Abstract
Familial Mediterranean fever (FMF) is the most common genetic autoinflammatory disease, but it has been considered a rare disease in Japan. We herein describe five patients with FMF who were diagnosed both clinically and genetically at a single Japanese institute. A genetic investigation of Mediterranean fever (MEFV) detected heterozygosity for the compound mutations L110P/E148Q (n=2) and L110P/148Q/P369S/R406Q (n=1), and heterozygosity for M694I (n=1) and S503C (n=1). Colchicine prevented febrile attacks and accompanying symptoms in four patients. One patient with an S503C mutation showed resistance. Physicians should be aware of the characteristic symptoms, as well as the more unusual symptoms such as headache, when diagnosing FMF.
Keywords: aseptic meningitis, familial Mediterranean fever, headache, Japan, periodic fever, inflammatory disease
Introduction
Familial Mediterranean fever (FMF) is the most common genetic autoinflammatory disease (1). FMF predominantly affects populations from the Mediterranean (2), but our nationwide survey identified a larger than expected number of such patients in Japan (3). The underlying cause of FMF is a mutation in Mediterranean fever (MEFV) (4). Genotype-phenotype correlations studied in Japan suggest that mutations within MEFV exon 10 are associated with a typical FMF phenotype, whereas mutations within exons 2 or 3 are associated with an incomplete FMF phenotype (5).
We herein describe our experience with five Japanese patients with FMF, who were diagnosed using clinical and genetic analyses in a single department of general internal medicine at a university hospital. All cases were diagnosed according to the Tel-Hashomer criteria (Livneh revision, 1997) (6). One patient with an MEFV exon 10 mutation presented with a typical FMF phenotype including periodic fever and abdominal pain. The remaining four patients with an incomplete FMF phenotype presented with uncommon FMF manifestations including headache. We discuss the importance of genotype-phenotype correlations in Japanese patients with FMF.
MEFV analysis
Genomic DNA was extracted from whole blood using the Promega WizardⓇ Genomic DNA Purification Kit (Madison, USA) according to the manufacturer's instructions. Polymerase chain reaction (PCR) was performed using forward and reverse primers for each MEFV exon, as described previously (5). Written informed consent for this case report and to perform genetic analyses was obtained from all patients.
Case Reports
Case 1
A 53-year-old Japanese male patient with periodic fever and bilateral ankle joint arthritis was referred to our department. He had experienced periodic fever attacks lasting 3 days, and had suffered bilateral ankle joint tenderness and swelling since his 20s. At the age of 51, he began to experience frequent febrile attacks (occurring within 3 days of fever, approximately every 3 weeks, with temperatures of around 38℃) with abdominal and anal pain and polyarthralgia. The bilateral knee and hip joints were also involved.
His physical examination was normal. His complete blood counts, liver function test, and urinalysis (no proteinuria) were also normal, but his serum creatinine level (Cr; 1.34 mg/dL) was slightly elevated. Use of a colon fiberscope found no abnormalities, including amyloid deposits, so a renal biopsy was not performed. An MEFV analysis detected a heterozygote M694I mutation in exon 10.
His abdominal pain was diagnosed as peritonitis, and his febrile attacks were consistent with typical FMF; they disappeared after colchicine treatment (1.0 mg/day).
Case 2
A 48-year-old Japanese female patient was taken to our hospital by ambulance with convulsions followed by a decreased level of consciousness. Since the age of 45, she had experienced periodic fever lasting 3 days with back pain, severe occipital headache with no signs of meningitis, and arthritis of the elbow and knee joints every month.
On admission, she was unconscious and had neck stiffness that indicated the possibility of meningitis. Brain magnetic resonance imaging (MRI) and an electroencephalogram showed no abnormal findings. A cerebrospinal fluid analysis revealed an opening pressure of 24 cm H2O, an elevated protein concentration (154 mg/dL), but cellularity (12/mm3) and glucose (79 mg/dL, serum glucose 167 mg/dL) within normal ranges. No microbes were detected by Gram staining, acid-fast staining, ink staining, cultivation, or PCR for acid-fast bacilli and Mycobacterium tuberculosis. Viral antigen tests for herpes simples virus (HSV)-1, HSV-2, and varicella zoster virus (VZV) were also negative. These findings were consistent with aseptic meningitis. She was referred to our department from the Department of Neurology for further examination for periodic fever.
