Saturday night fever: bizarre recurrence of fever attacks in a patient carrying a mutation in both the MEFV and TNFRSF1A genes.

Sirs, We read with interest the article by Granel et al. (1) who described an overlap syndrome between familial Mediterranean fever (FMF) and tumor necrosis factor (TNF)-receptor associated periodic syndrome (TRAPS) in a patient carrying a mutation in both the MEFV and TNFRSF1A genes, In his report Granel advocated the importance of a better evaluation of the role of the R92Q mutation in TNFRSF1A in patients with FMF and we would like to present some additional data. We report here another case of mutations in the MEFV and TNFRSF1A genes occurring together in a single patient. FMF is an autosomal recessive autoinfl ammatory disease caused by mutations in the gene MEFV and characterized by recurrent, self-limited febrile episodes with serositis, synovitis and, occasionally, skin involvement; attacks usually last less than 3 days (2). The diagnosis of FMF is based on clinical criteria, family history, and the patient’s response to colchicine; the demonstration of MEFV mutations is necessary to establish a defi nite diagnosis in suspected patients (3). Furthermore, genotyping may help to pin down the diagnosis in the case of atypical clinical signs and late onset (4). TRAPS is an autosomal dominant autoinfl ammatory condition caused by mutations in the gene TNFRSF1A encoding the 55-kD receptor for tumor necrosis factor-α (TNFα) and characterized by febrile episodes often lasting more than 7 days that can include abdominal pain, pleurisy, myalgia, conjunctivitis, periorbital oedema, skin rashes, and arthritis. Attacks usually last more than 7 days and fail to respond to colchicine, but are responsive to steroids (5). A 17-year-old female came under our observation for recurrent attacks of high fever (38o39.5°C) accompanied by diffuse abdominal pain, localized posterior thoracic wall pain, and arthritis with effusion in the right knee. She complained of weekly fl are-ups over the past 3 years, usually at the weekends, and mainly but not only during cold weather periods. She did not report any skin rashes and the duration of the symptoms was 1 to 2 days. The fever attacks were responsive to colchicine (administered for 40 days with rapid and dramatic improvement in her clinical status), but resistant to steroids and paracetamol. Circulating levels of serum amyloid A (SAA) (118.00 mg/L; normal value <10.0), interleukin-6 (IL-6) (42.2 pg/ml; n.v. <15) and interleukin-18 (IL-18) (384.98 pg/ml; n.v. <180.00) were elevated, while TNF-α (1.0 pg/ml; n.v. <15) and interleukin-1β (IL-1β) (1.5 pg/ml; n.v. <15) were normal. On the basis of the diagnostic criteria (3) the patient was suspected of having FMF and regular colchicine treatment was started. She was analysed for mutations in the MEFV, CIAS1 and TNFRSF1A genes. DNA analyses by PCR with specifi c primers, followed by sequencing on PCR products from both the patient and healthy controls identifi ed M694I and V726R mutations in MEFV and a low penetrance heterozygous mutation R92Q in TNFRSF1A (Fig. 1). She was diagnosed with an overlap syndrome between FMF and TRAPS and treatment with colchicine was continued at a dose of 1 mg daily, which brought about a resolution of the fever attacks and painful symptomology (thoracic, abdominal and articular pain) after 1 month and led to marked improvement in her pro-infl ammatory cytokine circulating levels: 2 months after starting treatment with colchicine, SAA levels were normal and IL-6 and IL-18 were signifi cantly decreased. A wide range of studies have shown that heterozygous carriers of MEFV mutations can be symptomatic and suffer from fever attacks and serositis (6, 7). Even asymptomatic MEFV heterozygotes have elevated acute phase proteins, such as CRP and SAA, compared to wild type subjects (8). Numerous studies demonstrate that the FMF phenotype is controlled by a number of factors: the MEFV itself, other genes, the patient’s sex, and hitherto undetermined populationspecifi c factors (6, 7). The R92Q mutation is described as a low penetrance mutation, and TRAPS patients with the R92Q mutation present less typical manifestations (9). FMF patients with a poor response to colchicine, such as the one Granel describes, have been reported even in the absence of TRAPS mutations (2). In our patient the duration of the fever attacks (no more than 2 days), the high levels of SAA, which are rarely associated with R92Q mutations, the responsiveness to colchicine, the monoarthritis, and the early onset of symptoms are all features supporting the FMF diagnosis (2), while the associated R92Q did not seem to infl uence the clinical presentation and management of the disease. It is also possible that symptoms associated with the R92Q mutation, and the subsequent modifi cation of the clinical picture, may develop later (5).