Drug-induced hypersensitivity syndrome with lupus manifestations due to mesalazine in a patient with ulcerative colitis

Abstract

Mesalazine is often used as first-line therapy for ulcerative colitis. Several reports have pointed to systemic adverse reactions associated with this drug. Most have evoked a drug-induced hypersensitivity syndrome, while some have described lupus syndromes but with limited clinical and varied biological features. A 75-year-old man presented with fever, dyspnoea, chest pain, polyarthralgia, and myalgia, following mesalazine introduction. Clinical symptoms and low-titre positive antihistone antibodies disappeared after mesalazine withdrawal without recourse to steroids. Pericardial effusion and 8F-fluorodeoxyglucose uptake on positron emission tomography/CT scan, and glomerular haematuria and proteinuria also disappeared. Cytokine-lymphocyte transformation tests showed a strong sensitisation pattern with interleukin-5 production. This case advances our knowledge of the mechanism of mesalazine-induced adverse effects, namely via drug-induced hypersensitivity with lupus manifestations, which we are the first to report.

Background

The clinical picture presented in this case is quite unique. The only other evidence for mesalazine-induced antinuclear-antibody-positive pericarditis, published in BMJ in 1992, did not describe other systemic symptoms or antihistone antibodies. The latter were described in several reports of mesalazine-induced arthritis, but without concomitant pericarditis or other systemic symptoms.

In addition to the unique clinical picture, we are the first to describe a drug-induced hypersensitivity syndrome with lupus manifestations. The case report, therefore, sheds light on the possible mechanisms behind mesalazine adverse effects.

Case presentation

Mesalazine, or 5-aminosalicylic acid (5-ASA), is one of several 5-ASA-containing compounds, including sulfasalazine, to be used as first-line therapy for ulcerative colitis (UC). Several reports have pointed to varied adverse drug reactions associated with 5-ASA-containing compounds, the most frequent being fever and gastrointestinal reactions with an incidence of approximately 5%. The mechanism behind these is thought to be a drug-induced hypersensitivity syndrome, with desensitisation therapy successful in up to 90% of cases, according to a recent study.¹ Much rarer adverse effects include pericarditis, eosinophilic pneumonia, pancreatitis, interstitial nephritis and hepatotoxicity.^2–4 Six reports have described mesalazine-induced lupus syndromes with predominant arthritis symptoms.^5–10

We present the case of a 75-year-old man patient taking mesalazine for UC, diagnosed 1 year previously. Initial colonoscopy revealed erythematous, erosive pancolitis with no ileal involvement. Notable medical history included cholecystectomy and sigmoidectomy after covered perforated diverticulitis more than 10 years ago.

Five months before admission to our hospital, the patient was hospitalised at two other institutions for fever, dyspnoea, chills and nonradiating chest pain. Pneumonia was diagnosed on both episodes, with negative blood and pleural fluid cultures. He was discharged after antibiotic treatment, with resolved fever but persistent inflammatory syndrome. Two weeks before the first admission, he had been started on sulfasalazine 1 g three times per day, which resulted in normalisation of faecal calprotectin levels (from >1000 µg/g to 304 µg/g) and improvement of UC symptoms (chronic diarrhoea with occult blood, abdominal pain and bloating). This treatment was, however, stopped due to acute kidney failure before the first hospital admission. Ten days before the second admission, he had been started on oral mesalazine 2 g once daily in order to avoid new UC flare-ups; this was replaced on admission by rectal mesalazine 1 g once daily. One month after discharge from the second hospital stay, outpatient chest CT revealed a new 4 mm pericardial effusion (figure 1).

Figure 1.

Timeline of clinical and biological manifestations with sulfasalazine/mesalazine introduction and withdrawal.

On admission to our hospital, he presented the same aforementioned symptoms in addition to polyarthralgia and myalgia. Blood tests showed persistent inflammation. Chest and abdomen CT showed no infectious focus. Rectal mesalazine was replaced by oral mesalazine 1.5 g once daily, due to suspicion of an UC flare-up. Rectosigmoidoscopy revealed minimal inflammation. We sought gastroenterologist advice, on which rectal mesalazine 500 mg once daily was reinstated in addition to oral mesalazine 1.5 g once daily as long-term UC treatment.

Endocarditis was excluded by transthoracic and transoesophageal echocardiography, the former showing infracentimetric pericardial effusion. Extensive viral and bacterial infectious work-up remained negative.

Whole body positron emission tomography/computed tomography (PET/CT) with 18F-fluorodeoxyglucose (FDG) showed infracentimetric pericardial effusion with increased FDG uptake of the pericardium and mediastinal lymph nodes. Non-infectious pericarditis with PR segment depression in left chest ECG leads was retained. We stopped oral but maintained rectal mesalazine 500 mg once daily, expected to exert a more local than systemic effect, on advice of the gastroenterologist. C reactive protein (CRP) normalised but mild leukocytosis persisted; eosinophilia (2.58 G/L, 14.5% of leucocytes) disappeared, and PR segment depression became less pronounced.

Two weeks after discharge for initial clinical improvement, the patient presented again with dyspnoea but no fever, persistent inflammation (leucocytes 14 G/L, CRP 147 mg/L) and elevated brain natriuretic peptide (BNP) levels (1200 ng/L). Rectal mesalazine was then discontinued.

Outcome and follow-up

Four weeks later, the patient presented in better condition and blood tests showed normal leukocytes with no eosinophilia, and normal CRP and BNP levels. Complete mesalazine discontinuation also resulted in urine analysis normalisation, with isomorphic erythrocytes at 3 M/L and protein-to-creatinine spot ratio (p/c) at 36 mg/g (previously dysmorphic erythrocytes and p/c at 518 mg/g), suggesting mesalazine-induced nephritis.

