We analyzed the appearance of individual manifestations chronologically including the prediagnostic phase (Table 3, Figure 1). Oral ulcer, the most common initial symptom, was found a mean of 7.5 ± 10.2 years before the diagnosis. Analysis of the cumulative rate of appearance of individual symptoms more clearly showed that oral ulcer appeared as a prediagnostic manifestation in the majority of patients. Of the 412 patients, 14% had suffered from oral ulcers for more than 20 years before being diagnosed with BD. Although other symptoms also preceded the diagnosis, the incidence of genital ulcer, skin symptoms, and eye involvement rapidly increased near the "zero point," representing the time of diagnosis. The findings indicate that these symptoms started to develop just before the diagnosis in the majority of patients. On the contrary, serious organ lesions of the CNS and the gastrointestinal and vascular systems developed later in most cases, although the lesions were found as the initial symptoms in some patients. These findings were compatible with previous observations.
We focused on patients with serious forms of manifestations, because the disease has a broad spectrum of clinical diversity and severity. First, we compared patients with eye involvement, a serious manifestation among the typical symptoms in BD, and those without eye involvement. Male sex and HLA-B51 were more frequent in patients with eye involvement than in those without (Table 4). CNS involvement was significantly associated with ocular lesions, while the other extraocular manifestations were less frequent in patients with ocular lesions than in patients without ocular lesions. These findings were consistent with previous reports.8,13,28
We also examined patients who were categorized as having the special types of BD-the neurologic, vascular, and intestinal types, all of which can be lethal or leave irreversible deficits. Because these symptoms developed later in general, as shown in Figure 1, we examined whether some of the major symptoms were associated with the development of the special types of BD. There was no difference in the frequencies of the 4 major symptoms between the patients with the special types and the other patients (Table 5). The number of the 4 major symptoms as initial manifestations showed no variation. However, analyses of the individual special types revealed that eye involvement occurred more frequently in the neurologic special type, and less frequently in the gastrointestinal and vascular special types (see Table 4).
It has been suggested that the severity of BD is becoming milder than previously found in patients in Japan. Therefore, we investigated whether clinical phenotypes have altered or not. First, we compared the frequency of individual symptoms between the patients diagnosed before 2000 ("previous group") and those diagnosed after 2000 ("recent group") (Table 6). There was no difference except for a higher incidence of ocular lesions in the previous group than the recent group. However, it is hard to compare the raw data of the 2 groups directly, because the observation period was significantly longer in the previous group (15.3 ± 9.5 yr) than in the recent group (3.3 ± 2.4 yr). Additional symptoms may have developed later, after completion of the study, especially in the patients with short observation periods, as shown in Figure 1. Therefore, we adjusted the observation period in the previous group to the first 3.3 ± 0.7 years (previous group adjusted is group B in Table 6). The difference in the frequency of eye involvement vanished after the adjustment. Instead, arthritis, gastrointestinal involvement, and vascular involvement were more often found in the recent group during the comparable observation periods.
We also compared disease phenotypes based on combination patterns of the 4 major symptoms (oral ulcer, genital ulcer, eye involvement, and skin involvement) between the patients diagnosed before 2000 (previous group) and those diagnosed after 2000 (recent group) (Table 7). Patients with complete BD, who had all 4 major symptoms, were the most common subtype in the previous group, while the frequency was significantly decreased in the recent group (previous group, 40% vs. recent group, 23%; p < 0.01). On the other hand, the combination of oral ulcer, genital ulcer, and skin involvement was the leading subtype in the recent group (36%). There was a significant increase in the number of patients with oral ulcer and eye involvement (previous group, 3% vs. recent group, 8%; p < 0.05). However, these differences disappeared when the observation period was adjusted, except for a higher frequency of complete BD in the previous group.
