Co-existence of Hashimoto's thyroiditis with familial Mediterranean fever: is there a pathophysiological association between the two diseases?

Abstract

Familial Mediterranean fever is an autosomal recessive disease characterized by periodic attacks of fever and polyserositis, while Hashimoto's thyroiditis is the most common cause of hypothyroidism. We suggest that common autoimmune mechanisms may underlie both disorders, describe their clinical co-existence in a patient, and discuss a possible causal link between them.

Introduction

Familial Mediterranean fever (FMF) is also called recurrent polyserositis. The salient features of this disease include brief recurrent episodes of peritonitis, pleuritis and arthritis, which are usually associated with fever [1]. There are reports that FMF may be an autoimmune disorder [2,3]. Hashimoto's thyroiditis (HT) is part of a spectrum of autoimmune thyroid diseases [4], characterized by the destruction of thyroid cells by various cell- and antibody-mediated immune processes [5]. In view of the autoimmune features common to FMF and HT, we feel it reasonable to consider the possibility of a pathophysiological association between them.

Index case

A 21-year-old woman was admitted to the Internal Medicine Outpatient Clinic, Dumlupinar University School of Medicine, with a history of intermittent abdominal pain, fever and weakness. These symptoms had been recurring monthly or bi-monthly for approximately 1 year. The patient's history included reports of high levels of thyroid-stimulating hormone (TSH) for 16 months, and treatment with levothyroxine for about 10 months. On admission, the patient stated that she had not been on any therapy for 2 months. The physical findings were as follows: hypotension (90/60 mmHg), tachycardia (110 beats/min), fever (38·2°C) and general abdominal tenderness. Laboratory studies revealed elevated levels of C-reactive protein, fibrinogen and TSH, and an erithrocyte sedimentation rate (Tables 1 and 2). Abdominal ultrasonography and direct abdominal X-rays were normal. The patient was placed on colchicine, 3 mg/day, and her symptoms subsided.

Table 1.

Laboratory findings at admission of patient.

Haematology		Blood chemistry				Urine analysis
WBC (mm3)	9·30	Glucose (mg/dl)	68	TP (mg/dl)	7·3	Glucose	(–)
RBC (mm3)	5·20	CRP (mg/l)	43·99	Albumin (mg/dl)	4·2	Protein	(–)
Htc (%)	41·8	Na (mEq/l)	140	BUN (mg/dl)	17	OB	(–)
Hb (g/dl)	14·0	K (mEq/l)	4·7	Creatinine (mg/dl)	0·8	2–3 WBC
Plt (mm3)	326	Cl (mEq/l)	99	AST (IU/l)	17	1–2 RBC
Sedimentation	50	Ca (mg/dl)	9·4	ALT (IU/l)	14
Fibrinogen (g/l)	235	P (mg/dl)	4·5	LDH (IU/l)	113

WBC, white blood cells; RBC, red blood cells; OB, occult blood; CRP, C-reactive protein; TP, total protein; BUN, blood urea nitrogen. Reference range: sedimentation: 0–20 mm/h; fibrinogen: 1·5–4·0 g/l.

Table 2.

Analysis of thyroid hormone and antibody and serological analysis.

Hormone analysis		Thyroid antibodies		Serological analysis
TSH (UıU/ml)	12·77	Anti-TPO (IU/ml)	36·4 (< 2·1)		Pretreatment	Post-treatment
FT3 (pg/ml)	1·37	Anti-T (mIu/ml)	14·9 (0–4·9)	CRP (mg/l)	43·99	5·34
FT4 (ng/ml)	0·98	TRab (IU/ml)	0·01 (< 0·15)	RF (IU/ml)	6·12

TSH, thyroid-stimulating hormone; TPO, anti-thyroid peroxidase; CRP, C-reactive protein; RF, romatoid factor; FT, free thyroxine; Trab, thyrotropin receptor antibodies.

The patient's symptoms, signs, clinical and laboratory findings were consistent with FMF. DNA amplification analysis revealed a homozygous M694V mutation in the MEFV (Mediterranean fever) gene, confirming the diagnosis.

Additional studies were undertaken to determine the reason for the elevation in TSH. The results showed that free thyroxine (FT₄) and triiodothyronine (FT₃) were within normal limits (Table 2) but the titres of thyroid antibodies (anti-thyroid peroxidase (anti-TPO) and anti-thyroglobulin (anti-TG) were high, and ultrasonography of the thyroid gland revealed slight heterogenity (Fig. 1). These findings were consistent with HT.

