Abstract
Azathioprine hypersensitivity syndrome is a rare but potentially severe side effect of azathioprine use. It has a variable and non-specific presentation making it difficult to distinguish from sepsis or disease relapse. High clinical suspicion is therefore required for recognition and prompt cessation of azathioprine for symptom resolution. Herewith two cases of severe azathioprine hypersensitivity syndrome are described, one in association with Sweet syndrome. Both presented with vague symptoms 2 weeks after commencing azathioprine for antineutrophil cytoplasmic antibody vasculitis. The differentials of sepsis and disease relapse were considered prior to cessation of azathioprine which resulted in a dramatic improvement in both cases. These cases highlight the diagnostic challenge azathioprine hypersensitivity syndrome presents. It should be suspected when there is a temporal relationship to drug initiation, with absence of infection or serological evidence of disease relapse.
Keywords: renal system, unwanted effects / adverse reactions
Background
Azathioprine is an immunosuppressant agent used to treat a variety of autoimmune diseases. It is a prodrug which is converted into 6-mercaptopurine (6-MP) and then onto active metabolites. The active metabolite 6-thioguanine nucleotide is responsible for the cytotoxic activity of azathioprine and induces dose-dependent side effects including myelosuppression and hepatotoxicity.1
Azathioprine hypersensitivity syndrome (AHS) is a rare dose-independent side effect. It has a variable presentation which may include fever, arthralgia, abdominal pain, nausea, vomiting or cutaneous involvement.2 This variable presentation can make azathioprine hypersensitivity syndrome difficult to distinguish from sepsis or disease relapse and therefore high clinical suspicion is required for prompt diagnosis.
Case presentation
Case 1
A 48-year-old man with no significant medical history was diagnosed with antineutrophil cytoplasmic antibody (ANCA) vasculitis (microscopic polyangiitis) in April 2018 after presenting to his general practitioner with left flank pain and decline in renal function. The p-ANCA titre was 1:640 and the antimyeloperoxidase antibodies titre was 199iu (reference range <20.; Bio-flash systems, Werfen, Australia). A renal biopsy confirmed pauci-immune crescentic and necrotising glomerulonephritis. He was treated with pulsed methylprednisolone over 3 days followed by 6 months of 150 mg oral cyclophosphamide and oral prednisolone initially 50 mg with slow wean to 5 mg. After 6 months, cyclophosphamide was ceased and switched to azathioprine at a dose of 200 mg/day (2 mg/kg).
Two weeks after starting azathioprine, the patient presented to his local hospital with non-specific symptoms of myalgia and headache. On presentation, he was found to be febrile and tachycardic. C reactive protein (CRP) was 197 mg/L (<5.0) and white cell count 6.9× 109/L (4.0–11.0) with creatinine 1.62 mg/dL (0.68–1.24). Broad spectrum antibiotic cover with piperacillin/tazobactam was started and he was transferred to the referral hospital. Further investigations including blood and urine cultures as well as imaging revealed no infective focus. He developed diarrhoea and a new rash covering the arms and legs. This was blanching and not painful or pruritic. A punch biopsy was performed and reported a focus of leukocytoclastic vasculitis noted in the upper dermis.
The patient’s condition deteriorated with persistent fevers, episodic hypotension and an acute kidney injury. Creatinine rose steadily to a peak of 13.35 mg/dL and he became anuric requiring dialysis via temporary catheter. Further investigations as to possible infective source were negative including rickettsia, leptospirosis, mycoplasma, HIV, hepatitis, quantiferon, cytomegalovirus, Epstein-Barr virus serum studies as well as additional cultures of sputum, urine and stool. ANCA myeloperoxidase titres were not raised. Thiopurine methyltransferase (TPMT) level was in the normal range at 0.88 μunits/g of Hb (0.5–1.0) prior to commencing azathioprine. A renal biopsy was performed which revealed tubulointerstitial inflammation with eosinophils and isolated small granulomas associated with disrupted tubules and intratubular neutrophils. The findings were most consistent with a drug-induced interstitial nephritis.
A diagnosis of azathioprine hypersensitivity syndrome was strongly suspected. The patient was given 500 mg intravenous methylprednisolone followed by 25 mg oral prednisolone daily and azathioprine was ceased. The symptoms rapidly improved and he was discharged with clinic review. Prednisolone was gradually weaned to 5 mg/day and 360 mg mycophenolate two times per day introduced for maintenance. A repeat creatinine 6 weeks later showed return to baseline at 1.75 mg/dL.
