cation, such as sodium (Na) or meglumine, in organic substltuents is a hydrogen atom on the which case the contrast agent is ionic. If the benzene ring, then the contrast is excreted by the hydrogen atom in the acid group is replaced kidneys. instead by amine-carrying hydroxyl groups, then There is a major difference in osmolality and the compound is nonionic (Fig. 2). viscosity between the ionic and nonionic contrast When there is only one benzene ring in the agents and between the monomeric and dimeric compound, the contrast agent is a monomer; contrast agents. Examples of these are listed in when there are two benzene rings in the com­ Table 2. Osmolality is defined as milliosmoles per pound, the contrast agent is a dimer (Fig. 3). The kilogram of water; osmolarity is defined as mil­ compound with 'two benzene rings contains twice liosmoles per litre of solution. In water, the os­ the amount of iodine in the molecule compared malility is equivalent to the osmolarity. Because to the monomer. Monomeric and dim eric contrast of the dissociation of the cation and the anion in agents can be ionic or nonionic, depending on solution, the ionic contrast agent obviously has a much higher osmolarity than does the nonionic 1 Address reprint requests to Dr Chuang, Division of Neuroradiology, contrast agent. Hence, the other names of the Department of Diagnostic Imaging, The Hospital for Sick Children, 555 two contrast agents are high osmolar contrast University Avenue, Toronto, ON M5G IX8 Canada. medium (HOCM) (ie, ionic) and low osmolar con­ Index terms: Contrast media; Pediatric neuroradiology trast medium (LOCM) (ie, nonionic), respectively. AJNR 13:785-791, Mar/ Apr 1992 0195-6108/ 92/ 1302-0785 In terms of viscosity, the larger the molecule, © American Society of Neuroradiology the more viscous is the contrast agent. Therefore, 785 786 AJNR: 13, March/April1992

ionic monomers (Table 3). The viscosity, how­ Structural characteristics of X-ray contrast media ever, changes with temperature (10). The higher Baste structure St ructural elements Signification the temperature, the less viscous the contrast

Benzene Basic structure becomes. For this reason, nonionic contrast 0 agents are warmed to body temperature for intra­ COOH I Iodine atomes Radiopague component venous or intraarterial injections, so as to make the injection easier and more rapid. COOH Acid group Sallformation or 6 21 acid-amide bound Because the nonionic contrast agents have a 3 Water-solubility R'2¢ ~ R,' lower complication rate than the ionic contrast R,; R2 Organic Reduction of toxicity agents in intravenous and intraarterial usage, they I substituents Influence on liophilia are the only contrast agents that can be used R, Organic Influence on eliminatron intrathecally for myelography and ventriculogra­ substituent phy (11-13). With the advent of MR imaging, there are now fewer indications for myelography Fig. 1 . Basic structure of iodinated contrast agent; benzene and ventriculography; however, in cases of iso­ ring; ionic monomer. lated tethered cord (or tight filum terminate syn­ drome) and in patients who have difficulty in Ionic and non-ionic monomeric contrast media cooperating, myelography is still a more useful --·------diagnostic tool (14). Ventriculography is still use­ Ionic contrast media Non-ionic contrast media ful in analyzing the communication and compart­ mentalization of the ventricular system and cis­ Kat• Na · -ca tion, Meg • -cation - N(R)2 Amine carrying hydroxyl groups terns (15). - R Organic radical - R Organic radical --- The dosage used for ventriculography is ap­ proximately 1-2 mL of isotonic nonionic contrast coo- Kat• CO· N(R)2

TABLE 1: Low osmolar contrast medium commonly used in North , ,I - I American R- OC- HN'D ~ 'NH - CO-R i R - OC-~'Dj~/ ' 'NH - CO- R Contrast Medium Manufacturer - I . I Rl_ ___ _l Hexabrix-loxaglate Mallinkrodt lohexoi·Omnipaque Sterling-Winthrop lopamidol-lsovue Squibb Fig. 2. Nonionic monomer contrast chemical structure; hydro­ loversoi-Optiray Mallinkrodt gen atom in acid group replaced by hydroxyl group.