Although the proportion of patients who met the ISG criteria3 was higher in the previous group (93%) than in the recent group (83%), that difference was also abrogated by adjustment of the observation period in the previous group (85%). The ISG criteria-based diagnosis was made after the adjusted observation period in 23 patients (8%) in the previous group. Although the findings suggest that the Japanese criteria20 contribute to earlier diagnosis of the disease, it is difficult to compare diagnostic performance between the ISG criteria and the Japanese criteria in the current study, in which the pathergy test was conducted in only 120 of 412 patients (29%), and a positive test was found in 54 patients, 50% in male, and 42% in female.
We investigated the appearance of individual symptoms chronologically and cumulatively, including the prediagnostic phase, in 412 patients with BD who visited 2 university hospitals over 16.5 years. All patients were Japanese except 3 (1 Chinese, 1 Korean, and 1 Syrian).
The frequencies of individual clinical features reported in patients with BD vary among the different studies. These variations result from differences in ethnic groups, geographic regions, and study designs, including the diagnostic criteria used. Sources of the subjects are also important when the data are interpreted. In general, hospital-based studies, especially those conducted by a single department, tend to recruit patients with severe disease manifestations, such as inflammatory eye disease, CNS involvement, and vascular involvement,24 leading to selection bias of the patients. In contrast, population-based studies include patients with mild symptoms, characterized mainly by recurrent bouts of aphthous ulcerations and skin lesions.6 To minimize the selection bias that accompanies hospital-based studies, we designed the current study to enroll all BD patients who visited any hospital department, including ophthalmology, dermatology, neurology, gastroenterology, and rheumatology. Indeed, the frequencies of individual clinical features found in the current study were comparable to those reported in previous population-based epidemiologic studies in Japan,21,25 except for slightly lower frequencies of arthritis and gastrointestinal involvement than in previous studies.14
Hospital-based studies provide a unique opportunity to review the full spectrum of disease manifestations,24 although there are several limitations because of the retrospective observation. In this study in particular, it is worthwhile to analyze the clinical manifestations that appeared before the diagnosis was established. Clinical courses in the prediagnostic phase are suggestive for early diagnosis of the disease. Most patients had presented with any of the BD symptoms for more than 8 years on average, until the diagnosis was established based on the Japanese criteria.20 Notably, some patients had been suffering from oral aphthosis, which is the most frequent initial symptom, for more than 20 years before the diagnosis. On the other hand, the present study also showed that oral ulcers were absent in 30% of the patients at the disease onset, although almost all of the patients experienced the symptoms during the entire clinical course. The findings are consistent with those in previous hospital-based studies, which reported that 19%-27% of patients with BD did not present with oral ulcers at disease onset.26 These patients having no oral ulcers were never diagnosed as having BD based on the ISG criteria,3 in which oral ulcers are considered a prerequisite for the diagnosis of BD. Indeed, as shown in Table 7, 23 of 279 (8%) patients in the previous group satisfied the ISG criteria more than 3.3 years after satisfying the Japanese criteria for the diagnosis. The ISG criteria are not always satisfactory for detecting the early phase of BD in daily clinical practice,3 although they are very useful for classifying established BD patients for research purposes.
Nevertheless, it may be precocious to emphasize the superiority of the Japanese criteria to the ISG criteria concerning early diagnosis of BD based on the current study, since the pathergy test was conducted in only 120 of 412 (29%) patients. The Japanese criteria do not include the test as an essential item, but the test is listed as a reference. This is the most important reason for the low frequency of pathergy tests for BD patients in Japan.
Although prevalence rates of the positive pathergy test in patients with BD vary among studies,30 a number of studies have shown a decline in the rate of positive skin pathergy tests (SPTs) since the mid-1980s.11 Presumably, widespread use of disposable needles, instead of a blunt (reusable, sterilized) needle may contribute to the decrease in the prevalence of positive pathergy tests since the mid-1980s. This is the case for Japan. Previous epidemiologic studies reported that a positive pathergy test was found in 75% of patients in 1972, but only 44% in 1991.25 In 2003, Nishiyama et al22 reported that the rate of positive SPTs was 53.0% (67.3% in men, 42.4% in women). Similarly, SPTs were positive in 35.9% of BD patients from Korea,10 and 33.7% from Germany.2 Prevalence figures from the United Kingdom and other parts of Europe are much lower. Thus, the clinical significance of pathergy tests in the diagnosis of BD seems limited when compared with previous studies.
