Article Text

PDF

Cefuroxime induced lymphomatoid hypersensitivity reaction
  1. S A M Saeeda,
  2. M Bazzab,
  3. M Zamana,
  4. K S Ryattb
  1. aManor Hospital, Walsall, UK: Department of Elderly Care, bDepartment of Dermatology
  1. Dr M Bazza, Department of Dermatology, Walsall Manor Hospital, Walsall, West Midlands WS2 9 PS, UK

Abstract

An 84 year old women developed erythematous blotchy erythema and purpuric rashes over the lower limbs three days after being started on intravenous cefuroxime for acute diverticulitis. A skin biopsy specimen showed a mixed infiltrate of lymphoid cells and eosinophils; many of the lymphocytes were large, pleomorphic, and showed a raised mitotic rate. Immunohistochemistry showed the infiltrate to be T cell rich, with all the large cells being CD30 positive. Typical mycosis fungoides cells, marked epidermotropism, and Pautrier's abscesses were not seen. The rash disappeared 10 days after cessation of cefuroxime and the patient remained asymptomatic 15 months later. This apparent cutaneous T cell lymphoma-like reaction is best described as lymphomatoid vascular reaction. The drug induced immune response with an atypical cutaneous lymphoid infiltrate mimics a cutaneous pseudolymphoma.

  • cefuroxime
  • atypical cutaneous lymphomatoid infiltrate
  • cutaneous T cell lymphoma

Statistics from Altmetric.com

Case report

An 84 year old women, with established diverticula disease, presented as an emergency with a short history of fever, acute abdominal pain, and diarrhoea. On examination she was found to be toxic, febrile, dehydrated, with a tachycardia of 110/min and blood pressure 80/40 mm Hg. The left iliac fossa was tender. Haemoglobin, full biochemistry profile, including serum amylase, chest radiography, plain abdominal radiography, and an abdominal ultrasound scan were normal. Her white cell count was raised at 20.4 × 109/l with marked neutrophilia at 14.2 × 109/l.

A provisional clinical diagnosis of acute diverticulitis with septic shock was made and she was started on intravenous (IV) fluids, IV metronidazole 500 mg eight hourly, and IV cefuroxime 750 mg three times a day in a sequential manner. Three days later and after five doses of IV cefuroxime, purpuric and erythematous macular rashes developed over the lower limbs. Cefuroxime was considered to be responsible and was discontinued; metronidazole was continued for another week. The rashes were treated with flucinolone acetonide 0.00625% cream. Further investigations at this stage showed normal immunoglobulins, antinuclear factor, antineutrophilic cytoplasmic antibodies, and autoantibody profile. A haematoxylin and eosin stained section of representative skin biopsy showed a mixed dermal infiltrate of lymphoid cells and eosinophils, marked red blood cell extravasation indicative of ongoing small vessel damage, minimal focal vacuolar interface change and some basement membrane hyalinisation, both within the papillary dermis of the inflamed skin and in the immediate vicinity. The lymphoid cells were predominantly lymphocytic, with many of these being large and polymorphic and showing hyperchromatic and irregular nuclei and mitoses (fig 1). Epidermotropism was only mild and focal and there were no Pautrier's abscesses. Immunohistochemistry showed the infiltrate to be T cell rich, with virtually all of the large cells being CD30 positive (fig 2). Direct immunofluorescence showed fine linear deposition of IgG along the dermoepidermal junction.

Figure 1

High power view, haematoxylin and eosin stain, demonstrating large, pleomorphic lymphoid cells with mitotic activity and lymphocytes in the background (original magnification × 40).

Figure 2

CD30 immunohistochemical stain demonstrating CD30 positive cells in the infiltrate (original magnification × 20).

Pyrexia, diarrhoea, tachycardia, and hypotension improved rapidly with supportive measures and the rashes began to fade quickly and cleared within seven days. She was completely asymptomatic and clear of all rashes when reviewed more than 15 months after the initial episode.

