Sulfa Drug Discombobulation

Revised and reviewed 5 April 2014

aka Immunological Discombobulation 002

You’ve just diagnosed your patient with congestive heart failure and are about to give him a dose of frusemide, when you remember he has an allergy to sulfa drugs.

“Hmm, doesn’t frusemide contain a sulfa moiety?”, you ask yourself as you scratch your head, feeling slightly immunologically discombobulated.

[NB. the International nonpropriety name for frusemide is furosemide]


Q1. What is frusemide’s mechanism of action?

Frusemide is a loop diuretic.

Frusemide inhibits NKCC2, the luminal Na-K-2Cl symporter in the thick ascending limb of the loop of Henle. This results in decreased reabsorption of Na, K and Cl and concomitant diuresis. Excessive frusemide use is characterised by a hypokalemic hypochloremic metabolic alkalosis.

The effect of loop diuretics of the Na-K-2Cl symporter also abolishes the gradient required for the reabsorption of Mg and Ca. Thus hypomagnesemia and hypocalcemia can occur with frusemide use, though the latter is less common as Ca is actively reabsorbed in the distal convoluted tubule.

Q2. Does frusemide contain a sulfa moiety?

Yes, it sure does — frusemide is a ‘sulfa drug’.

This is it’s chemical structure:

… and this is the structure of sulfamethoxazole, the most widely used sulphonamide antibiotic:

Q3. What other non-antibiotic drugs are ‘sulfa drugs’?

There are lots of sulfa drugs around (this list is not exhaustive!) — some of them may surprise you!


  • carbonic anhydrase inhibitors — acetazolamide
  • loop diuretics — e.g. frusemide, budenoside
  • thiazide diuretics — e.g. hydrochlorothiazide, indapamide, metalozone


  • chlopropramide
  • glipizide
  • tolbutamide

Rheumatological agents

  • sulfasalazine
  • probenicid
  • celecoxib (NB. rofecoxib is not a sulfa drug)

Other agents

  • sumatriptam
  • topiramate
  • ibutilide
  • sotalol
  • dapsone

Q4. What reactions may occur in patients with a ‘sulfa drug’ allergy?

Adverse reactions to sulphonamide antibiotics are relatively common (~3%), but only about 3% of these are true hypersensitivity reactions.

The most common manifestation of hypersensitivity is a fever or a maculopapular rash that develops 7 to 14 days after initiating the offending agent. However, there are several life-threatening manifestations of hypersensitivity to sulfa drugs. These include:

Unfortunately, there is no reliable test for sulfa drug allergy.

Q5. Can you give frusemide to a patient with a ‘sulfa drug’ allergy?

Yes, but…

Frusemide, like other non-antibiotic sulfa drugs lack the specific moieties present in antibiotic sulfa drugs (sulphonamides) that are thought to mediate hypersensitivity. These moieties are:

  • the N1 heterocyclic ring — causes type I hypersensitivity reactions.
  • the N4 amino nitrogen — results in reactive metabolites that cause either direct cytotoxicity or an immunologic response.

The best evidence available concerning sulfa drug cross-reactivity is probably Strom et al’s (2003) retrospective cohort study. These investigators studied 969 patients who had allergic reactions to sulfonamide antibiotics, and 19257 patients who had not, who subsequently received sulfonamide nonantibiotics. Using a very broad definition of ‘allergy’ patients allergic to sulfonamide antibiotics were more likely than nonallergic patients to react to sulfonamide nonantibiotics (9.9% vs 1.1%). However, there was an even higher rate among patients with documented penicillin allergy (14.2%)! Thus it is more likely that a reaction is related to a greater predisposition to allergic reactions in general among these patients rather than any specific sulfa hypersensitivity.

Expert opinion suggests that cross-reactivity between antibiotic sulfa drugs and non-antbiotic sulfa drugs is extremely unlikely.

Incidentally, there has only been one reported case of anaphylaxis to frusemide, and it could not be determined if the sulfa moiety was responsible.

Ponka (2006) summarises the approach to the problem of non-antibiotic sulfa drugs in patients with ‘sulfa drug allergy’ a bit like this:

  • Is there a history of severe or life-threatening sulfa allergy?
    YES —> do NOT give a non-antibiotic sulfa drug except in an emergency and there are no available alternatives.
    NO —> ask the next question…
  • Is there a history of allergy to non-antibiotic sulfa drugs?
    YES —> do NOT give a non-antibiotic sulfa drug except in an emergency and there are no available alternatives.
    NO —> give a ‘test dose’ in a monitored setting after appropriate patient counseling, or use an available alternative.

Always seek the opinion of an allergist/ immunologist first.

Some additional comments:

Ponka’s approach is cautious. From what we know of the pharmacology mediating antibiotic sulfa drug hypersensitivity and the limited evidence available, there is no reason to suspect that there should be cross-reactivity with non-antibiotic sulfa drugs. However, absence of evidence is not the same as evidence of absence…

I am not aware of any established protocols describing how to give a ‘test dose’ in this situation. I suspect that the risk is sufficiently low (when there is no history of a severe sulfa drug reaction, and no history of a reaction to a non-antibiotic sulfa drug) that the usual treatment dose could be given followed by a period of close observation. A more cautious approach (as suggested by Daman Langguth, see comments below) would be to give 1/10th of the usual dose.

