Analysis of epidemiologic studies in temporal arteritis is hindered by the variability of diagnostic criteria and differences in population bases in the various studies. In general the incidence is relatively high in northern regions including the United States, Scandinavia, and European countries, whereas a much lower incidence is reported in southern European countries. Genetic factors are probably important in view of the predilection of temporal arteritis to occur in Caucasians and the association with the HLA-DR4 genotype, at least in some studies.
Baldursson and colleagues reported a total population-based study of the epidemiology of temporal arteritis in Iceland between 1984 and 1990 (Baldursson et al 1994). During the study interval the authors identified 133 confirmed cases of new temporal arteritis in Iceland; 94 were women and 39 men, giving a female-to-male ratio of 2.4. The age- and sex-adjusted incidence of temporal arteritis was calculated to be 27 in 100,000 per year overall, 36 in 100,000 per year for women, and 18 in 100,000 per year for men. As expected, the age incidence increased into the eighth decade of life.
The mean age at diagnosis was 71.9 years (range 50 to 97), with no significant difference between men and women. Only 19 patients (14.3%) had visual symptoms, of which only 1 developed permanent blindness of 1 eye, but there was information on visual symptoms in only 63.9% of the cases. Also, this was not a longitudinal study of the cohort at risk for visual loss, so this information probably cannot be taken as representative of the disease.
Machado and others at the Mayo Clinic published data indicating a rising incidence of temporal arteritis during the interval from 1950 to 1985 (Machado et al 1988). The overall incidence in Minnesota during that interval was lower than the Icelandic figures cited above, 17 in 100,000 per year, with age-adjusted rates 3 times higher for women (23.4 in 100,000 per year) than for men (7.4 in 100,000 per year). This is despite the similar ethnic mix in Iceland and Olmsted County, Minnesota, suggesting a role for environmental and other factors. The trend toward rising incidence over the interval was true only for women (35.4 in 100,000 per year in 1980 to 1985); the incidence for men dropped significantly (6.5 in 100,000 per year in 1980 to 1985), having reached a peak in the mid-1970s (15.4 in 100,000 per year in 1970 to 1974). The authors believe that this rise in incidence may be partly explained by changes in disease awareness among physicians and by other extraneous factors, but that a biological change is likely occurring as well. They point out the interesting fact that the preponderance of women increases in the older age groups, unlike other rheumatic and inflammatory diseases, where the predilection for women decreases with advancing age. Petersdottir and colleagues also found that the annual incidence increased significantly with time (P<0.001) for both men and women in a study of biopsy proven temporal arteritis between 1976 and 1995 in Goteborg, Sweden (Petursdottir et al 1999). In that study the average incidence was 22.2/100000 inhabitants over 50 years of age (women 29.8, men 12.5). Significant annual fluctuation was also found, with peak incidence in late winter and autumn. These seasonal fluctuations and the rising incidence suggested to the authors that an extrinsic triggering factor, such as infection may be involved. Gonzalez-Gay and colleagues found a rising annual incidence of biopsy-positive temporal arteritis in northwestern Spain between 1981 and 2005 but did not find peaks or valleys in the annual incidence or any seasonal variation in incidence (Gonzalez-Gay et al 2007). On the other hand, in Israel, Bas-Lando and colleagues found fluctuation in the annual incidence from 1980 to 2004 with 3 distinct peaks 8 to 10 years apart—evidence of a possible infectious etiology, the authors suggest (Bas-Lando et al 2007).
The highest incidence figures come from South Norway where 322 patients with either temporal arteritis or polymyalgia rheumatica were identified between 1987 and 1994. Two hundred fifty-six patients had polymyalgia rheumatica and 66 had temporal arteritis, so the majority of the high incidence was attributable to polymyalgia rheumatica, which the authors felt was due to the design of their study with higher ascertainment rates for polymyalgia rheumatica than previous studies. The annual incidence of temporal arteritis in persons aged 50 years or older was 29.0 per 100,000 (39.9 in women and 16.3 in men), figures that are fairly similar to earlier series.
In a case control study of 400 patients with giant cell arteritis, Duhaut and colleagues found that histories of smoking and previous atherosclerotic disease were significant "risk factors" for developing giant cell arteritis in women but not men (Duhaut et al 1998). Women who smoked had an overall 6-fold increased risk for arteritis compared with age- and sex-matched controls; the relative risk rose to 17-fold among heavy smokers. Duhaut and colleagues also carried out a case control study (285 incident female cases of temporal arteritis and 186 age-matched, population-based female controls) seeking to determine if alloimmunization during pregnancy might predispose to the disease later in life. These workers found, on the contrary, that the incidence of pregnancy was significantly lower in cases compared with controls "(nulliparous: 21.55% vs. 12.90%; and 4 or more pregnancies: 16.25% vs. 27.42%; Wilcoxon rank sum test: P = 0.0019)." The authors suggest that the hyperestrogenic state that accompanies pregnancy may have a prolonged protective effect on arteries. The transfusion histories were the same in the 2 groups, and there was no evidence that alloimmunization plays a role in the later development of temporal arteritis (Duhaut et al 1998). A case control study from Norway involving 398 patients with temporal arteritis or polymyalgia rheumatica and 1592 controls found no increased frequency of malignant tumors in patients versus controls (Myklebust et al 2002).
Studies of the clinical patterns and incidence of temporal arteritis in Spain and Israel prior to about 1980 and subsequent to that date have shown, in both countries, that the average age at presentation has risen along with the overall incidence. Also, the proportion of patients with "classic" symptoms has fallen, and atypical presentation has become more common. An attractive explanation of these changes is that the disease has become more widely recognized and is being diagnosed earlier in the course than prior to 20 years ago (Nesher et al 1999; Gonzalez-Gay et al 2001).
Despite the trend for the disease to occur only in the elderly, individual case reports document more or less typical temporal arteritis in young people. One of these rare individuals presented with a tender pulsatile aneurysm involving the frontal branch of the temporal artery on his forehead. He had a history of chronic asthma, juvenile rheumatoid arthritis, and positive serum ANA. Biopsy demonstrated lymphocytic infiltration of the temporal artery that was most intense at the internal elastic lamina, intimal hypertrophy, and sparse multinucleated giant cells (Pipinos et al 2006).