TABLE 2: Osmolality and viscosity of non ionic contrast agents

Contrast Agent Viscosity at Concentration Osmolality

37°C (cps) (mg lodine/mL) (mOsm/Kg H20) Optiray 320 5.8 320 702 (loversol) lohexol 6.3 300 672 lopamidol 4.7 300 616

lohexol 2.0 180 408 Optiray 160 1.9 160 355 (loversol) lopamidol 1.4 128 290

TABLE 3: Viscosity of contrast agents of 300 mgiJ mg at 37°C

Contrast Viscosity Type Fig. 3. Dimer contrast agent with two benzene rings; ionic or Sodium trizoate 2.4 Ionic monomer nonionic dependent on the acid group. Meglumine trizoate 5.0 Ionic monomer Meglumine ioglicate 6.0 Ionic dimer the most viscous contrast agents are the nonionic lohexol 5.7 Nonionic monomer dimers, as opposed to the ionic dimers and the lopamidol 4.6 Nonionic monomer lotrolan 9.0 Nonionic dimer nonionic monomers. The least viscous are the AJNR: 13, March/April 1992 787

TABLE 4: Complete myelography dosage table the nonionic group. In intravascular usage, ad­ verse reactions appear to depend on three factors, Concentration mL/ kg mg 1/mgl namely chemotoxicity, osmotoxicity, and low toxicity (6, 21, 22). The chemotoxicity effect Less than 2 mo 180-210 1/2 2 mo-2 yr 180-210 2-4 causes red cell crenation and rigidity with direct 3 yr-7 yr 180-210 4-8 effect on cell membranes and organelles causing 8 yr-12 yr 180-210 7-1 2 endothelial damage in vessels. There is also re­ 13 yr-17 yr 180-210 8-14 lease of vasoactive substances from cells (6). (The vasoactive substances include histamine, ser­ TABLE 5: Cerebral angiography-contrast dosage tonin, fibrolysin, kallikreins, prostaglandins, leu­ kotrienes, complements, and bradykinin.) The Under 10 10-20 20-40 Over 50 osmotoxicity or osmotic effect causes pain during kg kg kg kg arteriography, vasodilatation with hypotension, Internal carotid artery 2-4 cc 4-6 cc 6-7 cc 7-8 cc and rigidification of red blood cells. The osmotox­ External carotid artery 1-2 cc 2-3 cc 3-4 cc 4-5 cc icity of the contrast agent is governed by a ratio Common carotid artery 4-5 cc 5-7 cc 6-8 cc 9-10 cc Vertebral artery 2-3 cc 3-5 cc 5-7 cc 6-8 cc that comprises the number of iodine atoms over the number of particles in an ideal solution, and is equal to the imaging effect of contrast medium agent. The dosage used for myelography is listed over the osmotic effect of contrast medium. The in Table 4. The dosage varies with the patient's higher the ratio, the lower the osmotoxicity. age rather than weight because the size of the Adverse reactions can be classified as mild or subarachnoid space tends to vary with age rather severe: mild reactions include urticaria, general­ than with weight (16). ized pruritus, sneezing, rhinitis, nasal stuffiness, We have done a comparative study of 180 mg coughing, and lacrimation; severe reactions in­ % b versus 210 mg % b, in myelography and clude wheezing, dyspnea, laryngospasm, status found that if same amount of Iodine is given, ie, asthmaticus, subglottic oedema, angioneurotic larger volume of 180 mg % b as compared to oedema, anaphylactic shock, and cardiovascular 210 mg % b, there is little appreciable difference collapse. Laryngotracheal edema and urticaria are in the quality of the myelogram (17); however, if part of the immune response, although no spe­ the same volume is given, then the more concen­ cific antibodies lgG or lgE have been isolated (23). trated 210 mg % b does provide a better myelo­ The cardiovascular effect is probably related to graphic study. This is not the case in CT myelog­ chemotoxicity as well as to osmotoxicity. In ad­ raphy. dition to endothelial damage, hyperosmolality Contrast dosage for cerebral angiography is causes decrease in smooth vascular muscle tone listed in Table 5. For good visualization of the as well as cardiac contractility (6). arterial, capillary, and venous phases, the total Hypotension may be secondary to a vasovagal volume of contrast medium should be given in response from anxiety, but may also result from less than 1.5 seconds. The concentration of the the vasodilatory effect of contrast agent. The contrast medium, however, can be substantially HOCM cause an increase in the circulatory vol­ less in digital angiography than in the conven­ ume, with an increase in peripheral blood flow tional film technique; thus, isotonic concentra­ and a decrease in systemic resistance resulting in tions of contrast can be used in digital studies. reduced blood pressure and hypotension (6). The Some authors have shown by animal experi­ hemodilution effects are due to the shift of extra­ ments that, for intravascular usage, the LOCM is cellular water into the blood stream, which further less neurotoxic than HOCM (18-20). However, contributes to hemodynamic perturbation asso­ the comparison is made with the