It is controversial whether BD is getting milder in Japan. Yoshida et al31 reported that the clinical outcomes of BD, especially judging by the number of ocular attacks per year and visual acuity, improved during the 1990s compared with the 1980s. In their study, the number of new BD patients with eye involvement in the 1990s decreased by approximately 20% from that in the 1980s. In the current study, we compared clinical features in BD patients stratified according to diagnosis before or after 2000. Because the observation period was significantly longer in the previous group (before 2000) than in the recent group (after 2000), we compared the data after adjusting the observation period in the previous group. The results clearly demonstrate that the frequency of complete BD has declined, although the frequency of eye involvement was not changed in the limited observation period. On the other hand, arthritis, gastrointestinal involvement, and vascular involvement were more often found in the recent group during the comparable observation periods.
Our findings do not necessarily lead to the conclusion that BD is getting milder, although the number of patients who are typically classified as having complete BD is decreasing. It is hard to interpret these data by a single factor. Although severe clinical manifestations, such as eye and neurologic lesions, are associated with patient characteristics, including male sex and younger onset,19,28,29 the male-to-female rate and age at diagnosis of BD were not altered in the present study (data not shown). Physicians and patients may be more sensitized to BD now than in previous studies. Earlier recognition of the disease may be helpful for early detection of minor or less frequent symptoms in patients. Early therapeutic intervention or early transfer to a center of expertise may also contribute to the reduced number of patients with the complete type of disease.
The cause of BD is still unknown, although both genetic and environmental factors are implicated in innate and/or acquired immune dysfunction, leading to the development of clinical manifestations of the disease. Some of the factors are related with disease phenotypes and severity.4,5,27 It is unlikely that genetic backgrounds affect the recent changes in clinical manifestations of BD in Japan, since Japanese populations have very homogenous genetic backgrounds. Indeed, all the patients enrolled in the present study were Japanese except for 3 patients. On the other hand, environmental factors involved in the disease would have been altered during the last 2 decades. Improved public hygiene may be one of these factors. Microbial pathogens have been proposed to trigger autoimmune responses against heat shock protein through a molecular mimicry fashion, leading to Th1-mediated inflammation in BD.16 In particular, the pathologic roles of streptococcal antigens, which exist as bacterial flora in the oral cavity, have been discussed. A close association between periodontal status and disease severity has been shown.4 Therefore, it is likely that improved oral hygiene contributes to the suppression of microbial pathogen-related immune responses, which are involved in the disease manifestations, resulting in less severe manifestations. If socioeconomic development associated with improved public hygiene is implicated in the frequency and severity of BD, similar epidemiologic changes will occur in other BD-prevalent countries in the near feature.
In the current study we also analyzed associations among clinical backgrounds and individual manifestations. Most of the results were consistent with previous reports. Male predominance was found in patients with eye involvement1,7,24 and CNS involvement,29 whereas genital ulcers29 and arthritis7 were more frequent in female patients. Significant association of ocular lesions with HLA-B51 and CNS involvement was also found, again. A 2010 report12 showed that men who develop BD at a young age have an increased risk of developing severe disease, even if they are free of major organ involvement during the initial years.
In the current chronologic analysis, we found that serious organ lesions in the CNS and the gastrointestinal and vascular systems appeared as late manifestations of the disease, as shown previously.1,18 If progression to serious organ involvement could be predicted in advance by particular pre-existing manifestations and clinical backgrounds, it would be possible to identify a subgroup that requires more intensive treatment. However, in the present study we did not find any significant factors in the pre-existing clinical features to predict more aggressive disease, except that patients with eye involvement are more likely to develop neurologic manifestations. Hamuryudan et al12 also reported that being free of major organ involvement at onset is not a sign of favorable outcome.