Case 2
A 63-year-old man was diagnosed with ANCA vasculitis (pauci-immune crescentic and necrotising glomerulonephritis) on renal biopsy in 2015 after presenting with microscopic haematuria and acute kidney injury with a creatinine of 1.81 mg/dL. Prior medical history included dyslipidaemia and non-obstructive coronary artery disease. He was treated with steroids and 100 mg cyclophosphamide then switched to 100 mg azathioprine daily after 6 months. TPMT level was 1.04 μunits/g of Hb (0.50–1.00) prior to commencing azathioprine.
After 2 weeks, he presented to the emergency department with subjective fevers, rigours and headache. A full septic screen was performed and broad-spectrum antibiotic cover with intravenous ceftriaxone was initiated with no obvious source identified. Over the next 5 days, his condition progressively worsened with persistent high fever, tachycardia, a painful swollen right eye and a maculopapular rash throughout the body, including the palms and soles. He developed an acute kidney injury with creatinine rising to 3.35 mg/dL from a baseline of 1.31 mg/dL. Antimicrobials were escalated to meropenem and valaciclovir to cover for possible periorbital cellulitis and mucormycosis. An MRI scan revealed unilateral thickening and infiltration of the ocular muscles with periorbital oedema. A skin biopsy was consistent with a neutrophilic dermatosis and in the clinical setting was in keeping with Sweet syndrome.3 4 Forty milligrams of oral prednisolone and cessation of azathioprine resulted in a dramatic and rapid recovery with creatinine improved to 1.71 mg/dL on discharge. Steroid was weaned by 5 mg/week to 20mg, then by 2.5 mg/week to 10mg, then by 1 mg/week. Five hundred forty milligrams of mycophenolate was later commenced. Onset of typical symptoms shortly after commencing azathioprine with rapid improvement following cessation in this case was in keeping with AHS and Sweet syndrome.
Differential diagnosis
These cases were diagnostic challenges due to the non-specific symptoms on presentation.
One differential to consider is disease relapse, which can present with fevers, raised inflammatory markers and a deterioration in renal function as seen in these cases. The first patient had normal ANCA myeloperoxidase titres and renal biopsy findings were in keeping with a drug-induced interstitial nephritis. In the setting of an elevated titre during active disease, a negative ANCA has a good negative predictive value for disease relapse. In a study of 37 patients with ANCA-associated vasculitis flares, only 3 (8%) were ANCA negative during relapse.5
Sepsis is another key differential diagnosis when encountering a febrile, immunosuppressed patient with high inflammatory markers. It is important to identify the source of suspected sepsis, with initial steps involving thorough history and examination followed by relevant investigations. Neither case had any localising features and were thoroughly investigated to exclude infection and relapse. Of note, the second patient had periorbital oedema raising the possibility of periorbital cellulitis. Despite broad spectrum antibiotic treatment, he continued to clinically deteriorate.
Sweet syndrome (or acute febrile neutrophilic dermatosis) is an autoinflammatory condition that is idiopathic in the majority of patients but can occur in association with malignancy or be drug induced. The rash of Sweet syndrome can vary but typically involves the sudden onset of tender papules and vesicles and some larger plaques or nodules. Fever and general malaise can occur. The eyes are reported as the most common system involved after the skin; ocular acute febrile neutrophilic dermatosis is described as presenting as a painful red eye. In the context of fever and tachycardia, the painful swollen eye of the patient was initially considered to be periorbital cellulitis and mucormycosis rather than a feature of an azathioprine reaction. The failure to improve on antibiotics prompted reconsideration of diagnosis and rapid improvement was seen on cessation of azathioprine with skin biopsy revealing a neutrophilic dermatosis in keeping with Sweet syndrome.
Outcome and follow-up
Cessation of azathioprine led to complete resolution of the clinical syndrome in both cases. Renal function improved in the subsequent weeks. The first patient remains on low-dose prednisolone with gradual weaning and 4-monthly review planned. The second patient has remained stable and attends renal clinic yearly for regular routine follow-up. Monitoring includes creatinine, blood pressure and urine checks every 3 months, and ANCA levels every 6 months.