same dosage ciated with the use of HOCM (6). On the other and iodine concentration; adjustments or experi­ hand, contrast-induced nephropathy does not ments were not made for comparing isotonic correlate well with the volume of contrast me­ LOCM and HOCM to see if osmolality is really dium injected or with HOCM versus LOCM (24). the key element in the factor. Decrease in renal blood flow and glomerular fil­ As mentioned previously, adverse drug reac­ tration rate can lead to proteinuria or anuria (25). tions are more common and may be more severe Premedication with steroids has not been with the ionic contrast agents. The incidence of shown to prevent damage; rather, it suppresses ionic contrast reaction is 4.17% versus 0.69% in the symptoms of the adverse reactions (6). This 788 AJNR: 13, March/ April 1992 is why we have chosen to use nonionic contrast 1) monitor vital signs frequently; 2) ensure patient agents for all intravascular studies instead of only airway; 3) observe for aspiration; 4) oxygen when in high-risk patients, despite the apparent cost necessary; 5) establish intravenous route; and 6) benefit. administer specific drug therapy (Table 6). Ionic contrast media cannot be used in my­ It is as important not to overreact. Most of the elography because of high neurotoxicity. Even medications listed in Table 6 do not cause harm­ the first nonionic contrast agent-metrizamide­ ful effects in a single dose; the dose of adrenalin, is more neurotoxic than the newer contrast however, is more critical and it should be admin­ agents, such as iohexol, iopamidol, or ioversol istered carefully. (26-29). Despite different reports by different authors, no significant differences have been shown to exist among the newer nonionic con­ Paramagnetic Contrast Agents trast agents (11, 20, 29-34). The three basic types of MR contrast agents The incidence of adverse drug reactions follow­ are paramagnetic, supraparamagnetic, and ferro­ ing intrathecal usage of nonionic contrast agents magnetic. The supraparamagnetic and ferro­ varies from 26% to 44% (35, 36). The com­ magnetic agents reduce T2 relaxation much more monest reactions are headache and vomiting, than T 1 relaxation and come in the form of although it is often uncertain whether the reaction particles. Therefore, they are unfit for neurora­ is caused by the lumbar puncture or the contrast diology usage (intravascular injections) (Table 7) agents. No seizures have been recorded in our (39). In addition, reducing T2 relaxation is a experience with the newer nonionic contrast disadvantage, because it causes a decrease in agents but have been reported by others (37, 38). signal intensity on spin-echo images (39). With metrizamide, the incidence of seizures is The only contrast agents presently usable for approximately 1 per 300. neuroradiology imaging are the paramagnetics. Four rules provide guidelines for the tolerance The paramagnetic ions used have magnetic di­ of intravascular contrast agents (6): Rule I, Use pole moments 1,000 times thatof protons (40). of CM with a ratio of 3 or higher almost com­ pletely eliminates pain during arteriography; Rule TABLE 6: Medications II, The lower the number of particles in the solution (higher ratio), the lower the number of Chlortripolon: oral, IM, IV 0.2-0.3 mg/ kg carboxyl groups; the higher the number of hy­ Maximum dose-20 mg droxyl groups in relation to the number of iodine atoms in the CM molecule, the higher its intra­ Diphenhydramine: IV venous tolerance will be; Rule III, Low neurotox­ (Benadryl) I mg/ kg icity in the subarachnoid space requires absence 50 mg = 1 cc maximum dose-75 mg q 6 hr of carboxyl groups and the presence of many hydroxyl groups evenly distributed around the Decadron: IV CM molecule; and Rule IV, A high venous CM 0.2 mg/ kg concentration in arteriography is obtained when Maximum dose-10 mg diffusion of CM is slowed down by using large Solu-cortef: IV molecules (dimers), and osmotic dilution is de­ 2 mglkg given over at least creased by using CM with low osmotic effects 10 min (ratio 3 or ratio 6). Applying the above principles to the properties Adrenalin: 1:1000 S.C. of the nonionic contrast agents, it can easily be vial 0.01 mg/ kg 0.01 ml/kg understood why the nonionic contrast agents have a lower toxicity and a higher patient toler­ Diazapam: IV ance. up to 0.2 mg/ kg Despite all precautions adverse reaction can OR still occur. Initial and immediate management of Adrenalin: 1:10,000 IV these patients may then be crucial to patient 0.01 mg/ kg survival. The treatment of the adverse reaction 0.1 mg/ kg often rests in the hands of the radiologist and management should consist of the following (22): Maximum q. 15 mins X 2 doses AJNR: 13, March/ April 1992 789