Fig. 1.

Thyroid ultrasonography shows slight heterogeneity in the thyroid gland.

Discussion

The FMF is an inherited multi-system disease with recurrent painful attacks affecting the abdomen, chest or joints, and is often accompanied by fever and sometimes a skin rash [6–8]. It is the most common and best-understood of the hereditary periodic fever syndromes, and is inherited in an autosomal recessive fashion [6–9]. FMF is caused by a mutation in the MEFVgene, which is located on the short arm of chromosome 16 [9]. The disorder is most prevalent in people of Armenian, Sephardic Jewish, Levantine Arabic and Turkish ancestry [7–10].

The FMF could be described as a disorder of inappropriate inflammation. In other words, an event that in a normal situation might cause a mild or unnoticeable inflammation may lead to a severe inflammatory response in someone with FMF. Episodes of FMF are associated with inflammation of sheets of serous membranes covering the organs inside the abdominal cavity (peritonitis), the chest cavity (pleurisy) and joints (arthritis) [6]. During an attack, large numbers of neutrophils migrate into the affected area and cause painful inflammation and fever. The MEFV gene encodes for a protein called pyrin (so-named because of the predominance of fever), which is designated as marenostrin by the French FMF Consortium [6,10]. This protein is expressed mainly in neutrophils [11] in the serosal linings of the peritoneal, pleural and synovial spaces, i.e. in areas of the body most at risk of FMF-related symptoms. Pyrin may be an inhibitor of the C5a chemotactic factor [12] and, perhaps, of interleukin (IL)-8 [13] and suppressor T cells [14]. Accordingly, individuals with normal pyrin may be able to deactivate the target chemical factor when it is produced in response to an inflammatory stimulus. Some studies hold that FMF may be a systemic autoinflammatory disorder [2,15–17]. Several studies have reported the co-existence of FMF with other autoimmune disorders, such as the overlap syndrome, Behçet's syndrome and polyarteritis nodosa [18–21].

The HT is the most common type of thyroiditis elicited by an aggressive and destructive autoimmune attack. Two types of HT have been identified: an atrophic variety, perhaps associated with human leucocyte antigen D-related 3 (HLA-DR3) gene inheritance, and a goitrous form associated with HLA-DR5. Studies of autoimmune hypothyroidism in monozygotic twins have shown that the concordance rate is low [22]. Non-major histocompatibility complex class II genes have been been implicated recently in susceptibility to HT. Data that have accumulated to date suggest an association between cytotoxic T cell antigen-4 – a major negative regulator of T cell-mediated immune functions – and several autoimmune diseases, including HT. New studies have appeared that bear on the zinc-finger gene in autoimmune thyroid disease (AITD), the susceptibility region gene (ZFAT), the TG gene and the protein tyrosine phosphatase-22 (PTPN22) gene. In addition, most patients with HT have high anti-TG and anti-TPO titres [23].

The main threat posed by the autoinflammatory diseases is that they overlap with, or mimic, autoimmune diseases [24]. Several studies have shown that FMF and HT share certain inflammation markers, such as tumour necrosis factor (TNF)-α and interferon (IFN)-γ[25–31]. These studies have demonstrated that in vitro stimulation of peripheral blood mononuclear cell with both proinflammatory cytokines upregulated the expression of MEFV[28,29,31]. In another study, T cells co-expressing IFN-γ and TNF-α were found in significantly higher frequencies in HT patients with high TPO antibody titres than in healthy donors. Therefore, the consensus was that these factors are probably responsible for thyroid cell damage and/or death in HT [32]. In addition, other studies have noted significant cytokine (IL-6, IL-10, IL-12, IL 18) elevations in FMF and HT [33–39].

Conclusions

Our observations suggest that similar pathophysiological mechanisms underlie FMF and HT, and that cytokine expression in FMF may provoke an autoimmune response that may set the stage for disorders such as HT. Because FMF and HT may be mediated by shared cytokines, it seems reasonable to suggest that further research is needed to ascertain whether, in fact, there is a causal association between FMF and HT.

Disclosure

There is not any conflict interest or financial problem among the authors.