Finally, it is rare that a sulfa drug ever needs to be given as a true emergency.

Q6. What alternatives are there to frusemide?

Ethacrynic acid

This drug is a non-sulfa containing loop diuretic. The dose is 50-200mg once or twice daily (50mg has a similar diuretic effect to 40mg frusemide). Like frusemide it is potentially ototoxic and hypokalemia is the main side effect — and it also causes diuresis over about 6 hours (hence the tradename ‘lasix’ for frusemide: ‘lasts six hours’).

Rare side-effects of ethacrynic acid are:

  • GI toxicity (nausea, vomiting, diarrhoea, GI haemorrhage)
  • CNS effects: hallucinations, inappropriate behaviour, mania
  • It is an ADEC category C drug with respect to pregnancy due to risk of neonatal thrombocytopenia and electrolyte disturbance.

Also, it should be noted that the potassium-sparing diuretics (triamterene, spironolactone, and amiloride) do not contain a sulfa-moiety and are also an option.



Journal articles

  • Ponka D. Approach to managing patients with sulfa allergy: use of antibiotic and nonantibiotic sulfonamides. Can Fam Physician. 2006 Nov;52(11):1434-8. PMID: 17279201; PMCID: PMC1783707.
  • Strom BL, Schinnar R, Apter AJ, Margolis DJ, Lautenbach E, Hennessy S, et al. Absence of cross-reactivity between sulfonamide antibiotics and sulfonamide nonantibiotics. N Engl J Med. 2003;349:1628–1635. PMID: 14573734
  • Wulf NR, Matuszewski KA. Sulfonamide cross-reactivity: is there evidence to support broad cross-allergenicity? Am J Health Syst Pharm. 2013 Sep 1;70(17):1483-94. doi: 10.2146/ajhp120291. PubMed PMID: 23943179.

FOAM and web resources

Print Friendly


  1. says

    This whole drug cross-interactivity shebang is fascinating…

    I suspect people who are ‘penicillin allergic’ are like ‘sulfa allergic’ people -- they just have higher rates of drug reactions that maybe nonspecific rather than involve true immunological cross-reactivity.

    A recent review suggests that penicillin allergy crosses over with cephalosporin allergy only about 1% of the time -- not 10% as I was taught in med school. Also, the cross reactivity is only to first and second generation cephalosporins or those with a similar ‘R1 group’. Thus, we should be able to ‘safely’ give sick patients with a history of ‘penicillin allergy’ a dose of ceftriaxone.

    Campagna JD, Bond MC, Schabelman E, Hayes BD. The Use of Cephalosporins in Penicillin-allergic Patients: A Literature Review. J Emerg Med. 2011 Jul 8. [Epub ahead of print] PubMed PMID: 21742459.

  2. daman langguth says

    Allergic cross reactivity is a difficult field as is diagnosis and testing generally in drug allergy. IgE reactions to drugs probably more likely than say Type IV (a-d) hypersensitivity.

    IgG anaphylaxis is real and probably explains some drug reactions.

    Cross reactivity from pen to ceph seems more likely in children than adults, according to published data, though there are considerable ascertainment biases in most studies.

    Sulpha reactions may be IgE or non-IgE (all the rest), with SJS being most worrying.

    With frusemide, most patients who are allergic (type IV generally) also cross react to other diuretics and desensitization is possible though may not be successful. Luckily true allergy to most things is rare.


    if patient says they are allergic: 1) they wont be 99/100 but 2) 1/100 might be lethal

    History of reaction is absolutely critical (ask re mucosal involvement etc)

    If they don’t need the drug, give a different one that is very different

    If they do need the drug, ask if they really need it again!!

    If they do, and history is not suggestive, test dose or challenge OK.

    Desens regimens are possible and work (probably because person not actually allergic).

    (Clinical Immunologist)

    • says

      Great comment Daman!

      Its important to remember:
      1. first, do no harm
      2. it is rare that a sulfa drug really needs to be given to provide benefit in an emergency (e.g. APO = CPAP and GTN, not frusemide initially)

      So unless the history sounds very unworrying it rare that we’d need to give a test dose of a sulfa drug in the ED. How exactly would you suggesting administering a test dose of a sulfa drug? Just the usual dose in a monitored setting with resuscitation capabilities? Or a reduced dose? Or something else? I suspect I’d be conferring with my friendly on call immunologist first anyway…


  3. daman langguth says

    Test doses usually 1/10 of a standard minimal dose that works (as u can see this is empiric, but widely used and would be standard).

    When reactions are delayed like most sulpur drugs, test doses not possible, but drug patch testing (not an ED thing may be useful…may be).

    If there is a history of immediate hypersensitivity then say 4mg of IV frusemide cf 40mg. and yes in a resus type environment.

    If history classic, can initiate a desesn protocol IV (like for penicillin which is published in several texts).

    And yes, spread the blame from the outset!!