TABLE 7: MR contrast agents

Paramagnetic Supra paramagnetic Ferromagnetic Chelate Particles Particles T1 ! T2 ! !!! !!!!

Ph ysica l form Chela tes dissolved Fe aggregates as Fe aggrega tes as in solution small particles large pa rticles

Number of paramagnetic 10'0 10'2 atoms per molecule

Examples Fe EDTA Dextran plus Microspheres plus Mn EDTA Magnetite Magneti te Gd DTPA

This affects the relaxation rate of the protons in NJ) Gd-D03A is currently undergoing trial in the vicinity. North America. The formula for the relaxation rate is as follows Gd-DTPA (Magnevist) has an extremely low (40): incidence of adverse drug reactions, ie, lower than nonionic iodinated contrast agents (Table 8) (45). 1/Tl obs = 1/T1 o + 1/T], p It is not certain whether a nonionic Gd compound where T1 obs =observed relaxation time; T1 o = would further reduce the rate of contrast reaction. intrinsic value; and T 1 p = contribution contrast Contrast reactions reported with Gd-DTPA in­ agents makes to the relaxation rate. clude headache, nausea, and vomiting, which are The increased relaxation rate is proportional to the most common symptoms (45). Other reac­ the concentration of the paramagnetic agent and tions include hypertension, local burning sensa­ to square of the magnetic moment and inversely tion, and hiccups (45). Sensations of warmth or 6 proportional to r , where r is the distance between cold at the injection site are not included, since the paramagnetic center and the protons to which these may just represent temperature of the in­ it is bound. jected solution. The incidence of drug reactions Paramagnetic agents affect both T 1 and T2 varies from 0. 7% without known history of al­ relaxation rates, although they affect T 1 at a lergy to 2.6% with known history of allergy (45). much lower concentration. Multiple paramag­ The overall incidence is approximately 1.46% 3 (45). There is no significant difference between netic agents have been tested and Gd + (gadolin­ ium) had the largest magnetic moment (Table 7) pediatric and adult population, and the reported (41 ). Gadolinium is attached to a chelating agent pediatric incidence is 1.21 %. to decrease toxicity, as well as to increase the On the biochemical level, there is a reported effectiveness of the paramagnetic centre (42-44). incidence of 10%-20% transient increase in Therefore, the selection of useful MR contrast serum concentration of ions and less often with agents is dependent on the following factors (40): bilirubin as well (45). In neuroradiology practice, 1) paramagnetic ion of a high spin number and a the use of both iodinated contrast agents for CT and paramagnetic contrast agents for MR relies large electron-spin relaxation time; 2) small ra­ mainly on breakdown of the blood-brain barrier. dius; 3) large complexes to retard molecular mo­ tion; 4) chelating agents for increasing stability and reducing toxicity; and 5) paramagnetic con­ TABLE 8: Adverse drug reactions comparing Gd-DTPA and LOCM centration > 100 ,urn. Other considerations in­ Symptoms LOCM Gd-DTPA clude biodistribution, metabolism, and cost (40). No. % No. %

The chelating agent currently being used by Nausea/vomiting 2363 1.40 57 0.42 Berlex is diethylenetriamine pentacetic acid Local wa rmth/ pai n 1635 0.97 55 0.4 1 (DTPA). The compound is ionic, although non­ Allergy-like skin reactions 1548 0.92 14 0.104 ionic Gd contrast agents are being developed by Allergy-like mucosa l reactions 698 0.4 1 7 0.052 Flush 271 0. 16 8 0.059 other drug companies, eg, Squibb's (Princeton, 790 AJNR: 13, March/ April 1992

