Delayed adverse reaction to sodium ioxaglic acid-meglumine

ARTICLE

Most adverse reactions produced by iodinated contrast media (ICM) occur within a brief period following administration of the agent and are considered to be toxic in nature [1]. The allergic reactions that have been described are of the immediate hypersensitivity type [2-4]. In contrast, delayed reactions to these agents have rarely been reported [5-8]. We now report the case of a patient who had a maculopapular rash with fever, hepatic cytolysis, rhabdomyolysis, eosinophilia and an elevated serum total IgE level caused by the iodinated radiographic contrast medium (ICM) ioxaglic acid-meglumine.

Case report

A 63 year old housewife had been admitted with an acute posterior-inferior myocardial infarction. Her medical history was otherwise not significant. Treatment with atenolol and lysine acetylsalicylate was instituted and a coronary arteriography was performed with ioxaglic acid-meglumine (Hexabrix^®, Guerbet, France). One week later, she underwent endoluminal coronary angioplasty, which included injection of the same ICM. The next morning she experienced intense shivering and generalized malaise, spiked a fever to 39.2° C, and became hypotensive. Several hours later she was found to have a nonpruritic maculopapular rash beginning on her face and right buttock. Within the next 48 hrs, the rash had extended to cover her entire body, while sparing the mucous membranes. She was transferred to the Infectious Disease Service. The total white blood count was normal but eosinophilia was slightly elevated (600/mm³), with an elevated erythrocyte sedimentation rate (52 mm at 1 hr). Her IgE level was elevated, 1,593 kU/l. Blood cultures were negative. However, the initial diagnostic impression was that of sepsis and she was therefore started on vancomycin, netilmicin and cefotaxime. The patient improved rapidly, the fever and the rash resolving completely. She was discharged from the hospital on 100 mg atenolol and 250 mg lysine acetylsalicylate, both once a day. Three days later she reappeared at the hospital with a temperature of 39.8° and recrudescence of the skin rash. She had marked eosinophilia (1,400/mm³) and elevated liver enzymes (ALAT, 116 IU/l; ASAT, 125 IU/l; alkaline phosphatase, 374 mIU/l). Muscle enzymes were elevated, lacticodehydrogenase being 116 mU/ml and aldolase being 125 mU/ml; creatinine phosphokinase was, in contrast, within normal limits. She was found to have a significantly elevated level of circulating immune complexes, 107 mg/l (normal < 25 mg/l). IgEs were at 1,479 kU/l. Histological examination of a skin biopsy revealed a predominantly perivascular lymphocytic infiltrate without fibrinoid necrosis. Atenolol and lysine acetylsalicylate were discontinued and she was placed instead on acenocoumarol, 2 mg/day, and amlodipine, 5 mg/day. The inflammatory syndrome rapidly disappeared and the hepatic enzymes and the eosinophil count returned to normal.

Because of the strong suspicion of drug induced reaction, she was admitted to our Immuno-Allergology Service two months later for further evaluation. Total IgE was still elevated, 1,649 kU/l. The white blood count and the C-reactive protein level were within normal limits. A test for serum antinuclear antibodies was positive at 1/256, and for anti-mitochondrial antibodies at 1/128. Tests for anti-native DNA, anti-histone and anti-smooth muscle antibodies were negative.

Skin prick tests (SPT) followed by intradermal tests (IDR) with lysine acetylsalicylate, atenolol, heparins, vancomycin, netlmycin, cefotaxime and ICM [Hexabrix^® (160 mgl/ml), Telebrix^® (sodium ioxotalamate - 300 mgl/ml) and Iopamiron^® (iopamidol - 300 mgl/ml)], Radioselectan^® (sodium amidotrizoate-meglumine-146 mg/ml), Omnipaque^® (iohexol - 180 mg/ml) beginning with a dilution of 10^{- 3} and progressing up to the undiluted preparations, patch-tests were carried out [2]. In 20 controls, prick-tests, intradermal tests patch-tests were constantly negative. Immediate skin tests were negative at 15 min. At 48 hrs, indurated erythematous papules of 10 mm of diameter were observed at the previous Hexabrix^®, Telebrix^® and Iopamiron^® skin test sites (Fig. 1). Patch-tests with ICM were negative. Skin tests with other drugs were negative.

