Regarded as the most common and best understood of the hereditary periodic fever syndromes, familial Mediterranean fever (FMF) is a recessively inherited disease of episodic fever with some combination of severe abdominal pain, pleurisy, arthritis, and a characteristic ankle rash. The flares typically last for up to 3 days at a time, and most patients are completely asymptomatic between attacks; if untreated with prophylactic colchicine, some patients later develop amyloidosis and renal failure. The recent cloning of the FMF gene on the short arm of chromosome 16p, and the subsequent finding that its tissue expression is limited to granulocytes, has helped to explain the dramatic accumulation of neutrophils at the symptomatic serosal sites; the wild-type gene likely acts as an upregulator of an anti-inflammatory molecule or as a downregulator of a pro-inflammatory molecule. For nearly half a century, FMF was thought to cluster primarily in non-Ashkenazi Jews, Arabs, Armenians, and Turks, although the screening of the 8 known mutations in an American cohort has identified substantial numbers of people from the Ashkenazi Jewish and Italian populations in the United States who also have this disease. Nevertheless, the symptoms often go unrecognized and patients remain undiagnosed for years, not receiving the highly efficacious colchicine therapy; their histories often include multiple laparotomies, laparoscopies, and psychiatric evaluations. The combinations of clinical manifestations among FMF patients are quite heterogeneous, but our American cohort did not establish any connections between individual mutations and specific clinical pictures--as is seen in other diseases like cystic fibrosis, in which distinct genotypes target certain organ systems. Specifically, the data from our American series are insufficient to evaluate the hypothesis that the M694V/M694V genotype confers a more severe phenotype, or increases the risk of amyloidosis; but both our data and the recent literature (160) indicate that amyloidosis can occur in FMF patients with only 1 copy, or no copies, of the M694V mutation. It appears that specific MEFV mutations are probably not the sole determinants of phenotype, and that unknown environmental factors or modifying genes act as accomplices in this disease. Although we hope the discovery of the FMF gene will allow the diagnosis of FMF to become genetically accurate, the reality is that both clinical and genetic tools must still be used together unless mutations are identified on both of a patient's chromosomes. Physicians should be careful not to rule out the diagnosis in patients of high-risk ethnic backgrounds just because of atypical clinical features, as our data indicate that MEFV mutations are sometimes demonstrable in such patients. At the same time, physicians cannot yet rely solely on a genetic diagnosis because we have not yet identified a sufficient spectrum of mutations, and it is not currently feasible to examine every patient's full DNA sequence for the entire gene; screening an ethnically consistent and clinically positive patient for the 8 known mutations frequently identifies a mutation on only 1 chromosome, and genetic analysis of other classic cases will often reveal none of the 8 mutations. Still, our data suggest that ethnic background is an important predictor of finding 1 of the presently known mutations, and the knowledge of ancestries atypical for FMF can suggest the diagnosis of other hereditary periodic fever syndromes. As the list of FMF-associated MEFV mutations is expanded, and/or new sequencing technologies permit more rapid screening, the value and interpretation of genetic testing for FMF will become more straightforward. Moreover, as the pathophysiology of this disorder becomes less of a hypothesis and more of an understood entity, it is likely that treatment options will broaden beyond the use of daily prophylactic colchicine. (ABSTRACT TRUNCATED)