Cross-Reactivity Between Sulfonamides and Loop or Thiazide Diuretics: Is it a Theoretical or Actual Risk?

Abstract

This is a case report of furosemide use in a woman with acute renal transplant rejection and history of anaphylaxis to sulfonamide and a review of the literature.

Allergists are consulted about the question of hypersensitivity to loop or thiazide diuretics in patients with a history of sulfonamide allergy because it is thought that there is a risk of cross-reactivity from common sulfonamide constituents of these compounds. In text references and drug-package inserts, reference is made to this risk. For example, the package insert for furosemide states that “patients allergic to sulfonamides may also be allergic to furosemide” [1]. Therefore, in clinical practice, the use of thiazide or some loop diuretics is often avoided in patients with a prior history of sulfonamide allergy. The literature on this topic, however, is sparse and not well documented. We report an instructive case of a woman with a history of anaphylaxis to sulfonamides who subsequently received furosemide while undergoing acute renal transplant rejection.

Case Report

A 30-year-old woman was the recipient of a cadaveric renal transplant because of end-stage renal disease secondary to insulin-dependent diabetes mellitus. She tolerated the transplant well, and for 3 years was maintained on cyclosporin at 4 mg/kg/day without complications.

At age 33, she developed an acute onset of fever, hematuria, headaches, progressive weight gain, and pain in the allograft area. She was admitted to the hospital with elevated creatinine levels and presumed organ rejection. She was given two doses of 3 mg/kg intravenous solumedrol 12 h apart on hospital day 1. A single dose of 1.3 mg/kg intravenous furosemide was administered after the patient became oliguric on day 2. She was then transferred to our institution for further management.

After transfer, she was given a 5 mg/kg intravenous loading dose of solumedrol and started 1 mg/kg oral prednisone. In addition, tacrolimus (FK506) rescue was given at 0.075 mg/kg per day. A renal biopsy was performed and showed widespread polymorphonuclear infiltrate with many eosinophils and a lymphocytic tubulitis consistent with acute rejection.

Twenty years previously, the patient had anaphylaxis after taking one oral dose of sulfamethoxazole and trimethoprim. She immediately developed urticaria and shortness of breath. At the emergency room she was hypotensive. Her symptoms resolved several hours later after treatment with epinephrine, diphenhydramine, steroids, and intravenous fluids. She had no further use of the medication and had no history of other drug allergies. There was no family history of medication allergies.

On transfer her complete cell blood count (CBC) was normal. Because of her history of sulfonamide allergy, loop and thiazide diuretics were avoided and she was given ethacrynic acid at 1 mg/kg intravenously daily to improve her urine output. After 2 weeks of therapy, she developed pancytopenia, a known complication of ethacrynic acid, which was then discontinued. Within 2 weeks, her CBC returned to normal in all indices.

Because of her sensitivity to the ethacrynic acid, her history of sulfonamide allergy, and evidence of numerous eosinophils on the renal biopsy, an allergy consultation was requested. Of particular concern was an appropriate diuretic to use for this patient. The drug of choice was furosemide, but there was concern about her history of sulfonamide allergy and the eosinophilia on renal biopsy which was taken within 2 days of a single dose of intravenous furosemide given at the referring medical center.

After review of the clinical history and pertinent literature, we offered the opinion that furosemide may be safely tried in this patient [2–4]. We concluded that it was unlikely that furosemide caused the eosinophilia on renal biopsy since eosinophilia is a pathologic finding consistent with acute allograft rejection [5, 6]. Furosemide was given repeatedly during the admission with no evidence of acute allergy, rashes, or eosinophilia.

Discussion

It has long been taught that loop and thiazide diuretics pose a theoretical risk of cross sensitivity in a patient with allergy to sulfonamide antimicrobials due to their common SO₂-NH_x structure (see Figure 1) [1]. In an extensive review of the literature, we found only four published cases of possible cross-reactivity between loop or thiazide diuretics and sulfonamides [2–4]. Table 1 summarizes these cases. All of these cases involved only a circumstantial reaction between reactions to drugs in the two categories.

Figure 1.

Structural formulas for the sulfonamide antimicrobials and diuretics used in the reported cases. SO₂-NH_x groups are highlighted.

Table 1.