Histological examination of a biopsy of the Hexabrix^® skin reaction site revealed discrete spongiosis of the epidermis associated with slight lymphocytic exocytosis. The basal layer contained numerous apoptotic keratinocytes and frequent mitotic figures. The superficial dermis was edematous and marked by the presence of a perivascular inflammatory infiltrate composed mainly of lymphocytic cells. No mast cells or eosinophils were observed. Immunohistological examination of paraffin-embedded sections of the site was made using anti-CD3 (Novo, UK), -CD20, -CD68 and CD45-Ro (Dako, Denmark) antibodies, and on frozen sections using anti-CD1a, -CD4, -CD8 (Immunotech, France), -IgA, -IgG (Dako), -IgM (Silenus, Australia), -C1q, -C3 and -C4 (Behring, Germany) antibodies. The infiltrate was predominantly composed of CD8⁺, CD4⁺, CD45Ro⁺ cells. CD8⁺ were present in greater numbers than CD4⁺ cells. The infiltrate contained some CD68⁺ macrophages and some rare CD1a⁺ cells. A significant number of these CD1a⁺ cells (Fig. 2) were found to be close to the rare CD8⁺ cells within the epidermis. There was no staining with anti-IgA, -IgG, -IgM, -C1q, -C3 and -C4.

When seen 6 months and again one year later, the patient appeared completely normal, eosinophilia had disappeared. IgEs were at 440 kU/l. On both occasions, the liver enzyme levels remained elevated, and M2 anti-mitochondrial antibodies were at a dilution of 1/1,280. The anti-M2 specificity was confirmed on a Western blot.

Discussion

In the present case, once the initial clinical impression of septicemia following angioplasty had been ruled out, a drug-induced immuno-allergic reaction was considered to be the most likely cause. Different drugs were regarded as potential candidates for the etiological agent: the beta-adrenergic blocker atenolol, the sodium ioxaglate-meglumine, heparin, antibiotics and lysine acetylsalicylate. The protracted duration of the signs and symptoms occurring after the short remission at first suggested that atenolol and/or lysine acetylsalicylate was responsible for the recrudescence of the patient's problem. However, skin tests to these drug were negative. In view of the clearly positive intradermal reactions to ICM, we were then led to focus on those agents. The fact that the patient had been given a dose of Hexabrix ^® when she underwent the first coronary procedure eight days before suggested that she had been sensitized by this agent. The immunological specificity would appear to be directed against the iodinated component, i.e., the ioxaglate molecule of Hexabrix^®, meaning that cross reactions with either of the other two ICM, which do not contain meglumine, would be possible.

The delay in the onset of the intradermal skin test reactions to the ICM does not favor an IgE-dependent type of immediate hypersensitivity. A delayed hypersensitivity type was suspected. A late phase reaction of an IgE-mediated reaction could be discussed [5]. The histology showed a mononuclear cell infiltration, such as that described in drug delayed type hypersensitivity [9-11]. The immunohistochemistry of this infiltrate revealed T lymphocyte infiltration, the majority of which being activated T cells, with predominantly T suppressor cells. The immunohistology excluded the role of immunoglobulins and complement, which ruled out an immune complex vasculitis.