Laboratory tests were normal except for an elevated white blood cell count (WBC; 12,530 /μL) and C-reactive protein (CRP; 9.32 mg/dL). An MEFV analysis showed a compound heterozygous mutation (L110P/E148Q) in exon 2. She was diagnosed with an incomplete type of FMF and was treated with colchicine (0.5 mg/day), which eradicated her febrile attacks and headache.
Case 3
A 40-year-old male had periodic fever, polyarthralgia, headache, and abdominal and back pain. He had suffered a recurrent high fever (>38℃) since he was 28 years old. His periodic fever lasted for 4 to 7 days and occurred at a frequency of once every 2 to 3 months. Each febrile attack caused a headache in the occipital region together with posterior cervical spine pain, but there were no signs of meningitis. His bilateral wrist and ankle joints showed severe inflammatory changes, such as pain on exercise, swelling, and warmth. Slight swelling of the elbow joints was also recognized.
His physical examination was unremarkable. Laboratory tests were normal, except for an elevated WBC (10,930 /μL), CRP (1.27 mg/dL), and serum amyloid A (SAA; 189.7 μg/mL). An MEFV analysis revealed a heterozygous mutation in exon 2 (L110P/E148Q). Both abdominal and back pain were diagnosed as peritonitis and pleuritis, respectively. He was diagnosed with an incomplete type of FMF. He was treated with colchicine (1.0 mg/day), which led to the disappearance of the febrile attacks.
Case 4
A 25-year-old male presented with high fever with bilateral pleuritic chest pain, respiratory distress, polyarthralgia, and occipital headache. His febrile attacks occurred every 2 weeks and involved a fever of 38.0 to 38.5℃ which lasted 3 days. The febrile attacks were always accompanied by headache with no meningeal signs. Physical examination was unremarkable except for swelling of the bilateral knee, hip, and ankle joints.
Laboratory tests revealed the presence of high acute phase reactants (CRP, 10.64 mg/dL; SAA, 844.4 μg/mL) and leukocytosis (WBC, 30,400 /μL) during the febrile attacks. Other laboratory data were normal. An MEFV analysis revealed heterozygous mutations in exon 2 (L110P/E148Q) and exon 3 (P369S/R408Q).
Chest symptoms were diagnosed as pleuritis. He was diagnosed with an incomplete type of FMF and was treated with colchicine (0.5 mg/day), which promptly reduced the febrile attacks. Complete elimination of the attacks was achieved by increasing the colchicine dose to 1.5 mg/day.
Case 5
A 47-year-old male was referred to our department with a fever of no apparent cause. He had recognized periodic fever (>40℃) lasting for 3 days every 2 months since the age of 25. At the age of 42, the frequency of the periodic fever increased (to an interval of almost every 10 days), and was accompanied by occipital headache, nausea, diarrhea, abdominal pain, and polyarthralgia. Arthralgia was detected in the bilateral knee, hip, and elbow joints. Nausea also occurred occasionally, and there were no signs of meningitis.
He was admitted to another hospital at the age of 47, and was examined by enhanced computed tomography, gallium scintigraphy, and upper and lower gastrointestinal fiberscope. However, no abnormality was detected. Therefore, he was referred to our department for further examination. Physical examination was unremarkable. He had high acute phase reactants (CRP, 1.97 mg/dL; SAA, 24.5 μg/mL) and leukocytosis (WBC, 14,600 /μL) during the febrile attacks, whereas other laboratory data including urinalysis revealed no abnormal findings. An MEFV analysis showed a heterozygote S503C mutation in exon 5. Because his recurrent fever occurred with or without nausea, he was diagnosed with an incomplete type of FMF. He was treated with colchicine, which markedly reduced the number of febrile attacks but did not completely eliminate them. The dose of colchicine was gradually increased to 3.5 mg/day.
The clinical features of the five patients with FMF are summarized in Table 1. The mean age at the time of diagnosis was 42.6 years, and the mean period between disease onset and disease diagnosis was 14.0 years. No patient had a family history suggesting FMF, nor were they children of consanguineous marriages. MEFV mutations were detected in all patients, with a heterozygote mutation in two patients and compound heterozygote mutations in three. These mutations included: M694I (exon 10), E148Q and L110P (exon 2), P369S and R408Q (exon 3), and S503C (exon 5). MEFV genetic analysis was approved by the Ethics Committee of Nagasaki Medical Center (No. 21003, 2009). A diagnosis of FMF was made when the patient had one or more of the major criteria, or two or more of the minor Tel-Hashomer criteria (6). The diagnostic criteria of FMF in five patients are shown in Table 2. Based on these criteria, the patients were divided into two groups: typical FMF (n=1) and incomplete FMF (n=4). The average period from pretreatment to posttreatment was 46 days.