Published cases of putative sulfonamide-diuretic cross sensitivity

Case	Age (years) and sex	Case
1	56, F	Papular rash developed 17 days after one dose oral sulfadimethoxine. One year later developed papular pigmented rash after oral chloramidobenzole. Four weeks later developed rash after oral acetazolamide. Two years later developed rash after oral hydrochlorthiazide [2]
2	66, F	Exanthem developed shortly after sulfamethoxypyridazine. Five years later developed blisters 24 hours after oral quinethazon [2]
3	55, M	Patient was on glisoxepid, a oral hypoglycemic. One tablet of furosemide caused nausea, edema, vomiting, and diarrhea. Addition of glibenclamid, an oral hypoglycemic caused worsening symptoms. Skin tests by scratch weakly positive to sulfamethoxazole but no documented challenge done [3]
4	68, F	Indapamide caused fixed drug eruption and was confirmed with repeat oral challenge. Oral challenge with sulfamethoxazole and sulfadiazine were positive but were negative to furosemide [4]

In case 1 the subject developed a papular rash 17 days after one oral dose of sulfadimethoxine. One year later, a papular pigmented rash developed after oral chloramidabenzole and acetazolamide. Two years later a similar papular rash developed while taking oral hydrochorothiazide [2]. In case 2, an exanthem developed shortly after sulfamethoxypyridazine. Five years later, blisters developed after oral quinethazon, a thiazide diuretic [2]. These reactions were not consistent with immunoglobulin E (IgE)-mediated hypersensitivity. They could reflect T-cell cross-reactivity, in a delayed-type hypersensitivity reaction. The cases are also consistent with multiple independent drug allergy. No immunologic or challenge studies were performed. In case 3, a patient on oral glisoxepid (an oral hypoglycemic), edema and gastrointestinal symptoms developed after one tablet of furosemide. Glibenclamide, another oral hypoglycemic, was started with worsening of symptoms. Skin tests were weakly positive to sulfamethoxazole, but no oral challenge was done [3]. The weakly positive skin test is difficult to interpret since most investigators have failed to elicit positive skin tests to sulfamethoxazole even in patients with good history of IgE-dependent reactions [5]. It is also interesting to note that the patient tolerated the oral hypoglycemic glisoxepide but did not tolerate glibenclamide, which both have internal SO₂-NH_x moieties (see Figure 2). In case 4, a fixed drug eruption developed on the oral diuretic, indapamide. Subsequent oral challenges to sulfonamide antimicrobials were positive, but the more structurally similar furosemide failed to elicit a response to oral challenge [4]. The patient could have had an independent allergy to the sulfonamide antimicrobial.

Figure 2.

Structural formulas for some oral hypoglycemics. SO₂-NH_x groups are highlighted.

Figure 1 compares the chemical structures of the sulfonamide antimicrobials with the thiazide and loop diuretics reported in these cases [1–3]. The common structural component of a SO₂-NH_x moiety is the theoretical basis for presumed cross-reactivity between such compounds [6].

The use of thiazide or loop diuretics in the general population is quite common, especially in geriatric, renal, and cardiac patients, and allergy to sulfonamide antimicrobials has been estimated to have a prevalence of only 5% [6]. It seems reasonable that if there were appreciable cross-reactivity between these medications, clearer cases of clinical cross-sensitivity would be more numerous. It remains possible that many such cases go unreported. Furthermore, what the published cases do illustrate is that putative cross-sensitivity was most often manifest as cutaneous rashes and late onset signs suggestive of T-cell pathogenesis rather than IgE antibody-dependent reactions.

Celecoxib is the first of a new family of nonsteroidal anti-inflammatory drugs (NSAIDs) that selectively inhibit cycloxy-genase 2 while sparing cycloxygenase 1, and thas clinically been shown to have less side effects of gastrointestinal bleeding than currently available NSAIDs [7]. Celecoxib also has a SO₂-NH_x side group [7] (see Figure 3). While clinical experience is still limited, if we accept the lack of substantial evidence for cross-reactivity among common SO₂-NH_x, structures, we might anticipate that the actual risk of cross-reactions with this medication may also be small.

Figure 3.

Structural formula for the new cycloxygenase 2 inhibitor, celecoxib. The SO₂-NH_x group is highlighted.

Another instructive point in our current case is that while eosinophils on pathologic biopsy may suggest drug hypersensitivity, alternative explanations must also be considered. Acute renal transplant rejection can manifest eosinophil infiltrates, as we believe is illustrated in this case [8, 9]. How eosinophils participate in allograph rejection is not clear [8].

In conclusion, this case reports an uneventful treatment with furosemide in a patient with a history of anaphylaxis to a sulfonamide antibiotic. A review of the literature suggests that while there is a theoretical concern about cross-reactivity among sulfonamide antibiotics, loop or thiazide diuretics, and oral hypoglycemics, the actual risk will likely be small. Further immunochemical research will be needed to define more clearly the frequency and degree of cross reactivity attributable to various drugs with common SO₂-NH_x structures.