Because of the sensitivity of MR, the dosage of 4) brain death; and 5) aneurysm and arteriove­ contrast medium given is much less than in CT. nous malformations. Our regular pediatric dose for CT has been 3 mL/kg of 300 mg b/mL up to a maximum of Future Developments 100 mL. The contrast is given in a bolus, since it is extremely difficult to infuse rapidly. The net In the iodinated contrast agents, new nonionic effect by the bolus is as good as a rapid infusion. dimers are currently being developed. Berlex (Ce­ For MR scans, the recommended adult dose is dar Knolls, NJ) has introduced iotrolan for clinical 0.1 mmol/kg. We feel this is too low for children trial in North America (48), while Sterling-Winth­ and that 0.25 mmol/kg is more appropriate. rop has introduced iodixanol. The advantages of Experiments have shown that use of an increased the nonionic dimers include lower osmolality (iso­ dose does give better enhancement of lesions, tonic intravascular injection) with high-iodine with little increase in normal tissue and areas of dose delivered. The contrast is of low toxicity, edema (46). highly hydrophilic with low protein-binding ca­ The indications for contrast agents in CT and pacity and, therefore, has a low incidence of MR are similar, although CT usually has one or contrast reactions (49, 50). The disadvantage will two additional uses. The commonest use in CT be that of high viscosity. and MR are infection, tumor and tumor-like con­ In MR, emphasis has been on low osmolar ditions, vascular diseases, and neurodegenerative agents with increased tolerance such as Gd-DTPA disorders. CT also uses contrast media routinely bismorpholide, high relaxivity such as Gd-DTPA for seizure disorders and, on rare occasions, to polylysine, and tissue-specific uptake from Gd­ localize the tentorium and the falx for anatomical DTPA derivatives that are lipophilic (51). For purposes. example, porphyrins are endogenous chelating Gd-DTPA is expensive. Recently, we looked agents and metalloporphyrins tend to accumulate into the cost-benefit ratio of Gd-DTPA for tumor preferentially in tumor tissues (46). Nitroxide spin diagnosis. In no case has T2 missed an abnor­ labels are synthetic paramagnetic organic mole­ mality, although T 1 MR with Gd-DTP A often cules with one or more unpaired electrons and shows the lesion better than T2. However, there can be designed to lodge in cell membranes or to are some cases in which T2 is better than Tl penetrate intracellular compartments (46). with Gd-DTP A. In a few cases, Gd-DTPA brings about no enhancement (47). This often happens Conclusion . in low-grade glioma where there is little blood­ brain barrier breakdown. Other than pituitary We are in constant search for ideal contrast tumors, we have found little value in precontrast agents that manifest the following properties: 1) Tl studies. Therefore, in order to make Gd-DTPA low toxicity; 2) high degree of efficacy; 3) wide more cost effective, we have proposed skipping margin of safety; 4) tissue or compartment spe­ the Tl precontrast study in selected cases. cific distribution; 5) rapid clearance; 6) high de­ gree of stability (in vivo and in vitro); 7) good solubility in water; and 8) uncomplicated synthe­ Xenon CT sis. Inhalation of inert xenon gas mixed with oxy­ Let us hope that our search continues at a gen has been used for studying perfusion of the goodly pace. brain. The theory of inhalational xenon CT is based on Fick 's Principle and that xenon is a References freely diffusible contrast agent. As xenon is car­ 1. Bettman MA, Morris TW. Recent advances in contrast agents. Radio/ ried by the blood flow to the brain, the changes Clin North Am 1986;24:347-357 seen on CT in Hounsfield units is proportional to: 2. Bagg MNJ, Horwitz TA, Bester L. Comparison of patient responses 1) percent concentration of inhal~d xenon gas; 2) to high- and low-osmolality contrast agents injected intravenously. AJR 1986;147: 185-187 xenon solubility in arterial blood; 3) xenon blood 3. Bettmann MA, Geller SC, McCJennan B, et al. Current use of low brain proportional coefficient; and 4) local cere­ osmolality contrast agents: results of a survey. AJR, 1989; 153: bral blood flow. 1079-1083 It is the cerebral blood flow we are interested 4. Bettmann MA. Guidelines for use of low-osmolality contrast agents. Radiology 1989; 172:901- 903 in measuring; indications for xenon CT are: 1) 5. Bettmann MA. Ionic versus nonionic contrast agents for intravenous occlusive vascular disease; 2) seizures; 3) trauma; use: are all the answers in? Radiology 1990;175:616-618 AJNR: 13, March/ April1992 791