Reports of delayed reactions to ICM are rare. However, they have been reported as occurring in 2 to 5% of cases in which these agents have been used [7, 12]. On the other hand, cases of delayed skin reactions have been documented [6, 8, 13]. It is known that beta-adrenergic blockers can potentiate the severity of immediate hypersensitivity reactions to ICM [14-18]. Reynolds et al. [19] reported the case of a woman who was being treated with atenolol and hydralazine and who presented with a cutaneous vasculitis 24 hrs after having undergone intravenous urography with iopamidol. Another case of cutaneous vasculitis with rapid evolution to the Stevens-Johnson syndrome occurred 48 hrs after an injection of iohexol in a patient who was also receiving these same two drugs [20]. The respective roles of the beta-blocker and of hydralazine remain uncertain. In the above-noted two cases, the presence of antinuclear antibodies suggested a pre-existing lupus-like syndrome, which could have amplified the severity of the cutaneous reaction to the ICM. Another possibility is that the beta-blocker might modify the hepatic metabolism of the ICM. Our patient had a recurrence of the skin reaction as well as hepatic and muscular disturbances when she had been receiving atenolol for only four days. We wonder if the beta-blockers do not also play a role in delayed hypersensitivity reactions to the ICM.

The evidence of hepatic involvement during the course of our patient's adverse reaction suggested the possibility of hepatic toxicity due to the ICM [21-23]. The mechanism of such an effect remains uncertain. Scholz et al. [23] suggested that it was a dose-dependent toxicity. Other authors [24] have suggested that the mechanism of this hepatotoxicity is one of hypersensitivity. However, the persistence of hepatic cytolysis for several months after the onset of the adverse reaction, plus the existence of significant levels of anti-mitochondrial antibodies (1/128 dilution), suggested that primary biliary cirrhosis already existed in these patients [25, 26]. Drug allergies occur very often during the course of autoimmune diseases, especially systemic lupus erythematosus, less often in primary biliary cirrhosis [27]. Another hypothesis is that a drug itself can induce an autoimmune disease [28]. Drug-induced autoimmune hepatitis has been reported, for example anti-smooth muscle induced by minocycline [29], anti-LKM2 antibodies induced by tienilic acid, with anti-LKM1 antibodies induced by halothane, and with anti-MAO antibodies induced by isoniazid [30]. Indeed, hepatotoxicity induced by the ICM meglumine iodipamide and accompanied by the presence of antinuclear and antimitochondrial antibodies has been reported [24]. Whether asymptomatic autoimmune hepatitis preceeded the reaction to ioxaglic acid, or was induced by ioxaglic acid hypersensitivity cannot be stated.

CONCLUSION

REFERENCES

1. Katayama H, Yamaguchi K, Kozuka T, Takashima T, Seez P, Matsuura K. Adverse reactions to ionic and nonionic contrast media. A report from Japanese committee on the safety of contrast media. Radiology 1990; 175: 621-8.

2. Laroche D, Aimone-Gastin I, Dubois F, et al. Mechanisms of severe immediate reactions to iodinated contrast material. Radiology 1998; 209: 183-90.

3. Kanny G, Maria Y, Mentre B, Moneret-Vautrin DA. Case report: recurrent anaphylactic shock to radiographic contrast media. Evidence supporting an exceptional IgE-mediated reaction. Allergie & Immunologie 1993; 25: 425-30.

4. Bush WH, Swanson DP. Acute reactions to intravascular contrast media types, risk factors, recognition, and specific treatment. AJR 1991; 157: 1153-61.

5. Delayed allergy-like reactions to X-ray contrast media. European Radiology 1996; 6 (suppl.): 1-24.

6. Brockow K, Kiehn M, Kleinheinz A, Vieluf D, Ring J. Positive skin tests in late reactions to radiographic contrast media. Allerg Immunol 1999; 31: 49-51.

7. Panto PN, Davies P. Delayed reactions to urographic contrast media. Br J Radiol 1986; 59: 41-4.

8. Schick E, Weber L, Gall H. Delayed hypersensitivity reaction to the non-ionic contrast medium iopromid. Contact Dermatitis 1996; 35: 312.

9. Kauppinen K, Alanko K, Hannuksela M, Maibach H. Skin reactions to drugs. In: Maibach HI, ed. Dermatology: clinical and basic science. Boca Raton: CRC, 1998: 177.

10. Hertl M, Merk HF. Lymphocyte activation in cutaneous drug reactions. J Invest Dermatol 1995; 105: 95s-8.

11. Warrington RJ, Silviu-Dan F, Magro C. Accelerated cell-mediated immune reactions in penicillin allergy. J Allergy Clin Immunol 1993; 92: 626-8.