Table 1.
Demographic and Clinical Features of the Five FMF Patients.
| Patient | Sex | Age (yr) | Onset (yr) | Delay (yr) | Febrile attack | Accompany symptoms of FMF | MEFV mutation | Pretreatment | Posttreatment | Colchicine (mg/day) | Treatment resistance | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Fre (mo) | Dur (day) | Serositis | Arthritis | Headache | WBC (/μL) | CRP (mg/dL) | SAA (μg/dL) | WBC (/μL) | CRP (mg/dL) | SAA (μg/dL) | ||||||||
| 1 | M | 53 | 20 | 33 | <1 | 3 | Peritonitis | poly | - | exon10 (M694I/-) | 7,270 | 5.72 | N.A | 6,410 | 0.17 | <10.0 | 1.0 | - |
| 2 | F | 48 | 45 | 3 | 1 | 3 | Pleuritis | poly | + Neurological disorder | exon2 (L110P/ E148Q) | 12,530 | 9.32 | N.A | 5,800 | 0.14 | N.A | 0.5 | - |
| 3 | M | 40 | 28 | 12 | 2-3 | 4-7 | Peritonitis Pleuritis | poly | + | exon2 (L110P/ E148Q) | 10,930 | 1.27 | 189.7 | 6,480 | 0.10 | <10.0 | 1.0 | - |
| 4 | M | 25 | 25 | 0.2 | 0.5 | 3 | Pleuritis | poly | + | exon2 (L110P/ E148Q) exon3 (P369S/ R408Q) | 30,400 | 10.64 | 844.4 | 4,610 | 0.02 | <10.0 | 1.5 | - |
| 5 | M | 47 | 25 | 22 | 2 | 3 | Peritonitis | poly | + | exon5 (S503C/-) | 14,660 | 1.97 | 24.5 | 12,030 | 0.18 | <10.0 | 3.5 | + |
Age: Age at diagnosis, Onset: Age at first time of FMF clinical feature, Delay: Delay in diagnosis of FMF, Fre: Frequency, Dur: Duration
Table 2.
Criteria for the Diagnosis of FMF in All Five Patients.
| Case | 1 | 2 | 3 | 4 | 5 | |||
|---|---|---|---|---|---|---|---|---|
| Major Criteria | ||||||||
| Typical attacks of: | ||||||||
| 1 | periotonitis (generalized) | + | - | - | - | - | ||
| 2 | Pleurisy (unilateral) or Pericarditis | - | - | - | - | - | ||
| 3 | Monoarthritis (Hip, Knee, or Ankel joint) | - | - | - | - | - | ||
| 4 | Fever alone | - | - | - | - | - | ||
| 5 | Incomplete abdominal attack | - | - | - | - | |||
| Minor Criteria | ||||||||
| Incomplete attacks involving either or both of the following sites | ||||||||
| 1 | Chest | - | + | + | + | + | ||
| 2 | joint | - | + | + | + | + | ||
| 3 | Exertional leg pain | - | - | - | - | - | ||
| 4 | Response to colchicine | + | + | + | + | - | ||
| Type of FMF | Complete | Incomplete | Incomplete | Incomplete | Incomplete | |||
Definitive diagnosis: 1 major criterion or 2 minor criteria.
Typical attacks are defined as recurrent (>/=3 of the same type), febrile (rectal temperature of 38°C (100°F) or higher), and short (lasting between 12 hours and 3 days).
Incomplete attacks are defined as painful and recurrent, differing from typical attacks in 1 or 2 features, as follows:
The temperature is normal or lower than 38°C (100°F).
The attacks are longer or shorter than specified (but no shorter than 6 hours or longer than a week).
No signs of peritonitis are recorded during the abdominal attacks.
The abdominal attacks are localized.
The arthritis is in joints other than those specified (hip, knee, or ankle).