In summary, in the present study we found that male sex and HLA-B51 are predisposing factors to eye involvement, as previously reported. It is hard to predict the development of serious organ disease based on pre-existing clinical manifestations, except that ocular symptoms are associated with a high frequency of neurologic lesions. Moreover, the most typical cases of BD, those with all of the 4 major symptoms, have been decreasing in Japan, although the reason remains uncertain. Further assessment may allow us to detect early predictors of more aggressive disease in the future.
The authors thank Mr. Tom Kiper (Yokosuka, Japan) for his review and invaluable suggestions in preparing the manuscript.
1. Alpsoy E, Donmez L, Onder M, Gunasti S, Usta A, Karincaoglu Y, Kandi B, Buyukkara S, Keseroglu O, Uzun S, Tursen U, Seyhan M, Akman A. Clinical features and natural course of Behcet's disease in 661 cases: a multicentre study. Br J Dermatol
2. Altenburg A, Papoutsis N, Orawa H, Martus P, Krause L, Zouboulis CC. [Epidemiology and clinical manifestations of Adamantiades-Behcet disease in Germany-current pathogenetic concepts and therapeutic possibilities]. J Dtsch Dermatol Ges
3. Anonymous. Criteria for diagnosis of Behcet's disease. International Study Group for Behcet's Disease. Lancet
4. Arabaci T, Kara C, Cicek Y. Relationship between periodontal parameters and Behcet's disease and evaluation of different treatments for oral recurrent aphthous stomatitis. J Periodontal Res
5. Aramaki K, Kikuchi H, Hirohata S. HLA-B51 and cigarette smoking as risk factors for chronic progressive neurological manifestations in Behcet's disease. Mod Rheumatol
6. Azizlerli G, Kose AA, Sarica R, Gul A, Tutkun IT, Kulac M, Tunc R, Urgancioglu M, Disci R. Prevalence of Behcet's disease in Istanbul, Turkey. Int J Dermatol
7. Bang DS, Oh SH, Lee KH, Lee ES, Lee SN. Influence of sex on patients with Behcet's disease in Korea. J Korean Med Sci
8. Barra C, Belfort Junior R, Abreu MT, Kim MK, Martins MC, Petrilli AM. Behcet's disease in Brazil-a review of 49 cases with emphasis on ophthalmic manifestations. Jpn J Ophthalmol
9. Cakir N, Dervis E, Benian O, Pamuk ON, Sonmezates N, Rahimoglu R, Tuna S, Cetin T, Sarikaya Y. Prevalence of Behcet's disease in rural western Turkey: a preliminary report. Clin Exp Rheumatol
. 2004;22(4 Suppl 34):S53-S55.
10. Chang HK, Cheon KS. The clinical significance of a pathergy reaction in patients with Behcet's disease. J Korean Med Sci
11. Dilsen N, Konice M, Aral O, Ocal L, Inanc M, Gul A. Comparative study of the skin pathergy test with blunt and sharp needles in Behcet's disease: confirmed specificity but decreased sensitivity with sharp needles. Ann Rheum Dis
12. Hamuryudan V, Hatemi G, Tascilar K, Sut N, Ozyazgan Y, Seyahi E, Mat C, Yurdakul S, Yazici H. Prognosis of Behcet's syndrome among men with mucocutaneous involvement at disease onset: long-term outcome of patients enrolled in a controlled trial. Rheumatology (Oxford)
13. Houman MH, Neffati H, Braham A, Harzallah O, Khanfir M, Miled M, Hamzaoui K. Behcet's disease in Tunisia. Demographic, clinical and genetic aspects in 260 patients. Clin Exp Rheumatol
. 2007;25(4 Suppl 45):S58-S64.