12. Rydberg J, Charles J, Aspelin P. Frequency of late allergy-like adverse reactions following injection of intravascular non-ionic contrast media. A retrospective study comparing a non-ionic monomeric contrast medium with a non-ionic dimeric contrast medium. Acta Radiol 1998; 39: 219-22.

13. Good AE, Novak E. Fixed eruption and fever after urography. South Med J 1980; 73: 948-9.

14. Greenberger PA. Effects of beta-adrenergic and calcium antagonists on the devlopment of anaphylactoid reactions from radiographic contrast media during cardiac angiography. J Allergy Clin Immunol 1987; 80: 698-702.

15. Hamilton G. Severe adverse reactions to urography in patients taking beta-adrenergic blocking agents. Can Med Assoc J 1985; 133: 122.

16. Lang DM, Alpern MB, Visintainer PF. Risk of anaphylactoid reactions from intraveinous radioiodinated urographic contrast media in patients receiving beta adrenergic blockers: a case control study. J Allergy Clin Immunol 1990; 85: 229.

17. Lang DM, Alpern MB, Visintainer PF, Smith ST. Increased risk for anaphylactoid reaction from contrast media in patients on beta-adrenergic blockers or with asthma. Ann Intern Med 1991; 115: 270-6.

18. Lang DM, Alpern MB, Visintainer PF, Smith ST. Elevated risk of anaphylactoid reaction from radiographic contrast media is associated with both beta-blocker exposure and cardiovascular disorders. Arch Intern Med 1993; 153: 2033-40.

19. Reynolds NJ, Wallington TB, Burton JL. Hydralazine predisposes to acute cutaneous vasculitis following urography with iopamidol. Br J Dermatol 1993; 129: 82-5.

20. Goodfellow T, Holdstock GE, Brunton FJ, Bamforth J. Fatal acute vasculitis after high-dose urography with iohexol. Br J Radiol 1985; 59: 620-1.

21. Stillman AE. Hepatotoxic reaction to iodipamide meglumine injection. JAMA 1974; 228: 1420-1.

22. Sutherland LR, Edwards MD, Medline A, Wilkinson RW, Connon JJ. Meglumine iodipamide (Cholographin^®) hepatotoxicity. Ann Int Med 1977; 86: 437-9.

23. Scholz FJ, Johnston DO, Wise RE. Intravenous cholangiography. Radiology 1975; 114: 513-8.

24. Motoki T, Ikeuchi M, Hirano M, Yatsuji Y, Murao S. A case of meglumine iodipamide hepatotoxicity. Am J Gastroenterol 1979; 72: 71-4.

25. Klein R, Huizenga JR, Gips CH, Berg PA. Antimitochondrial antibody profiles in patients with primary biliary cirrhosis before orthopic liver transplantation and titres of mitochondrial antibody-subtypes after transplantation. Hepatol 1994; 20: 181-9.

26. Koh WH, Dunphy J, Whyte J, Dixey J, Mchugh NJ. Characterisation of anticytoplasmic antibodies and their clinical associations. Ann Rheum Dis 1995; 54: 269-73.

27. Petri M, Allbritton J. Antibiotic allergy in systemic lupus erythematosus: a case-control study. J Rheumatol 1992; 19: 265-9.

28. Pirmohamed M, Kitteringham NR, Breckenridge AM, Park BK. Detection of an autoantibody directed against human liver microsomal protein in a patient with carbamazepine hypersensitivity. J Clin Pharmacol 1992; 33: 183-6.

29. Dadamessi I, Leduc I, Duché A, et al. Hépatite auto-immune et syndrome lupique secondaires à la prise de minocycline. Rev Méd Interne 1999; 20: 930-3.

30. Beaune P, Dansette PM, Mansuy D, et al. Human anti-endoplasmic reticulum autoantibodies appearing in a drug-induced hepatitis are directed against a human liver cytochrome P-450 that hydroxylates the drug. Proc N Acad Sci 1987; 84: 551-5.