Discussion
Azathioprine is a prodrug which is metabolised initially into 6-MP then further via one of three competing pathways, one of which involves conversion by thiopurine methyltransferase (TPMT) to form inactive 6-methylmercaptopurine. Several single nucleotide polymorphisms of the TPMT gene on chromosome 6 have been identified which result in reduced TPMT activity and the likelihood of adverse events.6 Azathioprine hypersensitivity syndrome can occur independent of TPMT levels and has a variable presentation which can be life threatening.7 8 It often mimics features common in sepsis or a relapse of the original disease being treated, making early diagnosis challenging.9 10
A recent study11 examined a cohort of patients with ANCA-associated vasculitis who were receiving azathioprine for maintenance therapy and found that 25 of 290 (9%) experienced azathioprine hypersensitivity with a median of 14 days between start of azathioprine and symptom onset. Common features included fever, malaise, arthralgia, rash and a rise in CRP and white cell count. Therefore, the diagnosis should be suspected when there is a clear temporal relationship to initiating the drug, often within the first 2 weeks, together with an absence of infection or serological evidence of relapse.
Both the cases described demonstrate the potential severity of azathioprine hypersensitivity syndrome and the rapid improvement which can be seen on drug cessation.
Raised awareness of azathioprine hypersensitivity syndrome among medical professionals is key to aiding prompt recognition of what can be a very vague presentation. It should be considered as a differential in any patient who presents unwell following commencement of azathioprine.
Learning points.
Azathioprine hypersensitivity syndrome is a rare but potentially catastrophic side effect of azathioprine use that requires high clinical suspicion to diagnose.
Presentation may mimic sepsis or disease relapse and typically occurs within 2 weeks of commencing azathioprine.
Suspect azathioprine hypersensitivity syndrome where there is a temporal relationship to drug initiation and absence of infection or serological evidence of disease relapse.
Early recognition and cessation of azathioprine is key to minimise morbidity and mortality.
Footnotes
Contributors: TM: Lead author involved in planning, research, consenting and write up of report. SD: Guidance in planning and revision of drafts. MM: Revision of drafts. CL: Consenting and pharmacology perspective in planning and research.
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests: None declared.
Provenance and peer review: Not commissioned; externally peer reviewed.
References
- 1.Aleissa M, Nicol P, Godeau M, et al. Azathioprine hypersensitivity syndrome: two cases of febrile neutrophilic dermatosis induced by azathioprine. Case Rep Dermatol;9:6–11. 10.1159/000454876 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bidinger JJ, Sky K, Battafarano DF, et al. The cutaneous and systemic manifestations of azathioprine hypersensitivity syndrome. J Am Acad Dermatol 2011;65:184–91. 10.1016/j.jaad.2010.04.041 [DOI] [PubMed] [Google Scholar]
- 3.Sweet Syndrome - NORD (National Organization for Rare Disorders), 2019. Available: https://rarediseases.org/rare-diseases/sweet-syndrome/
- 4.Acute febrile neutrophilic dermatosis | DermNet NZ, 2019. Available: https://www.dermnetnz.org/topics/acute-febrile-neutrophilic-dermatosis/
- 5.Boomsma MM, Stegeman CA, van der Leij MJ, et al. Prediction of relapses in Wegener's granulomatosis by measurement of antineutrophil cytoplasmic antibody levels: a prospective study. Arthritis Rheum 2000;43:2025–33. [DOI] [PubMed] [Google Scholar]
- 6.Asadov C, Aliyeva G, Mustafayeva K. Thiopurine S-methyltransferase as a pharmacogenetic biomarker: significance of testing and review of major methods. Cardiovasc Hematol Agents Med Chem 2017;15:23–30. 10.2174/1871525715666170529091921 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Brown G, Boldt C, Webb JG, et al. Azathioprine-Induced multisystem organ failure and cardiogenic shock. Pharmacotherapy 1997;17:815–8. [PubMed] [Google Scholar]
- 8.Fields CL, Robinson JW, Roy TM, et al. Hypersensitivity reaction to azathioprine. South Med J 1998;91:471–4. 10.1097/00007611-199805000-00013 [DOI] [PubMed] [Google Scholar]
- 9.Sinico RA, Sabadini E, Borlandelli S, et al. Azathioprine hypersensitivity: report of two cases and review of the literature. J Nephrol 2003;16:272–6. [PubMed] [Google Scholar]
- 10.Greite R, Deutsch K, Bräsen JH, et al. Azathioprine hypersensitivity syndrome in anti-myeloperoxidase anti-neutrophil cytoplasmic antibody-associated vasculitis. Clin Kidney J 2019;12:89–91. 10.1093/ckj/sfy038 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hessels AC, Sanders JSF, van de Ven AAJM, et al. Azathioprine hypersensitivity syndrome in a cohort of antineutrophil cytoplasmic antibody-associated vasculitis patients. J Allergy Clin Immunol Pract 2019;7:1004–9. 10.1016/j.jaip.2018.10.018 [DOI] [PubMed] [Google Scholar]