14. Ideguchi H, Suda A, Takeno M, Kirino Y, Ihata A, Ueda A, Ohno S, Baba Y, Kuroiwa Y, Ishigatsubo Y. Neurological manifestations of Behcet's disease in Japan: a study of 54 patients. J Neurol
15. Kaneko F, Nakamura K, Sato M, Tojo M, Zheng X, Zhang JZ. Epidemiology of Behcet's disease in Asian countries and Japan. Adv Exp Med Biol
16. Kaneko S, Suzuki N, Yamashita N, Nagafuchi H, Nakajima T, Wakisaka S, Yamamoto S, Sakane T. Characterization of T cells specific for an epitope of human 60-kD heat shock protein (hsp) in patients with Behcet's disease (BD) in Japan. Clin Exp Immunol
17. Krause I, Yankevich A, Fraser A, Rosner I, Mader R, Zisman D, Boulman N, Rozenbaum M, Weinberger A. Prevalence and clinical aspects of Behcet's disease in the north of Israel. Clin Rheumatol
18. Kural-Seyahi E, Fresko I, Seyahi N, Ozyazgan Y, Mat C, Hamuryudan V, Yurdakul S, Yazici H. The long-term mortality and morbidity of Behcet syndrome: a 2-decade outcome survey of 387 patients followed at a dedicated center. Medicine (Baltimore)
19. Mendes D, Correia M, Barbedo M, Vaio T, Mota M, Goncalves O, Valente J. Behcet's disease-a contemporary review. J Autoimmun
20. Mizushima Y. Recent research into Behcet's disease in Japan. Int J Tissue React
21. Nishiyama M, Nakae K, Kuriyama T, Hashimoto M, Hsu ZN. A study among related pairs of Japanese patients with familial Behcet's disease: group comparisons by interval of disease onsets. J Rheumatol
22. Nishiyama M, Nakae K, Takahashi M, Kuriyama T. A study of clinical features and the worsening factors among patients with Behcet's disease in Japan. Dokkyo J Med Sci
23. Pandrea A, Rudinskaya A, Klein B, Krebs T. What does it take to diagnose Behcet disease? J Clin Rheumatol
24. Pipitone N, Boiardi L, Olivieri I, Cantini F, Salvi F, Malatesta R, La Corte R, Triolo G, Ferrante A, Filippini D, Paolazzi G, Sarzi-Puttini P, Restuccia G, Salvarani C. Clinical manifestations of Behcet's disease in 137 Italian patients: results of a multicenter study. Clin Exp Rheumatol
. 2004;22(6 Suppl 36):S46-S51.
25. Sakane T, Takeno M, Suzuki N, Inaba G. Behcet's disease. N Engl J Med
26. Salvarani C, Pipitone N, Catanoso MG, Cimino L, Tumiati B, Macchioni P, Bajocchi G, Olivieri I, Boiardi L. Epidemiology and clinical course of Behcet's disease in the Reggio Emilia area of Northern Italy: a seventeen-year population-based study. Arthritis Rheum
27. Soylu M, Ersoz TR, Erken E. The association between HLA B5 and ocular involvement in Behcet's disease in southern Turkey. Acta Ophthalmol (Copenh)
28. Tugal-Tutkun I, Onal S, Altan-Yaycioglu R, Huseyin Altunbas H, Urgancioglu M. Uveitis in Behcet disease: an analysis of 880 patients. Am J Ophthalmol
29. Tursen U, Gurler A, Boyvat A. Evaluation of clinical findings according to sex in 2313 Turkish patients with Behcet's disease. Int J Dermatol
30. Varol A, Seifert O, Anderson CD. The skin pathergy test: innately useful? Arch Dermatol Res
© 2011 Lippincott Williams & Wilkins, Inc.
31. Yoshida A, Kawashima H, Motoyama Y, Shibui H, Kaburaki T, Shimizu K, Ando K, Hijikata K, Izawa Y, Hayashi K, Numaga J, Fujino Y, Masuda K, Araie M. Comparison of patients with Behcet's disease in the 1980s and 1990s. Ophthalmology