6. McCiennan BL. Ionic and non ionic iodinated contrast media: evolution 30. Hoe JWM, Ng AMN, Tan LEA. A comparison of iohexol and iopamidol and strategies for use. AJR 1990; 155:225-233 for lumbar myelography. C/in Radio/ 1986;37:505-507 7. Steinberg EP , Anderson GF, Powe NR, et al. Use of low-osmolality of 31. Macpherson P, Teasda le E, Coutinho C, et al. lohexol versus iopamidol contrast media in a price-sensitive environment. AJR 1988; !51 : for cervica l myelography: a randomised double blind study. Br J 271-274 Radio/ 1985;58:849-85 I 8. Wolf GL, Arenson RL, Cross AP. A prospective trial of ionic vs 32. Ringel K, Kuehn J . A double-blind study comparing the safety, nonionic contrast agents in routine clinical practice: compari son of tolerability, and efficacy of ioversol 320 and iopamidol 300 in cerebral adverse effects. AJR 1989; 152:939-944 angiography. Invest Radio/1988;24:S48-S51 9. Wolf GL. Safer, more expensive iodinated contrast agents; how do 33. Smith DC, Yahiku PY , Maloney MD, et al. Three new low-osmolality we decide? Radiology 1986;59:557- 558 contrast agents: a comparative study of patien t discomfort. AJNR I 0. Halsell RD . Heating contrast media: role in contemporary angiogra­ 1987;9: 137-139 phy. Radiology 1987;164:276-278 34. Va lk J , Crezee FC, de Slegte RGM , et al. lohexol 300 mg 1/ml versus 11. Davies AM, Evans N, Chandy J. Outpatient lumbar radiculography; comparison of iopamidol and iohexol and a literature review. Br J iopamidol 300 mg 1/ml for cervical myelography double blind trial. Radio/ 1989;62: 716-723 Neuroradiology 1987;29:202-205 12. Hammer B, Lackner W. lopamidol, a new nonionic hydrosoluble con­ 35. Provinciali L, Signori no M, Giovagnoli AR , et al. Lumbar myelography trast medium for neuroradiology. Neuroradiology 1980;19:119-121 with iopamidol: a methodological approach to the investigation of 13. Si mon JH , Ekholm SE, Kido DK, et al High-dose iohexol myelogra­ side effects. Neuroradiology 1988;30:528-533 phy. Radiology 1987;163:455-458 36. Skalpe 10, Nakstad . Myelography with iohexol (Omnipaque) ; a clinical 14. Chuang S, Hochhauser L, Harwood-Nash D, Armstrong D, Burrows report with special reference to the adverse effects. Neuroradiology P, Savoie J . The tethered cord syndrome. Paper presented at XIII 1988;30: 169-174 Symposium Neuroradiologicum, Stockholm, Sweden, 1986 37. Lipman JC, Wang A, Brooks ML, et al. Seizure after intratheca l 15. Chuang S, Fitz CR , Harwood-Nash, D. The use of metrizamide administration of iopamidol. AJNR 1988;9:787-788 ventriculography in pediatric hydrocephalus. Paper presented at the 38. Maly P, Bach-Gansmo T , Elmqvist D. Risk of seizures after myelog­ Society for Pediatric Radiology, Sa n Francisco, CA, 1981 raphy; comparison of iohexol and metriza mide. AJNR 1988;9: 16. Harwood-Nash DC, Fitz CR . Myelography: neuroradiology in infants 879-883 and children. Vol 3. St. Louis: Mosby, 1976:1125-1166 39. Wolf GL. Contrast media in MRI. Contrast Media in MRI. International 17. Babyn P, Chuang S, Armstrong D, Burrows PE, Harwood-Nash DC. Workshop, Berlin , February 1-3, 1990:15- 17 Comparative evaluation of iohexol myelography using 210 mgi/ m l 40. Gadian DG , Bydder GM. The mechanism of contrast enhancement: and 180 mgi/ ml in children. Paper presented at Society for Pediatric clinica l examples using GD-DTPA. In: Contrast agents in magnetic Radiology, San Antonia, CA, 1989 resonance imaging. 