Discussion

Cephalosporins are widely used for the treatment of septicaemia, pneumonia, meningitis, biliary and urinary tract infection, and peritonitis. Both the old and new third generations of cephalosporins are generally well tolerated at standard recommended dosage.1 The sequential cefuroxime regimen was chosen as it has been shown not only to be an effective treatment for severe infections but also easier to administer.2 The incidence of adverse effects is approximately 2.5%,3 the most common being gastrointestinal—for example transient, mild to moderate nausea, diarrhoea, increase in serum transaminases and alkaline phosphatase, and very rarely Clostridium difficile induced pseudomembranous colitis. Other adverse reactions include central nervous system complaints—for example headache, dizziness, somnolence, confusion, hyperactivity, and reversible encephalopathy4 and transient haematological changes—for example leucopenia, thrombocytopenia and eosinophilia, serum sickness reaction, and renal failure.5 Cutaneous reactions are observed less frequently but urticaria, angio-oedema, erythema multiforme, exanthema, pruritus, pustular eruptions, fixed drug eruptions, glossitis, oral ulceration, and vaginitis have been described.6 Henoch-Schönlein purpura with systemic manifestations, including renal involvement7 and acute generalised exanthematic pustulosis8 have recently been reported. Phenytoin,9 carbamazepine,10phenothiazine,11 angiotensin converting enzyme inhibitors,12 β-blockers,13antidepressants,14 and benzodiazepines15 have already been reported to cause atypical cutaneous lymphoid infiltrates, mimicking pseudolymphoma but cephalosporins, to our knowledge, have not been previously linked with this tissue response. Unlike the blotchy macular erythema and purpuric rashes as observed in our case, the more common mode of presentation was with solitary erythematous or purple nodules, a few infiltrated plaques or multiple infiltrative papules .11-15 Some diagnostic confusion, both clinical and histological, occurs most commonly with mycosis fungoides and lymphomatoid vascular reaction (table 1).

Table 1

Comparison between microscopical features of Mycosis fungoides and lymphomatoid vascular reaction

The precise mechanism of this drug induced lymphomatoid vascular reaction is unknown. However, the β-lactam ring of cephalosporins is well known to impart a degree of instability to the cephalosporin molecule; the ring can open up, conjugate with proteins, become a hapten and immunogenic, and thus induce a delayed type of hypersensitivity reaction. Indeed, moderate spongiosis, keratinocyte necrosis, papillary dermal oedema, the presence of angiocentric infiltrate of atypical lymphocytes, absent or minimal vessel wall damage, minimal and focal exocytosis of cells in the epidermis overlying the inflamed dermal papillae observed in this case are well recognised features of delayed type hypersensitivity reaction. Such a delayed type reaction with a cutaneous T cell lymphoma-like morphology has been described as lymphomatoid vascular reaction.16Most of the large lymphocytic cells in the infiltrate stained CD30 positive. In the context of neoplasia, CD30 is positive in Reed-Sternberg cells of Hodgkin's disease, the cells of lymphomatoid papulosis and in anaplastic large cell lymphoma (Kil lymphoma). This has been shown to carry an excellent prognosis.10 17

The close temporal association of starting cefuroxime and the appearance of the rash, the cutaneous T cell lymphoma-like histology with more features of a lymphomatoid vascular reaction than mycosis fungoides and rapid resolution of rash on cessation of the suspected drug, all favour lymphomatoid hypersensitivity reaction to cefuroxime as the most likely diagnosis. This reaction pattern, not previously described with cephalosporins, appears to be a benign reaction, with there being no relapses for more than a year after cessation of the original offending drug. Our patient remains well and clear of her rash as well as showing no sign of developing any cutaneous or systemic lymphomatous pathology 15 months after her discharge.

Acknowledgments

We are grateful to Dr D Slater, Consultant Histopathologist, Rotherham General Hospital and Dr C Allen, Consultant Histopathologist, Walsall Manor Hospital for their advice on skin histology.

References

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.