All patients suffered from self-limited, recurrent bouts of fever with a period varying from bi-weekly to quarterly. In addition to periodic fever, three had abdominal pain suggesting the existence of peritonitis, two had chest and/or back pain indicative of pleuritis, five had clinically diagnosed polyarthralgia, and four had headache. One patient experienced convulsions and reduced consciousness upon febrile attacks, leading to a diagnosis of aseptic meningitis. Colchicine resistance was shown in one patient.
Elevated levels of CRP were noted in all patients, and leukocytosis was observed in four during febrile attacks; we did not measure the levels of other inflammatory cytokines. No clinical features of amyloidosis such as cardiomegaly, hepatomegaly, or proteinuria were found in any cases. Upper gastrointestinal endoscopy and total colonoscopy were performed in all cases, and no evidence of amyloidosis was found by either observation or biopsy.
Discussion
FMF is a disease characterized by periodic fever, serositis, and synovitis (1). In the present study, we identified five Japanese patients with FMF. According to the commonly used Tel-Hashomer criteria (6), one of these patients carrying an MEFV exon 10 mutation was diagnosed with typical FMF, while the remaining four patients with MEFV non-exon 10 mutations were diagnosed with incomplete FMF. No laboratory tests specific for FMF are available (7); however, all five patients showed increased acute phase reactants, such as CRP, during their febrile episodes. A complete response to colchicine for preventing febrile attacks was demonstrated in all five patients, although the colchicine response was slightly reduced during the 2-year treatment in one patient with the MEFV S503C exon 5 mutation. Clinical manifestations and courses of FMF patients with the S503C mutation have rarely been reported (8), so further investigations are necessary to identify the clinical characteristics and therapeutic responses of FMF patients with MEFV exon 5 mutations including S503C.
One of the most striking features of the patients in our report is the advanced age of their FMF diagnosis. The typical interval between the first FMF symptoms and the start of colchicine treatment was around 12 years, during which time uncontrolled inflammation might have caused amyloid A (AA) amyloidosis (9). This is the most important factor determining the prognosis of FMF. However, although both the diagnosis and therapeutic intervention were delayed in our patients, no clinical symptoms or findings suggestive of AA amyloidosis were identified. Indeed, our survey of Japanese FMF patients reported frequencies of AA amyloidosis of 3.6% (10), which is lower than that seen in Mediterranean populations (11). We did not observe MEFV genotypes including M694V, the risk allele for amyloidosis (12), in our five Japanese cases of FMF, which may have partly contributed to the protective effects against amyloidosis.
Another observation of our case report suggests that MEFV non-exon 10 polymorphisms may be responsible for or associated with additional clinical manifestations that do not meet the typical FMF symptoms (13, 14), including aseptic meningitis. Various nervous system findings have been reported in the course of FMF; however, these findings are rare and have usually been documented as case reports. Neurological involvements associated with FMF include optic neuritis, demyelinating lesions, posterior reversible encephalopathy syndrome, and recurrent aseptic neuritis (15-17). Headaches were common in the current case report, occurring in four out of five patients, all of whom had MEFV non-exon 10 mutations and an incomplete FMF phenotype. Because colchicine treatment eradicated the headache in these patients in addition to the febrile attacks, headache could therefore be a symptom of recurrent aseptic meningitis.
The main limitation associated with the current case report was the small number of patients from a single institute and thus some possible bias, such as with patient selection. Additionally, we did not perform either a cerebrospinal fluid analysis or brain MRI in any of the patients. Furthermore, it is conceivable that the coexistence of headache and MEFV non-exon 10 polymorphisms may have been incidental. However, the headaches all resolved after colchicine treatment, indicating that they could have been a true clinical manifestation of FMF.
Recently, more case reports of FMF patients have been published in Japan. Ogita et al. (18) recently reported seven Japanese cases of FMF which all carried MEFV mutations and responded well to colchicine treatment. Kinoshita et al. (19) presented a case of Mollaret meningitis with an MEFV mutation, which showed a similar clinical course to our case 2. The authors suggested that FMF should be included in the differential diagnosis of recurrent meningitis.
In conclusion, we herein described our clinical evaluation of Japanese patients with FMF. Although the central nervous system is rarely affected in patients with FMF, neurological involvement such as aseptic meningitis is possible in those with an incomplete FMF phenotype. In our cases, most patients were controlled by colchicine treatment; therefore, early diagnosis and therapeutic intervention are warranted even in Japanese patients. This treatable disease should not be left undiagnosed because severe complications, such as amyloidosis, can sometimes occur. Therefore, special efforts should be made to diagnose FMF, even for uncommon ethnic backgrounds including Japan.
