• amoxicillin allergy;
  • drug allergy;
  • Epstein–Barr Virus infection;
  • infectious mononucleosis;
  • patch test

Amino-penicillins are a major cause of delayed type reactions to penicillins. Furthermore, 70–100% of patients receiving amino-penicillins during a florid Epstein–Barr virus (EBV)-infection develop a maculopapular rash (1), the percentage being 14% for penicillin G, cloxacillin or tetracycline (2). This phenomenon may also occur with herpes simplex-, cytomegaly-, HI-virus- and, probably, many other viral infections. Hypotheses concerning the pathomechanism are (i) that the immune system of patients with infectious mononucleosis has decreased tolerance and/or (ii) enhancement of immune reaction to certain drugs or its metabolites (3). The general opinion documented even in high ranked review articles on drug eruptions (4) as well as allergy text books is that in most cases an EBV-associated amoxicillin rash is reversible and, as a consequence, a differentiated allergy diagnostic test is not considered or recommended (4).

To test the hypothesis that some EBV-infected patients in contrast might develop a ‘true’ allergic drug reaction to amino-penicillins, 41 Dermatology outpatients presenting with drug eruptions following the intake of amino-penicillins were consecutively investigated for additional acute infectious mononucleosis (fever, tonsillopharyngitis, lymphadenopathy). In case of suspicion, anti-EBV-IgM and -IgG antibodies were investigated. After complete resolution but not earlier than 3 months after the diagnosis of acute EBV-infection had first been confirmed, allergy diagnostic tests consisting of IgE-detection against amoxicillin, ampicillin, penicillin and cefaclor with CAP-FEIA (Phadia, Freiburg, Germany), skin prick tests and patch tests with amoxicillin trihydrate, cefotaxim sodium and benzyl penicillin 10 000 IE with readings after 48 and 72 h were performed. Eight out of 41 had a florid infectious mononucleosis at the time of the drug eruption, five of eight patients had positive patch tests to amoxicillin and two of five additionally to penicillin (Table 1). One patient out of five was challenged with verum and developed a rash after 48 h which resolved under systemic treatment with anti-histamines and topically applied glucocorticosteroids without the necessity for hospitalization. After resolution, the oral challenge test with a cephalosporin was well tolerated.

Table 1.   Patients with EBV-associated drug rash to amoxicillin: synopsis of the allergy diagnostic test results
Patient no.SPTPatch test AMXPatch test penicillinPatch test cefutaximChallenge AMXChallenge penicillinChallenge cephalosporinChallenge refused
  1. Results written in italics describe those individuals who most probably only have a transient EBV-associated drug intolerance.

  2. EBV, Epstein–Barr virus; SPT, skin prick test; AMX, amoxicillin; –, not done.

5AMX at 48 h+PositiveNegativeNegativeyes

Another amoxicillin patch test positive patient tolerated the oral challenge with penicillin, thereby revealing an isolated allergy to amoxicillin. As a challenge test would involve the risk for the patients to experience the same signs and symptoms of the previous drug reaction, the decision of four out of eight patients not to perform an oral challenge test was respected. Two of three patients, who were completely negative in the allergy diagnostic tests, agreed to an oral challenge with verum and revealed tolerance to the drug. It can be hypothesized that the third patch test negative subject may also have tolerated amoxicillin challenge. These three individuals may belong to the population, which only has a transient decrease in drug tolerance during EBV infection. Two amoxicillin patch test positive patients could be examined in a follow-up visit 2.2 and 1.5 years, respectively, after the first patch testing. Those patients revealed a patch test reaction similar in strength to the preliminary test result, which is indicative for a persistent delayed-type reaction to amoxicillin.

In conclusion, this study does not only provide additional evidence that true delayed-type allergic reactions to amino-penicillins may develop during a florid viral infection, but that they may be definitely more prevalent than previously assumed and published (5, 6). Additionally, follow-up investigations on two patients revealed the persistence of the delayed-type reactions to amoxicillin, an approach, which has to the best of our knowledge not been published before. These observations should encourage the investigation of patients with amino-penicillin-induced exanthema for both, EBV-infection and true drug allergy. Particularly young women should be recommended an oral challenge with cephalosporins and maybe penicillins because the latter are antibiotics which may be given safely during pregnancy.


  1. Top of page
  2. References
  • 1
    Leung AKC, Rafaat M. Eruption associated with amoxicillin in a patient with infectious mononucleosis. Int J Dermatol 2003;42:553555.
  • 2
    Partel BM. Skin rash with infectious mononucleosis and ampicillin. Paediatrics 1967;40:910911.
  • 3
    Partel A. Amoxicillin hypersensitivity reaction in a patient with infectious mononucleosis: a case report. Conn Dent Stud J 1986;6:1213.
  • 4
    Gruchalla RS, Pirmohamed M. Antibiotic Allergy. New Engl J Med 2006;354:601609.
  • 5
    Nazareth I, Mortimer P, McKendrick GD. Ampicillin sensitivity and infectious mononucleosis – temporary or permanent? Scand J Infect Dis 1972;4:229230.
  • 6
    Renn CN, Straff W, Dorfmüller A, Al-Masoudi T, Merk HF, Sachs B. Amoxicillin-induced exanthema in young adults with infectious mononucleosis: demonstration of drug-specific lymphocyte reactivity. Br J Dermatol 2002;147:11661170.