PET/CT performed 8 weeks after the first exam showed disappearance of pericardial FDG uptake and effusion, and decreased size and uptake of mediastinal lymph nodes.

Immunologic workup showed non-specific antinuclear antibodies (ANA) with nucleolar pattern, already present a year prior at 1:5000 titre, unchanged. ANA analysis had been performed at the time for abdominal pruritus and pretibial petechiae. Antihistone antibodies were positive at low titers after mesalazine introduction, and were negative after complete mesalazine discontinuation with unchanged 1:5000 ANA titres.

Three months after discharge, at full remission and normal eosinophil levels, cytokine-lymphocyte transformation test (LTT) assays were performed according to Mayorga et al.¹¹ These measure cytokine release on in vitro drug exposure, are highly specific (95%) and more sensitive (80%) than LTT assays (50%), which measure lymphocyte propagation via tritium-labelled thymidine incorporation.¹¹ They provide an alternative to drug provocation tests, the gold standard for the diagnosis of drug-induced hypersensitivity, which, however, cannot be performed in cases such as drug reaction, eosinophilia, systemic symptoms (DRESS), Stevens-Johnson or toxic epidermal necrolysis. In our patient, cytokine-LTT revealed interleukin-5 production on exposure to acetylsalicylic acid, mesalazine, sulfasalazine or sulfamethoxazole. Interleukin-13 was also detected on sulfamethoxazole or mesalazine exposure, while granzyme B was detected only with the latter. The patient was not previously known for an atopic disorder or sulfa allergy.

Discussion

Positive cytokine-LTT assays argue for a mesalazine drug-induced hypersensitivity syndrome, assuming primosensitisation after exposure to sulfasalazine. Further, the positive tests to other compounds are compatible with a multiple drug hypersensitivity syndrome seen in DRESS. Interestingly, only two reports have described mesalazine-induced DRESS, occurring after prior exposure to sulfasalazine or sulfamethoxazole, and steroid therapy was necessary for clinical improvement.^{12 13}

Clinical manifestations and eosinophilia resolved after mesalazine discontinuation without recourse to steroids in our case. A score of 4 on the basis of RegiSCAR criteria points to probable DRESS (table 1). The peculiarity of this case is that typical DRESS exanthema and viral activations (EBV, CMV, HHV6) were absent, nonetheless not excluding this diagnosis.

Table 1.

RegiSCAR DRESS criteria (<2: excluded; 2–3: possible; 4–5: probable; > or = 6: definite)

RegiSCAR criteria	Score: 4 (probable)
Fever >38.5°	Present
Eosinophilia	Present +2 (>1.5 G/L)
Internal organ involvement	Heart, lungs, kidney +2
Enlarged lymph nodes	Present,<1 cm
Atypical lymphocytes	Absent
Skin rash	Absent
Resolution in > or = 15 days	Present
Evaluation of other potential causes	Positive ANA

ANA, antinuclear antibodies; DRESS, drug reaction, eosinophilia, systemic symptoms.

Lupus or lupus-like syndromes induced by mesalazine have been described with very limited clinical and varied biological features. Only one report, published in 1992, described mesalazine-induced, ANA-positive pericarditis, with no other systemic symptoms. Nonspecific ANA were detected at 1/300 titre and persisted at a lower titre (1/100) 2 months after drug discontinuation. No antihistone antibodies were detected.⁵ Five reports have described ANA-positive mesalazine-induced arthritis with no other systemic symptoms, in four of which antihistone or anti-DNA antibodies were detected, decreasing after drug withdrawal.^6–10

No report has, to this date, described mesalazine-induced nephritis in the context of a lupus syndrome with ANA. One report has described association of sulfasalazine with lupus nephropathy in two UC patients, but persistent clinical features long after sulfasalazine withdrawal led the authors to evoke an underlying association between UC and lupus nephropathy.¹⁴

N-acetyltransferase (NAT) 1 is involved in mesalazine and sulfasalazine metabolism, while NAT 2 metabolises sulfasalazine. The patient’s NAT1 (*4/*11) and NAT2 (*4/*5) genotypes were of unclear significance and associated with decreased activity, respectively.¹⁵ Current evidence, however, shows that NAT1 and NAT2 genotypes do not predict response or toxicity to mesalazine and sulfasalazine in UC patients.¹⁶ This is consistent with low-dose oral and rectal mesalazine, whose bioavailability is comparable (around 20%), being shown to cause peri/myocarditis.¹⁷ These findings are not surprising, given the immune nature of the adverse effect mechanisms described.

The original clinical presentation of our patient with pleuropericarditis, nephritis, polyarthralgia and myalgia with positive anti-histone antibodies, and its resolution on mesalazine withdrawal, are highly suggestive of drug-induced lupus. However, the strong sensitisation pattern with interleukin-5 production leads us to retain the diagnosis of drug-induced hypersensitivity syndrome with lupus manifestations, which has never been described before. It is important to note that although rare, severe adverse drug reactions to mesalazine can occur not only with systemic but also with rectal applications, and should not be underestimated in daily clinical practice.

Learning points.

• Mesalazine adverse events encompass multiple systemic symptoms with diverse mechanisms, including drug hypersensitivity reactions and, more rarely, drug-induced lupus.

• In our patient, antihistone antibodies and positive cytokine-lymphocyte transformation tests point to drug reaction, eosinophilia, systemic symptoms with lupus manifestations.

• Severe adverse events can resolve on drug cessation without recourse to steroids.

• Rectal mesalazine is associated with the same adverse events as oral mesalazine and should not be underestimated in clinical practice.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s)