The authors state that they have no Conflict of Interest (COI).
References
- 1.Samuels J, Aksentijevich I, Torosyan Y, et al. Familial Mediterranean fever at the millennium. Clinical spectrum, ancient mutations, and a survey of 100 American referrals to the National Institutes of Health. Medicine (Baltimore) 77: 268-297, 1998. [DOI] [PubMed] [Google Scholar]
- 2.El-Shanti H, Majeed HA, El-Khateeb M. Familial Mediterranean fever in Arabs. Lancet 367: 1016-1024, 2006. [DOI] [PubMed] [Google Scholar]
- 3.Migita K, Izumi Y, Jiuchi Y, et al. Familial Mediterranean fever is no longer a rare disease in Japan. Arthritis Res Ther 18: 175, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.French FMF Consortium. A candidate gene for familial Mediterranean fever. Nat Genet 17: 25-31, 1997. [DOI] [PubMed] [Google Scholar]
- 5.Migita K, Agematsu K, Yazaki M, et al. Familial Mediterranean fever: genotype-phenotype correlations in Japanese patients. Medicine (Baltimore) 93: 158-164, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Livneh A, Langevitz P, Zemer D, et al. Criteria for the diagnosis of familial Mediterranean fever. Arthritis Rheum 40: 1879-1885, 1997. [DOI] [PubMed] [Google Scholar]
- 7.Ozen S, Demirkaya E, Erer B, et al. EULAR recommendations for the management of familial Mediterranean fever. Ann Rheum Dis 75: 644-651, 2016. [DOI] [PubMed] [Google Scholar]
- 8.Fujikura K. Global epidemiology of Familial Mediterranean fever mutations using population exome sequences. Mol Genet Genomic Med 3: 272-282, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Grateau G. The relation between familial Mediterranean fever and amyloidosis. Curr Opin Rheumatol 12: 61-64, 2000. [DOI] [PubMed] [Google Scholar]
- 10.Migita K, Uehara R, Nakamura Y, et al. Familial Mediterranean fever in Japan. Medicine (Baltimore) 91: 337-343, 2012. [DOI] [PubMed] [Google Scholar]
- 11.Kishida D, Nakamura A, Yazaki M, Tsuchiya-Suzuki A, Matsuda M, Ikeda S. Genotype-phenotype correlation in Japanese patients with familial Mediterranean fever: differences in genotype and clinical features between Japanese and Mediterranean populations. Arthritis Res Ther 16: 439, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ben-Chetrit E. Familial Mediterranean fever (FMF) and renal AA amyloidosis-phenotype-genotype correlation, treatment and prognosis. J Nephrol 16: 431-434, 2003. [PubMed] [Google Scholar]
- 13.Ryan JG, Masters SL, Booty MG, et al. Clinical features and functional significance of the P369S/R408Q variant in pyrin, the familial Mediterranean fever protein. Ann Rheum Dis 69: 1383-1388, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ben-Chetrit E, Peleg H, Aamar S, Heyman SN. The spectrum of MEFV clinical presentations-is it familial Mediterranean fever only? Rheumatology (Oxford) 48: 1455-1459, 2009. [DOI] [PubMed] [Google Scholar]
- 15.Schwabe AD, Monroe JB. Meningitis in familial Mediterranean fever. Am J Med 85: 715-717, 1988. [DOI] [PubMed] [Google Scholar]
- 16.Topçuoğlu MA, Karabudak R. Familial Mediterranean fever and multiple sclerosis. J Neurol 244: 510-514, 1997. [DOI] [PubMed] [Google Scholar]
- 17.Kalyoncu U, Eker A, Oguz KK, et al. Familial Mediterranean fever and central nervous system involvement: a case series. Medicine (Baltimore) 89: 75-84, 2010. [DOI] [PubMed] [Google Scholar]
- 18.Ogita C, Matsui K, Kisida D, et al. A retrospective analysis of 7 cases of familial Mediterranean fever. Jpn J Clin Immunol 40: 21-27, 2017. [DOI] [PubMed] [Google Scholar]
- 19.Kinoshita T, Matsushima A, Satoh S, et al. A case of colchicine-responsive Mollaret's meningitis with MEFV gene mutation. Clin Neurol 54: 124-129, 2014. [DOI] [PubMed] [Google Scholar]