1986:14-18 18. Bird RC , Drayer BP , Velaj R, et al. Safety of contrast media in cerebral 41. Knop FE, Frank JA, Dwyer AJ, et al. Gadolinium crytelates as MR angiography: lopamidol vs methylglucamine iothalamate. AJNR 1984;5:801-803 contrast agents. J Comput Assist Tomogr 1987; 11 :35-42 19. Michele! AA. Effects of intravascular contrast media on blood-brain 42. Brasch RC. An overview of contrast agents for magnetic resonance barrier: comparison between iothalamate, iohexol, iopentol and iodix­ imaging. In: Contrast agents in magnetic resonance imaging. 1986: anol. Acta Radio/1 987;28:329- 333 11-13 20. Pelz DM, Fox AJ, Vi nuela F, et al. A comparison of iopamidol and 43. Brasch RC , Bennett HF. Mechanisms of contrast enhancement for iohexol in cerebral angiography. AJNR 1987;9:1163- 1166 MRI. Contrast Media in MRI. International Workshop, Berlin , Febru ary 21 . Ansell G. Adverse reactions to contrast agents: scope of problem. 1-3, 1990:9-14 Invest Radio/1990;25:374-384 44 . Wei nmann H-J. Chelated contrast agents. Contrast agents in magnetic 22. Cohan RH , Dunnick NR. Intravascular contrast med ia: adverse reac­ resonance imaging. 1986:19-23 tions. AJR 1987;149:665-670 45. Niendorf HP, Dinger JC, Haustein J , et al. Tolerance of GD-DTPA: 23. Wa lker AC, Carr DH . Reactions to radiographic contrast media: an clinica l experi ence. Contrast Media in MRI. International Workshop, attempt to detect specific anti-contrast medium antibodies in the sera Berlin, February 1-3, 1990:31-39 of reactor patients. Br J Radio/ 1986;59:531-536 46. Haustein J , Niendorf HP , Schubeus P, et al. Dosage of GD-DTPA: 24. Gomes AS, Lois JF, Baker JD, et al. Acute renal dysfunction in high­ new resu lts. Contrast Media in MRI. International Workshop, Berl in, ri sk patients after angiography: comparison of ionic and nonionic February 1-3, 1990:41-49 contrast media. Radiology 1989; 170:65-68 47. Chuang S, Hele MJ, Kucharczyk W, Ho V. Cost effectiveness of GD­ 25. Miller DL, Chang R, Wells WT, et al. Intravascular contrast media : DTPA: MRI in pediatric CNS tumors. Paper presented at In ternational effect of dose on renal function. Radiology 1988; 167:607-611 Pediatric Radiology, Stockholm, Sweden, 1991 26. Bannon KR , Braun IF, Pinto R, et al. Comparison of radiographic 48. Sovak M , Ranganathan R, Speck U. Non ionic dimer: development and quality and adverse reactions in myelography and iopamidol and initial testing of an intrathecal contrast agent. Radiology 1982; 142: metrizamide. AJNR 1983;4:312-313 115-118 27. Moschini L, Manara 0, Bonaldi G, et al. lopamidol and metrizamide in cervica l myelography, side effects, EEG, and CSF changes. AJNR 49. Coutinho C, McGeorge A, Macpherson P, et al. lotrolan versus 1983;4:848-850 iopamidol for cervica l myelography: a randomized double-blind trial. 28. Trevisan C, Malaguti C, Manfred ini M, et al. lopamidol vs metrizamide Br J Radio/1988;6 1 :964-967 myelography: clinical comparison of side effects. AJNR 1983;4: 50. Dawson P, Howell M . The non-ionic dimers: a new class of contrast 306-308 agents. Br J Radio/1986;59:987-991 29. Witwer G, Cacayorin ED, Bernstein AD, et al. lopamidol and metriz­ 51. Weinmann H-J, Press WR, Raduchel B, et al. Characteristics of GD­ amide for myelography: prospective double-blind clinical trial. AJR DTPA and new derivatives. Contrast Media in MRI. International 1984; 143:869- 873 Workshop, Berlin , February 1-3, 1990:1 9-30